Work and Organizations in China After Thirty Years of Transition (Research in the Sociology of Work)

WORK AND ORGANIZATIONS IN CHINA AFTER THIRTY YEARS OF TRANSITION RESEARCH IN THE SOCIOLOGY OF WORK Series Editor: Lis...

Author: Lisa Keister

43 downloads 1261 Views 3MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

WORK AND ORGANIZATIONS IN CHINA AFTER THIRTY YEARS OF TRANSITION

RESEARCH IN THE SOCIOLOGY OF WORK Series Editor: Lisa Keister Recent Volumes: Volume 1:

Class Consciousness

Volume 2:

Peripheral Workers

Volume 3:

Unemployment

Volume 4:

High Tech Work

Volume 5:

The Meaning of Work

Volume 6:

The Globalization of Work

Volume 7:

Work and Family

Volume 8:

Deviance in the Workplace

Volume 9:

Marginal Employment

Volume 10: Transformation of Work Volume 11: Labor Revitalization: Global Perspectives and New Initiatives Volume 12: The Sociology of Job Training Volume 13: Globalism/Localism at Work Volume 14: Diversity in the Workforce Volume 15: Entrepreneurship Volume 16: Worker Participation: Current Research and Future Trends Volume 17: Work Place Temporalities Volume 18: Economic Sociology of Work

RESEARCH IN THE SOCIOLOGY OF WORK VOLUME 19

WORK AND ORGANIZATIONS IN CHINA AFTER THIRTY YEARS OF TRANSITION EDITED BY

LISA KEISTER Duke University, Durham, NC, USA

United Kingdom – North America – Japan India – Malaysia – China

Emerald Group Publishing Limited Howard House, Wagon Lane, Bingley BD16 1WA, UK First edition 2009 Copyright r 2009 Emerald Group Publishing Limited Reprints and permission service Contact: [email protected] No part of this book may be reproduced, stored in a retrieval system, transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without either the prior written permission of the publisher or a licence permitting restricted copying issued in the UK by The Copyright Licensing Agency and in the USA by The Copyright Clearance Center. No responsibility is accepted for the accuracy of information contained in the text, illustrations or advertisements. The opinions expressed in these chapters are not necessarily those of the Editor or the publisher. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN: 978-1-84855-730-7 ISSN: 0277-2833 (Series)

Awarded in recognition of Emerald’s production department’s adherence to quality systems and processes when preparing scholarly journals for print

CONTENTS LIST OF CONTRIBUTORS

ix

INTRODUCTION

xi

PART I: ORGANIZATIONS BRINGING MARKET TRANSITION THEORY TO THE FIRM Victor Nee and Sonja Opper

3

ASSET MANAGEMENT AND PRODUCTIVITY IN REFORM-ERA CHINA Doug Guthrie, Zhixing Xiao and Junmin Wang

35

TRANSACTION COSTS, SOCIAL INSTITUTIONS, AND THE DURATION OF INTERFIRM CONTRACTS IN CHINA Ling Yang and Xueguang Zhou

69

FOREIGN DIRECT INVESTMENT AND SOE PERFORMANCE IN CHINA: EVIDENCE FROM THE THIRD INDUSTRIAL CENSUS Yusheng Peng

105

OWNERSHIP AND INNOVATION DURING ECONOMIC DEVELOPMENT: CORPORATE OWNERSHIP AND STRATEGY FORMATION IN CHINA Lisa A. Keister and Randy Hodson

129

v

vi

CONTENTS

LOCALIZATION IN THE AGE OF GLOBALIZATION: INSTITUTIONAL DUALITY AND LABOR MANAGEMENT STRUCTURES IN CHINA’S FOREIGN-INVESTED ENTERPRISES Yang Cao and Wei Zhao

165

A MICRO–MACRO LINK DURING INSTITUTIONAL TRANSITIONS Mike W. Peng and J. Martina Quan

203

RECRUITING AND DEPLOYING SOCIAL CAPITAL IN ORGANIZATIONS: THEORY AND EVIDENCE Nan Lin, Yanlong Zhang, Wenhong Chen, Dan Ao and Lijun Song

225

PART II: WORK NETWORK RESOURCES AND JOB MOBILITY IN CHINA’S TRANSITIONAL ECONOMY Yanjie Bian and Xianbi Huang

255

DANWEI AND SOCIAL INEQUALITY IN CONTEMPORARY URBAN CHINA Yu Xie, Qing Lai and Xiaogang Wu

283

THE PHASE-OUT OF THE UNFIT: KEEPING THE UNWORTHY OUT OF WORK Dorothy J. Solinger

307

GENDER INEQUALITY IN THE CHINESE LEGAL PROFESSION Ethan Michelson

337

Contents

vii

FLEXIBLE WORK, FLEXIBLE HOUSEHOLD: LABOR MIGRATION AND RURAL FAMILIES IN CHINA C. Cindy Fan

377

THE RUBIK’S CUBE STATE: A RECONCEPTUALIZATION OF POLITICAL CHANGE IN CONTEMPORARY CHINA Bai Gao

409

LIST OF CONTRIBUTORS Dan Ao

Department of Sociology, The Chinese University of Hong Kong, Hong Kong

Yanjie Bian

Department of Sociology, University of Minnesota, Twin Cities, MN, USA; School of Humanities and Social Science, Xi’an Jiaotong University, Xi’an, China

Yang Cao

Department of Sociology, University of North Carolina at Charlotte, Charlotte, NC, USA

Wenhong Chen

School of Communications, University of Texas, Austin, USA

C. Cindy Fan

Department of Geography, University of California, Los Angeles, CA, USA

Bai Gao

Department of Sociology, Duke University, Durham, NC, USA

Doug Guthrie

Department of Management, New York University, Stern School of Business, New York, NY, USA

Randy Hodson

Department of Sociology, Ohio State University, Columbus, OH, USA

Xianbi Huang

School of Social Sciences, LA Trobe University, Melbourne, Australia

Lisa A. Keister


Qing Lai

Department of Sociology, University of Michigan, Ann Arbor, MI, USA

Nan Lin

Department of Sociology, Duke University, Durham, NC, USA ix

x

LIST OF CONTRIBUTORS

Ethan Michelson

Department of Sociology, Indiana University, Bloomington, IN, USA

Victor Nee

Department of Sociology, Cornell University, Ithaca, NY, USA

Sonja Opper

Department of Economics, Lund University, Sweden

Mike W. Peng

School of Management, University of Texas at Dallas, Richardson, TX, USA

Yusheng Peng

Department of Business and Economics, Brooklyn College, New York, NY, USA

J. Martina Quan

School of Management, University of Texas at Dallas, Richardson, TX, USA

Dorothy J. Solinger

Department of Political Science, University of California, Irvine, CA, USA

Lijun Song

Department of Sociology, Vanderbilt University, Nashville, TN, USA

Junmin Wang

Department of Sociology, University of Memphis, Memphis, TN, USA

Xiaogang Wu

Division of Social Science, Hong Kong University of Science and Technology, Hong Kong

Zhixing Xiao

Department of Management, China-Europe International Business School, Shanghai, China

Yu Xie

Department of Sociology, University of Michigan, Ann Arbor, MI, USA

Ling Yang

Department of Sociology, Stanford University, Palo Alto, CA, USA

Yanlong Zhang


Wei Zhao

Department of Sociology, University of North Carolina at Charlotte, Charlotte, NC, USA

Xueguang Zhou

Department of Sociology, Stanford University, Palo Alto, CA, USA

INTRODUCTION Thirty years of rapid development and economic change have created organizations and work relations in China that would have been unthinkable at the start of transition. In December of 1978, the Chinese Communist Party agreed with Deng Xiaoping to allow agricultural privatization, a stark contrast to the communes of Mao Zedong’s era. This change established the financial foundation that would lead to development in eastern, coastal cities and that would ultimately fuel an extraordinary transformation of China’s economy and its global position. As a result, organizational structures have changed, and new organizational forms have emerged. There have also been dramatic changes in the way work organizations behave and in the nature and implications of work. This volume provides a glimpse into the state of organizations and work at the 30-year mark. The contributors are top scholars in the field, including many who have observed and studied China’s transition for decades, who are drawing on some of the most up-to-date and innovative data sources available. The chapters are samples of the current work of these researchers that, taken together, provide a snapshot of the state of research on China’s organizations and work behaviors as transition enters its fourth decade. Part I addresses the nature and behavior of organizations. Victor Nee and Sonja Opper extend market transition theory, Nee’s theory that initiated a great deal of current debate about stratification outcomes during transition, to a firm-level analysis. Nee and Opper identify the advantages of broadening the theory’s scope to include organizations, extend the theory’s hypotheses to firms, and provide preliminary empirical evaluation of their claims. The remaining papers in this section identify and study important changes in particular organizations and types of organizations, starting with a pair of papers that address emerging economic institutions. Doug Guthrie, Zhixing Xiao, and Junmin Wang start the section by exploring the transformation of state offices from governing bodies to asset management companies. Their use of quantitative and case study data allows them to draw general conclusions about the implications of structural change for firms while providing insight into the mechanisms by which this movement occurs. Ling Yang and Xueguang Zhou provide insight into the emergence xi

xii

INTRODUCTION

of another economic institution: interfirm contracts. Specifically, they examine contract duration and provide evidence from more than 800 contracts from 620 firms. The following three papers address unique innovations and their impact on three types of organizations. Yusheng Peng addresses the impact of inward foreign direct investment (FDI) on the productivity of state-owned enterprises (SOES). Peng shows that the presence of FDI-related firms improves total factor productivity of SOEs located in the same city but not associated with the FDI. He concludes that his results are not simply technology spillovers but rather FDI-induced change. Keister and Hodson also study SOEs, but they use a different data source to study the effect of ownership type (SOE compared to collective enterprises and other nonstate firms) on the adoption of four innovations. Cao and Zhao move beyond SOEs to study labor management structures in China’s foreign-invested enterprises (FIEs). They investigate the adoption of two Chinese-style and two Western-style labor structures. They provide evidence that the two Chinese-style structures reduce tension and conflict between labor and management, but the two Western-style structures have little of this effect. They use their findings to highlight the power of the Chinese institutional environment and to discuss how organizational interest interacts with institutional forces. The final two papers in Part I both discuss linkeages between organizations and work, providing a transition to the second part of this volume. Mike Peng and Martina Quan deal explicitly with micro–macro links. They point out that the relationship between micro (e.g., interpersonal connections) and macro (e.g., interorganizational relationships, firm strategies, performance) is an important theme in research on management and organizations in China. They review related papers from leading journals and propose ideas for understanding how the network structure of managerial connections will develop in different phases of China’s transition. Peng and Quan conclude with a call for future research to make this relationship more central. Lin, Zhang, Chen, Ao, and Song do not conceptualize their work as a micro–macro link, but their approach is quite similar to that of Peng and Quan. Lin et al. study the recruiting and deploying of social capital to organizations. They start with the observation that social capital operates at both the macro (organization) and micro (individual) levels. They develop a pair of intriguing hypotheses to explain why and how organizations recruit and deploy social capital. The papers in Part II of this volume address issues related to work and work relations. Yanjie Bian and Xianbi Huang start the section with an

Introduction

xiii

exploration of how network resources affect job mobility. They draw on data from a five-city survey to study how social ties affect job change, search time, and job–worker matching. They show that those who change jobs who use information and influence networks increase both search time and job– worker matching, whereas those using only influence networks improve earning opportunities. The next four papers each address work and inequality in some fashion. The first two papers address the continuing role of the danwei (the work unit) in producing and maintaining inequality, and the second two explore the mechanisms that generate particular types of inequality. First, Yu Xie, Qing Lai, and Xiaogang Wu propose that the danwei continues to be an important source of stratification in contemporary China. They draw on data from three large cities to show that earnings and benefits are still related to characteristics of the work unit in ways that are similar to pre-reform patterns. In the next paper, Dorothy Solinger explores the fate of those who no longer have a connection to a danwei. She studies people who lost their jobs and those who ultimately became destitute after losing positions at a danwei. She draws on her in-depth research on China’s economy and labor practices to describe and evaluate contemporary labor policies and the fate of workers who are at the mercy of these policies. The next pair of papers provides insight into specific types of work-related inequality. Michelson notes that gender inequality has become more extreme in many occupations in China, and he sets out to uncover whether the same is true in the legal profession. He uses four sources of quantitative data to document that gender inequality in law mirrors patterns in other professions. His findings demonstrate that women are, indeed, enjoying more opportunities in law, but he shows that women still earn less income and are less likely to become law firm partners, two important indicators of job quality. C. Cindy Fan also addresses gender inequality, but her conclusions about male– female differences are part of a larger study of labor migration and household work. Fan studies work flexibility among rural migrants and the implications of this flexibility for household organization. She argues that migrants’ work flexibility necessitates flexibility at home, between genders and across generations. Her conclusions about the implications of migration patterns for the household division of labor provide unique insights into two distinct literatures as well as for the future of social and economic stratification in China. The final paper in the volume addresses the important linkage between politics and economics. Bai Gao starts by noting that observers of China can be surprised by the coexistence of an economy known for expanding trade and global production and an authoritarian political system that has

xiv

INTRODUCTION

escaped a global movement toward democratization. Gao reviews the literature on China’s prospects for democratization, and he argues that other observers have overlooked important strategic responses by China’s party state. He also contends that other researchers need to move beyond reliance on a simple conceptualization of the state as authoritarian or democratic and to incorporate the varied state responses in their conceptualization and analysis. Gao identifies six distinct sides (or faces) the China’s party state assumes, including authoritarian, neoliberal, developmental, predatory, refined socialist, and corporatist. He concludes that the coexistence of these six personae has allowed China to become a powerful economic entity while preventing it from moving toward democratization. In conclusion, I would like to acknowledge the organizations and people who contributed to the production of this volume. The Duke University Department of Sociology provided resources, and Zoe Morris and Claire Ferres of Emerald provided production and editorial support. Most of all, Yanlong Zhang provided excellent administrative and research support throughout the compilation of this volume.

PART I ORGANIZATIONS

BRINGING MARKET TRANSITION THEORY TO THE FIRM Victor Nee and Sonja Opper 1. INTRODUCTION Market transition theory has specified general mechanisms to explain change in the balance of power between political and economic actors in transition economies. These mechanisms drive the endogenous construction of informal institutions of a market society; moreover, it is within the context of an ongoing change in relative power that the formal institutions of the emerging market economy arise. The theory makes clear predictions on the declining value of political capital as a consequence of progressive marketization, which incrementally results in transformative change in the direction of more relative autonomy between the political and economic spheres, not dissimilar from established market economies (Kornai, 1995; Evans, 1995; Nee, 2000; Lindenberg, 2000; Ricketts, 2000). In sum, the predicted change in relative power between redistributors and producers explains not only bottom-up entrepreneurial activity, but also the emergence of a market economy in departures from state socialism. Although the evolving market transition debate has been fruitful in stimulating a robust research program in sociology and economics with over 25 empirical research reports published in peer-reviewed journals in sociology and economics (see Appendix A; King & Sze´lenyi, 2005; Keister, 2009; Keister, forthcoming), the potential of the theory is far from exploited.

Work and Organizations in China after Thirty Years of Transition Research in the Sociology of Work, Volume 19, 3–34 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1108/S0277-2833(2009)0000019004

3

4

VICTOR NEE AND SONJA OPPER

While the theory in its original formulation (Nee, 1989) sought to explain dynamic power shifts from redistributors to producers as a consequence of marketization, the emerging research program, however, almost exclusively followed the first set of derived hypotheses and its focus on determinants of household income. Subsequent research employed readily comparable ordinary least squares (OLS) models – the standard human capital model of income determination (Mincer, 1958, 1974) – adapted to include measures of political capital (Nee, 1989; Xie & Hannum, 1996). Empirical evidence, however, remained mixed and inconclusive and controversy over the fate of political capital in new market economies seems unresolved. This is reflected in the anomaly of a near even split between the articles reporting results consistent with the prediction of decline in value of political capital caused by competitive markets (Nee, 1989, 1991, 1996) and those that report the opposite (Walder, 1996, 2003; Parish & Michelson, 1996) or tender competitive hypotheses (Rońa-Tas, 1994; Bian & Logan, 1996). The aim of this paper is to revitalize the theory’s original focus on dynamic power shifts between redistributors and producers by extending the theory’s empirical application to a firm-level analysis. The firm as the ultimate generator of income provides the most direct approach to analyze the interplay between market power, political capital, and economic outcome. By contrasting firm-level transactions across institutional domains representing various levels of marketization and state control, it is possible to directly examine the effect of political capital on distinct economic outcomes. If observed political benefits are manifest predominantly in statecontrolled institutional domains, this would support market transition theory predicting a direct link between marketization and the value of political capital. However, if political capital is just as fungible in marketized as it is in state-controlled domains, this would lead to a decisive rejection of the theory’s claim. Instead, such a result would support the assertion of the theory’s critic that markets have no causal significance in enabling, motivating, and guiding economic action (Walder, 1996). The remainder of this essay proceeds as follows: First, we identify general advantages connected with a focus on firm-level studies and highlight some of the inherent problems of income attainment models. Section 3 then extends the original propositions of market transition theory to firm-level analysis. Section 4 provides some cursory evidence, and section 5 concludes. Overall, with descriptive and qualitative evidence, we show that the value of political capital is closely linked with the type of institutional domains in which agents use political connections to secure advantages.

Bringing Market Transition Theory to the Firm

5

2. FROM HOUSEHOLD TO FIRM-LEVEL ANALYSIS To highlight general advantages of firm-level analysis for tests of market transition theory, it is useful to begin with a review of inherent problems of studies applying income attainment models. Oberschall (1996) raised early doubts about whether income and income inequality can ‘‘give us any clue about the shape of emerging institutions’’ (Oberschall, 1996). Similarly, Guthrie (1997) warned that income-related studies on elite change ‘‘show no direct or concrete evidence about the fate of the hierarchy of former command economies’’ and suggested instead a focus on firm-level studies (Guthrie, 1999). The problem is income attainment models cannot reveal whether economic advantages captured by politically connected households or individuals come from their ongoing exchange in political markets or from persisting advantages in competitive markets. Moreover, the reliability of household-level research hinges critically on the use of reliable proxies of marketization to control for the quality of the corresponding household environment. We identify three measurement approaches widely used in the market transition literature: First, an intuitively convincing way to measure the extent of market allocation has built on the assumption that market exchange depends on the existence of private property rights (Kornai, 1990). A related approach to measurement has focused on the proportion of industrial output produced by private, collective, and state-owned enterprises (Nee, 1996, Nee & Cao, 1999); firm ownership (Parish, Zhe, & Li, 1995; Wu, 2002); the proportion of household income in nonagricultural production (Walder, 2002b); and the degree of privatization (Opper, Wong, & Hu, 2002). Second, time has repeatedly been used as a proxy for the duration of market transition. This approach assumes a linear progression in the development of a market economy, which may be true in the long term, but not necessarily in the shorter run. Moreover, the use of time as a proxy for marketization risks overlap with confounding causes such as regional business cycle, labor market fluctuations, capital investments, and locally restricted reform initiatives, which are difficult to control for due to severe data limitations at the local level. Finally, a whole host of other measures have been used, which measure concepts other than marketization. A number of tests of market transition theory relied on measures of economic growth (Xie & Hannum, 1996; Walder, 2002a, 2002b; Hauser & Xie, 2005) or structural change (Parish & Michelson, 1996; Walder & Zhao, 2006). Validity problems of these measures are palpable. Economic growth and structural change are macroeconomic performance measures, determined by

6


a complex set of factors including technology development, labor, and capital input, among which marketization may but need not play a decisive role. Given so many different measurements of the same underlying concept, it is obvious that some will correlate only weakly with market transition defined as the ‘‘decisiveness of the shift to reliance on the market mechanism in the allocation and distribution’’ of goods and services (Nee, 1989, p. 667). An enormous variation in correlation coefficients between the different measures of the extent of marketization underlines the severity of the measurement issues. This is seen in using the standardized marketization index constructed by the National Economic Research Institute (NERI) (Fan & Wang, 2003) to calculate correlation coefficients with measures most commonly used in the market transition literature. Using provincial-level data covering the period from 1997 to 2003 (see Table 1), marketization measured as the proportion of non-state industrial production shows the highest correlation coefficient with the marketization index (0.81). Also comparatively high is the correlation coefficient of rural income (0.72), indicating a closer overlap with marketization. However, the lower correlation coefficient of 0.62 between rural income and proportion of non-state industrial production indicates that both proxies capture different concepts connected with marketization. All remaining measures are only weakly correlated. The use of GDP-growth measure seems particularly problematic. Though Hauser and Xie (2005) defend the use of GDP-based measures with reported close correlations between GDP and marketization (NERI), this is true only for absolute values but not for growth rates. Finally, passage of time captures marketization inadequately. With a correlation coefficient of 0.14, repeated surveys over relatively short periods of time are unlikely to properly signal the effects of market transition on predicted outcomes. The consequences of the inconsistent and diverse measures of the key causal concept – marketization – are nontrivial. This is seen in the close correspondence between the choice of measurement and the validation of market transition theory (see Appendix A). Only one of the nine studies, which actually measures the extent and scope of marketization, rejects the hypothesis of decline in the value of political capital. On the other hand, out of 19 studies using the proxies of economic growth and structural measures or time or no measure of marketization, 16 studies reject the power-decline hypothesis. This yields an odds-ratio of 42.7, that is, the odds of confirmation are about 43 times higher when proxies of market transition are used instead of measures of structural change or development.

1.00

2

3

0.14

0.72

Walder (2002b)

Nee (1989); Bian and Logan (1996), Zhou (2000); Murdoch and Sicular (2000); Choi and Zhou (2001)

0.53

0.57

0.08

0.62

0.13

0.12

Xie and Hannum (1996); Hauser and Xie (2005) Walder (2002a)

0.04

0.003

0.13

0.02

0.09

0.03

0.11

0.01

1.00

4

0.22

0.16

0.20

1.00

5

0.15

0.91

1.00

6

0.16

1.00

7

1.00

8

Sources: Provincial data from National Statistical Bureau of China, China State Statistical Yearbook, 1998–2004; marketization index from Fan and Wang (2003).

Time 8 Year

Growth and development 5 GDP growth/per capita GDP growth 6 Per capita gross industrial output 7 Rural income

4 Non-farm labor ratio

1.00 0.81

1

Walder (2002b), Walder and Zhao, 0.03 0.004 1.00 2006 Parish and Michelson (1996) 0.14 0.02 0.78

Li, Meng, and Zhang (2006) Nee (1996), Nee and Cao (1999)

Studies

Correlation Matrix of Provincial-Level Measures, 1997–2003 (N ¼ 230).

Economic structure 3 Share of agricultural in GDP

Marketization 1 Marketization index 2 Non-state industrial production ratio

Measure/Marketization Proxy

Table 1.

Bringing Market Transition Theory to the Firm 7

8


Inadvertently, a large proportion of the contributions to the markettransition controversy in fact tested alternative theories on the association between economic growth and structural changes and the value of political capital. The consistent finding of these studies (Xie & Hannum, 1996; Walder, 2002a, 2002b; Hauser & Xie, 2005; Parish & Michelson, 1996) is that economic growth and structural change do not adversely affect the market value of political capital of the established elite. Our review underscores the need for more direct ways to study the connection between marketization and the value of political capital based on positional power in the government and communist party. We assert that a revitalization of the theory’s original focus on producers provides an alternative approach to examine the association between market transition and the valuation of political capital. A firm-level analysis does not critically depend on the choice of proxies for marketization. We contend that in order to determine whether the prediction of a decline of political capital in pricemaking markets is accurate, empirical tests need to focus on discrete economic transactions linked to well-defined institutional domains of the transition economy. In this way it is possible to discern with greater reliability whether political capital loses its direct advantage in transactions in market exchange or, alternatively, maintains or possibly even gains advantage in such transactions. Clearly, this approach promises a more direct way to explore whether political connections really help managers and entrepreneurs ‘‘to get more out of their effort because of their power and connections’’ (Walder, 1996, p. 1067).

3. MARKET TRANSITION THEORY: THE PRODUCER PERSPECTIVE In the early period of hypothesis testing, the original set of derived hypotheses and subsequent empirical applications focused on agricultural households as the new producer class of the 1980s. But following the expansion of producer activities beyond the agricultural sector, conditions of income generation in nonagricultural firms emerged as an important new application for empirical confirmation. We briefly reconcile market transition theory’s three interrelated theses with the expansion of the scope of marketization beyond the agricultural sector, and derive testable hypothesis that allow a direct application to industrial and commercial enterprises.


9

The market power thesis asserts that replacement of state bureaucratic allocation by market allocation involves a shift of power favoring direct producers relative to redistributors. This assertion is enormously consequential for understanding change in the institutional environment of firms. Almost imperceptibly, but accelerating following tipping points, selfreinforcing shifts in the institutional environment cause traditional stateowned enterprises of the old redistributive economy to lose market share to hybrid and private ownership forms (Nee, 1992, 2005). Furthermore, increasing competition and dependence on market outcomes raises the costs of political interference (Boycko & Shleifer, 1993; Opper et al., 2002; Fan, Wong, & Zhang, 2007; Nee, Opper, & Wong, 2007). Concurrent with these trends, greater organizational autonomy embedded in decentralized markets enable economic actors to construct informal arrangements that build from ground-up the informal institutions of a private enterprise economy. From informal lending arrangements to provide private capital for start-up firms to far-flung supply and distribution networks, informal economic institutions emerged to facilitate the expansion of private sector entrepreneurial activities challenging the state-directed economy from below. A direct extension of the market power thesis suggests that with marketization, the economic success of producers is increasingly independent of the involvement of redistributors. In a general formulation, firms will experience a decline in the value of their political connections. This, however, does not imply, as some have interpreted, a complete devaluation of political connections. After all, political capital is fungible in all types of economies, from transitional to mature market economies. In all market economies, political connections matter for firms lobbying to secure preferential treatment by government (Stigler, 1971; Krueger, 1974). It follows that in market economies, political capital, as a fungible form of capital, has greatest valuation in those institutional domains where government restricts economic activity. Analogous to the original derivation specifying income effects at the household level, we assert: Hypothesis 1. The more market exchange replaces the redistributive mechanism, the less the value of political capital relative to capital stemming from the capabilities and market performance of the firm. The market incentive thesis emphasizes changes in the structure of incentive stemming from market transition. With marketization, rewards are increasingly based on performance rather than the strength of political ties, which creates positive incentives for entrepreneurial activities and innovativeness. Further, as market competition intensifies, firms face

10


growing pressure to invest in capabilities in order to survive the withering competition. Whether entrepreneurial activity is for the sake of the fruits of success, or for success itself, in price-making markets rewards are based on the competitive sorting and matching of quality and price. It is thus the restoration of consumers’ and producers’ sovereignty in transition economies, which activates market incentives. The specification of the market incentive thesis is close to Baumol’s (1990) supposition that the most effective way to stimulate productive entrepreneurial activity is to diminish relative rewards to unproductive or destructive rent seeking and increase payoffs to productive entrepreneurial activity. While Baumol’s entrepreneurial theory is referring to within-system variation in market economies, the market incentive thesis emphasizes the transfer from a planned economy to a system primarily based on market exchange. As a direct extension of the incentive thesis, capability development of firms should be correlated with the extent of marketization and interfirm competition. The most important entrepreneurial response to market incentives is through innovation. In increasingly competitive markets, firms have incentives to innovate to extend their profit margin or to come up with new products, which help to escape competitive pressure until imitators come up with similar product or production technologies. In general, we expect: Hypothesis 2. The transition from state socialist redistribution to markets increases the value of a firm’s capability development. Finally, the opportunity thesis emphasizes the markets’ crucial role in enabling entrepreneurial activities. The opportunity thesis goes beyond the idea of resource availability allocated through markets. The price-finding mechanism signals disequilibria of supply and demand, wherein high or increasing prices indicate demand and attract new producers to establish new or neglected lines of production. The market mechanism also offers economic actors a means to assess potential opportunities from entrepreneurial activities as well as opportunity costs for failing to invest in productive activities (Hayek, 1978). The emergence of markets thus endogenously expands the opportunities for entrepreneurs and firms to identify new markets and prospects for profit making. Given the central role of free markets for opportunity identification, we derive the following hypotheses: Hypothesis 3. In transitions to a market economy, the development of free markets provides the opportunity structure for new market entry by direct producers.

11


4. EVIDENCE FROM A TRANSACTION-FOCUSED ANALYSIS At the most cursory level, a focus on China’s increasing diversity of organizational forms signals an ongoing devaluation of political capital. The graph in Fig. 1 uses nonagricultural employment data to illustrate the accelerating ownership diversification between 1990 and 2002. What is notable here is not only the rapid decline of the relative share of state-owned firms, but also the rapid growth of private ownership, both in the form of private or individual enterprises and also in the form of rural private and individual firms formally registered as township village enterprises. Ownership diversification per se, however, may not yet indicate that political capital is truly devaluating. Political capital, independent of the ownership form, might still facilitate access to resources controlled by the state, or in the case of new organizational forms confer legitimacy. For example, the so-called ‘‘red hat’’ firms in the early reform period were predominately private firms in the guise of township and village enterprises owned by local government (Huang, 2008). Confirmation of Hypothesis 1 therefore hinges on a closer review of the relation between political capital and firm performance.

100% individual and private enterprise 80%

TVEs (self-employed)

TVEs (private) 60% TVEs (collective) 40%

urban collectively owned enterprise

20%

individual and private enterprise TVEs (private) TVEs (self-employed) TVEs (collective) urban collectively owned enterprise limited liability company shareholding corp state owned enterprises

state owned enterprises

19

19

90 91 19 92 19 9 19 3 94 19 9 19 5 96 19 9 19 7 98 19 9 20 9 00 20 01 20 02

0%

Year

Fig. 1.

Employment by Organizational Form. Source: National Bureau of Statistics of China (various years).

12


To reliably confirm a possible link between the value of political capital and marketization, we follow Nee’s (1991, p. 279) assertion that the ‘‘value of personal connections with cadres is a function of the extent to which the allocation of resources y remains bounded by the redistributive economy.’’ Because firms operate simultaneously in a great variety of institutional domains, we assert that a comparative analysis of regions or even industrial sectors would still leave a great leeway for potential ambiguity on the link between marketization and the fungibility of political forms of capital. Instead, we suggest a transaction-focused approach to study more directly the association between marketization and positional advantages stemming from political connections. Such approach acknowledges institutional diversity within distinct economic systems and shifts focus to examine the nature of institutional domains in which economic actors compete and cooperate to secure rewards. If the value of political capital is unaffected or even increases for economic transactions in liberalized markets, we can infer that we will see rising from the ruins of state socialism a hybrid type of economy where the economic and political spheres remain blurred, and interventions by political actors constitute an integral part of the economic order. If, on the other hand, political capital is devalued because of marketization, then we can expect a devaluation of political capital, not dissimilar from established market economies (Nee, 2000). A central advantage of a transaction-focused analysis is the more nuanced view of the fungibility of political connections. Moreover, we respond to criticism that reference to ‘‘partial reform’’ when results fail to confirm a decline in value of political connections renders market transition theory ‘‘immune to falsification’’ (Rońa-Tas, 1994, p. 44). While earlier incomebased approaches left the escape hatch to attribute lack of confirmation to insufficient levels of marketization (Nee, 1991, 1996), a transaction-focused approach using comparative analysis of different institutional domains leaves no room for such interpretation. Simultaneously, the comparative assessment of the value of political capital in different institutional domains provides a more direct test of market transition theory’s contending perspective which called attention to the capacity of political actors to adapt and profit from new elite opportunities stemming from marketization (Staniszkis, 1991; Oi, 1992; Burawoy & Krotov, 1992; McAuley, 1992; Shirk, 1993; Walder, 1995, 2003; Parish & Michelson, 1996; Gerber, 2001). If strategically placed political connections continue to provide access to valuable business information, giving communist cadres a first-mover advantage in emergent markets (Rońa-Tas, 1994) or help to provide priority access to state assets, then a


13

transaction-focused approach should be able to identify the exact sources of such advantage. For illustration of a transaction-focused analysis, we choose four specific institutional domains, exemplifying different levels of market liberalization. Following our first hypothesis, we expect that political capital will devalue the more marketized the institutional domain of the respective economic transaction. The product market is clearly China’s most competitive market. Except for few restricted monopolies such as tobacco, power, telecommunications, and railway, market entry barriers are relatively low, allowing for easy market access particularly in those industries, which require only small amounts of start-up capital and simple production technologies. China’s industrial concentration ratios are low even by international standards. Price controls have been widely abolished. Whereas in the beginning of reforms in China, price controls applied to 93% of agricultural products and 100% of industrial production materials, the shares of price controls were down to 10% and 14%, respectively, by 2000 (Pei, 2006; 125). Market success thus critically depends on time to market, price-quality match, and quality of after-sales services. In parallel, local protectionism, a serious temporary problem during the mid 1980s, has weakened significantly (Li, Hou, Liu, & Chen, 2004) giving rise to rapid growth in interprovincial trade and competition. To measure a firm’s success in the product market, we build on a key feature of entrepreneurial competence, the firm’s ability to successfully launch new product lines. Specifically, we explore whether political connections are associated with a higher share of new products in total sales. In addition, we examine a continuum of three partly liberalized markets with increasing degrees of state power in resource allocation. We select the public electricity market as a state-controlled factor market. In this industry, government assumed direct responsibility for electricity production. Not until after 1999, under pressure of growing industry demand for electricity, did regional initiatives begin to experiment with non-state forms of production. The restructuring of the electricity sector proceeded quickly. By 2003, only 35% of electricity was generated by state-owned firms; already 25% of electricity was generated by foreign-invested companies, including investments from Hong Kong, Taiwan, and Macao (China Data Online). As a performance measure for transactions in the electricity market, we explore the price per kilowatt-hour that firms have to pay. It is well known that governments worldwide utilize government-owned banks to distribute political favors (Sapienza, 2004; Dinc, 2005). As a third

14


institutional domain we include the state-dominated banking sector. Political involvement in financial markets is particularly pervasive in China (Cull & Xu, 2000). In spite of market entry by non-state domestic and foreign banks, the four state-controlled commercial banks controlled about 70% of deposits and loans in 2003 (Datastream). In the same year, private firms and individuals received only about 1% of short-term loans of China’s state commercial banks, including the four state commercial banks, policy banks, and agencies of postal savings (China State Statistical Yearbook, 2005, p. 674). Based on the legitimate assumption that any firm has a latent demand for external finance (Lummer & McConnell, 1989; Uzzi, 1999), we examine whether politically connected firms actually enjoy better access to the formal credit market than their unconnected competitors. Finally we look at the role of political capital in the market for government contracts. Whether in mature market economies, or in China, the market for government contracts is inherently vulnerable to favoritism and bribery. Although the Chinese government has invested great efforts in streamlining public bidding procedures in line with international practice, many of our interviewees doubt free and fair competition among bidders. A young Hangzhou entrepreneur in the business of manufacturing products useful for highway construction projects relies entirely on contracts with local government. His firm submits bids throughout Zhejiang province following the standard guideline for government contracts. Although other manufacturers bidding for the same contract are known to him through public access listing, he occasionally looks into the background of the winning bid, and suspects that the firm won the competition because it has connections in local government. To explore the value of political connections in the market for government contracts, we review to what extent firm’s total annual sales volume accrues to government contracts. In addition to interviews with entrepreneurs in the Yangzi Delta which we conducted from 2004 to 2008, we use data from the World Bank Investment Climate Survey, covering a sample of 2,400 firms of mixed ownership forms in a total of 18 large cities surveyed in the year 2003 to explore the interplay between political capital and transaction outcomes in these distinct institutional domains. From the dataset, we select four different measures of political capital to assess the value of political connections. First of all, we identify whether firm managers hold a party position either as deputy secretary or party secretary. Secondly, we cover whether the government was involved in CEO-recruitment decisions. Further, we include the socalled xia-hai entrepreneurs as a distinct type of cadre entrepreneur. Finally, we control board membership of government officials. We noticed in a firm

15


that specialized in producing industrial pumps for irrigation and hydroelectric projects that one of its board members held a provincial government position. In another firm, the founding entrepreneur came from the same government agency that contracted his firm to do evaluation research for local government. Obviously, government appointed managers of privatized firms, government officials on the firm’s governance structure, and xia-hai entrepreneurs who formerly were cadres in local government constitute clear signals of the expected value of political connections. Table 2 summarizes sample mean comparison tests comparing the performance of firms with political capital with unconnected firms. The pattern we identify is consistent with Hypothesis 1. Political capital secures no advantages in the highly competitive product markets, but captures increasing advantages in weakly marketized institutional domains. This is consistent with what we learned in many face-to-face interviews with entrepreneurs manufacturing products for the consumer market. Even in very successful large firms, where the CEO was not only the party secretary

Table 2.

Sample Mean Comparison Tests: Markets for Private Goods.

Institutional Domain

Product Market

Performance measure

N

Share of new products in total sales

Electricity Market

Market for Government Contract

N

Access to bank loan

N

0.77 0.73

1,322 966

0.20 0.27

1,269 894

4.81 4.39

Government was involved in CEO appointment No 665 35.25 1,726 0.76 Yes 181 37.07 602 0.76

1,707 597

0.23 0.22

1,631 551

3.85 6.50

CEO is a former government bureaucrat No 815 35.56 2,200 Yes 34 34.62 140

2,177 138

0.23 0.17

2,060 130

4.19 11.08

CEO holds party position No 477 36.83 Yes 363 34.07

N

1,343 970

Price paid for one kw h

Credit Market

0.76 0.74

Government is represented on the Board of Directora No 363 34.98 836 0.79 Yes 183 36.14 349 0.70

824 352

0.26 0.35

785 332

Share of sales to government

4.49 4.90

Source: World Bank Investment Climate Survey. po0.10; po0.05; po0.01. a Sample includes firms only which have a board of directors (listed firms, limited liability companies, join stock companies).

16


of the factory’s party branch, but also active in local business and civic associations, the substance of their political involvement in government sponsored organizations and associations is often a merely ceremonial involvement to secure legitimacy, as opposed to reliance on political ties motivated by resource dependence. Many of these CEOs were approached for political office, only after they had successfully built up their firms and had become local celebrities and entrepreneurial role models. In the electricity market, only firms with government officials on the board of directors seem to secure lower electricity prices, while the remaining types of political capital are not associated with significantly lower electricity prices. In the two tightly state-controlled markets (the credit market and the market for government contracts), political capital plays a stronger role. Firms with CEOs who are actively holding party positions and firms with government officials serving as board members have better access to credit. Also, firms with political capital linked to government involvement in recruitment decisions and operated by cadre entrepreneurs win significantly more government bids than unconnected firms. While mean comparison tests only suggest general tendencies, regression analyses confirm the predicted pattern under inclusion of a standard set of control variables (firm size, firm age, industrial sector, and firm location) commonly used in firm-level analysis (see Appendix B). Consistent with Hypothesis 1, political capital is not connected with advantages in liberalized institutional domains, while politically connected firms capture significant economic benefits in governmentcontrolled institutional domains. Economic transactions of firms are naturally not limited to markets of private goods. Equally important, political connections could help firms create market value in their dealings with government authorities and regulators (Rońa-Tas, 1994; Parish & Michelson, 1996). Due to the continuing role of the state as the sole supplier, network advantages and political ties could easily secure preferential treatment in markets for public goods and regulatory markets. We include taxation, licensing, and the legal system as institutional domains, which have an inherent potential for rentseeking activities. All these domains have repeatedly been cited in the literature as key areas, where the political elite may enjoy vast opportunities to create ‘‘new market value for official discretion’’ (Walder, 2003, p. 901). Specifically, we review the firm’s access to tax exemptions, import and export licenses. We also include the perceived security of property rights to respond to the common notion that political capital may in the first place provide an insurance mechanism which lends firms legitimacy and allows for long-term planning security in the absence of rule by law.

17


Table 3 summarizes sample mean comparison tests. Among the four types of political capital reviewed in our tests, only firms with government officials serving as board members seem to consistently secure advantages (in three out of four transactions under review). In addition, CEOs who are holding a party position perceive a greater security of their property rights. These apparent advantages, however, are not consistently confirmed by regression analysis under inclusion of control variables (see Appendix C). Only firms with government officials on the board of directors are associated with a higher probability of holding an export license. Otherwise, the regression results do not suggest systematic advantages for politically connected firms in the regulatory market. Also the perceived higher security of property rights by firms with politically active managers disappears once we control for firm size and firm age. This signals that perceived property rights security is rather a matter of company legitimacy stemming from local market power than pure political affiliation of firm managers. This is consistent with information

Table 3. Institutional Domain Performance measure

Sample Mean Comparison Tests: Regulatory Market. Taxation

N

Enjoys tax exemption

CEO holds party position No 1,362 0.25 Yes 988 0.21

Licensing N

Holds an import license

N

Legal System Holds an export license

N

Likelihood that legal system will uphold property rights

1,300 937

0.21 0.19

1,158 873

62.89 65.83

Government was involved in CEO appointment No 1,755 0.26 1,639 0.11 Yes 611 0.17 575 0.06

1,681 569

0.24 0.12

1,518 526

64.05 65.15

CEO is a former government bureaucrat No 2,235 0.24 2,096 Yes 142 0.19 129

2,131 130

0.21 0.12

1,928 124

64.40 59.34

812 342

0.31 0.41

742 309

64.75 68.10

1,277 924

0.10 0.10

0.10 0.05

Government is represented on the Board of Directorsa No 845 0.31 790 0.15 Yes 356 0.39 339 0.19

Source: World Bank Investment Climate Survey. po0.10; po0.05; po0.01. a Sample includes firms only which have a board of directors (listed firms, limited liability companies, join stock companies).

18


collected in our field interviews. Many of the interviewed entrepreneurs feel that local governments tend to be more accommodating once firms have reached a critical threshold in terms of economic power and local influence. It should be noted that our statistical analysis did not include the procedural aspects of public service provision, such as costs or time spent on securing specific services. Several of our interviewees mentioned that political ties may help to secure faster service and easier information access, while they did not expect a different decision outcome to result from their political capital. Striking are the remaining results presented in Table 3, which indicate that politically connected firms may even suffer certain disadvantages in the regulatory market. Firms with politically inactive managers, and firms without government involvement in management appointment decisions, for instance, are on average more successful in securing tax exemptions than firms with political ties. Also, firms without government involvement in management recruitment and firms not run by cadre entrepreneurs are on average more successful in securing direct export and import licenses than their politically connected counterparts. In the cases of tax exemptions and export licensing, standard regression analysis confirms that politically connected firms are likely to fare worse (see Appendix C). With a general decrease of the value of political capital in market transactions, firms need to invest in other forms of capability development in response to increasing marketization. Given a close linkage between the external environment and a firm’s strategic response (Saloner, Shepard, & Podolny, 2001), the gradual replacement of the redistributive mechanism by market allocation and the resulting empowerment of economic actors combine to motivate strategic adjustments to the emergent market economy, which in turn undermine the previous institutional foundations of firm survival. The greater importance that firm managers attach to the development of firm capabilities is evidenced by rapid strategic adjustment processes. Widespread experimentation with new organizational forms, gradual divestiture of state ownership, and the emergence of new property arrangements illustrate the search for a better fit between firm strategy and external environment (Nee, 1992). Also rapidly increasing investments in research and development confirm a general shift in the firm’s assessment of capability development. By 2003, aggregate R&D expenditures had surpassed India’s and had increased to 1.3% from only 0.9% in 1999 (National Bureau of Statistics/Ministry of Science and Technology, 2005). Aggregate provincial-level data supports the linkage between markets and the development of firm capabilities. Fig. 2 shows a scatterplott of R&Dinput development and marketization at the provincial level (measured by

19 50000 40000 30000 20000 0

10000

100 0

50

innovation funds

150

number of contractual deals in technical markets

200


2

4 6 8 Marketization Index

10

2

4 6 8 Marketization Index

10

Fig. 2. Innovation Activities and Marketization, 1997–2003. Source: Fan and Wang (2003) and National Bureau of Statistics and Ministry of Science and Technology 2005.

the NERI-marketization index) for the period from 1997 to 2003. Consistent with Hypothesis 2, the transition from state socialist redistribution to a market allocation is accompanied by intensified capability development. Nee, Kang & Opper (forthcoming) confirm these mechanisms at the micro-level. Using a cross-sectional dataset covering more than 3,900 firm observations, they show that marketization increases interfirm competition, creates new opportunities for entrepreneurship, and subsequently motivates innovative activity. Their study yields two important findings: First, they confirm a close link between marketization and higher innovative activity by firms. Moreover, their study suggests that marketization is associated with a higher effectiveness of innovative activities and R&D networks. A final prediction of market transition theory points at the crucial role of the opportunity structure provided by markets. Empowerment of economic actors will be most rapid, where marketization opens opportunity structures for new entrepreneurial activities (Hypothesis 3). It is worth noting that the empowerment of direct producers by the law on the books need not precede

20


the emergence of entrepreneurial activities. Instead, we assert that marketization itself creates opportunities and corresponding social structures that endogenously trigger problem-solving mechanisms at the grass root level giving rise to a self-reinforcing process of empowerment of direct producers. We follow White’s (1981) conception of markets as self-reproducing social structures wherein market players establish a pecking order arranged by signals of perceived quality (White, 1981). Conceived, as such, built into decentralized markets are social mechanisms that enable economic actors to develop endogenously the norms and conventions of cooperation, exchange, and competition (Nee & Ingram, 1998; Greif, 2006). Through networks and embedded norms, China’s private entrepreneurs built institutional arrangements that enable them to compete effectively. This includes mutual lending agreements, joint technology development, and network-based horizontal structures linking manufacturers with private sector suppliers and distributors. Although private enterprises lacked the formal institutions – legal status and secure property rights – the informal institutional arrangements of entrepreneurship sustained rapid growth of the private economy. This bottom-up nature of China’s private firm development is clearly reflected by the spatial distribution of private firm development. Direct producers first emerged in rural areas, where the state did not control all distribution channels, and where survival outside of the state-dominated system was easier. In urban state-dominated markets, discriminatory rules and barriers to entry were effectively enforced. Hence, formally registered private companies first operated in isolated rural and peri-urban niche markets where local regulatory control was less restrictive. Not until 2003, when the private enterprise economy was fully established as the most dynamic sector of the Chinese economy, did the central government grant full constitutional recognition of the legitimacy of private ownership forms. Employment data covering the period from 1978 to 2006 (see Fig. 3) illustrate that the main locus of privately owned firms shifted from rural and peri-urban markets to urban China only after the government had formally granted legal equality in 2003 (prior to 1990, official data did not distinguish between rural and urban areas). Consistent with the rural origin of entrepreneurs, multiple surveys confirmed that the newly emerging class of private producers in the early stages of transition was not fueled by the privileged political elite. Instead, founders came from modest educational and class background, often without alternative career prospects as employees or bureaucrats in the state sector and typically without close relations with government officials (Zhang, 2007). By 1991, only about 12% of the rural entrepreneurs had held prior positions as factory or village leaders (Huang, 2008, p. 65). Clearly,

21

120 100 80 60 40 20

04 20

02

00

98

20

20

19

96

94

19

90

88

92

19

19

19

86

19

82

84

19

19

80

19

19

78

0 19

Number of employees (in million)


Year rural individual and private enterprise employment individual and private enterprise

Fig. 3. Individual and Private Enterprise Employment, 1978–2006. Source: National Bureau of Statistics of China, 1991–2006. Disaggregated Private Employment Data for Rural and Urban Sector is Not Available for the Years Before 1990.

entrepreneurship was initially a low status affair, as confirmed by a study on social status of entrepreneurs conducted in 1987, where entrepreneurs ranked 23rd out of 38 occupations (Chen, Li, & Matlay, 2006). Typically, rural revitalization of private production followed and accompanied the opening of free markets. Wenzhou municipality, the role model of individual private firm development in southern Zhejiang province, provides a typical example. Wenzhou benefited from decades of state neglect during the pre-reform era, when the municipality’s state and collective enterprises received only modest state investment appropriations (Whiting, 2000, p. 70). Total state investments reached barely 655 million RMB between 1949 and 1981, while the neighboring municipality of Ningbo had received 2.8 billion RMB (Huang, Zhang, & Zhu, 2008). This left Wenzhou with a relatively underdeveloped state-owned manufacturing sector employing only 8% of Wenzhou’s total workforce in 1978 (Wenzhou City Yearbook, 2004). Per capita income was 55 RMB, compared to the national average of 165 RMB. These conditions virtually forced the population to develop alternative sources of income generation, when market opportunities first opened up in the late 1970s. New producers focused initially on simple goods neglected by large-scale state production such as shoes, toys, textiles, etc.


.8 .6 .4 .2 0

Production share of state-owned firms in total industrial output

1

22

0

2

4

6

8

10

Marketization of product market (NERI)

Fig. 4. Provincial-Level Product Market Development and Relative Shares of State Production in Gross Industrial Output, 1997–2005. Source: National Economic Research Institute (2007).

Producers in Wenzhou were relatively quick to understand the crucial role of market places. It is not a coincidence that Wenzhou witnessed massive clustering of new enterprises. Such cluster effects were supported by the government’s lenient and liberal role in promoting local markets as venues for exchange of goods and information that fueled the entrepreneurial miracle. By 1985, the city registered already 472 market places, with 120 specialized factor markets Liu (1992, p. 297). Scatterplots of the development of provincial product markets and the corresponding share of nonstate industrial production value between 1997 and 2005 support the assumed opportunity effect of marketization (see Fig. 4).

5. DISCUSSION AND CONCLUSION Our transaction-focused approach opens the way for a quantitative comparative institutional analysis to examine the value of political capital as a context-bound outcome of different types of economic transactions in


23

distinct domains of China’s market economy. We show that for transactions in competitive markets, firms with political capital enjoy no significant advantage over firms that have not invested in political connections. We confirm market transition theory’s prediction that political capital persists as a fungible form of capital in markets where government restricts economic activity and controls access to scarce resources. Surprisingly, however, we cannot identify systematic disadvantages of politically unconnected firms in the regulatory markets. This indicates the emergence of a level playing field with respect to public goods provision, which is consistent with the view that China has managed to build a rational-legal government bureaucracy since the start of economic reform. Exceptions are certainly possible. Administrative decisions on public listings, for instance, seem to involve a great deal of political favoritism making market access for politically unconnected firms difficult. Overall, however, we infer from our results that in China’s emerging market economy firms that rely solely on unproductive rent seeking are unlikely to emerge as winners in the intense market competition. Whether as insurance or fungible form of capital, our transaction-focused approach opens the way for developing a quantitative comparative institutional analysis useful not only in studies of transition economies, but also in advanced market economies. Indeed, with respect to political capital, the deepening global financial crisis has increased the value of political connections to firms for largely the same underlying reasons. For example, the Wall Street Journal quipped in reflecting on the rapid increase in the value of corporate connections with Senator Tom Daschle, whose windfall profit of $5.2 million after leaving the Senate became a source of growing controversy in his confirmation hearings. What was of concern was not the failure to pay in a timely manner personal income tax. Instead, ‘‘The real story is the massive transfer of power and wealth now underway from the private sector to the political class. Mr. Daschle could make so much money and achieve such prominence because he was expected to be a central broker in that wealth transfer y. Had Mr. Daschle been confirmed, he would have been the most important man in a health-care industry expected to be $2.5 trillion in 2009, which is larger than the economy of France’’ (February 4, 2009, p. A12). Along a similar vein, the Wall Street Journal (July 22, 2009, p. A8) reported that authorities in Africa and Europe have opened separate investigation of corruption and dumping allegations in deals involving Nuctech Co., a firm closely connected to Hu Haifeng, the son of President Hu Jintao. The younger Hu was the former president of Nuctech and is now

24


the Party secretary of Tsinghua Holdings Co., its parent firm. Whether or not the investigations lead to penalties, the case illustrated the correspondence between state-owned assets and the privileges and advantages of the political elite in China’s market economy. Whether in China or in the United States, political connections are valued by firms in transactions that are directed toward securing competitive advantage to acquire resources controlled by the state. Our findings suggest that such advantages, however, are unlikely to be decisive for overall firm success in China’s intensely competitive market economy. Other empirical studies failed to uncover positive performance effects (based on return on assets, return on equity, or stock returns) for politically connected firms (Qi, Wu & Hua, 2000; Fan et al., 2007; Li, Meng, Wang, & Zhou, 2008). Two main reasons come to mind: First, transactions in state-controlled institutional domains often do not constitute the critical component for survival and profits when viewed from the perspective of the overall range of a firm’s business operations. Second, even if politically connected firms rely heavily on repeat transactions in state-controlled institutional domains, the firm’s capability development combined with management’s ability to detect and react to market opportunities are likely to constitute more decisive prerequisites to pass the market test.

REFERENCES Adamchak, D. J., Chen, S., & Li, J. (1999). Occupations, work units, and work rewards in urban China. International Sociology, 14, 23–441. Baumol, W. J. (1990). Entrepreneurship: Productive, unproductive and destructive. Journal of Political Economy, 98(October), 893–921. Bian, Y., & Logan, J. R. (1996). Market transition and the persistence of power: The changing stratification system in urban China. American Sociological Review, 61, 739–758. Boycko, M., & Shleifer, A. (1993). Privatizing Russia. Brookings Papers on Economic Activity, 2, 139–193. Burawoy, M., & Krotov, P. (1992). The Soviet transition from socialism to capitalism. American Sociological Review, 57, 16–68. Chen, G., Li, J., & Matlay, H. (2006). Who are the Chinese private entrepreneurs? A study of entrepreneurial attributes and business governance. Journal of Small Business and Enterprise Development, 13(2), 148–160. China Data Online. China Data Center. The University of Michigan. http://chinadataonline. org/ Choi, E. K., & Zhou, K. X. (2001). Entrepreneurs and politics in the Chinese economy: Political connections and rent-seeking. The China Review, 1(1), 111–135.


25

Cull, R., & Xu, L. C. (2000). Bureaucrats, state banks, and the efficiency of credit allocation: The experience of Chinese state-owned enterprises. Journal of Comparative Economics, 28(1), 1–31. Datastream. Thomson Reuters Datastream. www.datastream.com Dickson, B. J., & Rublee, M. R. (2000). Membership has its privileges. The socioeconomic characteristics of communist party members in urban China. Comparative Political Studies, 33, 87–112. Dinc, S. (2005). Politicians and banks: Political influences on government-owned banks in emerging markets. Journal of Financial Economics, 77, 453–479. Domanski, H., & Heyns, B. (1995). Toward a theory of the role of the state in market transition: From bargaining to markets in postcommunism. European Journal of Sociology, 36, 317–351. Evans, P. (1995). Embedded autonomy. States and industrial transformation. Princeton, NJ: Princeton University Press. Fan, G., & Wang, X. (2003). NERI index of marketization of China’s provinces. Beijing: National Economic Research Institute. Fan, J. P. H., Wong, T. J., & Zhang, T. (2007). Politically connected CEOs, corporate governance, and post-IPO performance of China’s newly partially privatized firms. Journal of Financial Economics, 84, 330–357. Gerber, T. P. (2000). Membership benefits or selection effects? Why former communist party members do better in post-Soviet Russia. Social Science Research, 29, 25–50. Gerber, T. P. (2001). The selection theory of persisting party advantages in Russia: More evidence and implications. Social Science Research, 30, 653–671. Gerber, T. P. (2006). Getting paid: Wage arrears and stratification in Russia. American Journal of Sociology, 111(6), 1816–1870. Greif, A. (2006). Institutions and the path to the modern economy: Lessons from medieval trade. Cambridge: Cambridge University Press. Guthrie, D. (1997). Between markets and politics: Organizational responses to reform in China. American Journal of Sociology, 102, 1258–1304. Guthrie, D. (1999). Dragon in a three-piece suit. Princeton, NJ: Princeton University Press. Hauser, S. M., & Xie, Y. (2005). Temporal and regional variation in earnings inequality: Urban China in transition between 1988 and 1995. Social Science Research, 34, 44–79. Hayek, F. A. (1978). Competition as a discovery procedure. In: F. A. von Hayek (Ed.), New studies in philosophy, politics, economics and the history of ideas. Chicago, IL: University of Chicago Press. Huang, Y. (2008). Capitalism with Chinese characteristics. Cambridge: Cambridge University Press. Huang, Z., Zhang, X., & Zhu, Y. (2008). The role of clustering in rural industrialization: A case study of the footwear industry in Wenzhou. China Economic Review, 2008(19), 409–420. Keister, L. A. (2009). Organizational research on market transition: A sociological approach. Asia-Pacific Journal of Management. Keister, L. A. (forthcoming). Market Transition. In: R. Wittek, T. Snijders, & V. Nee (Eds), The handbook for rational choice social research. New York: Russell Sage Foundation. King, L. P., & Sze´lenyi, I. (2005). Post-communist economic systems. In: N. Smelser & R. Swedberg (Eds), The handbook of economic sociology (pp. 205–229). Princeton, NJ: Princeton University Press, New York: Russell Sage Foundation.

26


Kornai, J. (1990). The affinity between ownership forms and coordination mechanisms: The common experience of reform in socialist countries. The Journal of Economic Perspectives, 4(3), 131–147. Kornai, J. (1995). Highway and byways: Studies on reform and postcommunist transition. Cambridge, MA: MIT Press. Krueger, A. O. (1974). The political economy of the rent-seeking society. American Economic Review, 64(3), 291–303. Li, H., Meng, L., & Zhang, J. (2006). Why do entrepreneurs enter politics? Evidence from China. Economic Inquiry, 44(3), 559–578. Li, H., Meng, L., Wang, Q., & Zhou, L.-A. (2008). Political connections, financing and firm performance: Evidence from Chinese private firms. Journal of Development Economics, 87(2), 283–299. Li, S., Hou, Y., Liu, Y., & Chen, B. (2004). Survey and analysis on local protections. China Development Review, 6(1), 89–93. Lindenberg, S. M. (2000). A market needs a state: Securing calculability and market-induced values in China. Journal of Institutional and Theoretical Economics, 156, 89–94. Liu, Y.-L. (1992). Reform from below: The private economy and local politics in the rural industrialization of Wenzhou. The China Quarterly, 130, 293–316. Lummer, S., & McConnell, J. (1989). Further evidence on the bank lending process and the capital market response to bank loan agreements. Journal of Financial Economics, 25, 99–112. McAuley, A. (1992). The economic transition in Eastern Europe: Employment, income distribution, and the social security net. Oxford Review of Economic Policy, 7, 93–105. Mincer, J. (1958). Investment in human capital and personal income distribution. Journal of Political Economy, 66(4), 281–302. Mincer, J. (1974). Schooling, experience, and earning. New York: National Bureau of Economic Research and Columbia University Press. Murdoch, J., & Sicular, T. (2000). Politics, growth, and inequality in rural China: Does it pay to join the party? Journal of Public Economics, 77, 331–356. National Bureau of Statistics of China (various years). China Statistical Yearbook. Beijing: China Statistics Press. National Economic Research Institute. (2007). NERI index of marketization of China’s provinces 2006 Report. Beijing: Beijing Economic Science Publishing. Nee, V. (1989). A theory of market transition: From redistribution to markets in state socialism. American Sociological Review, 54, 663–681. Nee, V. (1991). Social inequalities in reforming state socialism: Between redistribution to markets in state socialism. American Sociological Review, 56, 267–282. Nee, V. (1992). Organizational dynamics of market transition. Administrative Science Quarterly, 37, 1–27. Nee, V. (1996). The emergence of a market society: Changing mechanisms of stratification in China. American Journal of Sociology, 101(4), 908–949. Nee, V. (2000). The role of the state in making a market economy. Journal of Institutional and Theoretical Economics, 156, 64–88. Nee, V. (2005). Organizational dynamics of institutional change: Politicized capitalism in China. In: V. Nee & R. Swedberg (Eds), The economic sociology of capitalism (pp. 53–74). Princeton, NJ: Princeton University Press. Nee, V., & Cao, Y. (1999). Path dependent societal transformation: Stratification in hybrid mixed economies. Theory and Society, 28(6), 799–834.


27

Nee, V., & Ingram, P. (1998). Embeddedness and beyond: Institutions, exchange and social structure. In: M. Brinton & V. Nee (Eds), The new institutionalism in sociology (pp. 19–45). New York: Russell Sage Foundation. Nee, V., Kang, J. H., & Opper, S. (forthcoming). A theory of innovation: Market transition and property rights in China. Journal of Institutional and Theoretical Economics. Nee, V., Opper, S., & Wong, S. (2007). Developmental state and corporate governance in China. Management and Organization Review, 3(1), 19–53. Oberschall, A. (1996). The great transition: China, Hungary, and sociology exit socialism into the market. American Journal of Sociology, 101, 1028–1041. Oi, J. (1992). Fiscal reform and the economic foundations of local state corporatism in China. World Politics, 45, 99–125. Opper, S., Wong, S. M. L., & Hu, R. (2002). Party power, market and private power: CCP persistence in China’s listed companies. Research in Social Stratification and Mobility, 19, 103–136. Parish, W. L., & Michelson, E. (1996). Politics and markets: Dual transformations. American Journal of Sociology, 101(4), 1042–1059. Parish, W. L., Zhe, X., & Li, F. (1995). Nonfarm work and marketization of the Chinese countryside. The China Quarterly, 143, 697–730. Pei, M. (2006). China’s trapped transition: The limits of developmental autocracy. Cambridge, MA: Harvard University Press. Qi, D., Wu, W., & Hua, Z. (2000). Shareholding structure and corporate performance of partially privatized firms: Evidence from listed companies. Pacific-Basin Finance Journal, 8, 587–610. Ricketts, M. (2000). Comment on ‘the role of the state in making a market economy’. Journal of Institutional and Theoretical Economics, 156, 95–98. Rońa-Tas, A´. (1994). The first shall be the last: Entrepreneurship and communist cadres in the transition from socialism. American Journal of Sociology, 100, 40–69. Rońa-Tas, A´., & Guseva, A. (2001). Privileges of past communist party membership in Russia and endogenous switching regression. Social Science Research, 30, 641–652. Saloner, G., Shepard, A., & Podolny, J. (2001). Strategic management. Hoboken, NJ: Wiley. Sapienza, P. (2004). The effects of government ownership on bank lending. Journal of Financial Economics, 72, 357–384. Shirk, S. L. (1993). The political logic of economic reform in China. Berkeley and Los Angeles: University of California Press. Staniszkis, J. (1991). The dynamics of the breakthrough in Eastern Europe: The Polish experience. Berkeley and Los Angeles: University of California Press. Stigler, G. J. (1971). The theory of economic regulation. The Bell Journal of Economics and Management Science, 2(1), 3–21. Uzzi, B. (1999). Embeddedness in the making of financial capital: How social relations and networks benefit firms seeking financing. American Sociological Review, 64, 481–505. Walder, A., & Zhao, L. (2006). Political office and household wealth: Rural China in the Deng era. The China Quarterly, 186, 357–376. Walder, A. G. (1995). Local governments as industrial firms. American Journal of Sociology, 101, 263–301. Walder, A. G. (1996). Markets and inequality in transitional economies: Toward testable theories. The American Journal of Sociology, 101(4), 1060–1073.

28


Walder, A. G. (2002a). Income determination and market opportunity in rural China, 1978– 1996. Journal of Comparative Economics, 30, 354–375. Walder, A. G. (2002b). Markets and income inequality in rural China: Political advantage in an expanding economy. American Sociological Review, 67(2), 231–253. Walder, A. G. (2003). Elite opportunity in transitional economies. American Sociological Review, 68(6), 899–916. White, H. C. (1981). Where do markets come from?. American Journal of Sociology, 87, 517–547. Whiting, S. (2000). Power and wealth in rural China. A political economy of institutional change. Cambridge: Cambridge University Press. Wu, X. (2002). Work units and income inequality: The effect of market transition in urban China. Social Forces, 80(3), 1069–1099. Xie, Y., & Hannum, E. (1996). Regional variation in earnings inequality in reform-era urban China. American Journal of Sociology, 101(4), 950–992. Zang, X. (2002). Labor market segmentation and income inequality in urban China. Sociological Quarterly, 43(1), 27–44. Zhang, J. (2007). Marketization, class structure, and democracy in China: Contrasting regional experiences. Democratization, 14(3), 425–445. Zhou, X. (2000). Economic transformation and income inequality in urban China: evidence from panel data. American Journal of Sociology, 105, 1135–1174.

Parish and China, rural, Michelson (1996) whole nation

Domanski and Heyns (1995) Nee (1996)

Parish et al. (1995)

Hungary, national China, rural, Eastern two-thirds Poland, whole nation China, rural, whole nation

Rońa-Tas (1994)

1988

1982, 1987, 1991 1989–1990

1993

1989, 1991

China, rural, 1985 Fujian Province

Nee (1991)

Year

China, rural, 1985 Fujian Province

Data

Nee (1989)

Author Measures of Political Capital

Household income, Administrative employment position

Household income, Cadre status, employment cadre relations

Household income Cadre position, former brigade/ team position Household income Cadre position, former brigade/ team position Household income, Former cadre employment position Income, Administrative employment position in family Income Party membership

Dependent Variable

Regions (sorted by relative industrial output of private, collective, and state-owned firms) Regions (sorted by ratio of non-farm labor)

Repeated tests over time

Repeated regressions by firm ownership

No

No

Before market reform, after market reform

Measures of Marketization (Italics Indicate Indirect/Proxy Measures)

Operationalization

No, no

Yes, mixed

Yes

Yes, mixed

No, no

Yes

Yes

Confirmation of MTT (Decline of Political Capital)

APPENDIX A. MODEL SPECIFICATION OF STUDIES ON MARKETIZATION AND POLITICAL ELITE SURVIVAL Bringing Market Transition Theory to the Firm 29

Russia, national

China, urban, selected areas

1993

China, urban

Dickson and Rublee (2000) Gerber (2000)

Zhou (2000)

1988

China, Guangzhou

Adamchak, Chen, and Li (1999)

1998

1993

1998

1988, 1993

1988

China, urban, whole nation China, urban, Tianjin China, rural, whole nation

Xie and Hannum (1996) Bian and Logan (1996) Nee and Cao (1999)

Year

Data

Author Measures of Political Capital


Household income Party membership Regions (sorted by economic growth) Household income Party membership Different years, professional categories Household income, Cadre status Regions (sorted by employment relative industrial output of private, collective, and stateowned firms) Total monthly Party membership Subsamples for income workers in non-profit institutions and profit-making firms Household income Party membership, Dummy variables cadre status for firm ownership Income Party membership No measure of marketization, time (only one year after start of reforms) Income Party membership Repeated years

Dependent Variable

Operationalization

APPENDIX A. (Continued)

No

No

No

Yes

Mixed, yes

No

No


30 VICTOR NEE AND SONJA OPPER

China, national, urban

China, selected cities

Wu (2002)

1996

China, national, rural China, national, rural

1993

1988, 1995

1996

1993

2000

Russia, national

China, national

1993

1990–1993

China, rural, Shandong province Russia, national

Bian and Zhang (1996)

Walder (2002b)

Choi and Zhou (2001) Walder (2002a)

Rońa-Tas and Guseva (2001) Gerber (2001)

Murdoch and Sicular (2000)

Per capita industrial output Household income Cadre position Village level context (measured by average household income, nonagricultural development, proportion of private household income in nonagricultural production, wage-labor economy, wage employment) Individual income Party membership, Ratio of nonagricultural cadre position labor in non-state sectors over the state sector; ratio of foreign investments to total investments in fixed assets; industrial growth Individual income Party membership, Firm ownership (private administrative versus state) rank

Party membership Private sector, branch, professional category Party membership Professional categries, industrial branches Former cadre Repeated years

Household income Cadre position

Firm profits

Income

Income

Household income Party membership, Time (1990–1993) cadre

Yes

No

No

No

No

No

No

No

Bringing Market Transition Theory to the Firm 31

2000

1988, 1995

China, national

China, national urban

China, national

Russia

China, national, rural

Opper et al. (2002)

Hauser and Xie (2005)

Li et al. (2006)

Gerber (2006)

Walder and Zhao (2006)


Party membership Firm ownership (private, Yes versus collective and state ownership) Yes Marketization, privatization

Measures of Political Capital


Annual growth rate in No PC GDP growth (city level) Political Party membership, Provincial marketization Yes index participation of management firm managers experience in public firm CPSU-member Economic branch, dummy No, yes Per capita for firm type, regional household controls mean wages, income; unemployment, small contracted wage, business scale actual wage Household income Cadre rank, cadre Regions (sorted by No relations relative proportion of agricultural production)

Decision making power of party com. in firms Income

Income

Dependent Variable

Operationalization

Note: Yes indicates that results are consistent with market transition theory (MTT).

1996

1998

2002

China, Zhongshan 2000 City

Zang (2002)

Year

Data

Author

APPENDIX A. (Continued) 32 VICTOR NEE AND SONJA OPPER

33


APPENDIX B. POLITICAL CAPITAL AND MARKETS FOR PRIVATE GOODS Share of New Products in Total Sales Coeff. (Std. Dev.) Political capital CEO is party secretary Government involvement in CEO recruitment CEO is former government official Government official is BoD member Control variable Log of sales volume (lagged) Log of firm age Industrial dummies Region City dummies cons Model Adj. R2/pseudo R2 N

Electricity Price Access to Share of Per One kw h Credit Coeff. Government Coeff. (Std. (Std. Dev.) Contracts in Total Dev.) Sales Coeff. (Std. Dev.)

0.423 (0.156) 3.054 (0.793)

0.010 (0.295) 0.100 (1.572)

0.329 (4.072) 0.122 (0.747)

7.229 (0.977)

0.002 (0.060)

0.075 (0.262)

12.160 (14.673)

0.048 (0.018)

0.069 (1.797)

0.080 (0.707)

3.712 (5.249)

1.009 (1.512)

0.018 (2.590)

0.269 (14.093)

0.308 (3.401)

0.797 (0.420) 8.490 (1.506) 11.056 (0.613) 8.675 (0.725) 38.816 (3.442)

0.030 (1.495) 0.051 (0.617) 0.140 (1.035) 0.020 (0.198) 1.197 (5.705)

0.018 (0.247) 0.131 (0.618) 5.316 (13.597) 0.268 (2.939) 8.460

1.738 (4.606) 16.650 (26.144) 155.022 (117.354) 4.936 (6.396) 171.697 (183.498)

Tobit 0.014 523

po0.10; po0.05; po0.01.

OLS (city clustered) 0.065 1,136

Probit (city clustered) 0.191 1,119

1.515 (2.048) 3.531 (5.443)

Tobit 0.040 1,075

34


APPENDIX C. POLITICAL CAPITAL AND REGULATORY MARKETS Firm Enjoys Firm Holds Firm Holds Perceived Security of Property Rights Tax Exemption Export License Import (Percent) Coeff. Coeff. (Std. Coeff. (Std. License (Std. Dev.) Dev.) Dev.) Coeff. (Std. Dev.) Political capital CEO is party secretary Government involvement in CEO recruitment CEO is former government official Government official is BoD member Control variable Log of sales volume (lagged) Log of firm age Industrial dummies Region City dummies cons Adj. R2/pseudo R2

N

0.240 (2.024) 0.227 (1.725)

0.093 (0.812) 0.222 (2.152)

0.102 (0.659) 0.215 (1.080)

1.140 (0.247) 3.534 (0.521)

0.129 (0.554)

0.060 (0.241)

0.222 (0.687)

5.781 (0.541)

0.056 (0.685)

0.166 (1.733)

0.005 (0.044)

1.373 (0.281)

0.262 (12.292)

0.220 (5.855)

0.225 (6.155)

1.888 (1.537)

0.308 (3.407) 3.370 (7.889) 0.880 (4.312) 1.119 (7.369) 7.493

0.138 (1.726) 3.245 (7.050) 0.231 (1.087) 0.692 (6.354) 6.343

0.169 (2.148) 2.790 (4.129) 0.095 (0.548) 0.231 (1.335) 5.328

3.063 (0.939) 7.528 (0.487) 3.620 (0.463) 31.342 (1.777) 9.961 (0.416)

0.231 Probit (city clustered) 1,150

po0.10; po0.05; po0.01.



0.018 Tobit 1,008

ASSET MANAGEMENT AND PRODUCTIVITY IN REFORM-ERA CHINA Doug Guthrie, Zhixing Xiao and Junmin Wang INTRODUCTION In the spring of 1995, the Electronics Bureau of Shanghai [Shanghai Dianziju] changed its name to ‘‘Shanghai Electronics State-Owned Asset Management Company’’ [Shanghai dianzi guoyou zichan jingying gongsi]. As one official in the former Bureau explained, it had changed its name and its function: It was no longer set up to ‘‘govern’’ or ‘‘manage’’ [guan] Shanghai’s electronics sector; instead it was now an asset management company whose function was to manage the assets of the firms that it owned.1 At the time, the transformation seemed purely cosmetic. Calling itself an asset management company instead of a government bureau was one thing, but actually acting like an asset management company was quite another. Would firms under this former Bureau be any more productive as a result of the change? Would the work-life experiences of the people actually working in these firms change at all as a result? This experience was indicative of a new institutional shift in China’s gradual transition. The second half of the 1990s saw the proliferation of a new type of institution governing firm ownership in China’s transition


35

36

DOUG GUTHRIE ET AL.

economy: the State-Owned Asset Management Company (SOAMCs) [gouyou zichan jingying gongsi]. The logic of the SOAMCs was that, rather than selling off or privatizing state assets, governmental entities would put these assets into the SOAMCs and encourage these entities to act like institutional investors – managing the companies in their portfolios, pushing them to think about issues like operational efficiency, return on investment, etc. This new institutional form eventually reached the highest levels of the government with the founding of the State-Owned Assets Supervision and Administration Commission (SASAC), a central government agency that now owns a controlling stake in 151 of the largest, most powerful Chinese corporations. SASAC, which came into existence gradually over the course of the 1990s (officially ‘‘founded in 2003’’ but the institutions were in place and operating for the better part of the 1990s), was initially set to manage about 6.9 trillion Yuan in assets (about $866 billion); with several high-profile IPOs on the books – including PetroChina’s recent IPO, which gave it (briefly) a market capitalization of over $1 trillion, the value of SASAC’s assets is likely over several trillion USD. The emergence of the AMCs and SASAC is a crucial step in the gradualist model of economic reform process, yet there has been surprisingly little empirical assessment of the impact on performance of this key institution. More important, there has been little exploration on how these new institutions have affected work in Chinese firms. In this paper, we look at the transition of many state offices from governing bodies to asset management companies. We use a variety of data in this analysis: first, we use quantitative data to explore the impact of state institutional investment2 (as opposed to traditional state ownership) on firm performance; second, we employ an in-depth case study of one of China’s most famous joint stock companies to further address the question of the impact of state asset management; third, we rely on personal experiences and observations that have come through working with some of the key organizations that are in charge of asset management in China. Our argument has two layers to it: one methodological and the other substantive. On the methodological level, we argue that in order to really understand the changes in productivity in Chinese firms, we need to look deep into these firms to see what is actually changing with respect to the ways in which work is actually organized. On a substantive level, we argue that the government asset management movement is not purely cosmetic; this movement is, in fact, actually changing the behavior of Chinese firms and the nature of work in Chinese firms. As such, this movement is a middle road – a part of gradualism like many of the other institutional changes that have guided China’s gradual reform process.

Asset Management and Productivity in Reform-Era China

37

GRADUALISM AND THE EMERGENCE OF ASSET MANAGEMENT COMPANIES Many studies have addressed the issue of productivity as a measure of economic reform in emerging economies in general and with respect to China specifically. Since the reforms began, scholars have examined the issue from a macro-level perspective of productivity of industrial sectors or at the provincial level. During the 1980s and 1990s, economists and institutional advisors from the West advocated the rapid transition to market institutions as the necessary medicine for transforming communist societies. This line of reasoning is well known, with prominent figures producing many important theoretical and empirical assessments of the transformation process of many planned economies. The argument here was that private property is a necessary institutional foundation of a market economy, and thus communist societies making the transition to a market economy must privatize industry and other public goods. Private property, the argument goes, is the most efficient way to create the incentives that are necessary for a healthy, functioning market economy. Thus, privatization is seen as the cornerstone of economic efficiency and, by extension, profitability. The radical members of this school argued that rapid privatization – the so-called shock therapy or big bang approach to economic reforms – was the only way to avoid costly abuses in these transitional systems.3 While countries like Russia followed Western advice – constructing market institutions at a rapid pace, immediately removing the state from control over the economy, and rapidly privatizing property – China took its time in implementing institutional change in a gradual and incremental fashion. As the state has gradually receded from control over the economy, it also took the time to experiment with new institutions and to implement them incrementally within the context of existing institutional arrangements.4 In the 1980s, this amounted to localizing the managerial control over enterprises to provincial-, municipal-, and township-level governments.5 At the same time, without privatizing state-owned industrial organizations, the government allowed for the emergence of a private economy that would compete with the state-owned economy. This localization of control, along with the competition from the growing private sector, contributed to the creation of a dynamic rural industrial economy that was the driver of China’s dramatic growth in the first decade of economic reform.6 The localization also had the effect of creating competition among local areas over a variety of scarce resources, from attracting capital from the central government to attracting foreign direct investment.7 In the 1990s, China embarked on a new set of

38

DOUG GUTHRIE ET AL.

projects tied to the gradual reform. Among these were the aggressive creation of a rule-of-law infrastructure that would help shape the economy and society, and the creation of the Shanghai and the Shenzhen stock exchanges. On the face of it, it appears that the stock exchanges became the primary vehicles for privatizing the large state-owned organizations in China’s industrial economy. However, as we show below, even among the publicly traded companies of China’s domestic stock exchanges, the privatization process has been much more complex and much more gradual than previous studies have shown. The success of China’s reforms have illuminated the fact that gradualism is a workable model of economic reform. The success of gradualism can be attributed to a number of factors. First, as a variety of authors have argued, through gradual reform, the government retained its role as a stabilizing force in the midst of the turbulence that inevitably accompanies the transition from plan to market. Institutions such as the ‘‘dual-track’’ system kept large state-owned enterprises (SOEs) partially on the plan and, at the same time, gave them incentives to generate extra income through selling what they could produce above the plan in China’s nascent markets.8 Local governmental control gave local-level officials incentives to treat the firms under their jurisdictions like corporate entities. And gradual control allowed governmental jurisdictions the opportunity to ‘‘teach’’ successful market behavior to the firms under their control.9 Second, the government has gradually pushed ownership-like control down the government administrative hierarchy to the localities. As a result, the central government was able to give economic control over to local administrators without privatization. But with economic control came accountability, and local administrators became very invested in the successful economic reform of the villages, townships, and municipalities under their jurisdictions. This strategy was partly borne out of necessity: as the central government sought to gradually dismantle the redistributive economy, firms in the rural economy were the first to be cut off from funds from state coffers. However, local officials were also given free rein to generate income as they could. In a sense, pushing economic responsibilities onto local administrators created an incentive structure much like those experienced by managers of large industrial firms.10 SOEs were slower to see true reform than their counterparts in the rural industrial economy, but the push for these changes would come in the 1990s. Since then, substantial restructuring of stateowned industry has been central to the reform agenda.11 The key point here is that, although the transition away from the planned economy was a gradual process, managers in many of China’s SOEs were increasingly being


39

handed the main responsibilities that fit with the management of business organizations: although they did not possess the right to transfer assets, they increasingly had the rights to residual income flows, and the power and responsibility of managerial control and this gave them the power to change the ways in which their employees worked.12 Privatization and the Shanghai and Shenzhen Stock Exchanges Coinciding with the legislative changes of the 1990s was the founding of the Shanghai and Shenzhen stock exchanges. The first of these, the Shanghai Stock Exchange, opened for business on December 19, 1990, with the Shenzhen Stock Exchange opening shortly thereafter. By the end of 2004, the number of domestically listed companies in China had risen to 1,371 with a total market capitalization of 525.6 billion USD.13 Following the gradualist model, the Chinese government’s construction of the institutions that govern public ownership has been spread across the period. After a series of regulations such as ‘‘the Opinions on Standardizing the Joint Stock Limited Companies’’ and ‘‘the Provisional Regulations on the Administration of Issuing and Trading of Stocks’’, the Securities Law of the People’s Republic of China was adopted in 1998 at the Ninth National People’s Congress and took effect in July 1999, thus institutionalizing the legal basis for the standardized operation of listed companies.14 Finally, in 2001, the central government passed The Tentative Measures for Decreasing State Shareholding.15 Yet, as China has been systematically constructing the institutions of a publicly traded economy, even in the area of public ownership of listed companies, we must acknowledge the complexities of enterprise–state relations in the Chinese model, as the government’s receding from control over publicly listed state enterprises has, like every other institutional change in the Chinese economic reforms, been a gradual process. The companies listed on China’s domestic stock exchanges are becoming ‘‘privatized’’ in some ways. A typical ownership transformation for a SOE would allow the state to retain between 40% and 50% of the company’s shares; between 20% and 30% of the shares are designated for institutional shares; the remaining 30% of shares are designated for public consumption as free floating shares. Transforming SOEs through New State Institutions Even with the Company Law and the emergence of the Shanghai and Shenzhen stock exchanges, state ownership of publicly traded firms has

40

DOUG GUTHRIE ET AL.

declined only by degrees, as the state maintains a controlling interest in these firms in variety of ways. First, while public ownership has formally transformed ownership structures of many formerly SOEs, enterprise–state relations in the Chinese system have remained considerably complex, as the state maintains a controlling share in the publicly traded firm. Previous studies have shown the shifting organizational relationships of a typical publicly listed firm as well as the ownership categories that emerge over the course of the initial public offering process. Most often it is the case that a listed company is the strongest performing factory or group of factories in a larger group company [jituan gongsi].16 The state office overseeing the group usually has a heavy hand in deciding which part of the group will be spun off for the IPO. The state maintains the basic state–firm relationship with the remaining part of the group company – it remains on as the advisory administrative office, playing a significant role in the strategic decisions the firm or group makes as well as maintaining a hand as the partial residual claimant on firm profits.17 The IPO firm becomes a separate legal entity, subject to the Company Law and the Securities Law independent of the group. In some cases in which firms are not part of a group, the entire factory may go public; however, IPO processes more often involve a spin-off situation. In the initial stages of the IPO, the state administrative office and the group company will maintain control over between 20% and 60% of the shares, though in the early years of the stock exchanges these numbers were closer to 70%. The remainder of the shares is divided between various types of institutional and free-floating shares.18 Initially, the overlapping ownership structure of China’s publicly traded firms remained very localized, with owners concentrating on investing in firms that are either in their local capital markets (Shanghai or Shenzhen) or at their level of governmental administration (central or provincial). Over the course of the 1990s, these relationships have become much less localized.19 Second, over the course of the 1990s, the state has increasingly maintained a role in these firms through the emergence of SOAMCs and SASAC; we might also think of these as ‘‘institutional investors’’ as they are organizations set up (at least in theory) to manage a portfolio of assets. Although China has been careful to always follow its own path with respect to the transformation of the economy – especially in the area avoiding rapid privatization – the formation of the Shanghai and Shenzhen stock exchanges changed this equation. In order to attract institutional investors from abroad, the state needed to demonstrate a commitment to a declining presence in these publicly listed firms. It did so through the creation of SOAMCs. At this early stage, in practical terms, little changed for these


41

offices – they still employed the same people and they still maintained the same relationships with the firms under their jurisdictions in terms of managerial and ownership control. Today, there are literally hundreds of organizations in the Chinese economy that used to be government administrative offices and now call themselves asset management companies or that are owned by state offices (see Fig. 1). In the mid-1990s, the largest and most famous of these are the SOAMCs connected to each of the four main central government banks (set up to help these banks move bad loans off of their books); however, there are many more spread throughout the economy today. As with the SOAMCs owned by the banks, these smaller SOAMCs are usually owned by a local government administrative office. Today, however, the largest and most powerful of the asset management institutions is SASAC, the central government commission charged with transforming the largest and most powerful of the firms in the state economy.20 Fig. 1 shows an ideal-typical rendering of this transformation. Fig. 2 shows the layers of asset management companies in an actual organization of China’s domestic stock exchanges. We examine the extent to which the emergence of these new institutions has influenced the governance of SOEs in recent years.

QUANTITATIVE ANALYSIS: PUBLICLY TRADED CHINESE FIRMS Data and Methods This paper has two empirical parts: In the first part, we analyze the performance of Chinese publicly listed firms during 1994–2003. My starting point for data on Chinese listed companies is the WindDB database, which reports all of the financial data for publicly listed firms in China’s domestic economy (see Guthrie et al. [2008] for description of data). These data represent the complete set of domestically listed publicly traded firms listed on the Shanghai and Shenzhen stock exchanges (1,371 firms by 2003). From the database, we obtained the information of shareholders who hold 5% or more in each of these listed companies. With the goal of uncovering who actually owned shares in each company in a given year – as opposed to the listed shareholder – we also traced the owners of each listed shareholding entity through supplementary information including company’s annual reports, announcements, website information, stock market research by stock analysts, and so forth. In some cases, we uncovered up to six

Factory

Factory

SOAMC

Factory

SOAMC

Factory Factory

Factory

Factory

Factory

Factory

Factory

SOAMC

Factory

Factory

SOAMC

Factory

Factory

Village Govt.

Factory

Factory

Township Govt. Direct control over (Bureau) township and village enterprises (TVES)

Factory

District Company

Direct control over urban dist. firms District Govt. (generally urban collectives) (Bureau)

Factory

Municipal Company

Direct control over Mun. Govt. firms (generally med-scale SOEs); also policy control for sector within municipality

Factory

Factory

SOAMC

Factory

SOAMC

Provincial Govt. Direct control over Prov. Govt. firms (generally large-scale SOEs); (Bureau) also policy control for sector within province

Fig. 1.

Administrative Hierarchy of the Industrial Economy.

Figure based on interviews with industrial managers and government officials. See also Walder(1992, 1995) and Guthrie (1997, 1999) for further discussion of the “nested hierarchy”of state administration.

SOAMC

Factory

SOAMC

Central Govt. Direct control over Central Govt. firms (generally large-scale SOEs); also policy control for entire sector (Ministry)

42 DOUG GUTHRIE ET AL.

43


POTEVIO CORPORATION*

PUTIAN EASTCOM GROUP 51.59% State inst. shares

95%

*(Formerly Bureau of Youdian Gongye, under the Ministry of Youdian)

90%

HANGZHOU EASTCOM INDUSTRIAL CO., LTD 5%

ZHUHAI EASTCOM INVESTMENT CO., LTD. 5.73% Domestic institutional shares

EASTERN COMMUNICATIONS CO., LTD.

Fig. 2.

F-I

Tianjin Motorola

Ownership of Eastcom (2001).

identifiable layers of ownership ‘‘shells.’’ We coded each real shareholder (the ultimate owner as opposed to the shell). Due to missing data and differential reports among some shares/companies, the final available data for our study are from 328 companies in 1994 and 1,305 companies in 2003. In addition, these numbers differ from the numbers listed above (particularly in the case of 1,371 versus 1,305) because in a number of cases, A and B shares are listed as different companies, when in fact they are listings of the same company. Once correcting for this factor, our actual number of cases that are omitted due to missing data are 15 cases or about 1% of the population in 2003.21 Two approaches to the question of the impact of ownership have become common. One of these examines the effectiveness of privatization by comparing the performance of privatized firms with those that are still governed by state ownership (e.g., Boardman & Vining, 1989; Frydman et al., 1999); a second compares the performance of firms over time, as they make the transition from state owned to private (Megginson et al., 1994). We take an alternative approach to this problem. We look at the degree of state ownership in publicly traded firms. Conceiving of state ownership as a continuous variable of the percentage of shares owned by a given state

44

DOUG GUTHRIE ET AL.

office, we look at the impact that percentage ownership has on various performance outcomes. Because this variable is a continuous variable, it is not framed as a binary comparison of state owned or not; rather the notion of the degree of state ownership allows us to examine of question of whether an increase in state ownership has positive, negative, or neutral effects on a firm’s performance. We then look at firms in categories based on who is the dominant shareholder – a traditional state office, a state institutional office, or a private shareholder. This approach may make the variables appear to be categorical variables, but it is important to note that the underlying variables defining dominant ownership are continuous and thus give us a more fine-grained sense of differences in ownership structure among firms. In the models we present below, private and state institutional are compared to state-dominant ownership, which is the left-out category for these variables. Variables We focus on the profitability and operating margins of these organizations: dependent variables include net profits and the operating margins of the firm (Holz, 2002). Net profits are calculated as gross profits minus indirect costs (i.e., costs not directly attributable to production that were already removed from sales revenues to arrive at the gross profit figure).22 Performance or operating margins are calculated as profits divided by revenues.23 In addition to the dominant ownership variables, we include several control variables in the analysis. We control for volume of economic activity by including the natural log of sales. We use debt-to-asset ratio as a measure of firm distress. A dummy variable for firm listing allows us to control for differences between the Shanghai and Shenzhen stock exchanges (Shenzhen ¼ 1). Chinese firms are officially broken down into the following categories: agriculture, mining, manufacturing, electronics, construction, transportation, information and technology, trade (wholesale and retail), insurance, real estate social services, communications, and ‘‘others’’. The lion’s share of listed firms is categorized as ‘‘manufacturing’’ (57.9%). In addition, the domestic exchanges collapse these sectors into five primary categories: agriculture and mining, manufacturing, energy and construction, services, and others. We follow these categorizations comparing the four combined sectoral categories against the left out category of manufacturing. Finally, we also include dummy variables for each year. Table 1 presents the means, standard deviations, and definitions for the critical variables included in our analysis. As Table 1 shows, over the course of the first decade of publicly traded firms in China, these organizations had

45


Table 1. Variables Net profita

Operating margins Trading city

Means, Standard Deviations, and Definitions. Means

Standard Deviations

Definitions

67,400,000

391,000,000

Net profit ¼ gross profit minus indirect costs (i.e., costs not directly attributable to production that were already removed from sales revenues to arrive at the gross profit figure) Profits divided by sales Dummy variable (Shenzhen ¼ 1; Shanghai ¼ 0) Debt divided by assets Natural log of gross sales Multiple owners with more than 5% holdings (1 ¼ yes) Percentage of shares controlled by the 10 largest shareholders Dummy variable, state shares largest (1 ¼ yes) Dummy variable, state institutional investor largest (1 ¼ yes); (SASAC among others) Dummy variable, private institutional shares largest (1 ¼ yes) Year trend Dummy variable, agriculture/mining (1 ¼ yes) Dummy variable, energy/construction (1 ¼ yes) Dummy variable, services (1 ¼ yes) Dummy variable, others (1 ¼ yes)

.20 .45

.23 .50

.49 19.98 .52

.44 1.27 .27

41.39

13.89

.41

.49

State inst. inv. dominant

.49

.50

Private dominant

.04

.19

6.77 .03

2.51 .17

Energy/const.

.06

.23

Services Others

.28 .08

.45 .27

Debt to asset ratio Sales (ln) Mult. owners (W5% shares) Ownership concentration State dominant

Time trend Ag./mining

a

RMB.

revenues of 1.16 billion Yuan (B$143 million). The average net profit of these firms over the last decade was 69 million Yuan (B$8.6 million), and cash operating profits averaged 117 million Yuan (B$14 million). The average operating margin was 21%, and the average price-to-equity ratio was a surprising 111. These firms had an average percentage of state shares of 11%, an average percentage of state institutional shares of 13%, and an average of percentage shares of less than 1%. These numbers are somewhat misleading, however, because many of the firms have zeros in each of these categories, a fact that brings the overall averages down. The numbers in parentheses beneath the means are the average percentages excluding the

46

DOUG GUTHRIE ET AL.

zero categories; in other words, among firms for which the Chinese state is a principal, the state controls an average of about 35% of the shares. Finally, the average percentage of shares held by the top 10 shareholders for these firms is nearly 60%. We model the determinants of firm profitability and efficiency using both random effects models: Y it ¼ a þ c0 Z it þ b0 ðDOÞit þ di þ lt þ it

(1)

where Yit is the outcome for firm i, i ¼ 1, y, n, at time t, t ¼ 1, y, T, Zit a column vector of measured variables, c a column vector of regression coefficients corresponding to Zit, b the effect of dominant owner (DO)it, and eit a random error term with mean E(eit|Zit, (OC)it, di, lt) ¼ 0 and variance s2. In the random effects formulation, the di are independent and identically distributed random variables representing unmeasured time-invariant characteristics of the i(th) firm with mean 0 and variance s2d . Similarly, the lt are independent and identically distributed random variables with mean 0 and variance s2l that represent unmeasured characteristics that vary over time but are common to all firms. The error term is vit ¼ di þ lt þ eit, where the three error components are assumed to be independent of each other and independent of regressors. Because the errors vit are correlated, the generalized least squares (GLS) estimator is more efficient than the OLS estimator.

Results from Quantitative Analysis Table 2 presents the analysis of the net profitability (Model I) and operating margins (Model II) of the publicly traded firms on the Shanghai and Shenzhen stock exchanges. Models I and II show the effects of SASAC- and private-dominant ownership compared to state-dominant ownership. Model I shows support for the notion that firms with a SASAC-dominant shareholder are both more profitable and more efficient than those with traditional state office as the dominant shareholder: there is a statistically significant positive relationship between private-dominant ownership compared to state-dominant ownership and a smaller positive effect between of SASAC-dominant ownership. Thus, based on Model I, on average, for the population of China’s domestic publicly traded firms, controlling for other factors, firms with a dominant SASAC owner make about 14,600,000 Yuan in net profits and 10,200,000 Yuan in cash operating profits (about $1.8MM and $1.3MM) more than other publicly traded firms


47

Table 2. Random Effects Models for Profitability and Operating Margins, Chinese Publicly Traded Firms (Shanghai and Shenzhen Stock Exchanges), 1994–2003. Net Profitsa Shenzhen Sales (ln) Debts/assets Owner concentration Mult. owners (W5%) State inst. inv. dominant Private dominant

37.90 (24.40) 57.70 (2.75) 55.70 (4.60) 1.09 (.25) 13.00 (6.56) 15.20 (8.32) 16.40 (8.70)

Sector cont. Ag./mining

316.00 (63.00) 125.00 (52.00) 25.40 (28.00) 14.90 (52.00)

Energy/const. Services Others Time controls Yr. trend Yr. controls Constant N R2

5.34 (1.16) Included 1,080 (61.70) 9,101 .10

Operating Margins .003 (.004) .00 (.00) .03 (.005) .003 (.000) .001 (.004) .008 (.004) .02 (.01) .006 (.01) .19 (.09) .02 (.005) .001 (.008) .01 (.001) Included .02 (.04) 9,101 .37

po.001; po.01; po.05 (one-tailed tests). a

Coefficients dividend by 1,000,000.

in the domestic economy; private firms make about 16,300,000 and 12,600,000 Yuan (about $2.04MM and $1.6MM) more than other publicly traded firms. Model II examines whether the associations discussed above would also apply to firm efficiency (operating margins). These results were

48

DOUG GUTHRIE ET AL.

consistent with our above findings. There is a statistically significant positive relationship between having a private sector principal as the dominant owner and firm efficiency. Firms with dominant principals from the private sector do about 2% better in terms of operating margins than firms with a dominant state owner. There is also a statistically significant positive effect of having SASAC as the dominant owner, compared to state ownership – though the effect is less than half of that for private-dominant ownership. These results suggest that there is, in fact, a significant difference between SOMAC ownership and the traditional model of state ownership. That is a significant point in and of itself, though this argument has been made in other places (see, e.g., Guthrie et al., 2008). However, what we still do not have a clear picture of why SOMAC and SASAC firms are more productive than their state-owned peers. While some would contend that it is the strategic advantage of being a favored firm within the state system – and this may be true – we emphasize here the importance of SASAC’s commitment to changing incentives and changing the environments in which workers do their work. There is no doubt that SASAC firms are among the most favored firms that are still connected to the Chinese state; however, they have also been deeply committed to getting incentives right, emphasizing managerial principles, and helping firms learn how to compete on global markets.

CASE: CNPC, PETROCHINA, AND THE TRANSFORMATION OF A STATE-OWNED GIANT One weakness of the analysis presented above is that we do not have a clear sense of exactly where the productivity gains come from. For this, we turn to one of the more famous cases of SOE transformation in China today. While many observers credit PetroChina’s success to its duopoly status and special strategic position within the Chinese government, the company has also been revolutionized in terms of incentive structures that link directly to worker productivity. As many scholars have documented this case, we draw heavily here on past work that has been done on the case. In this case as with many other transforming SOEs, observers have often attributed the organization’s success to its close strategic relationship with the Chinese government. As one of the key organizations in a sector that is crucial to China’s growth prospects, the organization does enjoy a privileged position in the world’s third-largest economy. Success is all but assured, given this strategic advantage. However, this view misses just how much asset


49

management in China has changed the performance of Chinese firms. It also misses the fact that successful asset management can be directed by state organizations – perhaps not as successful as private institutional investors, but successful nonetheless. We place special emphasis here on PetroChina’s relationship to SASAC and to key foreign investors, like Warren Buffett, and on its approach to revolutionizing incentives for the workforce. In this section, we will explore PetroChina’s productivity, highlighting not just its strategic positioning in the marketplace (and with respect to the Chinese government) but also with respect to the changes in incentive structures workers have experienced in the workplace. Ultimately, productivity is dependent on putting the right incentives in place and on building a culture of productivity in the workplace, and PetroChina has been successful at doing just that. This discussion will serve as a springboard for a broader discussion of work and productivity in China’s transforming state sector. PetroChina is an exchange-listed subsidiary of the China National Petroleum Company (CNPC), which currently produces approximately 66% of China’s oil and gas. CNPC owns nearly 87% of the company, and the remaining equity shares are listed as H-shares in Hong Kong, ADRs in New York, and on the Chinese domestic exchange in Shanghai. As a subsidiary of a SOE, PetroChina enjoys a duopoly in the Chinese domestic market. It is also one of the world’s largest energy companies, and the largest oil and gas distributor and producer in China. In 2007, the company became the world’s largest publicly held firm when its market capitalization briefly topped $1 trillion following a float on the Shanghai Stock Exchange that saw its shares triple in value. In the same year, the company made the single biggest discovery of oil within China in 50 years when it found an estimated 7.5 billion barrels of oil in a field at Bohai Bay. The company has stakes in over 16,620 service stations to provide retail distribution for its proven reserves of 11.6 billion barrels of oil and 53.5 trillion cubic feet of natural gas. The distribution network also includes 12 chemical plants and 26 refineries as well as 20,590 km of natural gas pipeline.24 It has focused on increasing the extent of its vertical integration as well as on expanding its operations abroad. Currently, PetroChina’s exploration and development businesses operate in 11 foreign countries and portions of the company are also involved in the production and marketing of crude oil and natural gas: refining, transportation, storage, and marketing of crude oil and oil products; production and marketing of primary and derivative petrochemicals; and transportation of natural gas, crude oil, and refined oil. Table 3 shows the productivity statistics of the PetroChina; there is no denying that the corporation is extremely successful in terms of efficiency and profitability.

50

DOUG GUTHRIE ET AL.

Table 3.

Productivity at PetroChina, 2004–2008.

Efficiency Ratios

Profitability Ratios

Cash Conversion Cycle Revenue per Employee

25 245,051

Net Income per Employee Average Collection Period Receivables Turnover Day’s Inventory Turnover Ratio Inventory Turnover Inventory/Sales Accounts Payable/Sales Assets/Revenue

42,735 25 0 0 0 10.6 4.84 0

Net Working Capital Turnover Fixed Assets Turnover Total Assets Turnover

25.24 1.01 0

Revenue per $ Cash Revenue per $ Plant

12.75 1.09

Revenue per $ Common Equity Revenue per $ Capital Invested Selling, General & Adm/time (SG&A) as % of Revenue SG&A Expense as % of Revenue – 5YEAR AVG Research & Development (R&D) as % of Revenue R&D Expense as % of Revenue – 5YEAR AVG

1.14 1.08 12.2 11

Free Cash Flow Margin Free Cash Flow Margin – 5YEAR AVG Net Profit Margin Net Profit Margin – 5YEAR AVG Equity Productivity Return on Equity (ROE)

4.1 0.53

Return on Equity (ROE) – 5YEAR AVG Capital Invested Productivity Return on Capital Invested (ROCI) Return on Capital Invested (ROCI) – 5YEAR AVG Assets Productivity Return on Assets (ROA) Return on Assets (ROA) – 5YEAR AVG Gross Profit Margin Gross Profit Margin – 5YEAR AVG EBITDA Margin – LTM EBIT Margin – LTM Pre-Tax Profit Margin

22.6

17.4 21.3 1.14 19.9

1.08 18.8 21 0 13.7 15.4 62.1 57.6 26 20.2 24.5

0

Pre-Tax Profit Margin – 5YEAR AVG Effective Tax Rate

30.2 24

0

Effective Tax Rate – 5YEAR AVG

26.9

Source: http://www.advfn.com/p.php?pid ¼ financials&symbol ¼ NYSE%3APTR

A Brief History of CNPC/PetroChina In 1988, the Chinese government disbanded its Ministry of Petroleum Industry and formed three state-owned companies to oversee and modernize its petroleum-related activities. The China Offshore Oil Corporation (CNOOC) was formed to handle offshore production, the China Petrochemical Group


51

(Sinopec) focused on refining and petrochemical production, and the CNPC took over all onshore exploration and production of oil and gas. The reforms leading to the abolishment of the Ministry of Petroleum Industry represented the first major steps the Chinese government took toward reform of heavy industries. Up until that point, most economic reform efforts were primarily focused on agricultural, light manufacturing, and commercial industries. The government had been more conservative with companies engaged in the manufacture of steel, machines, trucks, and other heavy industrial goods because of their importance to the economy as a whole. During the 1980s, heavy industries were the largest employers and providers of social services to urban residents. They were also the primary source of revenue at many different levels of the government. While the government made some incremental reforms in the energy industry by shifting production from government ministries to SOEs, the large number of people that were dependent upon on these companies for their livelihoods made the perceived consequences of any potential errors very serious. As a result, the newly formed SOEs continued operating as if they were still parts of the Ministry of Petroleum Industry because the government was willing to overlook the systematic failure of such companies to turn a profit due to overstaffing and low productivity (Wu, 2002).25 In the case of CNPC, the weaknesses in productivity were dramatic. To put the productivity in perspective: When a CNPC delegation visiting international oil companies found that the Norwegian state-owned oil company Statoil employed 80,000 people and produced 130 million tons of oil annually. CNPC had comparable annual production of 140 million tons, but at the time it employed around 1.4 million workers (Rowe, 2006). While not all of these workers were directly involved in oil exploration or production activities, CNPC was still employing over 11 times as many people to produce a comparable amount of oil. Some of the difference in staffing between the two companies can be explained by the fact that it was Statoil’s standard practice to outsource any aspects of its business not directly related to oil exploration or production. CNPC, on the other hand, used in-house divisions to provide all of its construction services, well logging, drilling, and engineering services. Such a large and varied organization was made even more difficult to manage profitably because the company’s legal structure afforded senior executives relatively little influence over their regional counterparts. While CNPC and its subsidiaries were technically operating as one company, in reality the level of integration between the various entities was very low.

52

DOUG GUTHRIE ET AL.

1998 Reorganization of the Oil and Gas Industries By 1997, it was evident that the social welfare burdens and organizational inefficiencies at the state-owned oil companies were threatening their ability to provide adequate income for their employees while continuing to grow economically. CNPC had been experimenting with the possibility of creating oil companies with international operations within the Chinese national petroleum companies. Many at CNPC believed a focus on international operations could provide the impetus needed for the company to make a turn around. Opening up to external pressures would drive reforms forward necessitating a return to the company’s core business and enabling the executive team to do away with the antiquated management system. The first step the government took was to remove the artificial functional segments that had been created when the production of oil was overseen by government ministries. This was seen as a crucial step toward achieving the government’s twin goals for the reorganization: first, to separate the industry oversight and business management functions, and second, to better position CNPC and Sinopec Group as efficient, competitive companies.26 To establish new market segments, the government organized a series of asset transfers between CNPC and Sinopec Group that significantly increased vertical integration at both companies. CNPC transferred to Sinopec many of its oil and gas production interests in the Southern and Eastern China, and in return it received 15 petrochemical plants and refineries in Northern and Western China. The Ministry of Chemical Industry transferred its oil and gas interests to CNPC, and the petroleum distribution companies owned by provincial-level governments were split between CNPC and Sinopec. These changes resulted in a regionally segmented industry, and government continued its industry-level reform efforts by shifting all industry oversight roles previously filled by CNPC and Sinopec to the newly formed State Bureau of Petroleum and Chemical Industry.

Problems at CNPC The management at the newly restructured CNPC realized that to become an efficient and competitive organization after operating for so many years as a government ministry would require a radical transformation. While CNPC had been established in 1988 to improve the competitiveness of China’s oil and gas industries, in the 10 years leading up to the 1998 reforms the company had been unable to make a break with its past as a government


53

ministry. Though no longer formally part of the government, it had continued to face many of the same problems and inefficiencies the Ministry of Petroleum Industry had encountered before it was disbanded. When CNPC was formed as a SOE it inherited responsibility for nearly all aspects of its employees’ lives. This imposed both operational and financial burdens, as the lack of market mechanisms made it necessary for CNPC to establish and operate its own social services such as hospitals, schools, court systems, and police departments. In fact, many of the oil fields the company operated were effectively self-contained cities presided over by its general manager, with around 500,000 employees working in social services positions unrelated to oil exploration or production. The danwei system made it all but impossible to fire workers, and CNPC’s payroll had ballooned to 1.54 million employees and 400,000 retirees by the early 1990s. Additionally, workers had ample opportunities to take advantage of the system. For example, Smyth, Qingguo, and Jing (2001) found that at Fushun CNPC, employees could apply for li gang status (leaving their post) when they reach age 45 for women or age 50 for men, and over 7,500 employees had secured such status in 1999. These workers continued to receive 80% of their salaries as well as all social benefits given to normal employees. The system is even more generous at some subsidiaries such as the CNPC’s Fushun Catalytic Plant, where li gang workers also receive private medical and property insurance. Programs of this nature exacerbated the burden of providing social services to the point that nearly all of the revenues from the company’s oil businesses were used to provide social services. The company’s dysfunctional financial system essentially placed each regional division in charge of its own money, thereby removing much of its incentive to grow profits across the firm. This left CNPC’s executive management unable to effectively implement strategic initiatives, and it gradually assumed the role of administrator to ‘‘powerful subordinates who operated their own businesses’’ (Rowe, 2006, p. 6). In this environment, many regional managers began making ill-advised investments in selfbenefiting pet projects often those unrelated to oil exploration and production. As government financial support continued to decline throughout the mid-1990s, it became increasingly clear to many at CNPC that unless they acted quickly the firm could be facing some very serious problems. Executives realized it was no longer feasible to rely on the government for financing, and the company began searching for new sources of capital. By 1997, the company’s government financing was for all intents and purposes zero, but it had still not been successful in securing the needed capital. Later that year, a company analysis predicted that, assuming

54

DOUG GUTHRIE ET AL.

the price of oil would remain relatively stable, CNPC would be unable to fulfill one-third of its required capital in each of the upcoming 3 years. In response, the company’s president Zhou Yongkang formed a 10-person team headed up by Gong Huazhang that was tasked solely with raising enough new capital for the company to continue its development. The team’s analysis quickly revealed that, as an intact company, CNPC would almost certainly be unable to raise sufficient capital for three key reasons: first, the company’s outdated management practices, social burden, and low efficiency effectively disqualified it from holding an international equity offering; second, because of its vast size it was also highly unlikely that it could raise enough through an offering on the still-immature domestic market; and third, very few banks were willing to lend any significant amount to CNPC because of its high debt/equity ratio and its projected capital shortfalls. Managing the Reform of CNPC27 CNPC first began exploring the idea of creating an international oil company within the model of the Chinese oil and gas industry when its senior management established the Enterprise Reform Office in 1992. The management team knew that it was in need of a serious restructuring to eliminate the multiple legal entities and levels of management, refocus the company on its core businesses, expose the company to external pressures to drive reform, and increase the efficiency with which the company used its resources. Shortly after a state government reform proposal was drafted in 1995, China International Capital Corporation and Goldman Sachs Asia approached CNPC with a proposal to provide advice regarding the restructuring and potential IPO. (CICC is a joint venture between four companies including the Construction Bank of China and Morgan Stanley.) Shortly thereafter the Enterprise Reform Office made a formal proposal to CNPC’s president that the company should split itself into three sectors and then hold an IPO for the oil and gas division. In 1997, the proposed changes were approved by Zhou Yongkang, the president of CNPC, and CICC, and subsequently brought before the State Council for approval. The government voiced strong support for the deal and encouraged CNPC as well as other large SOEs to enter the international capital markets. Coinciding with the government’s broader reforms in the oil and gas industry, the Ninth People’s Congress named Zhou Yongkang the Minister


55

of National Land and Resources. His successor at CNPC, Ma Fucai, continued to push the restructuring plans forward, eventually creating the Public Listing Preparation Team, led by Jiang Jiemin, in February 1999. The Public Listing Preparation Team identified four steps for the company’s restructuring: first, CNPC’s social functions would be returned to government; second, CNPC’s noncore services would be consolidated into a company still owned by CNPC; third, transfer all the core oil and gas operations to a new subsidiary (PetroChina); and fourth, to publicly list PetroChina in the United States, Europe, and Hong Kong. Returning Social Functions to Government: The first step of this plan presented several potential pitfalls. Because the company was relying on many different local governments to take over its social services, CNPC needed to negotiate the transfer terms with each government individually. This was often difficult because the governments wanted assurances that taking over CNPC’s social functions would not increase their financial burdens, especially because the elimination of the multilevel legal entity system would shift a significant portion of the tax revenue from CNPC away from local governments to the central government. To avoid these problems, CNPC secured temporary subsidies from the central government for the transferred social services organizations. Because these subsidies would expire when the restructuring was complete, many local governments were still worried about lost tax revenue. Because the central government strongly supported the reform efforts, it eventually agreed to leave the tax structure of CNPC unchanged despite the restructuring. Noncore Services: To maximize efficiency at PetroChina, it was decided that the noncore services operated by CNPC would not be publicly listed, but would instead remain a part of the parent company. To determine which of its businesses would be considered parts of noncore services, CNPC’s senior management identified three different categories of noncore businesses. The management team classified all of the businesses as companies focused on oil engineering/technical/production services and property management, companies only engaged in oil engineering/technical/production services, or companies involved only in property management and logistics. The goal for the noncore services businesses was to become an internationally competitive provider of petroleum engineering and technical services, which necessitated laying off redundant workers. The major challenge at this step of the restructuring was managing the expectations and perceptions of the employees who would be staying in the unlisted noncore services sector. Many of them felt as though the senior management team was listing the best part of the company and effectively deserting its remaining employees.

56

DOUG GUTHRIE ET AL.

To counteract this sentiment, CNPC privatized some of these businesses and transferred ownership to the employees as a form of compensation. This process was mutually beneficial for senior management because it meant that the contracts of employees who had received privatized assets could be terminated, thus reducing the financial burden on CNPC. Core Services – PetroChina: To establish PetroChina as a separate entity, all noncore social services as well as five chemical production facilities remained part of CNPC, all core oil and gas exploration and production businesses were transferred to the newly formed company. CNPC believed it was important that PetroChina have a complete value chain, and in late 1999, the company took over RMB 144.4 billion of CNPC’s debt and 480,000 employees, as well as 13 oil and gas exploration and production businesses, 2 research institutions, production-sharing contracts, 1 pipeline company, 15 petrochemical and refinery businesses, and 21 marketing companies. Because of changes made during the government’s 1998 reform, the State Council named CNPC as an authorized investment organization. This status meant that CNPC would receive investments from the central government that it would in turn invest in its subsidiaries. This arrangement gave CNPC the power to oversee this restructuring because it now had ownership stake, not just administrative oversight, of its subsidiaries. Another event that year, the plummeting price of oil, also influenced the structure of the emerging PetroChina subsidiary. After a failed attempt to impose a low-cost strategy across CNPC, the senior management realized there was little potential for successful administrative solutions to the company’s problems, and in response CNPC reorganized management of its businesses along four business segments: exploration and production, refining and marketing, petrochemicals and marketing, and natural gas. This move represented an effort to base the structure and management of its businesses on the model used by international oil and gas companies. The PetroChina IPO: After CNPC received approval for its plans from the State Council in 1999, it initially planned to take PetroChina public that same year. However, several unexpected challenges caused the company to delay the IPO until April 2000. One of these problems was that the low price of oil, coupled with lack of knowledge and interest in Chinese SOEs on the part of international capital markets and investors, meant that demand for the newly listed shares would be uncertain. To boost demand, CNPC sought out substantial investments from companies it believed would be strategically beneficial. The most notable of these investments was the purchase of 10% of the available shares by BP.28 A separate issue was


57

related to the opposition of the IPO because of PetroChina’s interests in Sudan. Initially these concerns were being voiced only by a small group of NGOs, but as the publicity surrounding the deal intensified, it was decided that CNPC should continue to hold these controversial investments. After making these changes, CNPC went ahead with the IPO on April 6, 2000 and raised US $2.456 billion through listing H-shares in Hong Kong and ADSs in Europe and the United States. While shifting ownership of its controversial projects in Sudan enabled the IPO to go forward, PetroChina has faced continued criticisms by groups that claim transferring ownership back to the parent company does not excuse PetroChina’s activities in Sudan. The campaign led by these groups was successful in pressuring several large investors such as state pension funds and the Harvard endowment fund to divest themselves of PetroChina shares. Much of the opposition to PetroChina’s IPO, however, stemmed from the fact that 86% of the company is owned, and controlled, by a Chinese national SOE.29 Many potential investors and analysts raised concerns about the company’s ability to act in the best interests of its public shareholders in the event that the CCP’s interests diverged from those of CNPC or PetroChina. A separate but related issue concerned many potential business partners as well. The widespread opposition to the IPO by labor unions and NGOs raised the possibility that doing business with PetroChina might lead to labor problems or a customer backlash. Such problems actually materialized for BP Amoco and Talisman of Canada, as both found themselves the targets of divestment and boycott movements aimed at pressuring the companies into ending their business dealings with PetroChina (Hiebert & Saywell, 2000). Despite the challenges the government’s involvement posed to PetroChina’s IPO, the company’s overall relationship with the government has been very beneficial. While plans for reform at the SOE were first voiced as early as 1992, it was not until the government reorganization of the oil and gas industry in 1998 that PetroChina became a viable international competitor. Up to that point, each separate function of the Chinese oil and gas industry had been administered independently. This meant that no one SOE had control of a complete value chain, and as a result, none of the companies were able to achieve the increased efficiency possible through vertical integration. Since the industry was reorganized along geographic lines, PetroChina has become a vertically integrated oil and gas concern with businesses stretching from exploration to retail distribution of gasoline. PetroChina’s relationships with the Chinese central government have helped it grow to become one of the largest oil companies in the world, accounting for roughly 66% of

58

DOUG GUTHRIE ET AL.

China’s oil and gas production as well as 40% of its refining. Because of strong support for efforts to reform SOEs, PetroChina has benefited from its close collaboration with the central government. For example, its duopoly status allows the company to enjoy a dominant position in the marketplace for several refined products and petrochemicals. The CCP attaches a great deal of importance to energy policy, and as a result PetroChina has been able to bypass red tape to form strategic partnerships with companies and governments abroad.30 The announcement, in late November, of the company’s leading role in a joint venture between itself, Royal/Dutch Shell, and the government of Qatar provides an excellent example of these benefits. The deal, which is expected to exceed the size of the 4.3 billion USD Shell–CNOOC joint venture, is expected to be approved in part because of the Chinese central government’s desire to strengthen its ties with Qatar, the world’s largest exporter of natural gas.31 In addition to the Qatar joint venture, PetroChina is also expected to benefit from an expanded development capacity after its pending acquisition of CNPC’s interest in CNPC Exploration & Development Co. Because the deal will establish CNPC, a major SOE, as an asset wholly owned by PetroChina in more than 10 countries, would likely not receive approval without close government ties. There is still a risk. While PetroChina has thus far benefited from working closely with the central government, both the rapid cooling of China’s economy and the company’s relatively spotty environmental/safety record have increased the risk of the central government intervening or regulating the market in a disruptive manner. The declining economy has brought to the surface what some investors believe is a conflict of interest between PetroChina and its state-owned parent company. These investors believe that the large number of assets transferred between CNPC and PetroChina at bargain prices indicate that CNPC sees PetroChina not necessarily as an independently operating investment asset, but instead as a back door to global capital markets.32

Managing Assets, Incentives, and Learning from Investors Abroad PetroChina’s initial public offering was backed by Goldman Sachs, UBS, and CICC. It was also coached along by McKinsey, Pricewaterhouse Coopers, and a number of foreign law firms. The cynical view of this team of firms was that they simply brought the legitimacy that top investment banks,


59

and consulting and accounting firms could bring to a project like this and that legitimacy would be necessary for the IPO road show that PetroChina would embark upon. However, the reality is that the Chinese learned much more from the firms they engaged in restructuring PetroChina, as they studied the officials governing the firm worked to create the incentive systems that would make the company more efficient. As Dyck and Huang (2004, p. 4) point out, ‘‘In the roadshow, aside from questions about future strategy, the reform that garnered the most attention was the innovative compensation scheme that offered strong incentives for managers to focus on stock performance and, it was hoped, provided a mechanism to refocus management from volume-based to profit-based goals.’’ And the structure of these incentives was specific and closely tied to cost reduction, attainment of profit targets, and return on capital. All employees from department general managers and above would be compensated through a combination of stock options, performance bonus, and salary. The approach was nothing short of revolutionary in the Chinese context: it fundamentally altered the approach to the link between performance and compensation that was heretofore accepted practice in China. The use of stock options as an incentive for senior executives and high-level skilled employees in Chinese SOEs has not been without controversy inside of China. And it is still a system under development. Some reports have indicated that the Ministry of Finance was only appointing senior executives who understood that they would never exercise their options; at the very least, it was taboo to discuss the issue or publicly acknowledge that options were being exercised (Wen & Ming, 2009). New official policies followed in 2006 – a regulation released by SASAC called the Measures for State-Owned Overseas Listed Companies – which stated that options should not exceed 10% of a state-owned corporation’s capital base; the regulation also instituted a 2-year wait for exercising options. (It is worth noting here that, while such restrictions may seem overly controlling to US observers, the recent crisis of capitalism in the United States, and, in particular, the pondering of runaway executive compensation, raises the question of whether the Chinese approach is a more balanced way to handle incentivebased pay.) The IPO was initially met with high-profile resistance in the market, as powerful institutional investors such as TIAA-CREF and CalPERS vowed to steer clear of the stock based on human rights, and environmental and transparency concerns. However, other institutional investors saw opportunity: as a holder of nearly 10% of the available shares, Berkshire Hathaway is now one of the company’s key principals. Its holdings are

60

DOUG GUTHRIE ET AL.

dwarfed by CNPC’s stake – about 40% of available shares – but both investors are principals with significant resources tied up in PetroChina. CNPC has a specific set of economic goals for PetroChina that fit with the developmental goals of the Chinese government and likely diverge somewhat from the economic models typical for a principal like Berkshire Hathaway. However, while Berkshire Hathaway almost certainly has very different ideas about development, profitability, and performance than CNPC, it is also certainly the case that, with a significant stake in PetroChina, Warren Buffett exerts some level of influence over the firm’s decisions about the implementation of efficient business practices.

DISCUSSION AND CONCLUSIONS We began this paper with the question of whether the emergence and proliferation of asset management companies in China has changed worker productivity. In the early phases of this project – as in the cases of the transforming bureaus and the cases of the original AMCs that removed bad debt from the banks’ books – a cynical view of these emerging institutions is warranted. However, a decade and a half later, we can begin to see that these institutions are having an impact. Our quantitative data showed that over the course of a decade’s worth of data from the Shanghai and Shenzhen stock exchanges, firms with dominant owners who were domestic SOMACs did better than firms with dominant owners that were traditional state offices, yet worse than the firms controlled by private investors. This is perhaps what we would expect with the gradualist model: the new institutions are somewhere in between traditional state and private actors in terms of how they affect performance. We then attempted to dig deeper into the mechanisms that lie behind rising productivity by looking at the details behind the restructuring of CNPC and PetroChina. This is a famous case, so what we have to add here is not entirely new. However, it is also a case that many have remained skeptical about in terms of why productivity has changed. To this end, we emphasize the views of Rowe (2006) who have explored the extent to which PetroChina has set up different incentive systems for managerial and executive employees. We further argue that these changes have helped to change the nature of work in transforming SOEs like PetroChina. As SASAC begins to emphasize the views that coprincipals like Warren Buffett emphasize, other asset management companies are following suit. Thus, while many have viewed the asset management movement as a smoke screen


61

of sorts – a way for the state to maintain hidden control over assets – the evidence also suggests that these state offices are thinking more and more like asset managers and making their employees more productive as a result.

NOTES 1. Interview with first author, Spring 1995. 2. By state institutional investment we mean the circumstances by which the state is still the owner of the shares of the firm in question, but this ownership runs through a state-owned asset management company rather than through a traditional state administrative office (e.g., a ministry or bureau). Technically, in the Chinese domestic capital markets, the government must own different types of shares to identify these different types of ownership. Traditional state offices must own shares called ‘‘state-owned shares’’ [guojia gu], while state-owned asset management companies own shares called state institutional shares [guoyou faren gu] (actually, the more literal translation would be something like ‘‘state-owned legal-person shares’’). 3. Kornai (1980, 1990), Sachs (1992, 1993, 1995a, 1995b), Sachs and Lipton (1990), Sachs and Woo (1994a, 1994b, 1997), Woo (1997, 1999), Woo, Parker, and Sachs (1997), Barro and Sala-i-Martin (1995), Blanchard et al. (1993), Cao, Fan, and Woo (1997), Parker Tritt, and Woo (1997), Fischer (1992), Fischer and Gelb (1991). 4. Walder (1995a, 1995b), Naughton (1995, 2007), Oi (1989, 1992, 1995); Rawski (1994, 1995, 1999), Chen, Wang, Zheng, Jefferson, and Rawski (1988), Jefferson and Xu (1991), Guthrie (1997, 1999, 2006). 5. This localization process was both managerial and fiscal and came in two general forms: (1) handover of management power of central enterprises to local governments; and (2) a tax-sharing scheme between central and local governments, whereby more of the revenues would flow directly to local governments. 6. Walder (1995a), Oi (1989, 1992, 1995). 7. Qian (1995), Qian and Roland (1998). 8. Naughton (1995). 9. For discussion of incentives for local-level officials, see Walder (1995a) and Oi (1989, 1992, 1995). For discussion of teaching market behavior, see Guthrie (2002, 2005). 10. Walder (1995a). 11. Guthrie and Wang (2007). By the mid-1990s, SOEs were increasingly being pushed to restructure their operations in fundamental ways, causing them to be treated and to behave more and more like business organizations in practice if not in legal form. Many firms were placed on independent budgetary systems [duli hesuan] and were cut off completely from the redistributive funds of central government coffers, and many were given full latitude to make decisions over how they would govern themselves in China’s emerging markets. By the mid-1990s, the government would formally define its role in the economy as ‘‘creat[ing] the conditions for all sectors of the economy to participate in the market competition on an equal basis,

62

DOUG GUTHRIE ET AL.

and guarantee enterprises from all sectors to be treated indiscriminately.’’ (See ‘‘the Decision on Several Issues for Establishing a Socialist Market Economy System’’ passed in the Third Plenary Session of the Fourteenth National Congress of the CCP in 1993.) 12. There are obvious and famous exceptions here – SOEs like Baoshan Steel were still closely monitored and supported by the state. For discussion, see Steinfeld (1998). However, while these organizations are often viewed as the markers of what is occurring in the Chinese reform process, this view is mistaken. While important to the economy, these largest state-owned organizations are not representative of the economy as a whole. Thus, while many scholars view these organizations are measures of the progress of China’s reforms, they are more accurately the exceptions that prove the rule. 13. Shanghai Stock Exchange [SSEa], Shanghai Stock Exchange Factbook, 2004 (Shanghai, China, 2005). Shenzhen Stock Exchange [SSEb], Shenzhen Stock Exchange Factbook, 2004 (Shenzhen, China, 2005) (Hertz, 1998; Gao, 2002). 14. People’s Republic of China, The Securities Law of the People’s Republic of China. Adopted by the Sixth Session of the Ninth National People’s Congress Standing Committee on December 29, 1998. Implemented on July 1, 1999. The law itself contains 12 chapters, covering a range of issues from stock issuance and stock transactions to the rules governing ownership and shareholding of publicly listed companies. The Company Law (People’s Republic of China, ‘‘The Company Law of the People’s Republic of China,’’ Adopted at the Fifth Meeting of the Standing Committee of the Eighth National People’s Congress on December 12, 1993, and effective as of July 1, 1994) laid the foundation for this standardization. This law governed the process of converting enterprises into shareholding companies and stipulated that companies funded by investing bodies of different ownerships were all equal under the law. More importantly, this law encouraged enterprises to build new corporate structures and standardize organizational bodies (mainly with regard to shareholder meetings, corporate boards and managers) in order to further block political interventions in the decision-making of enterprise, marking a fundamental shift in the organization of China’s industrial economy, as Chinese organizations can now apply for the status of the limited-liability company [youxian zeren gongsi] or a limited-shares company [gufen youxian gongsi] and thus assume the mantle of independent legal entities. 15. People’s Republic of China, Tentative Measures for Decreasing State Ownership. National People’s Congress (2001). 16. Doug, Xiao and Wang (2007). Under the planned economy, large state-owned factories generally reported directly to the ministries [bu] (central government), provincial bureaus [ting], or municipal bureaus [ju] that govern a given sector (Walder, 1995a; Guthrie, 1997, 1999). Over the course of the economic reforms, as the central, provincial, and municipal governments began to recede from the direct control over factories in their jurisdictions, factories in some sectors were placed into group companies, and coordinated economic decision-making as a coalition of organizations (Keister, 1998, 2000). 17. ‘‘Independent budgets’’ [duli hesuan] have meant that firms themselves are partial residual claimants on profits; however, among large state-owned firms, the administrative office still has access to this income.


63

18. The categories of stocks are State Shares [guojia gu], State Institutional Shares [guoyou faren gu], Non-State Institutional Shares [feiguoyou faren gu], Domestic Institutional Shares [jingnei faren gu], Foreign Institutional Shares [jingwai faren gu], Founder Shares [faqiren faren gu], Private Group Shares [dingxia faren gu], Public Group Shares [shehui faren gu], and Free Floating Shares [liudong gu]. 19. Doug Guthrie, Zhixing Xiao, and Jonghoon Bae. Networks connections in Chinese in China’s capital markets: The small world phenomenon and the rise of SASAC. Working Paper. NYU – Stern Department of Management and Organizations. 20. As with virtually all cases of institution building in reform-era China, the emergence of SASAC has been a gradual and experimental process, dating back to the first decade of the economic reforms, starting in 1988 with the State Council’s establishing of a national state-owned assets authority – the State-owned Assets Management Bureau [guoyou zichan guanli ju] to coordinate the management of all state-owned assets in the People’s Republic of China. Since that time, the organization has gone through many iterations until it was finally formalized as SASAC in 2003. 21. There are potential biases to consider with respect to these data. It is crucial to note here that in no case on the domestic exchanges from 1994 to 2003 does a firm, once it is listed, drop out of the sample. Thus, we do not have a problem here of censorship due to nonrandom dropping out of the sample. A larger concern, however, has to do with selection bias in the listing of a company. As we described above, companies are selected by their governing administrative office (zhuguan bumen) to be listed on the stock exchanges. They are usually part of a larger group or cluster of factories. As such, there is an obvious selection bias in the ways in which companies are chosen for their listing on the Shanghai and Shenzhen stock exchanges – state offices choose the best performing firms under their jurisdictions to be listed on the exchanges. And since we are examining firm performance, the very selection process itself is correlated with the dependent variables. However, we view this bias as less problematic than it might initially seem. The main reason for this is that we are dealing with a specific population of firms, and our comparisons are within this population. In other words, all firms suffer from the same selection bias, so there is a systematic bias that makes these firms comparable to each other. Thus, we are less concerned with the question of whether our firms are a representative sample of state-owned organizations in China – they are not. In this sense, we are more concerned with size, direction, and robustness of the effects we are analyzing than we are with statistical significance. But they are a population that is overwhelmingly selected through the same process and is therefore comparable within the sample. And because we are basically dealing with the entire population of firms, we are not concerned with the issue of sampling from within this population. 22. We should note that, while conventional approaches to models such as these would typically log the dependent variables – particularly in the cases of the two profitability variables – our models do not take this approach for two reasons. First, there are a significant number of cases that yield negative values on these outcomes, making the natural log of the outcome variables undefined in those cases. In order to test the importance of this issue, in addition to all of the models presented below, I also ran all of the models based upon the following transformation for the Net

64

DOUG GUTHRIE ET AL.

Profits: Y i ¼ ln½Y i þ ðjY c j þ 1Þ, where Yi represents either the dependent variable of Net Profits and Yc the largest negative value for Yi. In none of these models were there substantive differences in the magnitude or significance of the effects, so, for the sake of simplicity and interpretability of the results, I present results based on the untransformed variables. Second, the distributions of the outcome variables are not in the least skewed, with the exception of one outlier. I also ran all untransformed models omitting this sole outlier and found no substantive differences from the results we present here. All of these results are available upon request from the author. 23. This measure of operating margins (OM ¼ P/R, where P ¼ profits and R ¼ revenues) is tied to profitability (as P appears in the numerator of OM); however, because firms can produce equal levels of profit with different levels of revenues (depending on how efficient firms are in terms of their expenditures), it is useful to examine the ratio between these variables. 24. ‘‘PetroChina Company LTD,’’ Hoovers Company Information. 25. Wu (2002). 26. ibid. 27. Smyth, Qingguo, and Jing, ‘‘Labour Market Reform in China’s State-Owned Enterprises: A Case Study of Post-Deng Fushun in Liaoning Province.’’ 28. Wu, ‘‘A Study of Restructuring in the Chinese Petroleum Sector’’. 29. PetroChina (2007). 30. ‘‘PetroChina Company Limited,’’ in PetroChina Co., Ltd. SWOT Analysis (Datamonitor Plc, 2008). 31. ‘‘Qatar’s China JV Refinery above 200,000 Bpd,’’ Forbes.com, November 20, 2008. 32. Haruni (2008). Besides risks associated with the possibility of direct government intervention in PetroChina’s operations, the company is also significantly vulnerable to the consequences of the CCP’s macroeconomic policies. For example, throughout much of 2007 the central government kept retail prices for domestic gasoline and diesel in line with their targeted levels. As a result of these policies, PetroChina was forced to sell its refined products for as much as 2,000 RMB/ton less than the prices quoted in the Singaporean market. As international crude oil prices rose sharply, many of PetroChina’s refineries incurred substantial losses as a result of the price differential between the international price of crude oil and the regulated domestic prices for refined goods. PetroChina believes that whether or not it is able to realize its expansionary goals will depend in large part upon its ability to access and efficiently make use of limited resources, successfully position its products in increasingly competitive markets, and access global sources financial and human capital. The company used these assumptions as a basis for its general approach to strategic growth. PetroChina can only realize the benefits of vertical integration if it is able to maintain its access to raw materials. Because the company would like to ensure that its growth is sustainable over the long run, it has strong incentive to focus its attempts to gain resources around investments in new exploration projects to replenish its reserves as well as in energy sources that may one day become substitutes for fossil fuels. Additionally, the company places equal emphasis on the pursuit of natural gas and oil to help mitigate the price-level risk inherent to commodity-based industries through diversification.


65

REFERENCES Barro, R. J., & Sala-i-Martin, X. (1995). Economic growth. Cambridge, MA: MIT Press. Blanchard, O., Boycko, M., Dabrowski, M., Dornbusch, R., Layard, R., & Shleifer, A. (1993). Postcommunist reform: Pain and progress. Cambridge, MA: MIT Press. Boardman, A. E., & Vining, A. R. (1989). Ownership and performance in competitive environments: A comparison of the performance of private, mixed, and state-owned enterprises. Journal of Law and Economics, 32, 1–33. Cao, Y. Z., Fan, G., & Woo, W. T. (1997). Chinese economic reforms: Past successes and future challenges. In: W. T. Woo (Ed.), Economies in transition: Comparing Asia and Eastern Europe (pp. 19–40). Stephen Parker and Jeffrey Sachs. Cambridge, MA: MIT Press. Chen, K., Wang, H., Zheng, Y., Jefferson, G. H., & Rawski, T. G. (1988). Productivity change in Chinese industry: 1953–1985. Journal of Comparative Economics, 12, 570–591. China Daily. (2007). PetroChina branch fined for pollution. China Daily. Doug, G., Xiao, Z., & Wang, J. (2007). Aligning the interests of multiple principals: Ownership concentration and profitability in China’s publicly traded companies. Working Paper. Department of Economics, NYU-Stern School of Business. Dyck, A., & Huang, Y. (2004). PetroChina. Harvard Business School Case #9-701-040. Fischer, S. (1992). Privatization in Eastern European transformation. In: C. Clague & G. C. Rausser (Eds), The emergence of market economies in Eastern Europe (pp. 227–243). Cambridge, MA: Blackwell. Fischer, S., & Gelb, A. (1991). The process of socialist economic transformation. Journal of Economic perspectives, 4, 91–106. Frydman, R., Gray, C., Hessel, M., & Rapaczynski, A. (1999). When does privatization work? The impact of private ownership on corporate performance in the transition economies. The Quarterly Journal of Economics, 114(4), 1153–1191. Gao, S. (2002). China Stock Market in a global perspective. Research Report. New York: Dow Jones. Guthrie, D. (1997). Between markets and politics: Organizational responses to reform in China. American Journal of Sociology, 102(5), 1258–1303. Guthrie, D. (1999). Dragon in a three-piece suit: The emergence of capitalism in China. Princeton, NJ: Princeton University Press. Guthrie, D. (2002). China and globalization: The social, economic, and political transformation of Chinese society. New York: Routledge. Guthrie, D. (2005). Organizational learning and productivity: State structure and foreign investment in the rise of the Chinese corporation. Management and Organization Review, 1(2), 165–195. Guthrie, D. (2006). China and globalization: The social, economic, and political transformation of Chinese society. New York: Routledge. Guthrie, D., & Wang, J. (2007). Business organizations in China. In: H. Yeung (Ed.), Handbook of Asian business (pp. 99–121). Cheltenham, UK: Edward Elgar Publishing. Guthrie, D., Xiao, Z., & Wang, J. (2008). Multiple principals, ownership concentration and profitability in China’s publicly traded companies. Working Paper. Haruni, R. (2008). Petrochina to buy 50% stake in CNPC for $11.8b. Seeking Alpha. Hertz, E. (1998). The trading crowd: An ethnography of the Shanghai stock market. New York: Cambridge University Press. Hiebert, M., & Saywell, T. (2000). Market morality. Far Eastern Economic Review, 56.

66

DOUG GUTHRIE ET AL.

Holz, C. A. (2002). China’s industrial state-owned enterprises: Between profitability and bankruptcy. New Jersey: World Scientific Publishing Company. Jefferson, G. H., & Xu, W. (1991). The impact of reform on socialist enterprises in transition: Structure, conduct, and performance in Chinese industry. Journal of Comparative Economics, 15, 45–64. Keister, L. (1998). Engineering growth: Business group structure and firm performance in China’s transition economy. American Journal of Sociology, 104, 404–440. Keister, L. (2000). Chinese business groups: The structure and impact of interfirm relations during economic development. New York: Cambridge University Press. Kornai, J. (1980). The shortage economy. Amsterdam: North-Holland. Kornai, J. (1990). The road to a free economy. New York: Norton. Megginson, W. L., Robert, C. N., & van Randenborgh, M. (1994). The financial and operating performance of newly privatized firms: An international empirical analysis. Journal of Finance, 49, 403–452. Naughton, B. (1995). Growing out of the plan: Chinese economic reform 1978–1993. New York, NY: Cambridge University Press. Naughton, B. (2007). The Chinese economy: Transitions and growth. Cambridge, MA: MIT Press. Oi, J. C. (1989). State and peasant in contemporary China: The political economy of village and government. Berkeley, CA: University of California Press. Oi, J. C. (1992). Fiscal reform and the economic foundations of local state corporatism. World Politics, 45, 99–126. Oi, J. C. (1995). The role of the local state in China’s transitional economy. China Quarterly, 144, 1132–1149. Parker, S., Tritt, G., & Woo, W. T. (1997). Some lessons learned from the comparison of transitions in Asia and Eastern Europe. In: W. T. Woo, S. Parker & J. Sachs (Eds), Economies in transition: Comparing Asia and Eastern Europe (pp. 3–18). Cambridge, MA: MIT Press. PetroChina. (2007). PetroChina 2007 Annual Report. Qian, Y. (1995). Reforming corporate governance and finance in China. In: M. Aoki & H.-K. Kim (Eds), Corporate governance in transition economies: Insider control and the role of banks. Washington, DC: The World Bank. Qian, Y., & Roland, G. (1998). Federalism and the soft budget constraint. American Economic Review, 88, 1143–1162. Rawski, T. G. (1994). Progress without privatization: The reform of China’s state industries. In: V. Milor (Ed.), Changing political economies: Privatization in post-communist and reforming communist states (pp. 27–52). Boulder, CO: Lynn Reinner Publishers. Rawski, T. G. (1995). Implications of China’s reform experience. China Quarterly, 144, 1150–1173. Rawski, T. G. (1999). Reforming China’s economy: What have we learned? The China Journal, 41, 139–156. Rowe, W. G. (2006). Restructuring CNPC and the proposed listing of PetroChina. London, Ontario, Canada: Ivey Management Services. Sachs, J. (1993). Poland’s jump to the market economy. Cambridge, MA: MIT Press. Sachs, J., & Lipton, D. (1990). Poland’s economic reform. Foreign Affairs, 69, 47–66. Sachs, J. D. (1992). Privatization in Russia: Some lessons from Eastern Europe. American Economic Review, 80, 43–48. Sachs, J. D. (1995a). Consolidating capitalism. Foreign Policy, 98(Spring), 50–64.


67

Sachs, J. D. (1995b). Reforms in Eastern Europe and the Former Soviet Union in light of the East Asian experience. Journal of the Japanese and International Economies, 9, 454–485. Sachs, J., & Woo, W. T. (1994a). Experiences in the transition to a market economy. Journal of Comparative Economics, 18(3), 271–275. Sachs, J., & Woo, W. T. (1994b). Structural factors in the economic reforms of China, Eastern Europe, and the Former Soviet Union. Economic Policy, 9(18), 101–131. Sachs, J. D., & Woo, W. T. (1997). Understanding China’s economic performance. Working Paper #5935. Working Paper Series. National Bureau of Economic Research, Inc. Smyth, R., Qingguo, Z., & Jing, W. (2001). Labour market reform in China’s state-owned enterprises: A case study of post-deng fushun in liaoning province. New Zealand Journal of Asian Studies, 42–72. Steinfeld, E. (1998). Forging reform in China: The fate of China’s state-owned industry. New York: Cambridge University Press. Walder, A. G. (1995a). Local governments as industrial firms: An organizational analysis of China’s transitional economy. American Journal of Sociology, 101, 263–301. Walder, A. G. (1995b). The quiet revolution from within: Economic reform as a source of political decline. In: A. Walder (Ed.), The waning of the communist state (pp. 1–24). Berkeley, CA: University of California Press. Wen, X., & Ming, S. (2009). China’s execs sweating over stock options: After nine years and billions of yuan in cash-ins, China is reviewing stock options for state-owned company chiefs. Available at Caijing.com.cn. Retrieved in February. Woo, W. T. (1997). Improving the performance of enterprises in transition economies. In: W. T. Woo, S. Parker & J. Sachs (Eds), Economies in transition: Comparing Asia and Eastern Europe (pp. 299–324). Cambridge, MA: MIT Press. Woo, W. T. (1999). The real reasons for China’s growth. The China Journal, 41, 115–137. Woo, W. T., Parker, S., & Sachs, J. (1997). Economies in transition: Comparing Asia and Eastern Europe. Cambridge, MA: MIT Press. Wu, M. (2002). A study of restructuring in the Chinese petroleum sector. Cambridge, MA: Massachusetts Institute of Technology.

TRANSACTION COSTS, SOCIAL INSTITUTIONS, AND THE DURATION OF INTERFIRM CONTRACTS IN CHINA Ling Yang and Xueguang Zhou ABSTRACT Interfirm contracts are a ubiquitous economic institution in market economies. In this study, we examine the determinants of one important aspect of interfirm contracts – contract duration. We begin with Joskow’s (1987) study that demonstrated that contract duration is governed by mechanisms that economize transaction costs. Our study extends Joskow’s study in several ways: First, while Joskow’s study focuses on one particular area of extreme resource dependence, between the coal mine and the power company, we examine patterns of contract duration and their determinants across broader economic sectors, thereby providing a more general test of the key ideas in transaction cost economics. Second, we investigate the role of social institutions as a distinct mechanism underlying the design of contract duration, especially in terms of mitigating risks and transaction costs. Finally, by situating our study in China, we extend the research context beyond industrialized


69

70

LING YANG AND XUEGUANG ZHOU

market societies to a transitional economy where interfirm contracts are an emerging economic institution. The empirical study is based on the analyses of information on 877 contracts from 620 firms collected in two Chinese cities, Beijing and Guangzhou, in 2000.

INTRODUCTION Interfirm contracts are a ubiquitous economic institution in market economies. In many instances, the purchase of goods and their final delivery between two firms do not take place simultaneously in a spot market. A contract is a formal agreement on the promise and delivery in this transaction. The mechanisms that affect the design of contracts have been at the center of economic analysis in recent years. This study focuses on one important aspect of contract design – contract duration – and develops explanations of the observed patterns by considering transition cost factors and the role of social institutions in economizing such costs. What is the significance of contract duration for organizational analysis? Transaction cost economics provides the most elaborate theoretical explanation on this topic and is the starting point of our study. Consider the nature of formal contracts in an economic transaction. Central to any interfirm contract is the role of promises about future behaviors (e.g., payment, delivery of goods). Because of the separation in time between the promises and the final realization of that promise, an interfirm contract is plagued with such problems as uncertainties due to unforeseen contingencies, or risks of ‘‘hold-up’’ problems resulting from contract partners’ opportunistic behaviors. As such, interfirm transactions involve considerable transaction costs to which the design of contracts must respond. The form of legally binding contracts helps alleviate risks associated with such transactions by putting them under a formal governance structure and on the basis of legal protection. In this light, Joskow (1987) conceptualizes contract duration as indicating the role of contract assurance. That is, longer contractual duration implies more durable, formal assurance of repeated transactions. On the other hand, longer contract duration also means a longer time interval between a promise and the realization of that promise, which may expose the contractual parties to greater risks. Therefore, longer contract duration especially requires more durable, stronger interfirm relationships that sustain this type of transactions. Our task is to explain what are the economic as well as organizational factors that generate variations in such formal assurance?

Interfirm Contracts in China

71

In this light, then, contract duration is a critical aspect of contract design associated with both the transaction costs – risks of unfulfilled promises, costs of design, and implementing contracts – and the social institutions that may alleviate such risks. Interorganizational relationships have long been a central focus in organization research, from the early conceptions of organizational sets, organizational fields (see Scott, 2003 for a review of the early literature), to the recent studies of bank lending (Uzzi, 1999), strategic alliance (Baum, Calabrese, & Silverman, 2000; Burt, 1992; Gulati & Gargiulo, 1999), and venture capital investment (Sorensen & Stewart, 2001). In this study, we focus on a particular instance of interorganizational relationship – interfirm contracts; more specifically, we examine and explain variations in contract duration in interfirm relationships. Unlike other forms of interorganizational relationships, the contractual relationships analyzed in this study concern a specific resource exchange between two firms within a well-defined time framework. Once the goods are delivered and payment made, the contract relationship ends accordingly. We have two goals in this study: First, we examine the generalizability of transaction cost arguments, especially the proposition about the association between contract duration and transaction-specific investment developed by Joskow (1987) in a broader setting of economic transactions; second, we develop theoretical arguments on and empirically examine the role of social institutions – property rights and social networks – in the management of transaction costs. Because of the scarcity of information on formal contract, ours is one of very few studies that explicitly examine this set of issues in the literature. Our study is situated in the context of China’s transitional economy in recent years and we will make use of the institutional details of our research context to develop hypotheses and research design. The research context provides several additional advantages in exploring this set of issues. First, theoretical arguments on formal contracts have been developed largely in industrialized market societies. The evaluation of these lines of arguments in China’s transitional economy allows a more rigorous test, and a significant extension, of these theoretical ideas in a different economic setting. Second, the characteristics of the research context – the leading role of the state and local governments in China’s transition economy, the prevalence of social networks, and large-scale institutional transformations underway – all advance the frontiers of organization research. Our study is informed by over 30 detailed interviews and case studies and is centered on the analyses of contract information for 877 contracts from 620 firms collected in two Chinese cities, Beijing and Guangzhou, in 2000.

72


TRANSACTION COSTS, SOCIAL INSTITUTIONS AND INTERFIRM RELATIONSHIP: THEORETICAL CONSIDERATIONS Why do we observe considerable variations in the duration of interfirm contracts? An intuitive answer – and we concur – is that contract duration is contingent on the idiosyncratic features of the transaction under contract. Obviously, some transactions require longer time to complete or involve repeated exchanges over time, whereas others can be completed over a relatively short time interval or on the spot market. Our key research issue is: Are there some common economic or social mechanisms that affect variations in contract duration, beyond and above the specifics of each contract? Several theoretical logics in organization research point to distinct mechanisms in this regard. In this section, we review these theoretical arguments to shed light on the research issues in our study.

Transaction Costs and Social Institutions: Two Theoretical Propositions Williamson’s (1975, 1985) theory of transaction cost economics is most explicit in addressing the set of issues are associated with contract design in general and contract duration in particular. The key argument in transaction cost economics is that transaction costs associated with an economic transaction, and the efforts to minimize transaction costs are the fundamental forces that shape the characteristics of a contract. The logic is straightforward. Once both contract partners enter into a contract, their bilateral relationship is no longer subject to the pressures of market competition; thus, transaction costs due to opportunistic behaviors and unforeseen contingencies can be prohibitively high. Because transaction costs are a significant component of the costs in market transactions, contract partners strive to reduce transaction costs and gain economic efficiency through the design of contracts. Of various transaction costs, Williamson (1985) particularly emphasized the importance of asset specificity, that is, investment in contract-specific (or transaction-specific, or relationship-specific) assets. Contract-specific investment refers to the kind of investment that is especially designated for the fulfillment of this contract. Consequently, such investment would incur great losses if the contract could not be carried out. Clearly, asset specificity or contract-specific investment makes a firm especially vulnerable to ‘‘hold-up’’


73

problems. As Williamson puts it (1985, p. 30): ‘‘Transaction cost economics further maintains that the most critical dimension for describing transactions is the condition of asset specificity. Parties engaged in a trade that is supported by nontrivial investments in transaction-specific assets are effectively operating in a bilateral trading relation with one another.’’ For example, as Williamson (1985) argues, when a transaction incurs risks that cannot be safeguarded, the price for that transaction must be higher than the one without risks, reflecting the higher transaction costs involved. Along this line of arguments, economists have examined various aspects of contract design (see Shelanski & Klein, 1999 for a recent review). We summarize this line of argument in the following proposition: Proposition 1. The design of interfirm contracts varies systematically with the transaction costs associated with such contracts. Now consider the implication of this line of argument for contract duration – the focus of this study. From the perspective of transaction cost economics, the design of a contract reflects an effort to minimize transaction costs associated with the particular transaction under contract. Contract duration implies, as we noted before, the formal assurance of promises about future behaviors; therefore, we would expect that contract duration becomes longer, hence the assurance stronger, when transaction costs become larger. That is, the parties to the contract are likely to seek more stable and durable relations and assurance, hence longer contract duration, when contractspecific investment is involved. We should observe this association above and beyond the idiosyncratic features of particular transactions. In a rare and important study on this topic, Joskow (1987) explicitly addressed this very issue on the empirical ground. His study examined variations in the duration of contracts in the United States ‘‘between electric utilities and coal suppliers that were in force in 1979’’ and used the role of ‘‘relationship-specific investments’’ to explain such variations. In the transaction cost economics framework, he identified several salient asset specificities – that is, site specificity, physical asset specificity, and dedicated assets – involved in transactions between a coal mine and a power company. Because, from the transaction cost economics perspective, minimizing transaction costs is the paramount concern in the formation of interfirm relationships and in the design of contracts, Joskow predicted that contract duration is positively associated with these contract-specific investments. The findings in his empirical study of about 300 contracts are consistent with this prediction: The observed patterns of contract durations vary systematically and positively with relationship-specific investments.

74


Because of the rarity of information on interfirm contracts, to the best of our knowledge Joskow (1987) is the only systematic study of contract duration in the literature. Transaction cost economics provides the most elaborate and specific analyses of the research issues in this area; thus, we use this theoretical framework as the baseline of our study. Our first task is to examine and test these ideas and their empirical implications in a different and broad economic setting. Our second and related task is to investigate the role of social institutions in generating and explaining the observed patterns of variation in contract duration. If minimizing risks is of major concern in the design of contracts, then this line of reasoning opens doors to the role of social institutions such as ownership structures and social networks. A key insight in institutional theories of organization is the recognition that social institutions play a critical role in generating institution-based trust, reducing uncertainty, and managing organizational environments (DiMaggio & Powell, 1983; Zucker, 1986). Fligstein (2001) emphasized the role of political and social institutions that provide the very foundation for the emergence of integrated markets in Europe. The role of stable institutions is also explicitly recognized in Williamson’s (1985, 1996) work, where property rights and stable interfirm relations provide effective means to curb opportunistic behaviors in market transactions. In this respect, stable social networks can also be seen as important social institutions. Macaulay (1963) first called attention to the role of social relations in managing interfirm contracts and in resolving unforeseen contingencies in the implementation process. Granovetter (1985) developed the embeddedness argument to highlight the role of stable social networks that provide reliable and cost-effective information in facilitating economic transactions. This line of arguments has been further developed and empirically tested in a wave of social network studies (Baker, 1990; Fernandez & Weinberg, 1997; Uzzi, 1997; Uzzi, 1999). At the macrolevel, we can conceptualize durable social networks as social institutions that generate stable, predictable behaviors. For example, network-based social relations between bank managers and clients facilitated positive (but nonlinear) business relationships and access to financial capital (Uzzi, 1999); they provide bases for access to resources, variations in performance (Reagans & Zuckerman, 2001), formation of identity (Podolny & Baron, 1997; Rao, Davis, & Ward, 2000), and social interactions even among competitors (Ingram & Roberts, 2000). Similarly, the stable status order among investment banks, as shown in Podolny’s studies (1993, 1994), also portrays durable institutions that induce predictable behaviors in economic transactions (e.g., participation


75

in IPOs). In this sense, social institutions are dense, stable social relations coupled with equally durable, highly shared expectations. While insisting on the fundamental economic principles, economists also have increasingly recognized the role of social institutions (often phrased as ‘‘local conditions’’) that affect the forms and operations of interfirm contracts. For example, Taylor and Wiggins (1997) argued that transaction costs such as the ratio of setup to inspection costs play a critical role in the design of interfirm contracts, but solutions to quality control problems are mediated by particular institutional settings. They discussed how the institutional attributes in Japan and the United States contribute to the rise of two distinct systems: the ‘‘American’’ system often involves competitive bidding on each contract, large orders, and inspections, whereas the ‘‘Japanese’’ system involves repeat purchases from a supplier who earns a premium, small orders, and no inspections. Similarly, McMillan and Woodruff (1999) also highlighted the role of social relations in the acquisition of credits in Vietnam’s transitional economy. In this regard, sociological and economic arguments overlap to a considerable extent. Based on these considerations, we develop the following proposition: Proposition 2. Stable economic/social institutions (including social networks) play a significant role in reducing transaction costs in contractual relationships. The empirical implication is that the presence of social networks and stable institutional links is likely to mitigate risks in economic transactions, resulting in longer contractual durations. These two propositions summarize the main arguments derived from two distinctive theoretical logics. In many circumstances, they are complementary rather than competing arguments, as far as contract duration is concerned. But note that the logics of explanation differ from each other. In Proposition 1 (the economic view), contract duration is seen as formal assurance that minimizes relationship-specific transaction costs, whereas in Proposition 2 (the institutional view), contract duration is seen as plagued with potential hold-up problems that are only rescued and sustained through stable social institutions. In other words, in the former, contract duration is the institutional solution that minimizes transaction costs, whereas in the latter, contract duration is the behavioral consequence of the presence of social institutions. Moreover, because institutions tend to be more stable than and outlast specific transactions, except on rare occasions, institution-based arrangements cannot be the most efficient way of minimizing transaction costs for a specific

76


transaction, at least in the short run. Suppose that two large state-owned firms were negotiating two interfirm contracts with each other that involve different risks and stakes. Also suppose that, because both firms were backed by respective political authorities, they decided to forego safeguard devices for these contracts. That is, such a design relies on formal institutions and does not pay attention to the specific risks involved in either of the two transactions. But unless the institutional attributes in concern line up perfectly with both transactions, the general institutional conditions are not tailored to address the specific risks involved in either contract. Therefore, resorting to such general institutional conditions alone while ignoring the particularities of a transaction is likely to incur higher transaction costs (risks).

HYPOTHESES We now consider the empirical implications of these theoretical ideas and develop a set of hypotheses to guide our empirical study. The theoretical issues outlined above are general and present in all economic transactions, but the specific forms in which transaction costs occur and local solutions to these problems are closely related to the institutional context in which these transactions take place. Therefore, the design of contract is affected by the specific market and institutional conditions. The institutional details of the specific research context inevitably affect the formulation of specific hypotheses, the research design, choice of variables, and the interpretation of the results.

The Research Context This recognition brings our attention to the research context – China’s transitional economy in recent years. We now briefly introduce the research setting and use the knowledge of the institutional details to develop a set of hypotheses to guide our further analyses. Several characteristics of the institutional context are especially relevant to our study. First, since the 1980s, China has been undergoing major institutional transformation that substantively changed the landscape of the Chinese economy. The state sector has been shrinking over time and stateowned firms are being pushed into market competition for survival. Along with the decline of the state sector, there emerged a striving private sector


77

which has been expanding rapidly, especially since the 1990s. By 2000, the time of our data collection, market transactions become the most important mechanism in interfirm resource exchanges. For example, more than 70% of the over 600 firms in our data identified ‘‘market exchange’’ as the main channel for their firms’ acquisition of factor resources such as raw material supplies, financial resources, labor hiring in their firms, and the sale of their products. The rise of formal contracts was a direct consequence of the emergence of the market economy. In 1981, at the beginning of the economic reform, China enacted its first Economic Contract Law. In the subsequent years, separate legislations were passed with regard to transactions between domestic and foreign firms (1985) and technology transfer and cooperation (1987). In 1999, a new, comprehensive ‘‘contract law’’ took effect that provided, for the first time in the People’s Republic of China, a uniform legal framework for contracts in the area of economic transactions. Second, institutional legacies are still prevalent in China and they affect the formation of interfirm contracts. Of particular interest to our study is the persistence of a variety of organizational ownership: state owned, collective, private, foreign, as well as hybrid forms for those firms in transition from state firms to shareholding firms (Guthrie 1999; Keister 1998; Nee 1992). The importance of property rights can be seen in several ways: (1) There are different institutional links between firms of different property rights to political authorities; (2) property rights channel firms into different regulatory regimes; and (3) there are different incentive structures associated with distinctive ownership structures (Oi & Walder, 1999; Walder, 1995; Zhou, Zhao, Li, & Cai, 2003). Therefore, it is reasonable to expect that property rights play a critical role in inducing corresponding behavioral patterns and in the formation of interfirm relations. For example, firms of different ownerships are situated in distinctive institutional environments, corresponding to their distinct relationships to the local governments and to the regulatory environments. We can distinguish several regulatory regimes: (1) State firms are the closest to the government authority and under tighter scrutiny of the government agencies. Therefore, they develop similar norms and expectations and belong to a similar institutional environment. (2) At the other extreme are private firms who are most remote from the administrative fiat and regulatory influence. (3) Other types of firms – collective firms, hybrid firms – are between these two distinctive institutional environments. Therefore, institutional environments vary systematically with types of organization ownership in China’s emerging markets.

78


These considerations point to a strong association between firms’ property rights and their behaviors. Our interviews also revealed supporting evidence in this regard. One implication of our discussion is that the distinctive institutional environments impose different constraints on firms of different property rights. For example, as one manager at a state firm observed: State firms have much more restrictive accounting system. When we sell products, we must provide receipts to the purchasing side. If we did not provide receipts, our company would be able to keep all of the sale money. But the government agency often comes and checks on our books. If it is uncovered, who would shoulder the responsibility? Many private firms want to buy our products, and they pay cash and do not ask for receipts; in return, they want us to lower prices. But we dare not do that. A private firm would be more than willing to comply with these demands.

Because firms are situated in different institutional environments, firms of the same property rights tend to experience similar institutional legacies and face similar constraints; hence, they are likely to react in similar manners. Our interviews also show collaborative evidence in this aspect. The CEO of a private firm commented: I deal with both state firms and private ones. It is much simpler to deal with the private ones and you can count on the agreement you have reached with them. It is difficult to work with state organizations. They exhaust your mental strength. You have to ‘lubricate’ to get a deal done.

In contrast, a sales manager in a state firm took a very different view: To be honest, it is easier to communicate between state firms, because both are public enterprises. He (the partner firm) is likely to say politely to you ‘you still owe me money. When do you plan to pay me back?’ If this involves private firms, it could lead to a breakup. The state firm on the other side of the contract would understand that these problems are part of historical legacies, so he can gladly shake your hands and sit down with you for a meal together.

Finally, a salient characteristic of China’s transitional economy is the presence of high uncertainty and volatility in the business environment. Therefore, risks involved in interfirm transactions, especially in the design of contracts, become even more salient. Accordingly, we expect that social institutions and social network play an especially important role in the management of risks and in the transformation of reliable information (Bian, 1997; Boisot & Child, 1996). It is against this background that we now develop the implications of the theoretical ideas outlined in the preceding discussions.


79

Contract Duration and Transaction-Specific Investment As we argued before, contract duration implies stable relations that are often related to repeated transactions. Unlike transactions in spot markets, formal contracts not only involve legal protections, but also signal more intensive interfirm relations. Following Joskow (1987), we treat the length of contractual duration as indicating the assurance of economic transactions. That is, the longer the contract duration, the more protection it offers to the contractual partners in mitigating risks associated with contract-specific investment. Given the prevalence of market transactions and lateral interfirm resource exchanges in China’s economy today, it is straightforward to extend this set of arguments to the Chinese context. Proposition 1 above states that the specifics of a contract vary systematically with the transaction costs associated with a specific exchange under contract. An immediate implication of this logic is that contract duration should vary with the transaction costs associated with each transaction under contract. Indeed, this is the core finding in Joskow (1987): Relationshipspecific investments (i.e., site specificity, physical asset specificity, and dedicated assets) all contributed positively to contract duration. The rationale is that the prohibitively high transaction costs induce firms to adopt stronger formal assurance in the form of long-term contracts. However, Joskow’s study focused on an extreme type of transactions that involve long-term resource dependence and commitment between the coal mine and the utilities company, resulting in long-term contracts (an average of 13 years) in coal supply. In many similar cases, vertical integration is often the most efficient solution – that is, the merger between the coal mine and the utilities company – to overcome those formidable transaction costs in this type of transactions. To what extent is this line of argument applicable to other settings of economic transactions? If the logic of transaction cost economics is valid, we expect that this logic has explanatory power in other cases of interfirm contracts beyond the extreme setting of coal mine–utilities company relations. In this study, we extend Joskow’s study in two ways: First, as we know, most contract relations have shorter durations than the ones analyzed by Joskow and that contractual relations take place in other economic sectors beyond the coal mine–utilities industries. We focus on typical sale/purchase transactions in relatively short duration in a variety of industries. In so doing, we aim to test the generalizability of the transaction cost economics argument in a variety of contracts in other economic sectors. Second, we develop other measures of transaction costs (risks, resource dependence, and stake) and assess their relevance on contract duration.

80


In our study, we consider the following aspects of transaction costs: (1) contract-specific investment in terms of specific investment in human resources, information acquisition, equipments, training, and/or scale of production; (2) proportion of the contract value in the total sale by the firm; and (3) number of contractual partners. The last two aspects measure the extent of dependence, hence the transaction costs involved, between the contractual partners. In light of these transaction costs we have outlined, the transaction cost argument leads us to expect that contractual duration varies with the transaction costs involved in the contract. The higher the transaction costs, the longer the contractual duration. Specifically, we expect to find that: Hypothesis 1a. Contract duration increases with the specific investment associated with the contract. Hypothesis 1b. Contract duration increases with the proportion of contractual values. Hypothesis 1c. Contract duration decreases with the number of contractual partners. The Role of Institutions in Contract Duration As we argued before, organization researchers have long recognized the role of social institutions in mitigating risks, reducing uncertainty, and stabilizing interorganizational relations. We now discuss the implications of these ideas, as summarized in Proposition 2 above, for our research issues. In the Chinese context, one implication of Proposition 2 is that property rights play a significant role in mitigating risks and in offering contract assurance. Because firms of different property rights are situated in different regulatory regimes, and experience distinctive institutional environments, we expect that they develop corresponding behavioral patterns in response to different institutional pressures. For example, for many years state policies greatly curtailed the decision rights in state-owned firms, whereas there has been little regulation imposed on firms in the private sector. On the other hand, state policies imposed stronger restrictions on private firms regarding access to financial resources, but adopted favorable policies toward state-owned firms (Huang, 2003; Zhou, Cai, & Li, 2006). Not surprisingly, state firms are governed by significantly different set of regulatory rules than those in the non-state sectors. In this light, we should


81

expect systematic behavioral differences among firms of different property rights, and that these behavioral differences should be reflected in interfirm contracts. Because of the backing of political authorities that mitigates risks, for example, state-owned firms are more likely to sign long-term contracts than private firms do. Similarly, following the arguments developed by network researchers, we expect that stable social relationships also play a significant role in mitigating risks, transmitting effective information, and cultivating strong ties in managing interfirm relationships (Fernandez & Weinberg, 1997; Granovetter, 1985; Ingram & Roberts, 2000; Mizruchi & Stearns, 2001; Uzzi, 1999). Given the high uncertainty in China’s transitional economy, it is likely that firms rely heavily on social networks to obtain reliable information and in access to scarce resources (Bian, 1997; Boisot & Child, 1996; Zhou et al., 2003). Therefore, we propose that, Hypothesis 2a. There are significant differences across property rights in their effects on contract duration. Specifically, state-owned firms are likely to have longer contract durations than private firms do, other things being equal. Hypothesis 2b. Contracts that are based on social networks tend to have longer durations than those in the absence of network basis. A corollary of this line of reasoning is that economic transactions between firms of same property rights are likely to have lower transaction costs relative to those between firms of different property rights. There are several reasons for this speculation: First, because firms of same property rights share similar institutional ties and governance structures, and are in similar regulatory environment, they tend to develop isomorphic behaviors, a key proposition in DiMaggio and Powell’s (1983) argument. Second, firms of the same property rights share similar business environment and past experience, thereby inducing similar behavioral patterns. Here, we highlight the direct role of social institutions in reducing transaction costs such as risks, information costs, and safeguarding costs. As a result, a contract under the same transaction conditions (stake, number of partners, contractspecific investments) has lower risks or other transaction costs when contract partners are of same property rights, other things being equal. An empirical implication in this line of argument is that we should observe systematic variations in firm behaviors across property rights regimes in response to different regulatory pressures and institutional constraints. But between firms of the same property rights, similar regulatory pressures

82


induce isomorphic behaviors. These considerations give us the following two hypotheses: Hypothesis 2c. There is a higher rate of transaction between firms of the same ownership than that between firms of different ownerships. Hypothesis 2d. Because social institutions mediate the risks, the role of transaction-related factors is less important in contract duration between firms of same ownership than between firms of different ownerships.

RESEARCH DESIGN Data The data for this study were collected between 1999 and 2000 in two Chinese cities, Beijing and Guangzhou. Collecting information on business contracts presents several challenges. First, there is no ready information on the universe of contracts among firms from which we can design sampling procedures. Second, it is difficult to obtain access to firms, especially to information regarding business transactions. Because of these difficulties, we were unable to follow random sampling procedures to select our sample. Instead, we employed several sources to identify and interview respondents who have information on interfirm contracts related to their companies (usually top managers or those working in marketing or purchasing departments): (1) We used our personal ties (e.g., relatives, acquaintances, schoolmates, and colleagues) to introduce us to those who have knowledge of interfirm contracts. (2) We selected graduate students or trainees (some from industries) in our collaborating institutions to identify those working in industries that have knowledge about interfirm contracts. (3) We asked some government agencies (e.g., trade unions, government bureaus in charge of certain industries or businesses) to introduce us to managers they know (usually at business conferences) and invited them to cooperate with us in interviews. (Government officials were not involved in selecting specific firms or interviewees, nor were they present at the interviews.) (4) We attended business gatherings and conferences and interviewed those who were willing to cooperate with us. The majority of the cases were collected using the first two sources. The questionnaires were completed through faceto-face interviews by trained interviewers.


83

We asked each respondent to answer a series of questions on one or more (up to five) ‘‘formal or informal business agreements’’ in the firm about which he has relevant information. We intended to collect multiple contracts within a firm so that we can distinguish firm-level factors and contract-level factors on a rigorous statistical basis. In most cases, the interviewee provided only one contract, either because he or she had knowledge of only one contract, or because he or she did not want to provide additional information. Thus, we were able to collect a total of 877 contracts established by 620 firms.1 Compared with other studies of interfirm contracts that primarily focused on a single industry (see Masten, 1996), the contracts in our study cover a range of industries. This makes our findings less dependent on idiosyncratic industrial settings, but it also introduces contextual variations (especially variations across technologies, products, and market structures) that must be controlled for in statistical modeling. In addition, because of the difficulties accessing information on contractual relationships, our data collection process relied on availability and accessibility rather than on a random sampling scheme. We tried to minimize the problem of potential bias by adopting diverse ways of identifying and contacting potential respondents. The various sources used in data collection introduced randomness in the selection of respondents and firms that enter our sample, thereby reducing potential systematic bias. We conduct statistical analysis, including statistical inference and hypotheses testing, under the assumption that the observations are drawn based on the conventional random sampling procedure. But we must be cautious in generalizing our findings to the population and to other settings. Comparing our data with official statistics for Beijing and Guangzhou showed that state firms and hybrid firms are overrepresented in our sample, and collective firms are underrepresented. The firms in our sample are concentrated in heavy industry, light industry, commerce, and service sectors because of data accessibility. Also, because we asked the interviewees to describe one or more contracts, there is potential selection bias in the selection of certain type of contracts. One possibility is that large, formal contracts are more likely to be recalled. The distribution by the relative size of the contracts in our sample does not suggest systematic bias in this regard. Another possibility is that more recent contracts are likely to be presented; indeed, most contracts in our sample were signed since 1995. Appendix provides an assessment of the quality and representativeness of our sample. Our data set provides valuable information in an economic arena that has not been explored previously, and our assessment indicates that the data quality is reasonably good in terms of its representativeness of

84


major industrial categories and types of firms. However, given the limitations noted above, our study should be viewed as exploratory and our findings as suggestive rather than definitive. Dependent Variable The dependent variable is the duration of the contracts collected in our data. For each contract, we asked the respondent to provide information on the beginning date (month and year) and ending date (month and year) of the contract. Contract duration is calculated based on this information. The distribution of contract duration is skewed to the right, because of a few contracts with long durations. To avoid the effect of such extreme values, we use the logarithm of contract duration in statistical analysis. Independent Variables Based on our theoretical interest, we use three sets of covariates to explain variations in contractual duration under study. Measures of Transaction Costs We propose three indicators of transaction costs, all measured at contract level. (1) Percent of contract value in the total sale by the focal firm. This variable measures the stake, hence the risk, involved in the present contract. The larger the percentage, the higher the risk in the contractual relationship. (2) Number of contractual partners for the same product (service) produced (or purchased) by the firm. This variable provides information on the degree of resource dependence between contractual partners. The larger the number of contractual partners, the lower the resource dependence. If the number of partners is more than 10, this measure is truncated at 10 to avoid artificial effects of extreme values.2 (3) Contract-specific investment. We asked the respondents whether the contract involves any contract-specific investment in labor supply, facilities, information, training, technician, machinery, or other items. We further inquired how difficult it is to convert such contract-specific investment to other purposes if it is no longer used for this contract (3 ¼ very difficult, 2 ¼ difficult, 1 ¼ not difficult). Using this information, we constructed a variable that indicates the degree of special investment


85

involved in the form of the summation of the variables on ‘‘difficulty in conversion’’ along the dimensions outlined above. Measures of Institutional Attributes We measure the role of social institutions in several ways. First, we use a set of dummy variables to indicate firm ownership: state firms (the reference category), collective firms, hybrid firms, private firms, foreign firms, and a residual category of ‘‘other firms.’’ Firm ownership provides information on the institutional links of these firms to central and local governments and other institutional environments. State-owned firms have strong institutional ties to the political authority; private firms are most distant from government directives. Collective firms often have close ties with local authorities. ‘‘Foreign firms’’ include American and European firms as well as those from Hong Kong and Taiwan. Second, we also examine the effects of social institutions by analyzing patterns of contract duration between firms of the same property rights or of different property rights. Measures of Social Networks We use a dummy variable to indicate whether the firm had prior acquaintance with the contract partner before the current contract agreement (‘‘yes’’ ¼ 1). Such an acquaintance suggests a recurrent relationship. We also include in model estimation another variable measuring the duration of prior acquaintance. We asked the respondents which kind of channels they used to search for their contractual partners: (1) social networks – contract based on prior relations or through mutual friends, etc.; (2) government sponsorship – contracts that are brokered by the local government; (3) collegial referral – introduced by other firms or personnel in the same trade; (4) open information, such as advertisement or market performance; (5) selfinitiative, etc. We use a set of dummy variables to indicate the type of search channels that were most important in the formation of that contract. This set of variables captures different types of relations between contractual partners in the formation of the contracts. Control Variables We distinguish the following broadly defined industrial contexts: (1) heavy industry, (2) light industry (the reference category), (3) commerce (including the food industry), (4) construction (including transportation and

86


communication), (5) service, and (6) a residual category of ‘‘other industries.’’ Industrial context variables capture both institutional and contractual attributes. We also include the logarithm of firm sales (in f10,000) in the model estimation, which is used mainly for statistical control. We use an indicator variable (Guangzhou ¼ 1) to control for contextual variation between the two cities. Finally, a note on how we deal with missing values. We asked respondents to answer a variety of questions about the specifics of the contract and firmlevel attributes of both parties to the contract. Some respondents were unable to answer all questions; hence, there are missing values for some of the key variables. Given the dearth of information on contractual relationships, we want to use as much information as possible from the data we gathered. For this reason, we include those cases with missing values on three variables (firm sales, contract value as percent of the total sale, and number of contract partners) in our analyses by adopting the following strategy: We create a unique code for those cases with missing values on a variable and include this indicator variable in model estimation. This is analogous to allowing for a different regression line for, thus statistically controlling for, those cases with missing values on that variable. This strategy allows us to use information on other variables for this group of cases in our statistical analyses. Method Because contract duration (and its logarithm transformation) is a continuous variable, conventional multivariate regression model is appropriate. One complication is that, in some of the cases in our sample, a firm may have more than one contract, which may violate the assumption of statistical independence among the cases under the ordinary least squares (OLS) estimation. One way to deal with is to adopt a mixed regression model that estimates robust variance/covariance matrix adjusting for clustered observations. We compared the results from this model and those from the OLS estimation and found that the results from the two models are qualitatively the same. In this paper, we report the statistical results from the OLS estimation because of the ease in comparison and in interpretation.

RESULTS We first describe the basic characteristics of the contracts under study to familiarize the reader with the research settings and research issues. Fig. 1

87

Interfirm Contracts in China 80

freq.

60 40 20 0 0

10

20

30

40

50

60

number of month

Fig. 1.

Distribution of Contract Duration (in Month).

displays the distribution of contract duration in our sample. Only durations within 60 months are plotted in Fig. 1 to enhance display clarity. Many contract durations are clustered around 6 and 12 months, as shown in the bimodal pattern. Among the contracts we collected, the mean contract duration is 14 months, with the standard deviation of 16 months (the median is 10 months). That is, most contracts we study here have shorter durations, relative to those in Joskow’s data, and mainly involve sales of products. Table 1 reports other characteristics of the contracts and of the firms involved in these contracts. Regarding measures of transaction costs, on average each firm has about 6.4 contract partners for the same products. Only about a third of the contracts have special investment (31.2%). On average each contract accounts for about 13.5% of the total sale (purchase) for the focal firm. Overall, these characteristics of the contracts in our sample indicate a much less extent of resource dependence than that in Joskow’s data. This is not surprising, since Joskow’s data reflects an extreme case of resource dependence between contract partners. Among these contracts, more than half of them involve repeated transactions, with 56.4% having more than one transaction over the span of contract duration. There are a variety of property rights among the firms in our study. State firms (35.8%) had the largest presence in our data, followed by hybrid firms (23.8%), and private firms (20%). These firms and their products also cover a wide range of industries and economic sectors: The light industry has the largest presence (29.1%), followed by the commerce sector (21.9), and heavy industry (13.9%). These contracts were initiated through a variety of channels: Social network is the most prevalent channel (35.8%), followed by

88


Table 1.

Descriptive Statistics of the Variables Used in Analyses.

Variables

N

Mean

S.D.

Contract duration (in month)

808

14.1

16.9

Number of contractual partner Special investment (range 1–6) Firm sale (logarithm) Percent of contract value Special investment or not (1 ¼ yes)

758 840 724 705 840

6.4 0.8 7.5 17.3 0.7

3.6 1.5 2.5 53.6 –

Type of property rights State-owned firm Collective firm Hybrid firm Foreign firm Private firm

311 96 207 82 174

% 35.8 11.0 23.8 9.4 20.0

Channel of contract initiation Government sponsorship Collegial referral Social network Self initiative Open information

59 159 284 102 190

% 7.4 20.0 35.8 12.9 23.9

Economic sector/industry Heavy industry Light industry Commerce Construction Service Other sectors

130 254 191 72 82 78

% 14.9 29.1 21.9 8.3 9.4 8.9

Source: http://B/xzhou/project/contract/0402/descdur.lst

open information (23.9%), and collegial referral (20%). In contrast, government sponsorship plays a minor role (7.4%). In sum, the contracts we consider here are, on average, not salient in terms of resource dependence, nor do they involve large stakes or risks. But there are considerable variations across the contracts along these dimensions. These contracts cover a variety of firms with different property rights, across a wide range of economic sectors, and are initiated through a variety of channels. We believe that the variety of contracts in our data is more typical of the interfirm contracts in the marketplace than those in Joskow’s (1987) study, thereby allowing for a more general test of the theoretical ideas. It is worth pointing out that more diverse and short-term contracts


89

are likely to be more sensitive to idiosyncratic transaction factors under contract than to systematic effects of transaction cost considerations or the role of social institutions that we emphasize here. Therefore, these characteristics of our data make it more conservative in the hypotheses developed in the preceding discussions.

Explaining Sources of Variation in Contract Duration We now turn to examine the sources that generated the observed variation in contract duration. We estimated a set of models to identify distinctive mechanisms of theoretical interest. Table 2 reports the OLS parameter estimates in these models. The parameter estimates give the magnitude and direction of the effects of the corresponding variables on the logarithm of contract duration (log[duration]). The first three models (Models 1–3) in Table 2 report the estimates of the three separate sets of covariates for transaction costs, property rights, and social networks, respectively. Our aim here is to take a first look at the contribution of these covariates separately so as to establish a baseline for comparison later. Models 1, 4, and 5 are a set of nested models that allow us to evaluate the relative contribution of these sets of variables across models. In all these model estimations, we control for the location of industrial and economic sectors, firm size, as well as a dummy variable for city location. In Models 1–3, evaluating the three sets of variables of theoretical interest – transaction costs, property rights, and social network – separately, we find that Model 1 has the best model fit, which explains 8% of the variations, as indicated by the R2 at the bottom of Table 2. In contrast, Model 2 for property rights variables and Model 3 for social network effects account for only 4% and 5% of the variation, respectively. A closer look at the specific effects of the covariates indicates that the three variables of transaction costs have significant effects on contract duration, all in the hypothesized directions. First, ‘‘percent contract value’’ has a significant, positive effect on contract duration, indicating that contract duration plays a role in alleviating risks associated with larger stake in transactions. Second, contract duration decreases with the number of contract partners, as consistent with the Hypothesis 1b on the role of resource dependence. Finally, contracts involving special investment have longer contract duration. All these effects are consistent with Hypotheses 1a–1c. We speculated that there are significant variations in contract duration across types of property rights associated with firms; specifically, we

90


Table 2. OLS Parameter Estimates of Determinants of Contract Duration (Dependent Variable: Log[Contract Duration]). Variables

1

2

3

Intercept

0.480

0.267

0.292

Transaction costs Percent contract value Number of partners Special investment (1 ¼ yes)

0.373 0.016 0.160

Type of property rights Collective firm Hybrid firm Foreign firm Private firm Other type of firm

0.133 0.125 0.217 0.157 0.099

4

5

0.417

0.375

0.368 0.015 0.164

0.348 0.017 0.176

0111 0.129 0.195 0.150 0.154

0.107 0.138 0.185 0.129 0.120

Channel of contract initiation Government sponsorship Collegial referral Social network Self initiative

0.311 0.208 0.072 0.086

0.262 0.176 0.049 0.073

Social network ties Prior relationship Time of prior relation

0.122 0.000

0.111 0.000

Control variables Guangzhou Log (firm sale) Heavy industry Commerce Construction Service Other sector

0.037 0.038 0.132 0.115 0.059 0.115 0.024

0.000 0.018 0.180 0.089 0.093 0.118 0.112

0.001 0.007 0.083 0.083 0.139 0.107 0.117

0.021 0.038 0.186 0.124 0.051 0.071 0.060

0.022 0.030 0.154 0.125 0.092 0.032 0.111

R2 Number of cases

0.08 807

0.04 807

0.05 762

0.10 807

0.12 762

Notes: ‘‘State-owned firm’’ is the reference category for ‘‘type of property rights’’ variables, ‘‘open information’’ is the reference category for ‘‘channel of contract initiation’’ variables, and ‘‘light industry is the reference category for industries. Source: http://B/xzhou/project/contract/0402/dur3.lst po0.10; po0.05; po0.01.

hypothesized (2a) that state firms have longer contract durations, whereas private firms have shorter ones. The empirical pattern is partially consistent with this hypothesis, as indicated by the marginally significant (po0.06), shorter duration for ‘‘private firm’’ relative to state-owned firms (the


91

reference category). Moreover, Model 3 (see below) indicates that government-brokered contracts tend to have longer contract duration, which is consistent with Hypothesis 2a. But there are no statistically significant differences between collective firms, hybrid firms, and state firms in this regard. Interestingly, contracts involving foreign firms also have shorter durations. In other words, it seems that private and foreign firms belong to one regime, whereas state firms, collective firms, and hybrid firms belong to another regime in this regard.3 Model 3 estimates the effects of two network variables and a set of variables on search channels for contractual partners. The results show that contracts initiated on the basis of prior relationship tend to have longer contract duration, as consistent with Hypothesis 2b. But in the same model, the duration of prior acquaintance does not have a significant effect, nor does the dummy variable indicating network-initiated partnership. Overall, the evidence suggests that social network plays a role in ensuring contract duration, but its effect is not as systematic or salient as one would expect. Finally, because Models 1, 4, and 5 are nested to each other, we can assess distinctive contributions of the sets of covariates across these models. First, notice that R2 increases from 8% to 10% when variables on property rights are added in Model 4, which further increases to 12% when the set of variables on network/search channels are added in Model 5. F-test of model significance shows that both Models 4 and 5 improve significantly over their respective previous model at po0.05. That is, property rights and network ties make significant contribution to contract duration even after we control for industry location and transaction costs. Second, if we examine changes in the parameter estimates across these models, the parameter estimates of the three transaction cost variables have similar magnitudes across these models, indicating that their effects are robust even after we control for institutional factors included in the last two models. The role of ‘‘prior relationship’’ also shows a similar pattern. In contrast, the effects of the property rights variables have noticeable changes across these models, suggesting that property rights effects are confounded with the effects of other variables in the model. For example, ‘‘government sponsorship’’ in the ‘‘search channel’’ variables (in Model 5) may have captured the importance of property rights, since such sponsorship typically involve large, stateowned firms. To sum up, the findings show that transaction cost factors play a consistent and significant role in affecting contract duration, in the directions hypothesized by the transaction cost arguments. We also find some evidence that government sponsorship and prior network relations

92


contribute to contract duration, as consistent with our hypotheses about the role of social institutions. However, the effects of these institutional factors seem less salient compared with those of transaction cost factors.

The Role of Social Institution in Contract Duration: Further Exploration We have argued that property rights facilitate economic transactions because firms of same property rights share similar regulatory regimes, similar business experience, and hence similar behavioral patterns. If this proposition is sensible, we should expect to find, as hypothesized in 2c, a higher proportion of contractual partners between firms of the same property rights than between firms of different property rights, other things being equal. That is, a firm is more likely to form contractual relations with a firm of the same type of ownership, because similar norms, expectations, and behaviors would greatly reduce uncertainty or other transaction costs, other things being equal. To test this empirical implication, we estimated a log-linear model and analyzed the 2 2 table between both sides of contract partners across different property rights (state-owned, collective, hybrid, foreign, and private). For our purpose, we distinguish two types of contracts: (1) withintype contracts – those contracts in the cells along the diagonal – that are between firms of the same property rights (labeled ‘‘within-type’’); (2) crosstype contracts – those in the cells in the off-diagonal regions – between firms of different property rights (labeled ‘‘cross-type’’). We estimated a series of models to detect patterns of contractual partnership among firms of different ownerships. In Table 3, the ‘‘independence’’ model tests the null hypothesis that there is no association between property rights and the choice of contract partners. As the model statistic L2 shows, this model fits the data poorly (L2 ¼ 127.4, with 16 d.f.). The ‘‘Quasiindependence’’ model tests the null hypothesis that there are associations among firms of the same ownership type (cells along the diagonal), but there is no association across types of property rights (cells off the diagonal) in the firms’ choice of contract partners. This model fits the data well (L2 ¼ 15.5, with 11 d.f., 0.25opo0.10). As Table 2 shows, a quasi-symmetry model also fits the data well, which assumes that there are associations between firms of the same property rights, but across ownership types, the choice of partnership is symmetric to each other, from either side of the contract. These patterns show that (1) firms are more likely to form contracts with other firms of the same property rights; (2) net of the partnership patterns

93


Table 3.

Log-Linear Model of Contractual Partnership, by Ownership.

Model Independence Quasi-independence Quasi-symmetry ‘‘Immobility’’ index Diag 1 – state firm Diag 2 – collective Diag 3 – hybrid Diag 4 – foreign Diag 5 – private

L2

d.f.

127.4 15.5 9.5

16 11 6

0.824 0.706 0.688 0.995 0.926

Note: The 2 2 table is formed by the five types of property rights (state-owned, collective, hybrid, foreign, private) for both sides of the contracts. Source: http://B/xzhou/project/contract/0402/dur3.lst

between the same type of ownership (the diagonal cells), there are virtually no barriers for firms to form contracts across types of organizations. In other words, there is a significantly higher proportion of contracting between firms of same property rights, but there is no significant difference between firms across different types of property rights. For example, a state-owned firm prefers to contract with another state-owned firm, but it is indifferent between contracting with a collective firm or with a private firm (or a firm of any other property rights). The estimates of the ‘‘immobility’’ effects provide a measure of the strength of within-type transactions across these diagonal cells (calculated as the log[ fii/Fii], where fii refers to the observed frequency distribution in the data, and Fii the expected frequency distribution based on the model). The results show that there are higher than expected frequencies along all diagonal cells: Cells for foreign firms and private firms have the highest within-type exchange, followed by those for state firms, and collective firms. But the differences in ‘‘immobility’’ across these categories are minor in magnitude. The findings can be summarized as follows: There is clear evidence of disproportional within-type contractual relations, as consistent with the idea that similar institutional environment facilitates economic transactions. On the other hand, type of ownership does not serve as an important barrier in interfirm relationships across different property rights, as indicated by the finding of high permeability across firm ownership.4 A further implication of this line of argument is that, if economic transactions within the same type of property rights enjoy lower transaction

94


costs, we should expect that transaction cost factors play a less salient role in affecting contract duration in those contracts between firms of the same property rights than in those between firms of different property rights (Hypothesis 2d). Table 4 reports the statistical analysis that is designed to test this hypothesis. Here, we analyzed the effects of transaction cost factors on contract duration for within-type contracts and cross-type contracts separately, controlling for their industrial locations and firm size. As the statistics at the bottom of the table shows, R2 for these two models are 11% and 6%, respectively. Although one cannot compare R2 directly across two separate subsamples, the evidence is clear that the same model fits the cross-type contracts much better than the within-type contracts. This is consistent with our hypothesis that transaction factors play a more important role in contracts between firms of different property rights, whereas their role is less salient in contracts between the same types of property rights. A closer examination of the effects of the transaction cost factors indicates that, for contracts involving firms of different property rights, number of contract partners significantly decreases contract duration. That is, less resource dependence leads to less pressure for assurance in the form of contract duration. On the other hand, this variable has no significant

Table 4. OLS Parameter Estimates of Determinants of Contract Duration, by Transaction Type (Dependent Variable: Log[Contract Duration]). Variables

Model 1: Cross-Type

Model 2: Same-Type

Intercept

0.316

0.836

Percent contract value Number of partners Special investment (1 ¼ yes)

0.331 0.020 0.214

0.670 0.002 0.080

Log (firm sale) Heavy industry Commerce Construction Service Other sector

0.023 0.190 0.105 0.112 0.122 0.021

0.057 0.049 0.124 0.003 0.141 0.020

R2 Number of cases

0.11 500

0.06 298

Note: ‘‘Light industry’’ is the reference category for industry variables.

po0.10; po0.05; po0.01.

95


effect on contract duration between firms of the same property rights, indicating that resource dependence is less a problem for firms sharing the same property rights, presumably because their similar institutional environment and shared experiences have alleviated the risks involved. We observed a similar pattern for the effect of ‘‘asset specificity’’ variable. The presence of contract-specific investment significantly increases contract duration for partnership between different property rights (b ¼ 0.214, po0.01), but has no statistically significant effect on contract duration for contracts between firms of the same type of ownership. Contrary to our expectation, percent contract value has a much larger, significant effect on contract duration for contracts between firms of the same property rights than for contracts between those of different property rights. Notice also that firm size (log[firm sale]) also plays a significant role for within-type contracts. Without further information, we are unable to pin down precisely the causes of this phenomenon, but the observed patterns clearly suggest that contracts between firms of the same property rights and those between firms of different property rights respond to different types of transaction costs and in different ways. One tentative interpretation is that, for within-type transactions, ‘‘scale,’’ as measured by firm size and the magnitude of the stake (percent contract value), plays a larger role in determining contract duration, whereas cross-type transactions are more sensitive to ‘‘risks’’ involved as reflected in special investment and resource dependences. Overall, the determinants of contract duration show patterns that are noticeably different between within-type transactions and across-type transactions. This finding is consistent with our argument that firms of the same property rights are situated in similar regulatory environments, share similar experiences and behavioral patterns; consequently within-type transactions are more likely to reduce transaction costs, making these transaction cost-related factors less salient.

Sensitivity Analysis We have conducted additional analyses to check the sensitivity of our findings. First, we compared our OLS model estimates and a mixed model that allows for clustered observations and correlated error terms. The findings between the two models are very similar. Second, we also used an alternative measure of asset specificity (the degree of asset specificity) in the model estimation. The effect of the alternative variable is consistent with the one we used in our model. Third, we also estimated models by

96


excluding those contract durations that are extremely large (beyond 5 years) or short (near zeros). The patterns of the findings are still broadly consistent with the findings reported here. These analyses provide comforting evidence that the findings from our analyses are robust under a variety of model specifications.

DISCUSSION AND CONCLUSION Interfirm contracts are a ubiquitous economic institution in market economies. The design of contracts and the mechanisms that underlie the formation of contractual relationships have been at the center of institutional economics in recent years. Due to the scarcity of information on interfirm contracts, unfortunately, research in this area is rich in theoretical speculations and poor in empirical evidence. Our study contributes to this literature in two ways: (1) Theoretically, we considered mechanisms of economizing transaction costs as well as those of social institutions in mitigating risks and uncertainty in interfirm contracts; (2) empirically, we explicitly tested these theoretical ideas and their empirical implications using a sample of over 800 contracts in a variety of industries. Specifically, we explored one important aspect of the design of interfirm contracts – contract duration. We begin with Joskow’s (1987) observation that contract duration has particular theoretical significance and is governed by those mechanisms that economize transaction costs. Our study extends Joskow’s study in several ways: First, while Joskow’s study focused on one particular area of extreme resource dependence (between the coal mine and the power company), we examined patterns of contract duration and their determinants across a variety of economic sectors. As a result, we provided a more general test of the key ideas in transaction cost economics. Second, we incorporated the role of social institutions as a distinctive mechanism affecting the design of contracts, especially in terms of contract duration. Finally, by focusing on China’s transitional economy, we extended the research context beyond conventional industrialized market societies in which formal contracts are well-established institutions. We now assess the findings in our study. First, several analyses reported in this study all point to the salient role of transaction cost factors in generating the observed patterns of contract duration. These findings are robust under different model specifications and they are consistent with the predictions derived from transaction cost economics: higher stake involved in the contract (as measured by ‘‘percent


97

contract value’’), more resource dependence experienced by the firm (as measured by the ‘‘number of contract partners’’), and the presence of contract-specific investment, all contribute to longer contract duration, reflecting the efforts to secure formal assurance in interfirm relationships. This set of results are robust and salient and have reasonable explanatory power, as measured by model fit statistics, especially relative to other alternative factors such as property rights, and social network factors. Second, social institutions – property rights and stable social relationships – also contribute to ensuring long-term contractual relations. We find that government-sponsored contract partnership and contracts based on prior relations contribute to longer contract duration. In contrast, non-state firms – private firms and foreign firms – tend to have shorter contract duration, other things being equal. The role of social institutions is further supported in our findings that, in those contracts involving firms of the same property rights, transaction cost factors play a less salient, or even insignificant, role in contract duration (cf. Table 4). Our reasoning is that firms of the same property rights are situated in similar regulatory and institutional environments; they tend to share similar expectations and business experience, which induces similar behavioral patterns. As a result, we speculate that the transaction costs in such a partnership are lower than those involving firms of different property rights. This idea is further supported by the finding that the probability of forming contracts between firms of the same property rights is significantly higher than that between firms of different property rights (cf. Table 3). However, we should point out that, relative to the significant effects of transaction cost factors, the roles of social institutions are less systematic or salient in explanatory power. To bring in the research context, we observe that the institutional attributes (e.g., structures of property rights and the role of government sponsorship) and social networks play an important role in forging patterns of interfirm relationships in China’s transition economy, as evidenced in patterns of contract duration in our study. The significant effects of transaction cost factors notwithstanding their effects take place on the basis of existing industrial structures, and are mediated and mitigated by the infrastructure of social institutions. Therefore, a broad implication of our study is that these institutional legacies and social relationships are likely to contribute to the emergence of new economic institutions of contracts and other forms of interfirm relationships. Without a thorough understanding of the ‘‘local conditions’’ and social institutions, it would be difficult to explain, let alone predict, the future course of socioeconomic changes in transition economies.

98


In concluding our study, we want to discuss the limitation of our study and suggest future research agenda. In contrast to the previous study by Joskow (1987), ours focused on a large number of contracts that are relatively short term and cover a wider range of economic sectors and property rights. These characteristics introduced contextual variations that may have complicated the observed patterns reported in this study. On the other hand, a positive side of these features is that these diverse contexts, the variety of organizational forms, and shorter duration of contracts are likely to make idiosyncratic features of the contracts under study more salient and make it more difficult to uncover systematic effects of transaction cost factors and of social institutions, hence making the tests of our hypotheses more conservative. The important role of transaction cost factors revealed in this study raises important issues about future studies of interfirm relationships. Because transaction cost factors take effects at the contract level, this means that studies of interfirm relationships should direct attention to those areas beneath firms as a level of analysis and examine micro-processes that underlie specific transactions. For example, to explain interfirm relations in loan lending, strategic alliance, and resource exchange, we need to pay attention not only to firm-level characteristics but also to those attributes of the specific transaction that may have affected the stability and durability of the contractual relationship. This recognition also calls for further efforts to identify those institutional attributes that affect firm behaviors at the level of specific transactions. Social institutions and stable network relationships play some role in contract duration in our study, but their effects are less systematic or salient compared with transaction cost factors. Given the strong interest in these social institutions in organization research, whether such patterns have broad implications or merely reflect contextual sensitivity or the characteristics of the contracts in our study remains to be an important issue on future research agenda.

NOTES 1. These contracts are between the 620 focal firms in our data set and other firms that are not included in our sample. 2. Some firms have a large number of contractual partners. Because this variable is used to measure resource dependence in contractual relationships, we reason that the number of contractual partners beyond a certain threshold (more than 10) would indicate that the firm has very weak dependence on that particular partner and contract.

99


3. Since contracts involve firms on both sides, one may wonder if we should examine the effects of property rights on both contract partners. We did so in our preliminary analyses but did not find significant effects from property rights from the other side of the contract. A more refined pairing of firms’ property rights does not reveal patterns significantly different from the ones reported here. 4. Because different types of firms are disproportionally located across industries as we noted before, we need to ascertain that choices about contractual partners are not an artificial consequence of firms’ industrial locations. For this purpose, in our preliminary analysis, we estimated three-way log-linear models of focal firm ownership type (X), partner firm type (Y), and industry location (Z). Model (XY, XZ) which assumes that industrial location had no significant effect fits the data well. We further analyzed firm property rights using a logistic regression, controlling for additional contract-related factors, the patterns are similar to those reported here. Also note that the log-linear model estimated in Table 2 purges the effects of marginal distributions; therefore, these results are robust with respect to the distribution of ownership on the margins of the table.

ACKNOWLEDGMENTS This project is supported in part by a grant from the CCK Foundation. We thank the editor and the reviewer for their helpful comments and He Cai and Qiang Li for their collaboration on this research project.

REFERENCES Baker, W. E. (1990). Market networks and corporate behavior. American Journal of Sociology, 96, 589–625. Baum, J. A. C., Calabrese, T., & Silverman, B. S. (2000). Don’t go it alone: Alliance network composition and startups’ performance in Canadian biotechnology. Strategic Management Journal, 21, 267–297. Bian, Y. (1997). Bringing strong ties back in: Indirect connection, bridges, and job search in China. American Sociological Review, 62, 266–285. Boisot, M., & Child, J. (1996). From fiefs to clans and network capitalism: Explaining China’s emerging economic order. Administrative Science Quarterly, 41, 600–628. Burt, R. S. (1992). Structural holes. Cambridge: Harvard University Press. DiMaggio, P. J., & Powell, W. W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48, 147–160. Fernandez, R. M., & Weinberg, N. (1997). Sifting and sorting: Personal contacts and hiring in a retail bank. American Sociological Review, 62, 883–902. Fligstein, N. (2001). The architecture of markets: An economic sociology of twenty-first-century capitalist societies. Princeton, NJ: Princeton University Press.

100


Granovetter, M. (1985). Economic action and social structure: The problem of embeddedness. American Journal of Sociology, 91, 481–510. Gulati, R., & Gargiulo, M. (1999). Where do interorganizational networks come from? American Journal of Sociology, 105, 177–231. Guthrie, D. (1999). Dragon in a three-piece suit: The emergence of capitalism in China. Princeton, NJ: Princeton University Press. Huang, Y. (2003). Selling China: Foreign direct investment during the reform era. New York: Cambridge University Press. Ingram, P., & Roberts, P. W. (2000). Friendships among competitors in the Sydney hotel industry. American Journal of Sociology, 106, 387–423. Joskow, P. L. (1987). Contract duration and relationship-specific investments: Empirical evidence from coal markets. American Economic Review, 77, 168–185. Keister, L. (1998). Engineering growth: Business group structure and firm performance in China’s transition economy. American Journal of Sociology, 104, 404–440. Macaulay, S. (1963). Non-contractual relations in business: A preliminary study. American Sociological Review, 28, 55–67. Masten, S. E. (1996). Case studies in contracting and organization (p. 344). New York: Oxford University Press. McMillan, J., & Woodruff, C. (1999). Interfirm relationships and informal credit in Vietnam. The Quarterly Journal of Economics, 114, 1285–1320. Mizruchi, M. S., & Stearns, L. B. (2001). Getting deals done: The use of social networks in bank decision-making. American Sociological Review, 66, 647–671. Nee, V. (1992). Organizational dynamics of market transition: Hybrid forms, property rights, and mixed economy in China. Administrative Science Quarterly, 37, 1–27. Oi, J. C., & Walder, A. G. (1999). Property rights and economic reform in China. Stanford, CA: Stanford University Press. Podolny, J. M. (1993). A status-based model of market competition. American Journal of Sociology, 98, 829–872. Podolny, J. M. (1994). Market uncertainty and the social character of economic exchange. Administrative Science Quarterly, 39, 458–483. Podolny, J. M., & Baron, J. N. (1997). Resources and relationships: Social networks and mobility in the workplace. American Sociological Review, 62, 673–693. Rao, H., Davis, G. M., & Ward, A. (2000). Embeddedness, social identity and mobility: Why organizations leave NASDAQ and join New York Stock Exchange. Administrative Science Quarterly, 45, 268–292. Reagans, R., & Zuckerman, E. (2001). Networks, diversity and performance: The social capital of R&D units. Organization Science, 12, 502–517. Scott, W. R. (2003). Organizations: Rational, natural, and open systems. New Jersey: Prentice Hall. Shelanski, H., & Klein, P. G. (1999). Empirical research in transaction cost economics: A review and assessment. In: G. R. Carroll & D. Teece (Eds), Firms, markets, and hierarchies: The transaction cost economics perspective (pp. 89–118). New York: Oxford University Press. Sorensen, O., & Stewart, T. (2001). Syndication networks and the spatial distribution of venture capital investments. American Journal of Sociology, 106, 1546–1590. Taylor, C. R., & Wiggins, S. N. (1997). Competition or compensation: Supplier incentives under the American and Japanese subcontracting systems. American Economic Review, 87, 598–618.


101

Uzzi, B. (1997). Social structure and competition in interfirm networks: The paradox of embeddedness. Administrative Science Quarterly, 42, 36–67. Uzzi, B. (1999). Embeddedness in the making of financial capital: How social relations and networks benefit firms seeking financing. American Sociological Review, 64, 481–505. Walder, A. G. (1995). Local governments as industrial firms: An organizational analysis of China’s transitional economy. American Journal of Sociology, 101, 263–301. Williamson, O. E. (1975). Markets and hierarchies, analysis and antitrust implications: A study in the economics of internal organization. New York: Free Press. Williamson, O. E. (1985). The economic institutions of capitalism. New York: Free Press. Williamson, O. E. (1996). The mechanisms of governance. New York: Oxford University Press. Zhou, X., Cai, H., & Li, Q. (2006). Property rights regimes and firm behavior: Theory vs. evidence. In: A. S. Tsui, Y. Bian, & L. Cheng (Eds), China’s domestic private firms: Multidisciplinary perspectives on management and performance (pp. 97–119). Armonk, NY: M.E. Sharpe, Inc. Zhou, X., Zhao, W., Li, Q., & Cai, H. (2003). Embeddedness and contractual relationships in China’s transitional economy. American Sociological Review, 68, 75–102. Zucker, L. (1986). Production of trust: Institutional sources of economic structure 1840–1920. Research in Organizational Behavior, 8, 53–111.

APPENDIX. INFORMATION ON DATA QUALITY AND SAMPLE REPRESENTATIVENESS In this appendix, we provide additional information on our assessment of data quality and sample representativeness.

Sample Representativeness Our data were not drawn randomly from the population of firms (or contracts) in these two cities. To evaluate the representativeness of our data, we examined the distribution of firms with respect to (1) firm ownership and (2) industrial location. Table A1 compares the distribution of firms in our sample with official statistics for Guangzhou and Beijing. Note that the official statistics were not always consistent throughout our categories and we have made adjustments in grouping categories. The discrepancies in sample statistics between Table 1 and Table A1 is due to the fact that Table 1 is based on information at the time of contract formation, and Table A1 uses information at the time of interview. In Table A1, we also made adjustments to parallel official classification of industries. Detailed information on the comparison is available from the author.

102


Table A1.

Percentage Distribution of Firm Characteristics for the Sample Compared with Official Statistics.

Measure

Beijing

Guangzhou

Our Sample

Official Statistics

Our Sample

Official Statistics

Firm ownership State-owned Collective Hybrid Foreign Private

35 09 24 07 18

28 32 15 06 12

26 11 28 13 22

10 39 24 05 22

Industry Heavy industry Light industry Commerce Construction Service Other

18 21 20 07 12 22

11 11 28 06 12 32

12 38 24 10 07 08

06 27 24 06 14 23

Notes: Our data were gathered in 1999 and 2000. Official statistics are from the Statistical Yearbook of Beijing (1999) and Statistical Yearbook of Guangzhou (1999). See text for definitions of firm ownership and industry.

With regard to firm ownership, our sample overrepresents state firms and hybrid firms compared with the official statistics, and collective firms are underrepresented. The overrepresentation of state firms and hybrid firms reflects the fact that researchers and interviewers have better access to these types of firms. The proportions of private firms and foreign firms in our sample are comparable with official statistics. The underrepresentation of collective firms is somewhat surprising. Our speculation is that, because of the rapid changes in China’s organizational reforms, many officially labeled ‘‘collective firms’’ may in fact see themselves nowadays as private firms or hybrid firms when interviewed in our data. Hybrid firms include those whose property-right ownership are ambiguous and often involve both public (state-owned) and private (or foreign investment) ownership. This is characteristic of transitional firms: Some were previously state-owned firms but are in a process of being privatized; others are non-state firms that involve multiple sources of ownership (jointly financed by several sources).

103


With regard to type of industry, Beijing and Guangzhou show distinctive patterns in the official statistics, and these differences are also reflected in our sample. Heavy industry and light industry are overrepresented in our sample, relative to the official statistics, because firms in these industries tend to be large and provide better accessibility. The patterns across other industries in our sample are more or less comparable with official statistics, with some variation between these two cities and a higher percentage of ‘‘other industries’’ in official statistics. Because we did not use a random sampling procedure, the distribution of firms by type of ownership and industrial setting reflects firms’ accessibility. Overall, however, the distributions do not indicate any serious bias in our sample. The overrepresentation of state firms and hybrid firms and large firms (in heavy and light industries) also coincides with the fact that these firms tend to have durable, bilateral relations due to the nature of their products and institutional context. Other Selection Bias In addition to the issue of sample representativeness, there is also a potential bias in the selection of specific contracts by the interviewees. In our questionnaire, we explicitly asked about ‘‘formal or informal business agreements’’ so as to include both formal and informal contracts in our sampling frame. We also tried to use the term ‘‘business agreement’’ and ) which in Chinese often connotes avoid using the word ‘‘contracts’’ ( ‘‘formal contracts.’’ However, we had to rely on the interviewees to choose the specific contract(s) to report. A firm typically has many contractual partners. When an interviewer recalls a particular contract in our interview, the saliency of certain types of contracts may influence his or her selection of that contract for presentation, thus introducing potential selection bias. One potential bias is that large, more formal contracts are more likely to be selected, partly because such contracts are more salient and easier to recall, and partly because formal contracts are more ‘‘legitimate’’ to present to the outside world. To investigate this issue, we examined the distribution of the size of the contracts as measured by the percentage of the value of the contract in the total production (or sale) of the same products in the focal firm. The distribution by quartile is as follows: Q1, 2%; Q2, 10%; Q3, 25%; Q4, 100%, that is, about half of the contracts (Q2) made up less than 10% of the total sales of that product. And 75% of the contracts (Q3) made up 25%

104


or less of the total sales for the firm. Thus, most contracts in our sample are relatively small in their relative value to the focal firm. We cannot rule out the possibility that recall bias contributes to the high percentage of formal contracts in our sample. But the fact that there is a high proportion of formal provisions even for those contracts that involve small stakes is consistent with and reinforces our theoretical arguments that the formality of contracts is not conditioned on contract size. Hence, it is less likely that the high formality in contracts in our sample results from selection bias in recall. We recognize that our data collection methods are unconventional, and we do not have independent information to ascertain or rule out the potential bias introduced in the data collection process. Given these limitations, our study should be treated as an exploratory study and the findings reported should be interpreted as suggestive rather than definitive.

FOREIGN DIRECT INVESTMENT AND SOE PERFORMANCE IN CHINA: EVIDENCE FROM THE THIRD INDUSTRIAL CENSUS$ Yusheng Peng ABSTRACT Using the 1995 Third Industrial Census data in China, I explore the impact of inward FDI on the productivity performance of domestic SOEs. Multilevel analyses of city- and firm-level data show that presence of FDI-related firms (sanzi qiye) in a city significantly improves the total factor productivity of SOEs that are located in the same city but not affiliated with FDI. I interpret the effects as not only technology spillovers but also as FDI-induced institutional innovations and reforms.

$

This paper was presented at the International Conference on ‘‘Chinese Society and China Studies,’’ Nanjing University, China (October 24–26, 2008) and Social Science Workshop on Contemporary China at Yale University (February 16, 2009). I thank participants of these events. For helpful comments, I am particularly indebted to Deborah Davis, Philip Huang, Lisa Keister, Pierre Landry, and the anonymous reviewer for Research in the Sociology of Work.


105

106

YUSHENG PENG

INTRODUCTION China’s extraordinary economic growth during the past three decades has been largely driven by two forces: rural enterprises and foreign direct investment (FDI). It is not an overstatement to say that rural industrialization was the primary growth engine during the 1980s. During the first decade and half since Deng Xiaoping launched his programs of reform and opening-up in 1978, international borrowing was the main form of exposure to foreign capital. FDI was minimal in magnitude, mostly small-scale investments from overseas Chinese entrepreneurs in Hong Kong, Macau, Taiwan, and concentrated geographically in a few coastal ‘‘Special Economic Zones.’’1 By the 1990s, especially after Deng Xiaoping’s Tour of the South in 1992, inward FDI increased explosively, together with the robust growth of the rural industrial sector and an emerging private sector. The domestic private sector was a new player on the stage, small but dynamic, posed to play a major role in the future. The sudden massive influx of foreign capital and multinational corporations (MNCs) was the defining feature of the 1990s (see Fig. 1). In 1991, the total amount of FDI inflow to China was $4.4 billion, including those from Hong Kong, Macau, and Taiwan. The figure jumped to $11 billion in 1992, and then tripled again to 33.8 billion 2 years later. The robust growth trend continued until 1998, when a retrenchment set in probably due to the Asian financial crises. The growth trend picked up steam again after 2001 when China entered WTO. In 2006, the total value of actually utilized (vis-a`-vis contracted) FDI stood near $70 billion. Many scholars (e.g., Firebaugh, 1992, 1996; de Mello, 1999; Frankel & Romer, 1999; Lloyd, 1996; Campos & Kinoshita, 2002) have found that FDI helps improve the economic condition of the recipient nations, especially developing nations. Typically, a developing nation lacks not only physical capital stock such as infrastructure and factories, but more importantly management skills and technological know-how. This is what Romer (1993a) calls ‘‘object gaps’’ vis-a`-vis ‘‘idea gaps,’’ that is, gaps in ‘‘the knowledge used to create value in a modern economy.’’ FDI from advanced nations not only helps increase the host nation’s physical capital stock but also helps to bridge the ‘‘idea gaps’’ through the process of knowledge diffusion. It is the diffusion of ideas that I will focus on in this paper. On the micro level, when MNCs invest in a less-developed nation, either through acquiring equity shares of local firms, or forming joint ventures, or establishing a wholly-owned foreign enterprise, they often bring with them new technology (e.g., new brands, new machinery) or better management

Fig. 1.

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

80.0

2.2

2.3

3.2

3.4

3.5

11.0

37.5

45.3 45.5 40.3 40.7

46.9

Foreign Direct Investment to China in Billions USD (Actually Utilized Amount), 1985–2006. Source: China Statistical Yearbook, 2007 (NBS, 2008). Available at http://www.stats.gov.cn/tjsj/ndsj/

1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

2.0

4.4

27.5

33.8

41.7

52.7 53.5

60.6 60.3

69.5

FDI and SOE Performance in China 107

108

YUSHENG PENG

skills or both. Domestic employees will gain the new knowledge through onthe-job training or learning by doing. Productivity and efficiency grow as a result. Once transferred, however, the new knowledge and novel ways of doing things will easily spread beyond the boundaries of foreign-invested enterprises, thus creating positive externalities to be captured by other domestic firms. The spillovers may occur through imitation or pirating, employee mobility, or local outsourcing and subcontracting. FDI to China is generally viewed in a positive light, even though scholars may disagree with regard to the micro-mechanisms of its growth-promoting impacts (Pomfret, 1997; Lardy, 1995; Huang, 2003). Chinese government agencies at the central and local levels have been vehemently advocating, soliciting, and attracting overseas investment with preferential terms and encouraging the formation of joint ventures between domestic and foreign firms, with an emphatic and explicit intention to learn the ‘‘advanced’’ western technology and managerial practices (Lardy, 2002; Huang, 2003). Reform-minded officials and scholars believe that fast integration into the global economy will speed up China’s transition to a market system and the transformation of the state-owned enterprises (SOEs; Lin, 2001a, 2001b). There are few controversies regarding the direct and obvious benefits of FDI to the Chinese economy, such as bringing in physical capital, generating employment, and enlarging the tax bases. The interesting question is on the indirect and less obvious long-term and short-term impacts of FDI and massive entry of MNCs on domestic firms. In this paper, I focus on the impact of FDI on Chinese SOEs and ask specifically: Have SOEs benefited from productivity spillovers of FDI and the massive entry of MNCs? I will use a firm-level data set from the 1995 Third Industrial Census (ACMRC, 1999) and a city-level data set from urban statistical yearbooks to explore the relationship between SOE performance and FDI presence in Chinese cities. Multilevel analyses will show that FDI significantly improves the productivity performance of domestic SOEs unrelated to FDI. I interpret this finding as proof of positive FDI knowledge spillovers. Before moving onto empirical data analysis, I first review the relevant literature and develop some hypothesis.

LITERATURE REVIEW With dependency theories falling hopelessly out of currency (Firebaugh, 1992, 1996), endogenous growth models (cf. Romer, 1986, 1990, 1993b; Grossman & Helpman, 1991; Findlay, 1978a, 1978b) are gaining ground as

FDI and SOE Performance in China

109

the dominant model for examining the effects of globalization on lessdeveloped economies. The basic argument of endogenous growth theories is that long-run economic growth is achieved through technological advancement and accumulation of human capital. Advanced nations have a comparative advantage in technological innovation through R&D spending, whereas less-developed nations can reap the benefits of technological innovations through imitation and leapfrogging. New knowledge tends to generate positive externalities due to non-rivalry and non-excludability in its use. International trade and FDI speed up the diffusion process. As Grossman and Helpman (1991, p. 238) put it: [T]he most important benefit to a country from participating in the international economy might be the access that such integration affords to the knowledge base in existence in the world at large. Countries that trade in the world markets invariably learn a great deal above innovative products and about the novel methods that are being used to produce older goods. While it is true that agents in an economically isolated country might also acquire some such information by reading professional journals, speaking to foreign experts, or inspecting prototype products, it seems that the contacts that develop through commercial interaction play an important part in the international exchange of information and ideas. At the least participation in world markets would seem to accelerate greatly a country’s acquisition of foreign knowledge.

FDI-induced knowledge diffusion can occur either in the form of intended technological transfer between MNCs and their local subsidiaries in the host country or in the form of unintended knowledge spillovers. FDI usually entails some transfer of more advanced technology and other intangible assets. Theoretically, direct investment is chosen over other strategies, say, licensing, with an explicit intention to capture higher returns for intellectual property (Dunning, 1997; Caves, 1996). When MNCs decide to invest in production facilities in the host country, they need to train local engineers, managers, and workers in the use of the new technology and more efficient ways of marketing and management. Thus, wholly foreign-owned and joint-venture firms tend to have higher productivity than domestic firms. Even though MNCs will try to keep their advanced technology away from the inquisitive eyes of domestic firms, technological spillovers are hard to prevent. Knowledge spillovers may occur, for instance, through the following three channels. First, local firms may imitate foreign firms through reversed engineering or through simple observation. Imitation will be even easier when the ‘‘new ways of doing business’’ are nonproprietary knowledge in the investor’s home country. Second, knowledge may spillover through backward and forward linkages with local suppliers and customers. MNCs may demand a minimal standard from their upstream local suppliers

110

YUSHENG PENG

and may even help them upgrade technology in order to achieve the minimal standard. By supplying downstream purchasers in the host country with highquality products, the latter may also gain in productivity (Blomstrom & Kokko, 1998). Third, new knowledge from MNCs may spill over to unrelated domestic firms through employee mobility. That is, local competitors may strategically poach skilled workers from MNC subsidiaries (Fosfuri et al., 2001). FDI may also drive up the productivity level of the unaffiliated local firms through increasing competition pressures. In order to maintain their market share, local firms may be forced to be more efficient. They may upgrade their technology either through acquiring new technology in the open market or through mimicking foreign competitors and poaching skilled workers from MNCs. FDI-induced competition probably hurts local firms in the short run but may strengthen the host country’s growth potential in the long run. Although empirical analysis is a bit less elegant than economists’ mathematical equations, there is overwhelming evidence testifying to a positive correlation between FDI and productivity growth. There is a consensus in empirical findings that foreign-invested firms are generally more productive than domestic firms (e.g., Haddad & Harrison, 1993; Aitken & Harrison, 1999; Kokko, 1994; Blomstrom & Kokko, 1998; Lall, 1978; Blomstrom & Sjoholm, 1999; Byun & Wang, 1995). The controversy is really over whether this productivity advantage accounts for all the productivity gains from FDI. In other words, are there positive productivity spillovers from foreigninvested firms to domestic firms? Evidence on spillovers from foreign-owned firms to domestic firms is, however, less consensual (Go¨rg & Greenaway, 2004). On the one hand, data from Mexico (Kokko, 1994; Blomstrom & Kokko, 1998), Indonesia (Blomstrom & Sjoholm, 1999), and South Korea (Byun & Wang, 1995), India (Lall, 1978), Lithuania (Javorcik, 2004), and many other developing countries do yield supporting evidence for the spillover hypothesis. Even FDI between advanced nations, such as the United States and Japan, is reported to create positive externalities for domestic firms. For instance, Okamoto (1999) finds a small but significant technological spillover effect from Japanese-owned auto assembly plants in the United States to their U.S. counterparts. She believes that productivity growth in the United States auto industry from 1982 to 1992 was mostly due to increased competition. Using firm-level data for the Japanese manufacturing sector, Murakami (2007) reports that the entry of foreign-owned firms in an industry lowers the total factor productivity (TFP) growth of domestic


111

Japanese-owned firms in the short run, but raises it in the long run via competition pressure. On the other hand, several studies find an insignificant or even negative correlation between foreign investment and local firm productivity. For instance, Haddad and Harrison (1993) find no link between productivity growth of domestic firms in specific sectors with foreign investment in Moroccan data. Aitken and Harrison (1999) also fail to find positive spillovers from MNCs to domestic firms located in the same region in Venezuela, but they do find negative spillovers from MNCs to local firms in the same sector. They conclude that all gains from FDI seem to be captured by joint ventures. Konings (2001) reports negative spillovers in Bulgaria and Romania and insignificant positive spillovers in Poland. He concludes that the competition effect of MNC affiliates outweighs any positive technology effect. One consensus emerges from this myriad of inconsistencies: knowledge spillovers via FDI are not automatic and are possible only if certain conditions are met. For instance, it seems generally agreed that the technological gap between the home country and the host country should not be too large for spillovers to be feasible; and that the host country should have enough human capital that enables its workers to learn and to absorb new technology (Kokko, 1994; de Mello, 1997; Kinoshita, 2001).

Literature on FDI to China There is a growing academic interest in FDI to China, only matched by the growing size of inward FDI to China. With a few exceptions, for example, Huang (2003) discussed below, almost all researchers find a positive impact of FDI on China’s miraculous economic growth over the past three decades. The focus of research interests is on productivity spillovers from FDI to domestic firms. A large number of empirical studies on the topic can be found in both English language literature (e.g., Yao & Wei, 2007; Liu & Zou, 2008; Liu, 2008; Liu & Wang, 2003) and Chinese language literature (e.g., Yan, 2005; Zhang & Liu, 2004; Wang, Zhang, & Xu, 2005). The differences among these studies are usually in data sources and methodology. Most researchers use aggregated data at the national, regional or sectoral level, and thus the impacts of FDI on joint ventures, wholly foreignowned firms, and domestic Chinese firms often cannot be separated. For instance, Yao and Wei (2007) use a provincial-level panel data from 1979 to 2003 and estimate an augmented Cobb–Douglas production

112

YUSHENG PENG

function, with measures of FDI and export as shifters. FDI is measured by log share of FDI in total investment (i.e., the sum of FDI and domestic investment). They find that both FDI and export shifted provincial TFP upward. Even though Yao and Wei’s findings are solid, due to the aggregated nature of their data, we are not sure if the positive impact of FDI on TFP reflects knowledge spillovers or just productivity gains within the joint ventures themselves. Using quarterly national statistics from 1984 to 1995, Dees (1998) reports that FDI and machinery imports (proxy for technology spillovers) improve China’s labor productivity. Many researchers turn to analysis of industrial sectors and subsectors. Liu and Wang (2003), for instance, use the 1995 Third Industrial Census data aggregated at the sector level and find that the share of FDI in an industrial sector has a positive impact on the sector’s TFP. Liu and Zou (2008) also use sector-level data in the electronic industry (1997–2004, China Statistical Yearbook on High Technology Industry). They report that greenfield FDI and cross-border M&A have positive impacts on the innovation performance (ratio of new product sales over total sales) of indigenous Chinese firms. Liu et al. (2001) use 41 subsectors of the Chinese electronic industry in 1996 and 1997 to examine the overall effects of inward FDI. They find that FDI is associated with higher labor productivity. Buckley, Wang, and Clegg (2007) break down a sector-level data (2001) by ownership types, high-tech versus low-tech industries. They use OLS to estimate a simple Cobb–Douglas function and find that SOEs in both high-tech and low-tech sectors benefited from spillovers of FDI presence. Firm-level analysis is more effective in terms of pinpointing the spillover externalities. Using firm-level data in the manufacturing industries (1995– 1999), Liu Zhigang (2008) specifically tackles the task of capturing spillover effects from FDI (measured as percentage of foreign equity at the level of four-digit and two-digit industrial sectors). He finds that FDI increases the TFP of firms in the same industry and downstream industries, suggesting positive productivity spillovers.

FDI and SOE Reform Even though there is a growing body of empirical studies on the positive association between FDI to China and productivity growth in the Chinese economy, not all scholars are enthusiastic about FDI. For instance, Huang (2003) is very critical of FDI, not because he doubts the benefits of FDI, but because he believes that the Chinese government had been attracting foreign


113

investors with tax incentives and asset discounts at the expense of the indigenous private sector. Thus, Huang predicts that the large influx of FDI is unhealthy for China’s economic growth in the long run. Apparently, Huang does not believe that FDI-induced technology spillovers are large enough to justify ‘‘selling China’’ short. Supporters of China’s integration into the world economy have more than just technology spillovers in mind. Justin Lin (2001a, 2001b), for example, argues that China’s accession to the WTO in 2001, which leads to more FDI and competition, will spur economic reforms in the financial sector and SOEs. What will be spilled over through FDI and international trade are not just ideas of technical nature, such as new production technology, more effective marketing and sales strategies, or more efficient management practices, but also ideas of market institutions that are particularly important for a transition economy. By setting up subsidiaries in China, MNCs are setting up examples of genuine capitalist enterprises and their best practices. Increased competition will not only force local firms to upgrade their technology but also force officials to reform and local firms, especially SOEs, to adapt. Despite years of reform, SOEs still remained afflicted with inefficiencies and low productivity and became the most inefficient sector in the otherwise dynamically growing economy by the mid-1990s. In fact, SOE reform was the most difficult task of China’s reform efforts in that decade. The ailment of the SOEs and that of the financial sectors were closely related. In the final analysis, SOEs performed poorly because they faced soft budget constraints. In the classical socialist system, budgetary fund allocations to the stateowned firms are notoriously soft (Kornai, 1992). During the reform era, the state tries to harden budget constraints by replacing grant allocations with bank loans, but the state-owned financial institutions tend to dole out soft monies to SOEs (Walder, 1995; Steinfeld, 1998; Peng, 2001). Soft loans kept the underperforming SOEs afloat and the accumulation of nonperforming loans to SOEs threatened the stability of the state-owned financial system. It was the difficulties in SOE and banking reform that led Lardy (1998) to describe China’s reform as an ‘‘unfinished economic revolution.’’ In the late 1990s, major reform measures were carried out, including ‘‘keeping the large and releasing the small’’ that sold all the small and inefficient SOEs to domestic and foreign investors and imposed various corporate restructurings of the remaining large SOEs, and the ‘‘debit-equity swaps’’ in which the state wrote off a large amount of bad loans from the books of the SOEs. These reform measures have produced tangible results, but there are still many hurdles to overcome (Huang, Saich, & Steinfeld, 2005).

114

YUSHENG PENG

Justin Lin (2001a) reiterates that policy burdens were at the root of the poor performances of the SOEs. Most of the SOEs were in the heavy industrial sectors that were capital intensive and did not reflect China’s comparative advantage in cheap labor. In order to keep them from going under, the state had to distort market prices and extend cheap credits to them. During the reform era, SOEs continued to receive preferential treatment and evade market disciplining due to policy burdens and lack of a level playing field (Lin, Cai, & Li, 1998; Lin & Tan, 1999). When the reform enhanced the autonomy of managers in public enterprises, serious agency problems arose because the government, as owner, could not closely monitor the behavior of managers due to information asymmetry. In a fully developed market system, competition can overcome information asymmetry by making relative profit rates a reliable summary indicator of the manager’s performance. Thus, managerial incentive can be aligned with the owners’ interest by linking the manager’s compensation and future wages (reputation) with the firm’s financial performance. However, this mechanism is not working for China’s SOEs, because SOE profits or losses may not reflect managerial competence and effort due to policy burdens such as irrational price control, redundant and retired workers, and bureaucratic intervention. Unable to distinguish policy losses from mismanagement losses, the state had to assume full responsibility. This softened the budget constraints on the firms which in turn worsened the agency problem. Hence the vicious cycle of policy burdens, subsidies, agency problems, and bureaucratic intervention. Without a level playing field, managerial contracts in SOEs did enhance the managers’ profit incentive, but did not align that incentive to a focus on improving efficiency. Lin (2001a, 2001b) emphasizes that after China enters WTO, foreign banks will be allowed to operate in China and compete with domestic banks for deposits and credit services. The state-owned banks will be forced to operate on commercial principles and stop extending soft credits to inefficient state-owned firms. Officials will also have to learn to relieve SOEs of their policy burdens in order to create a level playing field. Extending Lin’s argument that WTO accession will spur financial and SOE reforms, I propose that FDI before the WTO accession had spurred SOE reforms through intensifying the competition and setting examples of competitive enterprises and market-oriented behaviors. In order to attract and compete for foreign investors, local governments also made efforts to create a more functional legal framework, build fairer market institutions, and reduce subsidies to the SOEs. Facing mounting competition, SOEs had to figure out how to make better and cheaper products in order to compete


115

in the market with those new foreign products that are produced abroad or next door. Competition may drive down the profit margins for SOEs, but drive up the latter’s managerial and technological efficiency. Therefore, FDI will not only bring technology spillovers to SOEs, but also serve as catalysis for SOE reform. Chinese cities vary greatly in terms of levels of FDI investment, quantity and quality of SOEs, and paces of reform by the municipal governments. In fact cities compete with each other for foreign investors. The presence of FDI in Chinese cities should generate positive externalities for nonaffiliated SOEs located in these cities either through incurring technology spillovers or through enticing efficiency-augmenting institutional changes. This leads to our central hypothesis. Hypothesis. SOEs located in cities with a high level of FDI should perform better in productivity than those located in cities with a low level of FDI.

METHODS Contextual analysis is appropriate here because it can combine firm-level variables and city-level variables into the same equation, thus enabling the test of the impact of city-level variables on SOE performance. Ordinary least square regression is inappropriate in this case because: (1) firms within the same city are inherently clustered, that is, subject to the same disturbances; and (2) treating city-level variables as the same as firm-level variables inflates the number of observations. Both situations would lead to an underestimation of standard errors in OLS. Thus, I specify a multilevel model in the following form. ln yij ¼ Aj þ a ln kij þ b ln l ij þ xij c þ ij

(1a)

Aj ¼ l þ fFDIj þ Cj j þ uj

(1b)

In Eq. (1a), yij stands for sales revenue of the ith firm in the jth city (i ¼ 1,2,3yn and j ¼ 1,2,3y560); kij for capital input; lij for total number of employees; xij stands for row vector of a set control variables including firm age, log ratio of redundant workers, dummy variables for industrial sectors, and ownership levels. With a constraint of a þ b ¼ 1 (i.e., constant

116

YUSHENG PENG

return to scale), this would conform to the standard Cobb–Douglas production function. a þ bW1 implies an increasing return to scales. Aj in Eq. (1a) is an intercept, measuring TFP in this case. Its subscript suggests that it is a random effect and varies from city to city. Eq. (1b) specifies TFP variation between cities as a function of presence of FDI and a set of control factors (C) such as city administrative levels, GDP per capita, log ratio of self-employment, city-wide ratio of SOE redundant workers, coastal location, etc. In this production function specification, the coefficient of FDI (and other shifter variables) is interpreted as the effect on TFP because input factors (labor and capital) are held constant. Both the firm-level error term (eij) and the city-level error term (uj) are assumed to be normally and independently distributed. The variance of uj captures the residual correlation between firms within the same city. The model can be estimated via the restricted maximum likelihood method in SAS/PROC MIXED procedure (for reference, see Singer, 1998; Kreft & de Leeuw, 1998). Note that given the cross-sectional data and a lack of instrumental variables, our estimates of capital elasticity a and labor elasticity b might be biased, because of the endogeneity problems (i.e., capital and labor supply may be influenced by sales). But this bias should not have much impact on the coefficient estimates for the contextual variables, such as city-level measures of FDI presence.

DATA AND MEASURES This study uses both firm-level and city-level data. The primary data source for firms comes from the 1995 Third National Industrial Census, which collects financial and labor information on all industrial SOEs nationwide (ACMRC, 1999). Such a comprehensive data set would allow me to examine the FDI on SOE performance in productivity and efficiency. The advantage of using 1995 industrial census data is that the survey was conducted before the shocks of major SOE restructuring and financial reforms, such as ‘‘keeping the large and releasing the small’’ that sold the small and inefficient SOEs, many to MNCs, and the ‘‘debit-equity swaps’’ that helped the bottom lines of many large SOEs. Thus, we have less of a selection bias. The original firm-level data set from the Third Industrial Census contains records for over 80,000 SOEs (Table 1). As we can see, most of these SOEs were owned by county-level governments in 1995. These were called the ‘‘local SOEs’’ that tended to be small and inefficient and were the first to go

117


Table 1. Level of Ownership

Distribution of SOEs by Levels of Ownership (1995).

Complete Census List of SOEs

Urban SOEs Selected for Analysis

Frequency Percentage Frequency Percentage

Central Province Prefecture County Other Total

4,252 9,596 22,892 49,536 304 86,580

4.9 11.1 26.4 57.2 0.4 100.0

Average capital (millions)

Average employment (person)

Average sales (millions)

3,561 6,733 17,990 13,273

8.6 16.2 43.3 31.9

1,130 268 136 49

2,615 1,018 669 292

2,403 659 348 144

41,557

100.0

214

772

509

during the ‘‘keeping the large and releasing the small’’ campaign. About 5% or slightly more than 4,000 SOEs were owned by the central government (or its various ministries). But these were the large ones and some of them were to become gigantic corporations. For the current analysis, I only selected those SOEs that were located in 560 cities (defined below) and had valid data on key variables. Only about half of the total number, or 41,557, are used in the following analysis. Note that the selected pool of SOEs has a smaller proportion owned by the counties, because most of the counties are classified as a rural county instead of a county-level city. Some central-, provincial-, and prefectural-level SOEs were dropped as well because they were located in rural areas. Table 1 also presents the basic financial and labor statistics of the selected SOEs by ownership levels. We can see that the size and scale of SOEs increase as one moves up the hierarchical ladder of ownership. As Walder (1992, 1995) points out, firms at different levels of ownership enjoyed different privileges in the central planning hierarchy, such as access to bank loans and subsidies. Firm-level variables are defined in the following: Log 1995 sales revenue, which is believed to be the most reliable output measure in China. Log total amount of capital by the end of 1995. Log total number of employees in 1995. Firm age measured as the duration (years) since founding (those founded prior to 1949 are counted as born in 1949).

118

YUSHENG PENG

Log share of redundant employees out of total number of employees. This variable and firm age are included as control because they indicate policy burdens and historical baggage faced by SOEs (Lin, Cai, & Li, 1998; Lin & Tan, 1999). Ownership levels of SOEs are divided into (1) central, (2) provincial, (3) prefectural, and (4) county or district. City-level data are either generated from the 1995 Third Industrial Census or taken from the 1993–1994 volume and the 1995 volume of the Urban Statistical Yearbooks of China (NBS, 1995, 1996). By the end of 1993, there were 570 cities in China, consisting of 3 provincial-level cities, 16 semiprovincial level cities, 183 prefecture-level cities, and 368 county-level cities (NBS, 1995). This is my definition base of cities. After deleting 10 cities due to mergers, missing values, and mismatches, 560 cities remain for the following analysis. These cities vary widely with regard to inward FDI and SOE performances. Examining the effects of FDI at the city level on SOE performance at the firm level will reveal the ‘‘spillover’’ externalities from FDI to domestic firms. The following measures are generated from the 1995 Third Industrial Census unless otherwise indicated.

Measuring FDI Penetration at the City Level Two alternative measures of FDI are employed in this paper. The first measure is the log share of the number of firms with FDI (sanzi qiye) within a city, including both wholly foreign-owned enterprises and foreign joint ventures, as percentage of the total number of both FDI-related firm and domestic SOEs. As the second alternative measure, I use the log share of foreign-invested capital stock over the total capital stock of both FDI-related firm and SOEs in the same city. Both measures are computed from the Third Industrial Census by aggregating all FDI-related firms, including those from Hong Kong, Macau, and Taiwan, but excluding those village-level sanzi firms.2 Both are lagged 1 year to 1994 and are logged to conform to the log-linearized form of the Cobb–Douglas production function. The following is a list of city-level variables that may influence SOE productivity and FDI and should be controlled for. Table 2 reports summary statistics for these variables.3

31.4 59.8

Mean STD

20,560 17,911

2,789 9,998 15,030 24,794 189,470 75 168

2 27 40 71 2,859

1995 Number of SOEs

41 132

0 2 7 27 1,708 2,588 10,412

0 40 178 921 135,248 9.7 15.0

0.8 1.0 3.5 11.3 91.0 8,672 28,761

0 400 1,453 5,591 335,434

1994 1994 FDI 1994 FDI 1995 Number of Capital Capital Employees FDI Firms (Million) Share (%) in FDI Firms

57,470 112,832

80 10,015 20,108 66,078 1,487,962

4.7 4.5

0.0 1.6 3.5 6.4 36.7

3.9 5.3

0.0 1.3 2.8 4.8 64.5

1995 1995 SOE 1993 SelfEmployees Labor Employment in SOEs Redundancy Rate (%) Rate (%)

Descriptive Statistics of China’s Cities in 1995 (N ¼ 560).

Note: Population, self-employment, and GDP data are taken from the Urban Statistical Yearbook (NBS, 1995) and the rest are calculated from the Third Industrial Census.

1 10 16 28 810

Minimum Q1 Median Q3 Maximum

1993 1994 GDP Population Per Capita (10,000) (Yuan)

Table 2.

FDI and SOE Performance in China 119

120

YUSHENG PENG

Administrative levels of Chinese cities are classified into four categories: provincial, semiprovincial, prefectural, and county. Log total number of SOEs in the same city, excluding those located outside of the city proper. This variable is aggregated from the Third Industrial Census. Log total employment within a city in 1993, taken from urban statistical yearbook (NBS, 1995). This variable measures the scale of the overall economy in a city. GDP per capita in 1994, taken from urban statistical yearbook (NBS, 1996). This is an indicator of overall level of economic development in a city. It is also a proxy for the general level of technological and institutional efficiency and thus serves as control for the initial level of general productivity in a city. Log share of total number of self-employed workers out of the total workforce as a measure of development of private economy in a city (NBS, 1995). This indicator together with GDP per capita serves as a control for different degrees of institutional reform in these cities. As we can see in Table 2, in 1995 the ratio of self-employment in Chinese cities was quite low, merely 4% on average. Log share of redundant workers out of total employment in the city as an indicator of policy burdens on SOEs. This is a city-wide measure of employment burden in addition to the firm-level measure. According to the 1995 Third Industrial Census, the redundancy rate in SOE workforce was not that serious, less than 5% on average (Table 2). By the late 1990s, however, a much higher percentage of SOE workers were laid off. Coastal city is a dummy variable that equals 1 if the city is along the coastal seaboard.

FINDINGS Table 3 presents the results of multilevel production function analysis of SOEs. Three models are estimated. Model 1 does not include FDI measures and is a base for comparison. Models 2 and 3 are identical except for their different measures of FDI. The following findings should be noted. First, Model 2 and Model 3 show that presence of FDI in a city does have significant and substantial impacts on the TFP of domestic SOEs in the same city. The finding is robust against different measures of FDI penetration, although it seems that the number of FDI-related firms (Model 2) have a larger impact than the share of FDI capital (Model 3). To put the

121


Table 3.

Multilevel Analysis of Log SOE Sales Revenue in China (Data from the 1995 Third National Industrial Census). Model 1

Model 2

Model 3

Coefficient t-ratio Coefficient t-ratio Coefficient t-ratio Intercept City-level variables (N ¼ 560) Log share of FDI firms (1994) Log share of FDI capital (1994) Log number of SOEs in the city Log total employment (1993) Log share of SOE redundant workers Log share of selfemployment (1993) Log GDP per capita (1994) Coastal city (yes ¼ 1) Administrative levels Provincial Semiprovincial Prefecture (County) Firm-level factors (n ¼ 41,557) Log capital Log labor Log share of redundant workers Firm age Ownership levels Central Provincial Prefectural County Industrial sectors Chemical Light industry Machinery Material (Metallurgy) Fit statistics 2 Res log likelihood AIC

1.662

40.1

2.044

6.32

0.131

6.82

2.133

6.56

0.085

6.23

0.092

2.86

0.067

2.08

0.011 0.024

0.4 2.21

0.025 0.028

0.94 2.61

0.024

1.92

0.025

1.96

0.358 0.004

13.07 0.13

0.369 0.011

13.54 0.38

0.254

7.15

0.151 0.152 0.062 0.000

0.67 1.51 1.59

0.176 0.152 0.055 0.000

0.94 1.58 1.32

0.095 0.088 0.015 0.000

0.51 0.92 0.36

0.306 0.754 0.111

64.87 118.98 37.19

0.303 0.757 0.110

64.35 119.65 37.06

0.303 0.757 0.110

64.48 119.63 37.03

0.003

8.83

0.003

8.73

0.003

8.77

0.298 0.105 0.019 0.000

12.74 4.94 1.07

0.304 0.112 0.027 0.000

13.08 5.28 1.48

0.307 0.113 0.029 0.000

13.21 5.34 1.61

0.181 0.038 0.220 0.310 0.000

7.28 1.68 9.63 11.52

0.180 0.040 0.221 0.309 0.000

7.25 1.78 9.68 11.46

0.181 0.039 0.221 0.309 0.000

7.25 1.76 9.7 11.48

119,494.4 119,498.4

119,187.2 119,191.2

119,195.1 119,199.1

122

YUSHENG PENG

coefficients in perspective, a double in the share of FDI-related firms in a city would lead to 9.5% (E20.1311) productivity increase for SOEs unaffiliated with FDI but located within the same city, other things equal.4 By comparison, a double in the share of FDI-related capital stock would lead to a 6% (E20.0851) productivity increase for unaffiliated SOEs, ceteris paribus. The larger coefficient for the number of FDI-related firms may suggest that domestic SOEs benefit more from contact points with FDIrelated firms than from the presence of large foreign capital stock per se. Since the SOEs in the analysis were not directly related to FDI, the effects can only be interpreted as positive externalities of knowledge spillovers. I believe that what has been spilled over from FDI-related firms to domestic SOEs are not only ideas of technical nature but also ideas about institutional innovation and institutional reform. Second, redundant SOE workers measured at the city level have a negative impact on the productivity of individual firms even after the redundancy rate is controlled at the firm level. This suggests that employment redundancy in SOEs has generated negative externalities in the whole city. A SOE is not only burdened by its own surplus laborers but also negatively affected by redundant workers in other firms within the same city. When the local government forces SOEs to shoulder employment burdens, normal market competition and disciplining cannot function and firms will have more excuses to be inefficient (Lin et al., 1998). Third, SOE productivity does not significantly differ across the administrative levels of the municipality. Instead, it does differ across different levels of ownership. In other words, who owns and controls (and cares for) the firm matter. It is obvious that one moves up the planning hierarchy, firm performance moves up too. This result seems robust against variations in model specification. Fourth, coastal cities do have an advantage in productivity, as demonstrated in Model 1. But once we control for city-level covariates, this advantage disappears. This result suggests that the most prominent spatial variation in productivity between coastal and inland areas is explained by the city-level variables included in the model, such as FDI penetration and initial levels of development. Finally, estimates of labor and capital elasticities seem to be well behaved. We find a slight increasing return to scale in all three models, which should be expected. Consistent with previous analysis of SOE data (Li, 1997; Bai, Li & Wang, 1997), labor’s share of output is about 30% and capital’s share about 75%.


123

DISCUSSION This study contributes to the growing literature on FDI to China in two aspects. First, it uses a very large data set from the Third Industrial Census. Despite the shortcomings of cross-sectional data, the census data provide the most comprehensive coverage of all industrial SOEs. Second, it uses multilevel models to specifically capture the spillover externality from FDIrelated firms to domestic enterprises. This approach eliminates the impact of direct knowledge transfer from MNCs or overseas Chinese investors to their subsidiaries or affiliated sanzi firms in China. Since we measure FDI at the city level, and observe its impact at the firm level on domestic SOEs not affiliated with FDI, we can be confident that the observed productivity gain reflects positive externalities generated by FDI. The Chinese government may have been well justified in its efforts to attract FDI with tax holidays and preferential treatment, because the benefits of FDI do spillover to domestic firms, including SOEs. My analysis of the 1995 Industrial Census data has shown that SOEs reaped significant productivity gains from the sudden influx of FDI. Existing theories of FDI and economic growth emphasize technological diffusion and knowledge spillovers. Through direct investment, MNCs bring not only physical capital into the host country but also new products, new technology, and new ideas about how to produce goods of higher quality and at lower prices, how to market products more effectively, and to manage more efficiently. These novel technology and new ideas will diffuse into domestic industries through imitation, observation, backward and forward linkages, and employee turnovers. China’s manufacturing sector, as well as the service sectors, has seen tremendous technology leapfrogging during the past three decades, especially the last 15 years. There is no doubt that the massive entry of MNCs and influx of inward FDI since the early 1990s played an important role in this process. I have emphasized that in a transition economy such as China, exposure to FDI, MNCs, and international trade and competition may have the double benefits of spurring institutional learning and institutional reform. From the evidence presented here, we can expect that China’s entry into WTO will help the transformation of SOEs and the shake up of the financial system. International competition may drive down profits for domestic firms in the short run but promote economic health and growth in the long run by creating a level playing field, weeding out inefficiencies, and overcoming reform hurdles.

124

YUSHENG PENG

NOTES 1. In August 1980, the central government announced to establish four Special Economic Zones: Shenzhen, Zhuhai, Shantou, and Xiamen. These were the four initial windows of China to the outside world. Later in 1988, Hainan province became the fifth Special Economic Zone. Foreign investment, mostly from Hong Kong, Macau, and Taiwan at the early stage, was allowed and encouraged within the perimeters of these special zones. In 1984, the central government designated another 14 coastal cities as ‘‘open’’ for foreign investment and international trade. These 14 ‘‘open cities’’ include Shanghai, Guanzhou, Fuzhou, Tianjin, Dalian, Qinhuangdao, Yantai, Qingdao, Ningbo, Wenzhou, Lianyungang, Nantong, Zanjiang, and Beihai. Restrictions against foreign investment in the inland provinces were gradually removed by the 1990s (cf., Lardy, 2002). 2. By official definition, sanzi firms refer to (a) wholly owned foreign enterprises, (b) joint capital ventures (hezi qiye) that require a minimum foreign investment of 25%, and (c) joint cooperative ventures (heying qiye) that define a looser form of cooperation and do not have to be a legal entity. 3. All city-level data are for city proper and do not include economic activities in the rural area but within the jurisdiction of a city. Some county-level cities are nested under prefectural-level cities. Data for the prefecture-level cities do not include these nested county-level cities. 4. Assume that there are two identical SOEs in two separate cities, with sale revenues y1 and y2, respectively. The two cities share the same endowments considered in our model except that the percentage of FDI-related firms equals to f in the first city and doubles that in the second city, that is, 2f. Then, according to Model 2 in Table 3 and other things equal, we have: y log 2 ¼ logðy2 Þ logðy1 Þ ¼ f logð2f Þ f logðf Þ ¼ logð2f Þ y1 Or, y2 ¼ 2f ¼ 20:131 1:095 y1 In other words, a double in the share of FDI-related firms would lead to a 9.5% increase in sales revenue. Because input factors (labor and capital) are held constant and other things equal, an increase in output reflects productivity gain resulting from FDI spillovers.

REFERENCES ACMRC (All China Marketing Research Co. Ltd). (1999). Quanguo di sanci gongye pucha ziliao (National Database of the Third Industrial Census). Beijing, China: All China Marketing Research Co. Ltd. Aitken, B. J., & Harrison, A. E. (1999). Do domestic firms benefit from direct foreign investment? Evidence from Venezuela. American Economic Review, 89, 605–618.


125

Bai, C.-E., Li, D., & Wang, Y. (1997). Enterprise productivity and efficiency: When is up really down? Journal of Comparative Economics, 24(3), 265–280. Blomstrom, M., & Kokko, A. (1998). Multinational corporations and spillovers. Journal of Economic Survey, 12, 247–277. Blomstrom, M., & Sjoholm, F. (1999). Technology transfer and spillovers: Does local participation with multinationals matter? European Economic Review, 43, 915–923. Buckley, P. J., Wang, C., & Clegg, J. (2007). The impact of foreign ownership, local ownership, and industry characteristics on spillover benefits from foreign direct investment in China. International Business Review, 16, 142–158. Byun, H.-Y., & Wang, Y. (1995). Technology transfer and multinational corporations: The case of South Korea. Journal of Asian Economics, 6(2), 201–216. Campos, N. F., & Kinoshita, Y. (2002). Foreign direct investment as technology transferred: Some panel evidence from the transition economies. The Manchester School, 70(3), 398–419. Caves, R. E. (1996). Multinational enterprise and economic analysis. London: Cambridge University Press. Dees, S. (1998). Foreign direct investment in China: Determinants and effects. Economics of Planning, 31, 175–194. Dunning, J. H. (1997). Alliance capitalism and global business. London: Routledge. Findlay, R. (1978a). Some aspects of technology transfer and foreign direct investment. American Economic Review, 68(2), 275–279. Findlay, R. (1978b). Relative backwardness, direct foreign investment, and the transfer of technology: A simple dynamic model. Quarterly Journal of Economics, 92, 1–16. Firebaugh, G. (1992). Growth effect of foreign and domestic investment. American Journal of Sociology, 98, 105–130. Firebaugh, G. (1996). Does foreign capital harm poor nations? New estimates based on Dixon and Boswell’s measure of capital penetration. American Journal of Sociology, 102(2), 563–575. Fosfuri, A., Motta, M., & Rønde, T. (2001). Foreign direct investment and spillovers through workers’ mobility. Journal of International Economics, 53, 205–222. Frankel, J. A., & Romer, D. (1999). Does trade cause growth? American Economic Review, 89(3), 379–399. Go¨rg, H., & Greenaway, D. (2004). Much ado about nothing? Do domestic firms really benefit from foreign direct investment? World Bank Research Observer, 19, 171–197. Grossman, G. M., & Helpman, E. (1991). Innovation and growth in the global economy. Cambridge, MA: MIT Press. Haddad, M., & Harrison, A. (1993). Are there positive spillovers from direct foreign investment? Evidence from panel data for Morocco. Journal of Development Economics, 42, 51–74. Huang, Y. (2003). Selling China: Foreign direct investment during the reform era. Cambridge, MA: Cambridge University Press. Huang, Y., Saich, T., & Steinfeld, E. (Eds). (2005). Financial sector reform in China. Cambridge, MA: Harvard University Asia Center. Javorcik, B. S. (2004). Does foreign direct investment increase the productivity of domestic firms? In search of spillovers through backward linkages. American Economic Review, 94, 605–627. Kinoshita, Y. (2001). R&D and technology spillovers through FDI: Innovation and absorptive capacity. Centre for Economic Policy Research Discussion Paper #2775. (www.cepr.org/ pubs/dps/DP2775.asp).

126

YUSHENG PENG

Kokko, A. (1994). Technology, market characteristics, and spillovers. Journal of Development Economics, 43, 279–293. Konings, J. (2001). The effects of foreign direct investment on domestic firms: Evidence from firm-level panel data in emerging economies. Economics of Transition, 9(3), 619–633. Kornai, J. (1992). The socialist system: The political economy of communism. Princeton, NJ: Princeton University Press. Kreft, I., & de Leeuw, J. (1998). Introducing multilevel modeling. London: Sage Publications. Lall, S. (1978). Transnationals, domestic enterprises, and industrial structure in host LDCs: A survey. Oxford Economic Papers, 30, 217–248. Lardy, N. (1998). China’s unfinished economic revolution. Washington, DC: Brookings Institution Press. Lardy, N. R. (2002). Integrating China into the global economy. Washington, DC: Brookings Institution Press. Lardy, N. R. (1995). The role of foreign direct investment in China’s economic transformation. China Quarterly, 1065–1082. Li, W. (1997). The impact of economic reform on the performance of Chinese State Enterprises, 1980–1989. Journal of Political Economy, 105(5), 1080–1106. Lin, J. Y. (2001a). WTO accession and financial market reform in China. The Cato Journal, 21(1), 13–19. Lin, J. Y. (2001b). WTO accession and China’s SOE reform. In: K. T. Lee, J. Y. Lin & S. J. Kim (Eds), China’s integration with the world economy: Repercussions of China’s accession to the WTO. Seoul, Korea: Korea Institute for International Economic Policy. Lin, J. Y., Cai, F., & Li, Z. (1998). Competition, policy burdens, and state-owned enterprise reform. American Economic Review, 88(2), 422–427. Lin, J. Y., & Tan, G. (1999). Policy burdens, accountability, and the soft budget constraint. American Economic Review, 89(2), 426–431. Liu, X., & Wang, C. (2003). Does foreign direct investment facilitate technological progress: Evidence from Chinese industries. Research Policy, 32, 945–953. Liu, X., & Zou, H. (2008). The impact of greenfield FDI and mergers and acquisitions on innovation in Chinese high-tech industries. Journal of World Business, 43, 352–364. Liu, X., Parker, D., Vaidya, K., & Wei, Y. (2001). The impact of foreign direct investment on labour productivity in the Chinese electronics industry. International Business Review, 10, 421–439. Liu, Z. (2008). Foreign direct investment and technology spillovers: Theory and evidence. Journal of Development Economics, 85, 176–193. Lloyd, P. J. (1996). The role of foreign investment in the success of Asian industrialization. Journal of Asian Economics, 7(3), 407–433. de Mello, L. R. (1997). Foreign direct investment in developing countries and growth: A selective survey. Journal of Development Studies, 34, 1–34. de Mello, L. R. (1999). Foreign direct investment-led growth: Evidence from time series and panel data. Oxford Economic Papers, 51, 133–151. Murakami, Y. (2007). Technology spillover from foreign-owned firms in Japanese manufacturing industry. Journal of Asian Economics, 18, 284–293. National Bureau of Statistics. (1995). Statistical yearbook of Chinese cities, 1993–1994 (Zhongguo chengshi tongji nianjian, 1993–1994). Beijing, China: Zhongguo Tongji Chubanshe.


127

National Bureau of Statistics. (1996). Statistical yearbook of Chinese cities, 1995 (Zhongguo chengshi tongji nianjian, 1995). Beijing, China: Zhongguo Tongji Chubanshe. National Bureau of Statistics. (2008). Statistical yearbook of China, 2008. Beijing, China: Zhongguo Tongji Chubanshe. Okamoto, Y. (1999). Multinationals, production efficiency and spillover effects: The case of the U.S. Auto Parts Industry. Review of World Economics, 135(2), 241–260. Peng, Y. (2001). Chinese villages and townships as industrial corporations: Ownership, governance, and market discipline. American Journal of Sociology, 106, 1338–1370. Pomfret, R. (1997). Growth and transition: Why has China’s performance been so different? Journal of Comparative Economics, 25, 422–440. Romer, P. (1986). Increasing returns and long-run growth. Journal of Political Economy, 94, 1002–1037. Romer, P. (1990). Endogenous technological change. Journal of Political Economy, 98, S71–S102. Romer, P. (1993a). Idea gaps and object gaps in economic development. Journal of Monetary Economics, 32, 543–573. Romer, P. (1993b). Two strategies for economic development: Using ideas and producing ideas. In: L. Summers & S. Shah (Eds), Proceedings of the World Bank annual conference on development economics 1992 (pp. 63–92). New York: The World Bank. Singer, J. D. (1998). Using SAS PROC MIXED to fit multilevel models, hierarchical models, and individual growth models. Journal of Educational and Behavioral Statistics, 24(4), 323–355. Steinfeld, E. (1998). Forging reform in China: The fate of state industry. Cambridge: Cambridge University Press. Walder, A. (1992). Property rights and stratification in socialist redistributive economies. American Sociological Review, 57(4), 524–539. Walder, A. (1995). Local governments as industrial firms. American Journal of Sociology, 101(2), 263–301. Wang, Q., Zhang, J., & Xu, W. (2005). Who benefits from whom? FDI in China’s manufacturing sector’’ (shui cong shui huoyi: Zhongguo zhizhaoye zhong de waishang zhijie touzi). Zhongguo Gongye Jingji, 2005(10), 23–31. Yan, B. (2005). A survey of foreign literature on FDI spillovers (guowai FDI yichu xiaoying yanjiu zongshu). Dangdai Caijing, 2005(4), 100–103. Yao, S., & Wei, K. (2007). Economic growth in the presence of FDI: The perspective of newly industrializing economies. Journal of Comparative Economics, 35, 211–234. Zhang, H., & Liu, H. (2004). The spillover and competition effects of DFI on China’s industries’’ (waizi yichu xiaoying yu jingzheng xiaoying dui zhongguo gongye bumen de yingxiang). Guoji Maoyi Wenti, 2004(3), 76–81.

OWNERSHIP AND INNOVATION DURING ECONOMIC DEVELOPMENT: CORPORATE OWNERSHIP AND STRATEGY FORMATION IN CHINA Lisa A. Keister and Randy Hodson ABSTRACT Innovation is critical to organizational survival, competitive advantage, and economic development. Yet the process by which innovative strategic behavior occurs is not well understood. This paper takes advantage of rapidly changing corporate governance structures and environmental conditions during China’s economic transition to explore the role of corporate ownership in shaping firm innovation. We argue that managers draw on internal strengths within external constraints to develop strategies and that the nature of corporate ownership determines the degree to which internal or external factors are salient. We capitalize on differences between Chinese state-owned enterprises (SOEs) and collective enterprises and other non-state firms (CNFs) in the adoption of firm strategies during transition. Analyzing data from 1994 to 1999 on 800 Chinese firms, we study the effect of ownership type on the adoption of four key organizational innovations and identify major strategic groups Work and Organizations in China after Thirty Years of Transition Research in the Sociology of Work, Volume 19, 129–163 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1108/S0277-2833(2009)0000019008

129

130

LISA A. KEISTER AND RANDY HODSON

that developed during reform. The findings provide important insight into the role of corporate governance in influencing strategy formation and adaptation, outcomes that are increasingly important in all economies.

Innovation in response to external shocks is essential for competitive advantage and organizational survival (D’Aveni, 1994; Richardson, 1996; Volberda, 1996), and it is a critical component of economic development. Organization response to change in transition economies offers a unique opportunity to study the process by which innovation occurs. In the past, firms in nearly all economies were able to establish dominance during extended periods of economic stability. Increasingly, however, firms face rapidly changing environments in which brief periods of advantage are interrupted by external shocks that threaten performance and survival (D’Aveni, 1994; Ilinitch, Lewin, & D’Aveni, 1998). While awareness of the importance of organizational adaptation has intensified, the factors that encourage or hinder adaptation are not well understood. The political and economic shocks that occurred in transition economies in recent decades facilitate exploration of this process because transition forced firms to adapt and innovate in nearly all aspects of their behavior. Early in reform, basic institutional arrangements governing economic behavior were dismantled, environmental uncertainty intensified, and managerial response was relatively rapid and visible. Research on Western firms suggests that either internal incentives and preferences (Dess & Beard, 1984; Keats & Hitt, 1988; Pfeffer & Salancik, 1978) or external pressures (Buchko, 1994; DiMaggio & Powell, 1983; Meyer & Rowan, 1977) determine innovative organizational response. In reality, both internal and external processes are likely important, but the relative weight of these influences and the conditions under which each becomes most relevant is unclear. An organizational trait that is likely to determine the relative salience of internal and external influences is corporate ownership. Corporate ownership, particularly its relationship to control of the firm, has been shown to shape various firm outcomes including equity capital and managerial employment contracts (Baysinger & Hoskisson, 1990), research and development (Baysinger, Kosnik, & Turk, 1991), and productivity and efficiency (Hill & Snell, 1989). It is usually difficult to study the effect of ownership on firm outcomes because there is little variation either across firms or within firms over time in most economies. However, during an extreme economic transition, the changing nature of firm ownership can highlight the role of ownership in shaping innovation.

Corporate Ownership and Strategy Formation in China

131

To study firm adaptation during the early part of China’s economic transition, we integrate ideas from strategic choice and institutional theory to develop a set of arguments about the adoption of a range of strategies during transition. We propose that managers exploit internal strengths within external constraints to develop a coherent strategy for the firm, and we argue that the nature of corporate ownership determines the degree to which internal or external factors shape innovation. In particular, we explore how differences between state-owned enterprises (SOEs) and collective enterprises and other non-state firms (CNFs) determine the adoption of a general firm strategy during transition. We focus on multiple elements of strategy rather than a particular strategy or innovation in order to understand the importance of strategic choice and institutional factors across a broad set of organizational actions. First, we study the effect of ownership on the adoption of four separate strategies: establishing a corporate form, a board of directors, a finance department, and a marketing department. Then, we identify the formation of major strategic groups within the two broad ownership groups (SOEs and CNFs). The strategic groups are clusters of firms with common post-transition strategies. Identifying the strategic groups allows us to identify and explore the most common general corporate strategies in the post-reform era. China’s transition is an ideal context to study innovation because reform was a major, identifiable external shock that began at a clear point in time. During transition, the emergence of organizations with radically different histories, structures, and incentives facilitates adjudicating between the various processes that shape innovation. Moreover, the reforms were extreme in their scope suggesting that the use of a given corporate strategy was likely influenced by changes begun in recent decades. At the same time, because China’s reform was relatively gradual and incremental compared to other transition economies, it is possible to isolate and observe firm responses. We first develop arguments regarding the effect of ownership on the adoption of four key innovations and the formation of strategic groups. We then use longitudinal (1994–1999) data on 800 firms in China to test our hypotheses.

CHINESE INDUSTRY BEFORE REFORM Prior to the start of economic reform in 1978, Chinese firms were all state owned and were part of an immense system of redistribution and central planning. In many ways, individual firms were branch plants of a single, albeit very large, enterprise. The hierarchical planning system included

132


multiple layers from central ministries to provinces, municipalities, counties, and townships, with each level redistributing resources within its jurisdiction. Central planning was less comprehensive in China than in the former Soviet Union, but in China, factories took fixed deductions for social support and housing and then remitted remaining profits to the state. Central planners then redistributed remitted profits according to long-term state plans. Production targets and mandates regarding production methods forced firms to conform to state-specified standards. Similarly, the government controlled labor resources. State agencies permanently assigned workers to factories based on firm needs, worker qualifications, and unofficial factors such as personal contacts. Factories could not fire workers and quitting was rare. Wages were set and tightly controlled by central authorities, workers did not receive material rewards for superior performance, and promotions were largely dictated from above. Soft budget constraints were a necessary component of the financial relationship between the state and enterprises. While there were some constraints on firm spending, these constraints were not wholly binding because the state could readily reallocate funds to cover additional expenditures. The state used its network of administrative bureaus to control resource flows throughout the economy and to redistribute resources from profitable firms to those that were not performing well. This virtually guaranteed firm survival, but it also created resource shortages and intense pressure for firms to increase production. While the firm depended on the state for all of its inputs, the state also depended on firms to provide scarce resources to other enterprises and to provide employees with jobs, housing, medical care, and other social services. State bureaus closely monitored many of the firm’s activities; however, the need to monitor a large number of firms created informational asymmetries, and managers responded by hoarding resources and bargaining for favorable treatment. Bargaining for scarce capital was acute and financing was highly uncertain because funding varied with state political whims and the personal allegiances of high-ranking officials.

Reform of Corporate Structures and Strategies Beginning in 1978, industrial reform began to change the way firms operated in China. The state gradually began to allow firms to retain profits, transferred some responsibility for firm inputs and outputs to managers, and otherwise reduced the degree to which enterprise activities were centrally coordinated. Although reformers initially intended to experiment with profit


133

retention in a small group of firms, the practices spread rapidly and widely. Reformers also sought to increase productivity by improving incentives for workers and by transferring responsibility for factory operations to managers. The ‘‘factory manager responsibility system’’ gave managers clear authority over the organization, and most enterprises adopted longterm managerial contracts (often tied to performance) that prompted the development of a labor market for managerial personnel. Central and local government control was not eliminated, nor did the reduction in control happen instantaneously. Rather, the state loosened its control gradually and, by many accounts, unevenly. Yet the loosening was clear, and this change prompted firms to respond or face being disbanded. In response, formerly state-owned enterprises adopted new structures, and collective enterprises and other non-state enterprises began to emerge. In both types of firms, the government still controls property rights. That is, the government decides when to hire and fire managers, controls assets, and decides when to close the organization or change its primary activities (Walder, 1995). The two types of firms, however, differ in the degree to which the government controls the firm’s activities. In the SOEs, the government level to which the firm answers has nearly total control. In the collectives, a local or provincial government typically shares ownership with other entities, such as individuals or a private company. Other non-state enterprises also began to emerge during China’s transition. Stock companies are those that are held publicly through one of China’s three stock markets. Joint ventures are owned by at least one Chinese interest (either governmental or private) and one foreign entity.1 Private companies are wholly owned by a Chinese entity, either an individual or another company. Because collectives are often owned at least partially by non-state entities, they tend to behave more like other non-state enterprises than SOEs. Thus, we use the term collectives and non-state entities (CNFs) to refer to this group of enterprises as a whole. During reform, both SOEs and CNFs adopted innovative corporate structures in response to changing government control of enterprises. An important change in this regard occurred in 1994 when the Company Law of China was adopted. Prior to 1994, there was no comprehensive law governing all enterprises. The company law allowed companies to be recognized as legal persons and officially recognized the existence of limited liability companies, joint stock companies, and other non-state forms of corporate organization. The majority of SOEs reorganized under company law, making them limited liability companies controlled solely by the state. Firms that incorporated after the implementation of company law were

134


recognized as legal persons. All private companies and the majority of SOEs and collectives had incorporated by the late 1990s. Many companies also established boards of directors, although this was much less common among wholly SOEs than enterprises with some private ownership. Many companies also established internal corporate structures that resembled Western structures, including marketing departments and legal departments, and enterprise groups (holding companies) formed that resembled Western conglomerates (Keister, 2000). While these innovations were common, there was considerable variation in adoption of the new strategies and structures.

FIRMS, ENVIRONMENTS, AND INNOVATION Strategy research undertaken primarily in developed market economies offers two broad explanations for innovative firm behavior: a strategic choice explanation and an institutional explanation. Strategic choice arguments emphasize the importance of internal factors such as the knowledge base of the managers and the firm, organizational routines, and related capabilities in creating innovation. From this perspective, firms’ unique knowledge and abilities allow managers to undertake new activities that, under the right set of conditions, can improve performance and the firm’s long-term competitive position. Central concerns are the identification of relevant capabilities and the explanation of how these capabilities are acquired and transferred (Ginsberg, 1988; McEvily & Marcus, 2005). Managerial innovation occurs when the firm has the necessary internal resources and capabilities to change, and the change appears to be in the firm’s advantage given available information. Penrose (1959), Nelson and Winter (1982), Barney (1986, 1991), and a host of others (Barnett, 1993; Dutton & Duncan, 1987; Ginsberg, 1988; Tichy, 1983) have shown that these largely internal processes shape the nature and timing of innovative actions. Yet there is also evidence that external environmental factors create incentives for altering organizational structures and strategic orientations. Institutional theories are particularly useful in articulating the nature of the relationship between a firm and external organizations and pressures (Buchko, 1994; DiMaggio & Powell, 1983; Meyer & Rowan, 1977). Institutional ideas are more commonly used to study similarities among organizations and organizational behaviors than as a lens for understanding structural change in organizations (Buchko, 1994; DiMaggio & Powell, 1983; Meyer & Rowan, 1977). However, in some of the works that specify the basic ideas underlying the approach, institutional factors identify both


135

process and property variables, which can usefully extended to discuss change (DiMaggio & Powell, 1983; Zucker, 1977). In these works, institutional theory distinguishes radical from incremental change (Dougherty, 1994; Greenwood & Hinnings, 1996), identifies the contextual conditions under which change occurs (Leblebici, Salancik, Copay, & King, 1991; Oliver, 1991), and identifies the conditions under which ideas are diffused among organizations (Haunschild & Miner, 1997; Strang & Meyer, 1993; Tolbert & Zucker, 1983). Managerial innovation may result, for example, from external normative pressure, a need for legitimacy, or simply exposure. In reality, the process by which organizations adapt to change is likely to be a function of multiple interacting influences including both organizational and environmental factors, and ownership may determine the degree to which internal and external factors are salient. During the extreme economic restructuring that occurred in China largely during the 1990s, the central government began to reduce its control of many firms leading to a decline in the importance of bargaining for resources and a hardening of budget constraints (Kornai, 1986; Naughton, 1992; Walder, 1986). As state control declined, firms were forced to develop new corporate strategies for addressing nearly every aspect of corporate operations. New organizational forms, such as business groups, emerged, but existing firms also changed in response to the environmental pressure (Chacar & Vissa, 2005; Keister, 2000). Managers implemented a host of innovative responses to address needs related to such diverse concerns as corporate structure, capital acquisition, investing, hiring and firing workers, obtaining inputs to production, and marketing. The approach that these firms took varied in important ways by ownership, and these differences may be instructive for understanding the process by which innovation occurs and the relative weight of internal and external influences on this process. In the next sections, we explore how ownership structure affects the relative salience of internal and external influences.

Corporate Ownership and Strategy Ownership structure shapes a great deal about organizational decision making, and during economic transition this influence can be particularly apparent because it affects both willingness and opportunity to change (Guthrie, 1997; Keister, 2000). Innovation during transition depends on both the firm’s desire and internal incentives to alter its behavior and opportunities to change the way business is done. While transition meant

136


change for most firms, many of China’s SOEs continued to receive soft loans and other types of supports from the central government. These firms were usually high priorities for the state because they were in a vital industry (e.g., defense, telecommunications), provided critical jobs, contributed vital resources to the state, or were otherwise deemed critical. Firms that were high priorities did not face the same uncertainties that other firms faced and were either not permitted or did not need to immediately adopt innovations that would prepare them to operate in markets. Consistent with institutional arguments, then, strong central state ownership was the dominant pressure affecting innovative response for these firms. Yet many SOE managers recognized that some innovation was in the long-term interest of the firm. Managers were aware that enterprise closures were likely to become more common with time, and many responded by ascertaining firm strengths and adopting appropriate, but limited, innovations. In particular, in order to appear legitimate in the new markets, they were eager to mimic corporate structures of Western firms, especially when such structures were encouraged by the state owners and firms that have adopted similar structures were regarded as pioneers in the reform. However, many SOE managers were also reluctant to undertake innovations that would bring more fundamental changes in firms’ employment and managerial structures. For SOEs, a reasonable compromise strategy was to adopt innovative corporate structures, particularly those that state officials promoted, while avoiding making dramatic changes to their employment practices or adopting other risky strategies. Conversely, private firms and other firms that were not controlled by the state were often subject to hard budget constraints from inception and were more inclined to be innovative to establish their positions in developing markets. Consistent with strategic choice arguments, CNFs assessed firm capabilities and adopted appropriate innovations. They tended to act more rigorously in experimenting with innovative managerial structures. However, these firms were also more likely to be operating within unique institutional constraints. In particular, developing markets were highly uncertain, and savvy managers may have taken steps to reduce the extent to which they assumed risk by being overly innovative. Compared to SOEs, CNFs were exposed to much less regulative pressure from governments. Managers of these firms were more likely to opt for practices that had been proven efficient, and avoid innovations with less clear benefits in the short run. In contrast to SOEs, CNFs were more likely to adopt innovative employment and managerial strategies that would affect their agility, take advantage of their unique traits, and allow them to compete. At the same time,


137

CNFs had little incentive to adopt innovative corporate structures, and because there was no external state pressure to adopt particular structures, these firms were less likely to use their resources on structural change.

Ownership and Strategic Orientation While understanding how ownership shapes the adoption of particular strategies is important, studying a firm’s overall strategic orientation can also provide insight into the relative importance of internal and external influences on firm’s behavior. Identifying strategic groups – sets of firms with common strategic orientations – is a useful analytic tool for characterizing overall strategic orientations. Western organization theorists have used the notion of strategic groups effectively to understand the nature and performance implications of market structures (Barney & Hoskisson, 1990; Mascarenhas, 1989; McGee & Thomas, 1986; Newman, 1978). Applying this notion to Chinese industry during transition will allow us to explore the effect of ownership structure on the general strategic approach taken by firms. That is, grouping firms based on strategic orientation can highlight variation and provide evidence of its causes. Indeed, it is likely that overall strategic orientation did vary in important ways with ownership structure during China’s transition. During the transformation, SOE managers focused on corporate structure to appease central government oversight organizations, and differences across enterprises centered on the degree to which the firm altered its structure if at all. In contrast, variation in overall corporate structure in CNFs was centered around the movement to markets. More innovative CNFs mimicked their Western counterparts and began to operate in markets. More conservative CNFs adopted strategies and structures that resembled SOEs. Our arguments about the different incentives and constraints faced by SOEs and CNFs suggest that SOE strategic groups varied in the degree to which they adopted innovative corporate structures, if they innovated at all. In contrast, CNF strategic groups are likely to vary in the degree to which they were market oriented versus oriented toward a redistributive system.

Government Ties and Strategy Because ownership can be difficult to specify during the economic transition from planning to markets, it is useful to consider how other indicators of

138


control also shaped firm behavior. In particular, the source of a firm’s funds in a transition economy is an important indicator of authority. In China, prior to reform, resources were controlled and redistributed by the central government as well as government offices in provinces, cities, counties, and even townships (Wong, 1987). Local governments controlled most physical and financial resources and played an important role in shaping the institutional environment in which firms were embedded. Although the development of product markets during reform allowed firms to be less dependent on administrative allocations, local governments remained important because of their continuing control of the allocation of important financial resources (Wong, 1987). During the early stages of reform, local governments received funds from firm profits and taxes as a result of fiscal decentralization. They also influenced the bank lending process via their administrative power over local banking systems. The allocative power of local governments was further expanded as they were granted authority to set profit-retention rates for local enterprises. During transition, their control of local product prices and commercial taxes also increased. The level of local government’s control of investment decisions and resources directly influenced the evolution of local markets. Local governments with smaller industrial bases often had stronger incentive to encourage firms to rely on markets (Walder, 1995). Compared with governments at higher administrative levels, local governments often had clearer financial incentives and constraints. The revenues of local governments were closely linked to firm financial performance, consistent with research on Western firms that suggests that access to critical resources (and historically valuable) shapes performance (Kraatz & Zajac, 2001). Because they had fewer entities to monitor and because there were fewer bureaucratic barriers to information flow, local governments were more aware of firm decision making and were more active in encouraging firms under their supervision to make pro-market decisions. Fewer informational barriers and clear incentives allowed local governments to make faster and more effective decisions concerning the operations of enterprises in their jurisdictions. Governments with larger industrial bases had to ensure that firms provided reliable sources of products, particularly industrial inputs, for other firms in the jurisdiction. These governments thus tended to care more about maintaining a stable supply of inputs and outputs rather than minimizing the marginal cost of production (Walder, 1995). Because local governments played a central role in permitting or even encouraging innovation, when the local government allocated funds to firms under its jurisdiction, they had greater interest in the firm’s financial


139

performance and encouraged innovation. As a result, firms that received funds from the local government were more likely to innovate in all areas. In contrast, firms that obtained the bulk of their funds from the central government often had closer relationships with national-level state agencies, and hence were more likely to suffer from problems related to soft budget constraints. They had little incentive to experiment with markets or to assume the risks associated with market exchange. That is, firms that received most of their funds from the central government would have been less innovative on all dimensions.

Ownership, Size, and Innovation Finally, variations among firms of different size within ownership groups may also provide further insight into the role of ownership in shaping innovation. Western literature provides evidence that firm size influences firm behavior, and size has historically been an important determinant of firm status and behaviors in China as well. Among SOEs, larger enterprises tended to be more protected by the state because they were more critical for national interest and because they often provided more revenues to the government and more jobs. For larger SOEs, the state was thus particularly salient and encouraged a focus on firm structure. For smaller SOEs, external pressure from the central government was less acute, and the level of protection they were able to secure from the government was also lower. As a result, managers in larger SOEs focused on reforming firm structure while those in smaller SOEs were either more likely to adopt market-oriented strategies or more likely to refrain from innovating at all. Similarly, larger SOEs avoided drastic changes to their product lines and investment strategies; smaller SOEs that had adopted market-oriented strategies were more likely to adopt innovative products and innovative investment strategies. In CNFs, we expect that size played a different role in terms of affecting the type of strategies firm managers selected. One of the unanticipated consequences of China’s market liberalization has been the successful emergence of the non-state sector. Rapidly growing non-state enterprises have come to be a crucial force driving China’s reform process as a whole to move forward (Naughton, 1996). CNFs that had expanded in scale during the reform were quicker in responding to changing market demands, and were more willing to take on risks. In order to maintain the level of growth, they also actively sought financial resources most available to them, such as bank loans. For these reasons, it is likely that larger CNFs were more

140


likely to adopt innovative products strategies and to obtain more financial resources. Because they are similar to smaller SOEs, smaller CNFs were either cautious about adopting drastic managerial innovations, or they tended to focus on more conservative approaches such as building up equity or retained earnings. The meaning of retained earnings affected how firms borrowed. Accounting and auditing standards in China were developing and becoming consistent with Western standards (Ji, 2002), and retained earnings were calculated as net profits accumulated in a business, as they are in the West (Lin, Yang, & Wang, 1998).2 For SOEs, retained earnings were equivalent to state funds because they were owned by the state and because they were used to determine tax rates (or remitted profits) in the early stages of reform (Xu, 1998; Yi, 1994). Like appropriated earnings in the West, much of a SOE’s retained earnings in China were earmarked for other purposes and not available for reinvestment. However, for CNFs, retained earnings could be used to cover current expenses, fund reinvestment, and nearly all other purposes for which retained earnings are used in the West.3 The expansion of local financial markets across China in the 1990s enabled firms to diversify their sources of funding to a large extent. However, while many actively sought external financial resources, such as bank loans, interfirm investment and stock market funds, others chose to rely heavily on internal resources. Studies on corporate borrowing strategies in Western firms indicate that the availability of capital is a key determinant of managerial autonomy (Mintz & Schwartz, 1985). When a firm depends on other organizations to provide its capital, the suppliers of capital will exercise power over the firm. Under these circumstances, unless the expected return from borrowing from outside makes it economically efficient to do so, firms will prefer to minimize their dependence on external funding and use retained earnings (Mizruchi & Stearns, 1994). Thus, smaller CNFs were less likely to adopt innovative strategies and more likely to focus on building equity and increasing retained earnings.

RESEARCH METHODS Data We use longitudinal data from 1994 to 1999 on 800 Chinese firms, collected with the assistance of the Chinese Academy of Social Sciences, to study the factors that encourage innovation and to explore how the types of innovation


141

differ by ownership. The sample includes 433 SOEs and 367 collective enterprises, joint ventures, private firms, and other types of non-state companies (CNFs). All firms are independent entities (i.e., there are no subsidiaries included). The questionnaire had two parts. Part one was a set of 92 questions regarding the structure of the enterprise, manager appointment, participation in reform, organizational and management structure, manager incentives, firm strategy, relations with government offices, investment strategies, and related information. The firm’s CEO or their representative in the management hierarchy answered these questions. Part two was a series of 141 quantitative questions regarding enterprise operations from 1994 to 1999. The firm’s accountant answered these questions. We collected information about firm output, production expenses, wages and labor, investments, assets and liabilities, costs, profits, and related traits. Table 1 provides an overview of the firms and their managers. As the table indicates, 26.8% of the firms were located in Jiangsu, 24.5% were located in Sichuan, 25.4% were in Jilin, and 23.3% were in Shanxi. Approximately one-half of the firms reported that they were medium-sized enterprises, with nearly 25% each reporting that they were large or small companies. Nearly half of the firms were under the jurisdiction of a municipal government, and nearly half (46.88%) had been formed within the last 20 years. A large number of the firms were also formed more than 40 years before the survey was conducted. Firm managers in this sample were typically between 50 and 59 years old and generally had a post-secondary technical education. The vast majority of the managers were either the secretary or vice secretary of the local Communist Party branch, although nearly 6% were not Party members.

Dependent Variables and Analysis Strategy We use two sets of analyses to examine our ideas empirically. We perform logistic regression to investigate the relationship between ownership and the adoption of four key innovations during 1994 and 1999. We then use cluster analysis to identify strategic groups separately for SOEs and CNFs in 1999. Ownership and Adoption of Select Innovations Our first models explore firm innovation in corporate structure, employment structure, internal structure, and investment. We model the effect of ownership structure on four dichotomous dependent variables: (1) the adoption of the corporate form, (2) establishing a board of directors,

142


Table 1.

Descriptive Statistics: Firms and Managers, 1999.

Firm Characteristics (%) Ownership State-owned (SOE) Collective-nonpublic (CNF) Collective Stock Joint venture Private Other

54.2 45.8 29.0 11.5 2.6 1.5 1.2

Age Less than 20 years 21–40 years Greater than 40 years

46.9 21.5 31.6

Size Large Medium Small

24.2 52.8 23.0

Level of subordination Central government Province Municipality County Other

6.2 6.8 49.8 7.4 29.8

Province Jiangsu Sichuan Jilin Shanxi

26.8 24.5 25.4 23.3

Age (years) 40–49 50–59 60þ

36.3 57.5 6.2

Tenure (years) 3 or fewer 4–5 6 or more

49.9 35.5 14.6

Education (highest level completed) Elementary or less Upper middle/2nd technical school Post-2nd technical school College degree or more

0.0 7.2 54.0 38.8

Party status Secretary or vice secretary Committee/branch member Ordinary member Not a member

74.0 19.8 0.5 5.7

Manager Characteristics (%)

(3) a marketing department, or (4) a finance department. We pooled 1994– 1999 firm data to create 4,800 firm-years, and we used fixed effects logistic models to generate parameter estimates (Greene, 1993). We focus on adoption of the corporation form because incorporation was an important policy reform promoted by the central government. This policy had two goals. First, incorporation was supposed to improve the management of state assets by restructuring the traditional state ownership system. Second, incorporation was expected to promote growth in the nonstate sectors by introducing a standard format of corporate governance for non-state firms to follow. SOEs tended to comply, while CNFs were reluctant to incorporate because the model did not explicitly protect their interests (Clarke, 2003; Lin, 2001). Despite the effort of the reformers to eliminate the distinction between state and non-state-owned firms by imposing this universal policy, adoption patterns varied in important ways


143

by ownership type. In our sample, 94.5% of SOEs and 51.2% of CNFs had incorporated by 1999. We also explored the use of boards of directors, marketing departments, and finance departments because these are critical elements of the governance structures of firms in both the West (Dalton, Daily, Ellstrand, & Johnson, 1998; Levine, 1997; Seashore & Yuchtman, 1967) and in postreform China (Davies & Walters, 2004; Guthrie, 1999; Peng, 2004; Tam, 2000; Tian & Lau, 2001). Moreover, each of these corporate structures would have been relatively new following reform in China, indicating that the structure was adopted during the transition. There are notable differences in the use of these structures by ownership type in our sample. Only 8% of SOEs established a board of directors, but almost a third of CNFs had formed a board. Among SOEs, 32.3% and 24.2% had marketing and finance departments, respectively. In contrast, the proportions of CNFs with these two departments were 15.8% and 8.2%, respectively. Thus, CNFs tended to establish boards of directors but SOEs were more aggressive in establishing marketing and finance departments. Strategic Groups Among SOEs and CNFs To examine variations in strategic group formation, we use a variation of cluster analysis on the 1999 portion of our strategy data. To adequately address different aspects of firm strategy, we select a diverse set of variables that represent important features of Chinese industry. We include 25 indicators of innovation in structure, finance, labor management, investment, and production. Table 2 presents descriptive statistics for the strategy variables separately for SOEs and CNFs. Because our concern is firm strategy, we define our strategy outcomes at the firm level and initially use cluster analysis to categorize firm-level groups without forcing industry boundaries. Consistent with prior research on strategic groups, we also acknowledge that within-industry variations are important. In subsequent analyses, we take industry into consideration and explore relevant industryspecific variation. Cluster analysis is a generic name for a variety of procedures mainly developed to group objects based on measures of similarity in such a way that the profiles in a particular group are more similar to each other than they are to those of other groups (Aldenderfer & Blashfield, 1984). Because our variables do not have equal variance and we have a relatively large set of variables to analyze, we perform a canonical discriminant analysis of the strategy variables before the clustering procedure. This analysis is a dimension-reduction technique related to principal component analysis and

144


Table 2.

Descriptive Statistics for Strategy Variables, 1999.

Strategy Components

SOE

CNF

Mean

S.D.

Mean

S.D.

Corporate structure Corporation Re-organized under company law Part of enterprise group or holding company Stock system

0.94 0.82 0.66 0.08

0.23 0.45 0.47 0.10

0.51 0.65 0.45 0.01

0.50 0.45 0.49 0.05

Employment structure Had board of directors Temporary employees (% of total employees) Managers (% of total employees) Sales/marketing staff (% of total employees)

0.08 14.49 5.04 2.50

0.27 9.54 2.18 2.48

0.31 84.63 9.86 5.00

0.46 17.85 5.42 5.08

0.32 0.33 0.02 0.01

0.46 0.47 0.13 0.08

0.15 0.15 0.10 0.07

0.36 0.35 0.30 0.26

Investment Had finance company Total investment/total assets Expanding production (% total investment) External to company (% total investment)

0.24 8.80 89.22 6.53

0.43 4.42 15.07 5.52

0.08 5.57 71.86 2.01

0.27 9.31 40.38 5.89

Source of funds (% of total revenues) Retained earnings Bank loans Loans and investment from other firms Central and local governments

86.55 11.91 0.95 0.53

22.98 21.49 5.80 4.57

76.13 11.39 1.51 0.32

36.50 24.11 9.00 3.58

Equity distribution (1 ¼ yes) Had significant change in equity distribution Acquired formerly state-owned assets

0.10 0.07

0.30 0.28

0.15 0.11

0.36 0.32

Products New product output/total output Intellectual property output/total assets

1.75 0.21

3.11 0.56

9.29 1.52

17.59 4.28

Internal structure Had marketing department Had legal department Contract responsibility system Factory manager responsibility system

Notes: Bolded strategies are dependent variables in analyses in Table 3. SOEs are state-owned enterprises; CNFs are collective enterprises and other non-state firms.

canonical correlation. It derives a linear combination of the variables that has the highest possible multiple correlation. The maximum multiple correlation is called the first canonical correlation (Can1). The second canonical correlation (Can2) is obtained by finding the linear combination


145

uncorrelated with the first canonical variable that has the highest possible multiple correlation. We used SAS to execute these analyses, repeating the process of extracting canonical variables until the number of canonical variables equaled the number of original variables. We then implemented the clustering procedures using these canonical correlations. The cluster function is called the within-group sum of squares or the error sum of squares (ESS) and is defined as: ESS ¼

n X 2 ðxi xÞ i¼1

where xi is the variable score of the ith observation in a cluster and x the mean of a variable in the cluster. The procedure starts with having each observation form its own cluster, in which the ESS ¼ 0. The method then works by joining those groups or observations that result in the minimum increase in the ESS. The procedure is repeated until all observations are grouped into a single cluster (Aldenderfer & Blashfield, 1984; SAS 1999). To determine how many clusters to retain, we used Ward’s method, which is designed to minimize the variance within clusters. We selected this method because it effectively identifies ‘‘true’’ underlying groupings in data sets and tends to produce results that are easily interpreted (Aldenderfer & Blashfield, 1984; Blashfield, 1976). We perform both the cluster analyses and the subsequent multinomial analyses on strategic group membership separately for SOEs and CNFs. We supplement these analyses with a multivariate analysis of variance (MANOVA) on the centroids defined by the strategy variables for each strategic cluster. This allows us to determine whether clusters are statistically distinct. We selected the cluster structures for SOEs and CNFs where MANOVA testing pointed to significant differences in the cluster centroids and where subsequent levels of aggregation resulted in nonsignificant differences between cluster means.4 Based on the results we obtained on and previous studies on firm strategic choices, we decided on a four-cluster solution for SOEs and five-cluster solution for CNFs in our sample.

Independent Variables In the logistic regression analysis of adoption of innovative strategies, we measure firm ownership with a dichotomous variable indicating whether or

146


not a firm was owned by the state. We measure ties to different levels of government hierarchy with three variables: a dummy variable indicating whether or not the firm was subordinated to the municipal government and two continuous variables indicating total funds the firm received from central and local governments. We log the two value measures to reduce skew. To control the impact of manager’s traits on firm behavior, we use two dichotomous variables indicating whether the manager (1) was a Party official or (2) graduated from college. We also included managers’ tenure to measure increased willingness to engage in risky, innovative behaviors that may accompany increased tenure and security in the CEO position. We control for retained earnings and profits (both logged) because more financially sound firms might feel ready to experiment with new managerial strategies more quickly than those that are less stable. Because industries developed unevenly and firms in different industries were exposed to different levels of normative and market pressures for innovation, we controlled for industry using dummy variables for the five main industry groups: chemical, textile, building materials, electronics, and machinery. We control for geographic location with dichotomous province indicators for Sichuan, Shanxi, and Jilin (with Jiangsu as the reference category). We control number of employees (logged) and firm’s age as larger and older firms may be more secure about creating more elaborate and complex corporate structures. Age squared controls for the possibility that the effect of age increases or diminishes at larger values. Finally, we represented time with a continuous indicator of year ranging from 1994 to 1999 to capture changes in firm behavior as China’s transition progress.

RESULTS Corporate Ownership Shapes Strategy Our results provide strong support for the argument that SOEs are more likely than CNFs to adopt innovations that were mandated by their state owners. Table 3 presents results from logistic regression analyses of the likelihood that a Chinese firm adopted, and of four representative strategy, innovations between 1994 and 1999. The results show that state ownership significantly increased the likelihood that a firm adopts the corporation form. This is consistent with our argument that SOEs were subject to a higher level of regulative pressure than CNFs and thus were more likely to respond to coercive pressure (coercive isomorphism). In addition, heightening normative

147


Table 3. Odds Ratios from Logistic Regression Analysis of Select Strategy Components on Ownership and Control Variables: 800 Chinese Firms, 1994–1999.

State-owned enterprise (SOE) Government ties Firm subordinated to city Central government funds (log) Local government funds (log) Manager traits Party official College graduate Tenure Control variables Retained earnings (log) Profits (log) Number of employees (log) Firm age Firm age squared Year w2

Corporation

Board of Directors

3.99

0.02

1.90

1.80

1.44 1.23

0.80 0.74

2.04 0.76

1.24 0.90

0.98

1.06

1.09

1.03

1.18 1.38 0.95

1.09 1.10 1.07

1.50 2.21 0.87

1.31 2.46 0.93

1.03 0.01 1.59

0.99 2.22 1.13

0.99 9.01 3.90

1.08 9.01 4.84

0.99 1.00 1.38 1,071.36

1.09 1.00 1.05 758.12

Marketing Department

1.02 1.12 1.32 1,433.993

Finance Department

1.01 1.00 1.25 1,377.93

Notes: Strategies used as dependent variables are representative of the multiple components of overall firm strategy. Analyses use 4,800 observations and 800 firms. Standard errors are available upon request. Also controlled but not displayed are measures of industry (chemical, textile, building materials, electronics, and machinery) and province. po0.05; po0.01; po0.001 (two-tailed tests).

demands from government agencies and other institutes to reform SOEs’ corporate structures also prompted managers to innovate. However, such innovations were often limited. Managers were cautious about possible interventions or conflicts of interests resulting from new structures. As we anticipated, SOEs were significantly less likely to form a board of directors. In addition, state ownership significantly increases the likelihood that a firm establishes finance and marketing departments consistent with our arguments that financial restructuring was one of the major components of the state’s proposals to reform SOEs (Jevons Lee, 2001).

148


Our results also provide strong support for the argument that firms not owned by the state were more likely to adopt structures such as a board of directors. It is likely that managers in CNFs were better able to recognize that an innovative structure can increase resources, legitimacy, and survival capabilities (Scott, 1995), and it is also likely that these managers were more able to act on this knowledge. The lesser likelihood that CNFs adopted the corporate form suggests that managers in non-state firms were less constrained by regulative pressures from governments. It also implies that managers may have perceived state policy encouraging incorporation as detrimental to the firm, or at least not beneficial. In addition to ownership, firms’ ties to different levels of government and managers’ characteristics also affected the type of innovations firms chose. Compared to governments at higher administrative levels, local governments had stronger financial interests in firms under their jurisdictions and were more likely to promote firm structures conducive to the development of local markets. Thus, we proposed that subordination to a municipality and receiving funds from local governments would be positively associated with adoption of innovative structures, whereas receiving funds from the central government would have the opposite effect. Our results yield some interesting findings. Subordination to a municipality and receiving funds from the local government are positively related to innovative structures in general, but municipal subordination has a negative association with the creation of boards of directors, and receiving local government funds has a significant positive effect on the adoption of the same structure. As expected, receiving large amount of funds from the central government has an inverse relationship with market-oriented innovations, but a strong positive relationship with state-initiated structural changes such as corporatization. Finally, it is interesting to note the relationships between some of the control variables and firms’ strategic choices shown in Table 3. Managers closely affiliated with the Party and those with higher levels of education were more likely to promote innovation. High profits were positively related to the adoption of a board of directors, marketing and finance departments. Retained earnings encouraged incorporation and developing a finance department, but retained earnings did not encourage the creation of a board of directors or a marketing department. Firm size (number of employees) and age were positively associated with firm innovation regardless of ownership type. This result is consistent with our finding in the subsequent cluster analysis of the strategic group formation among Chinese firms. The generally positive effect of year indicates an overall pattern of Chinese firms becoming more actively engaged in changing corporate management structure over time.

149


Ownership and Strategic Group Formation in Chinese Firms The strategic groups that emerge in our analyses suggest that there were indeed important differences in the overall strategic orientation of firms by ownership. In our strategic group analysis, four primary groups emerge among SOEs and five among CNFs. Figs. 1 and 2 illustrate these clusters and their relative distances from each other. Tables 4 and 5 report the mean values of the 25 strategy variables we included in the cluster analysis and general descriptive statistics for each of the four SOE and five CNF strategic

Group 2 Expand products, Diversify investment

Group 1 Reform firm structure

Group 3 Reorganize ownership, increase equity

Group 4 No innovation

Fig. 1. Plot of Four SOE Strategic Groups, 1999. Notes: The first two canonical variables calculated are plotted here to show cluster membership. The canonical variables are linear combinations of the strategy variables that summarize betweenclass variations. The first canonical variable (Can1) is the linear combination of the 25 strategy variables that provides the greatest differences (in terms of a univariate Ftest) between the group means. The second canonical variable (Can2) provides the greatest differences between group means while being uncorrelated with the first canonical variable. This plot shows that Can1 best discriminates between (1) cluster 1 and (2) clusters 2, 3, and 4. Can2 discriminates between clusters 1, 2, and 4. Cluster 3 loosely scatters around clusters 2 and 4. The first two canonical variables explain about 68% of the total variance.

150


Group 4 Innovative products, SOE-type structure

Group 3 Firm autonomy, But stalled

Group 1 Build equity

Group 2 Obtain bank loans

Group 5 Increase retained earnings

Fig. 2. Plot of Five CNF Strategic Groups, 1999. Notes: The first two canonical variables calculated are plotted here to show cluster membership of CNFs in our sample. This plot shows that Can1 best discriminates between (1) cluster 3 and (2) clusters 1 and 5, and (3) clusters 2 and 4. Can2 best discriminates between (1) clusters 2, 3, and 4 and (2) clusters 1 and 5. The first two canonical variables explain about 55% of the total variance.

groups we identify. Tables 4 and 5 also summarize the focus of innovation for each of the strategic groups. Among SOEs in our sample (Table 4), the first group (SG1) consists of firms that focus on reforming corporate structures by adopting the corporation form and establishing finance company, legal and marketing departments. A quarter of firms in this group are under direct supervision of the central government, and the vast majority were members of an enterprise group or a holding company. The second group (SG2) includes small firms that focus on expanding product range and diversifying investment options. The third group (SG3) is composed of medium-sized firms that experienced significant changes in equity distribution and actively acquired formerly state-owned assets. This group was also the most likely to adopt a board of directors. Strategic group 4 (SG4) consists of medium-sized firms that

151


Table 4.

Strategic Group Characteristics: SOEs, 1999. Strategic Group SG1

Focus of innovation

Reform firm structure

Number of firms Primary industries

140 Machinery, textiles, chemicals

Province (%) Jiangsu Sichuan Shanxi Jilin Subordination (%) Central ministry Province Municipality County and lower Firm age (mean) Total assets (mean, 1,000 yuan) Sales revenue (mean, 1,000 yuan) Employees (mean, 1,000)

SG2 Expand products, diversify investment 122 Machinery, chemicals, textiles

SG3 Reorganize ownership, increase equity 49 Machinery, chemicals, medical supplies

SG4 No innovation

122 Textiles, machinery, building material

27.14 23.57 23.57 25.71

45.08 17.21 11.48 26.23

51.02 20.41 12.24 16.33

31.97 21.31 20.49 26.23

25.00 13.57 60.00 1.42

4.10 7.38 83.61 4.92

6.12 14.29 77.55 2.04

5.74 15.57 70.49 8.20

41.14 495.84

39.79 97.68

41.79 149.18

38.82 123.31

544.39

103.11

199.96

157.92

4.57

0.86

1.32

1.72

0.02

0.00

0.00

0.06

0.00

0.00

0.01 0.91 0.01 0.58

0.00 0.86 0.02 1.00

0.00 1.00 0.00 0.72

0.00

0.59

0.00

Structural innovations (1 ¼ yes) 0.01 Factory manager responsibility system 0.01 Contract responsibility system Leasing 0.01 Corporation 0.95 Stock system 0.02 0.75 Re-organized under company law Equity distribution (1 ¼ yes) 0.11 Significant change in equity distribution

152


Table 4. (Continued ) Strategic Group SG1 Acquired formerly state-owned assets

0.06

SG3

SG4

0.00

0.36

0.01

0.01

0.52

1.00

0.01

0.42

0.01

0.01

0.01

0.01

0.02

0.08

0.02

0.01

0.04

0.01

16.04

13.71

12.93

4.21

4.55

4.38

10.21

0.99

9.53

3.16

4.99

4.33

4.14

89.66

88.73

91.19

88.66

1.56

1.96

0.71

0.93

0.01

1.98

0.02

10.90 0.77

6.56 2.87

12.59 0.61

Organizational structure (1 ¼ yes) 0.97 Part of enterprise group or holding company Had board of 0.10 directors Had finance 0.76 company Had legal 0.99 department Had marketing 0.99 department Workforce (% of total employees) Temporary 14.94 employees Sales and 4.38 marketing personnel Managers 9.40 Investment Total investment/ total assets Expanding production (% total investment) External to company (% total investment)

SG2

Source of funds (% of total revenues) 1.00 Central and local governments Bank loans 12.18 1.80 Loans and investment from other firms

153


Table 4. (Continued ) Strategic Group

Retained earnings Products Intellectual properties output/total assets New product output/total output

SG1

SG2

SG3

SG4

77.81

85.52

86.67

82.55

0.23

0.18

0.11

0.21

1.30

2.29

1.77

1.55

had no innovation on virtually all of the strategic dimensions. These groupings demonstrate that there were very clear groupings by size and choice of innovation among the SOEs. If SOEs adopted any innovations, they focused on reforming corporate structures, and they only varied in the degree to which they innovated. Large firms were most advanced in terms of adopting innovative firm structures, and medium-sized firms focused on increasing equity by acquiring formerly state-owned assets or refrained from innovating at all. An interesting finding is that small SOEs relied more on marketing through expanding product lines and investment. This suggests that perhaps these firms felt little pressure to adopt certain innovations but also had the flexibility to experiment with a potentially highyield changes. Consistent with our arguments, the strategic focus of CNFs was considerably different than SOEs, and size was an important factor underlying these differences. Table 5 summarizes our findings for the CNF strategic groups. SG1, SG3, and SG5 are all small firms and are found to have devoted little efforts in reforming firm structures, although SG1 does have a significant proportion of member firms (96.3%) that adopted boards of directors. In situations where small CNFs do choose to innovate, they tend to either focus on building equity or on increasing retained earnings. Results from Table 5 show that SG1 and SG5 are the two groups of small most firms engaged in reforming. Firms in SG1 had significant changes in their equity distribution, whereas firms in SG5 demonstrated strong interest in increasing retained earnings. SG3 is composed of firms that had actively adopted early reform strategies such as the factory manager responsibility system, but refrained from innovating in more recent years. SG2 and SG4

154


Table 5.

Strategic Group Characteristics: CNFs, 1999. Strategy Group SG1

SG2

SG3

Focus of innovation

Build equity

Obtain bank loans

Firm autonomy but stalled

Number of firms Primary industries (most common)

85 Building materials, medical supplies, chemicals

30 Machinery, coking, building materials

73 Paper, synthetic fibers, metal smelting

Province (%) Jiangsu Sichuan Shanxi Jilin Ownership type (%) Collective Stock Private Joint ventures and other Firm age (mean) Total assets (mean, 1,000 yuan) Sales revenue (mean, 1,000 yuan) Employees (mean, 1,000)

SG4 Innovative products, SOE-type structure 23 Machinery, electronics, building materials

SG5 Increase retained earnings 156 Textiles, building materials, electronics

7.06 5.88 42.35 44.71

13.33 30.00 26.67 30.00

20.55 35.62 24.66 19.18

13.04 13.04 13.04 60.87

18.59 40.38 28.21 12.82

2.35 74.11 7.06 16.47

90.00 0.00 6.67 3.33

90.41 2.74 1.37 5.48

4.35 78.26 0.0 17.39

87.18 7.69 1.28 3.85

11.71 87.78

13.90 435.31

7.86 38.52

17.65 513.37

11.41 100.46

52.80

182.18

14.98

203.81

38.68

0.79

2.90

0.39

2.55

1.10

0.02

0.38

0.02

0.01

0.01

0.47

0.02

0.02

0.01 0.99 0.00 0.64

0.14 0.01 0.02 0.08

0.01 0.05 0.04 1.00

0.01 0.98 0.02 0.83

0.01

0.01

0.66

0.03

Structural innovations (1 ¼ yes) Factory manager 0.01 response system 0.01 Contract responsibility system Leasing 0.02 Corporation 0.02 Stock system 0.01 Re-organized under 0.96 company law Equity distribution (1 ¼ yes) 0.44 Significant change in equity distribution

155


Table 5. (Continued ) Strategy Group SG1 Acquired formerly state-owned assets

0.03

Organizational structure (1 ¼ yes) 0.45 Part of enterprise group or holding company Had board of 0.96 directors Had finance 0.02 company Had legal 0.01 department Had marketing 0.01 department Workforce (% of total employees) Temporary 84.27 employees 4.47 Sales and marketing personnel Managers 8.93 Investment Total investment/ 6.86 total assets 66.14 Expanding production (% total investment) 3.49 External to company (% total investment) Source of funds (% of total revenues) Central and local 0.58 governments Bank loans 9.93 4.32 Loans and investment from other firms Retained earnings 58.59

SG2

SG3

SG4

SG5

0.53

0.06

0.35

0.07

0.96

0.02

0.95

0.47

0.01

0.01

1.00

0.05

0.49

0.03

0.37

0.02

1.00

0.01

1.00

0.01

1.00

0.01

1.00

0.01

88.13

86.98

72.63

87.28

3.54

6.17

7.72

4.51

9.62

10.00

10.95

10.30

3.18

6.07

2.33

5.30

72.40

65.38

83.47

76.12

2.63

1.25

0.28

1.70

0.01

0.01

0.01

0.30

22.17 0.22

3.64 3.67

7.11 0.62

10.28 2.45

55.38

66.68

32.17

68.60

156


Table 5. (Continued ) Strategy Group SG1 Products Intellectual properties output/total assets New product output/total output

SG2

SG3

SG4

SG5

2.77

1.75

1.16

4.18

0.72

8.80

15.10

7.67

11.67

9.13

include CNFs of larger size. Both groups of firms reformed their corporate structures, except that all SG4 firms adopted boards of directors, but none from SG2. Furthermore, firms in SG2 seem to emphasize obtaining bank loans as a major source of funds, whereas those in SG4 seem to be more R&D intensive and focused on developing more innovative products. Overall, the types of innovations CNFs adopt seem to diverge along the dimensions of either market oriented (develop new products, obtain bank loans, or increase retained earnings) or redistributive system oriented (follow SOE type of structure or refrained from innovating).

DISCUSSION AND CONCLUSION The organizational structure and strategy choices of Chinese firms were both complex and rapidly changing during the country’s economic transition. This background provides an ideal situation for evaluating theories of organizational structure and change in development. This complex and fluid situation also has great implications for both the future of China and, indeed, for the world economy as a whole. Two broad themes emerge from our analysis of the patterns of adoption of four key organizational innovations, one consistent with institutional analysis and one consistent with strategic choice analysis. Institutional analysis suggests that organizational structure and strategy will reflect pressures from internal and external actors. Patterns of adoption of key organizational innovations give clear testament to such forces. SOEs pervasively adopted corporatization in response to pressures from the central government. Also, again in response to government demands to


157

acquire capital on their own, they initiated finance departments. These patterns are common both to SOEs and also to firms receiving significant funds from the central government. Firms subordinate to city governments evidence a similar pattern. The latter firms also tended to develop marketing departments, partly in response to city demands to increase sales and revenues and to serve local markets. Interestingly, this pattern of corporatization and selective organizational reform for firms with strong external pressures is repeated for firms with powerful and entrenched management. That is, firms with managers who are party members or officials and who are more educated evidence this same pattern of corporatization and selective reform. Thus, pressure from either strong internal or external actors results in selective innovation focusing on reforming the internal structure of the corporation. What is missing from this pattern characteristic of strong internal and external actors is any attempt to develop internal boards of directors. It appears that strong actors feel no need to add other strong competing actors to the stage. Instead, they desire to make reforms that will encourage firm’s success and security, but which also preserve power for themselves. This pattern can thus be characterized as selective reform of corporate structure, but limiting corporate governance to exclude new actors. Clearly, institutional forces and logics have had a strong hand in molding this pattern of innovation and change. The second broad theme that emerges from our analysis involves the very large effects of profits on the adoption of reforms. Here, the pattern is significantly different from the pattern for firms under the influence of strong internal or external actors. High profits are negatively associated with corporatization but positively associated with the adoption of marketing and finance departments and the creation of boards of directors. Thus, it appears that the availability of significant new sources of money from profits produces organizational innovation, even without directives from government actors or powerful internal leaders. A second facet of this pattern is the role of firm age, management tenure, and organizational size in promoting organizational innovations that include boards of directors as well as finance and marketing departments. It appears that the most innovative firms are some of the larger and older CNFs, which have long confronted the problems of lack of adequate government budgetary support and which have developed alternative, more locally based, networks for meeting their financial and marketing needs. CNFs of significant size and age are most generally either municipal cooperatives or former SOEs that reorganized as CNFs under company law reforms. Such enterprises have either collective or joint ownership patterns

158


that may include private individuals, and also frequently include government units at the local or municipal level. Such mature enterprises represent an important and unique hybrid type of organization that is simultaneously public and private but that is not under the direct control of central governmental agencies. Such enterprises are some of the most innovative in the Chinese economy and are setting trends that will have a significant influence on determining the uniquely Chinese path to a market economy. This second pattern of high profits, in combination with mature leadership, principally among CNF’s, suggests corporate actors pursuing patterns of innovation based on the logic of strategic choice. The strategic groups emerging from the cluster analysis reinforce the image of SOEs being dominated by institutional forces and CNFs being guided by strategic choice. Among SOEs, reform of firm structure appears to be a guiding principal, with several different paths taken toward this goal, including the path of no action at all. Some attention to expanding products and diversifying investments is also evidenced, but these have been goals well articulated by central authorities and pressed on firms as agendas they should adopt. Among CNFs, three of the five clusters focus on financial concerns, as is typical of market-oriented actors. These concerns include building equity, securing bank loans, and increasing retained earnings – all key strategies for expansion and development in market-based economies. Such strategies are crucial for allowing autonomous development outside the planned directives of the central government and are a clear and direct response to the marketization of significant parts of the Chinese economy. Again, such actions are well conceptualized in terms of organizational actors following the logic of strategic choice. What future developments in the Chinese economy are suggested by these emerging patterns? It appears that large, established CNFs are leading the way toward reform and organizational innovation. Many of these firms, with roots as collective or SOEs, are transforming themselves into public–private hybrid organizations that give a unique character to the Chinese economy. Such firms are both profitable and innovative and provide an important impetus to change. At this point in the development of the Chinese economy, it appears that the development, growth, profitability, and innovative strength of such enterprises exceed both that of smaller and newer CNFs as well as the more constrained and conservative SOEs. Perhaps, the most basic lesson for China, however, is that there is considerable innovation occurring among all types of firms. We know that innovation encourages market development, facilitates the efficient allocation of resources, promotes


159

privatization, and prevents capital flight (Demirgucs-Kunt & Levine, 1996; Mizruchi & Stearns, 1994; Ratcliff, 1980). If these hold true in China, and there is reason to expect they will, the behaviors we study here are likely to translate into positive collective outcomes. The ongoing and highly dynamic development of the Chinese economy suggests significant research challenges for students of economic organization. Strategy choices and strategic groups can be expected to change over time and to remain highly fluid for at least the next several decades. This suggests the need to develop both longitudinal data and longitudinal taxonomies of strategic groups and how these develop, fuse, and divide over time. Multiple pathways are possible moving forward from the existing organization of the Chinese economy, and it is essential for researchers to engage in the difficult tasks of data collection and conceptual development necessary for understanding these pathways. Such conceptual work will need to consider both institutional factors and the logic of strategic choice. Application of these theories to the unique and highly dynamic Chinese economy may provide opportunities not only for better understanding China, but also for development and refinement of theories of organizational analysis.

NOTES 1. Others have considered different varieties of joint ventures in greater depth (Merchant and Schendel, 2000). However, our data do not allow us to consider the role that joint venture structure plays in this process. 2. While Chinese definitions of accounting terms are similar to Western definitions, there are still differences. The terms we use are rough equivalents. In China, retained earnings determined remitted profits, the proportion of earnings that firms gave to the state to be redistributed. Prior to reform, state mandates determined internal costs, depreciation, and the sale of fixed assets, and related costs. As Western accounting practices diffused and financial markets developed, these calculations became market driven (Ji, 2002). 3. There was some incentive for managers to hide retained earnings, but in the early stages of reform, administrative control was strict enough that most funds were accounted for (Keister, 1998). 4. We omit the MANOVA results to save space, but they are available on request.

ACKNOWLEDGMENTS We are grateful for extensive research assistance from Jin Lu. Grants from the National Science Foundation (SBR-9633121), the Luce Foundation, the

160


National Bureau of Asian Research, and the Mershon Center at the Ohio State University supported this research.

REFERENCES Aldenderfer, M. S., & Blashfield, R. K. (1984). Cluster analysis (Vol. 07–044). Beverly Hills: Sage Publications. Barnett, W. P. (1993). Strategic deterrence among multipoint competitors. Industrial and Corporate Change, 2, 249–278. Barney, J. B. (1986). Strategic factor markets: Expectations, luck, and business strategy. Management Science, 42, 1231–1241. Barney, J. B. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17, 121–154. Barney, J. B., & Hoskisson, R. E. (1990). Strategic groups: Untested assertions and research proposals. Managerial and Decision Economics, 11, 187–198. Baysinger, B., & Hoskisson, R. E. (1990). The composition of boards of directors and strategic control: Effects on corporate strategy. Academy of Management Review, 15, 72–87. Baysinger, B. D., Kosnik, R. D., & Turk, T. A. (1991). Effects of board and ownership structure on corporate R&D strategy. Academy of Management Journal, 34, 205–214. Blashfield, R. K. (1976). Mixture model test of cluster analysis: Accuracy of four agglomerative hierarchy methods. Psychological Bulletin, 83, 377–388. Buchko, A. A. (1994). Barriers to strategic transformation. In: P. Shrivastava, A. Huff & J. Dutton (Eds), Advances in strategic management (Vol. 10, pp. 81–106). Greenwich, CT: JAI Press. Chacar, A., & Vissa, B. (2005). Are emerging economies less efficient? Performance persistence and the impact of business group affiliation. Strategic Management Journal, 26, 933–946. Clarke, D. C. (2003). Corporate governance in China: An overview. China Economic Review, 14, 494–507. Dalton, D. R., Daily, C. M., Ellstrand, A. E., & Johnson, J. L. (1998). Meta-analytic reviews of board composition, leadership structure, and financial performance. Strategic Management Journal, 19, 269–290. D’Aveni, R. A. (1994). Hypercompetition: Managing the dynamics of strategic maneuvering. New York: Free Press. Davies, H., & Walters, P. (2004). Emergent patterns of strategy, environment and performance in a transition economy. Strategic Management Journal, 25, 347–364. Demirgucs-Kunt, A., & Levine, R. (1996). Stock market development and financial intermediaries: Stylized facts. World Bank Economic Review, 10, 291–322. Dess, G. G., & Beard, D. W. (1984). Dimensions of organizational task environments. Administrative Science Quarterly, 29, 52–73. DiMaggio, P. J., & Powell, W. W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48, 147–160. Dougherty, D. (1994). Commentary. In: P. Shrivastava, A. Huff & J. Dutton (Eds), Advances in strategic management (Vol. 10, pp. 107–112). Greenwich, CT: JAI Press.


161

Dutton, J. E., & Duncan, R. B. (1987). The creation of momentum for change through the process of strategic issue diagnosis. Strategic Management Journal, 8, 279–295. Ginsberg, A. (1988). Measuring and modelling changes in strategy: Theoretical foundations and empirical directions. Strategic Management Journal, 9, 559–575. Greene, W. (1993). Econometric analysis. New York: Macmillan. Greenwood, R., & Hinnings, C. R. (1996). Understanding radical organizational change: Bringing together the old and the new institutionalism. Academy of Management Review, 21, 1022–1054. Guthrie, D. (1997). Between markets and politics: Organizational responses to reform in China. American Journal of Sociology, 102, 1258–1304. Guthrie, D. (1999). Dragon in a three-piece suit: The emergence of capitalism in China. Princeton, NJ: Princeton University Press. Haunschild, P. R., & Miner, A. S. (1997). Modes of interorganizational imitation: The effects of outcome salience and uncertainty. Administrative Science Quarterly, 42, 472–500. Hill, C. W. L., & Snell, S. A. (1989). Effects of ownership structure and control on corporate productivity. Academy of Management Journal, 32, 25–46. Ilinitch, A. Y., Lewin, A. Y., & D’Aveni, R. A. (1998). Managing in times of disorder: Hypercompetitive organizational responses. Thousand Oaks, CA: Sage Publishers. Jevons Lee, C. W. (2001). Financial restructuring of state owned enterprises in China: The case of Shanghai Sunve Pharmaceutical Corporation. Accounting, Organizations and Society, 26, 673–689. Ji, X. (2002). Development of accounting and auditing systems in China. Brookfield, VT: Ashgate. Keats, B. W., & Hitt, M. A. (1988). A causal model of linkages among environmental dimensions, macro organizational characteristics, and performance. Academy of Management Journal, 31, 570–598. Keister, L. A. (1998). Engineering growth: Business group structure and firm performance in China’s transition economy. American Journal of Sociology, 104, 404–440. Keister, L. A. (2000). Chinese business groups: The structure and impact of interfirm relations during economic development. New York: Oxford University Press. Kornai, J. (1986). The soft budget constraint. Kyklos, 39, 3–30. Kraatz, M. S., & Zajac, E. J. (2001). How organizational resources affect strategic change and performance in turbulent environments: Theory and evidence. Organization Science, 12, 632–657. Leblebici, H., Salancik, G. R., Copay, A., & King, T. (1991). Institutional change and the transformation of interorganizational fields: An organizational history of the U.S. radio broadcasting industry. Administrative Science Quarterly, 36, 333–363. Levine, R. (1997). Financial development and economic growth: Views and agenda. Journal of Economic Literature, 35, 688–726. Lin, C. (2001). Corporatisation and corporate governance in China’s economic transition. Economics of Planning, 34, 5–34. Lin, Z. J., Yang, D. C., & Wang, L. (1998). Accounting and auditing in China. Brookfield, VT: Ashgate. Mascarenhas, B. (1989). Strategic group dynamics. Academy of Management Journal, 32, 333–352. McEvily, B., & Marcus, A. (2005). Embedded ties and the acquisition of competitive capabilities. Strategic Management Journal, 26(11), 1033–1055. McGee, J., & Thomas, H. (1986). Strategic groups: Theory, research, and taxonomy. Strategic Management Journal, 7, 141–160.

162


Merchant, H., & Schendel, D. (2000). How do international joint ventures create shareholder value? Strategic Management Journal, 21(7), 723–737. Meyer, J. W., & Rowan, B. (1977). Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology, 83, 340–363. Mintz, B., & Schwartz, M. (1985). The power structure of American business. Chicago, IL: University of Chicago Press. Mizruchi, M. S., & Stearns, L. B. (1994). Money, banking, and financial markets. In: N. J. Smelser & R. Swedberg (Eds), The handbook of economic sociology (pp. 331–341). Princeton, NJ: Princeton University Press. Naughton, B. (1992). Hierarchy and the bargaining economy: Government and enterprise in the reform process. In: K. G. Lieberthal & D. M. Lampton (Eds), Bureaucracy, politics, and decision making in Post-Mao China (pp. 245–279). Berkeley, CA: University of California Press. Naughton, B. (1996). Growing out of the plan: Chinese economic reform, 1978–1993. New York: Cambridge University Press. Nelson, R. R., & Winter, S. G. (1982). An evolutionary theory of economic change. Cambridge, MA: Harvard. Newman, H. H. (1978). Strategic groups and the structure-performance relationship. Review of Economics and Statistics, 60, 417–427. Oliver, C. (1991). Strategic responses to institutional processes. Academy of Management Review, 16, 145–179. Peng, M. W. (2004). Outside directors and firm performance during institutional transitions. Strategic Management Journal, 25, 453–471. Penrose, E. (1959). A theory of the growth of the firm. New York: Wiley. Pfeffer, J., & Salancik, G. R. (1978). The external control of organizations: A resource dependence perspective. New York: Harper and Row. Ratcliff, R. E. (1980). Banks and corporate lending: An analysis of the impact of the internal structure of the capitalist class on the lending behavior of banks. American Sociological Review, 45, 553–570. Richardson, J. (1996). Vertical integration and rapid response in fashion apparel. Organization Science, 7, 400–412. Scott, W. R. (1995). Institutions and organizations. Thousand Oaks, CA: SAGE. Seashore, S. E., & Yuchtman, E. (1967). Factorial analysis of organizational performance. Administrative Science Quarterly, 12, 377–395. Strang, D., & Meyer, J. W. (1993). Institutional conditions for diffusion. Theory and Society, 22, 487–512. Tam, O. K. (2000). Models of corporate governance for Chinese companies. Corporate Governance: An International Review, 8, 52–64. Tian, J. J., & Lau, C.-M. (2001). Board composition, leadership structure and performance in Chinese shareholding companies. Asia Pacific Journal of Management, 18, 245–263. Tichy, N. M. (1983). Managing strategic change. New York: Wiley. Tolbert, P. S., & Zucker, L. G. (1983). Institutional sources of change in the formal structure of organizations: The diffusion of civil service reform, 1880–1935. Administrative Science Quarterly, 28, 22–39. Volberda, H. W. (1996). Toward the flexible form: How to remain vital in hypercompetitive environments. Organization Science, 7, 359–374.


163

Walder, A. G. (1986). The informal dimension of enterprise financial reforms. China’s Economy Looks Toward the Year 2000, 1, 630–645. Walder, A. G. (1995). Local governments as industrial corporations: An organizational analysis of China’s transitional economy. American Journal of Sociology, 101, 263–301. Wong, C. (1987). Between plan and market: The role of the local sector in Post-Mao China. Journal of Comparative Economics, 11, 385–398. Xu, X. (1998). China’s financial system under transition. New York: St. Martin’s Press. Yi, G. (1994). Money, banking, and financial market emergence in China. New York: Westview. Zucker, L. G. (1977). The role of institutionalization in cultural persistence. American Sociological Review, 42, 726–743.

LOCALIZATION IN THE AGE OF GLOBALIZATION: INSTITUTIONAL DUALITY AND LABOR MANAGEMENT STRUCTURES IN CHINA’S FOREIGN-INVESTED ENTERPRISES Yang Cao and Wei Zhao ABSTRACT We examine, from an institutional perspective, labor management structures in China’s foreign-invested enterprises (FIEs). Focusing on the adoption of two Chinese-style and two Western-style labor structures, we seek to understand the influences of global capitalism vis-a`-vis those of China’s domestic institutions which themselves are undergoing marketoriented transformations. Our empirical analysis is based on data collected in a national survey of FIEs conducted in 1996. The results show that a significant proportion of FIEs localize their labor practices by adopting Chinese-style structures and that the extent of localization is shaped by both institutional processes and strategic considerations. We also find that the two Chinese-style structures effectively reduce tension


165

166

YANG CAO AND WEI ZHAO

and conflict between labor and management, whereas the two Westernstyle structures do not appear to serve such function. These findings underscore the potency of the Chinese institutional environment and reveal how organizational interest interacts with institutional forces, giving rise to a distinctively ‘‘local’’ mixture of managerial structures and practices in the era of globalization.

INTRODUCTION The closing chapter of the twentieth century had featured two prominent themes: the worldwide collapse of state socialism and the accelerating trend of economic globalization. While many countries continue to experience the profound impacts of these changes, few have been involved in both as deeply and directly as China. Since the inception of its economic reform in 1978, China has not only made significant progress in transitioning toward a market economy but has also become one of the largest recipients of foreign direct investment (FDI) and a vital component of the new global economic system. Together, the transition toward markets and the penetration of global capitalism have led to fundamental changes in the Chinese society. The dual emphases on marketization and global integration have been explicitly recognized by the Chinese government’s ‘‘reform’’ and ‘‘opendoor’’ policies. However, existing sociological literature tends to focus on the former, particularly on how marketization interacts with other societal forces to reshape stratification order (Nee, 1989, 1996; Walder, 2002; Zhou, 2000) and firm behavior (Guthrie, 1997; Keister, 1998, 2001; Zhou, Zhao, Li, & Cai, 2003). With few exceptions (Guthrie, 1999; Lee, 1995), little attention has been paid to the influx of global capital and the fast growing sector of foreign-invested enterprises (FIEs hereafter). This lopsided focus is problematic because foreign investment has become such a crucial driving force of China’s postsocialist transition. From 1979 to 2005, China had undertaken over half a million foreign investment projects and utilized a total of $622.43 billion of foreign capital (Chinese Statistical Yearbook, 1996, 2006). In 2004, FIEs employed 17.6 million workers and accounted for 31% of China’s industrial output and 57% of its total export (Chinese Commerce Yearbook, 2005). As the impact of foreign investment reaches far beyond the economic domain (Gallagher, 2005), the lack of sociological research on this topic becomes especially troublesome.

Localization in the Age of Globalization

167

In this paper, we take a step to fill in the gap by examining Chinese FIEs’ labor practice, a crucial aspect of their organizational operation. We focus specifically on the adoption of four labor management structures: the trade union, the workers’ congress, the suggestion system, and formal grievance procedures. The first two structures are indigenous to China and have been widely adopted by domestic enterprises, while the latter two are elements of modern human resource management commonly found in advanced market economies. Drawing on the neoinstitutional theory of organizations (Meyer & Rowan, 1977; DiMaggio & Powell, 1983) and the recently developed translation model (Czarniawska & Joerges, 1996; Zilber, 2002), we underscore the dual institutional pressures that Chinese FIEs face and formulate a series of hypotheses to explain both the differential adoptions and the differential impacts of these labor management structures. In contrast to previous research that highlights the diffusion of western practices in Chinese organizations (Guthrie, 1999), we point to a parallel process where foreign investors and FIEs are compelled to adapt to China’s distinctive institutional environment by incorporating Chinese-style labor management structures. As will be shown later on, the adoption of local structures is not just a matter of formality but has important implications for labor–management relations. The empirical evidence is drawn from a national survey of Chinese FIEs conducted in 1996, eighteen years after China embarked on its economic reform. Throughout the 1980s, FDI in China had grown steadily, although its annual volume remained only about 1% of the GDP. Since Deng Xiaoping’s famous southern tour speech in January 1992, there had been a dramatic increase, and FDI exceeded 6% of the GDP in 1994 before settling around 5% in the next a few years (Gallagher, 2005, p. 21). In light of this historical trajectory, our analysis of data from the 1996 survey captures the initial phase of the explosive growth and will hopefully provide a baseline for understanding subsequent change. To the best of our knowledge, this is the first sociological study that focuses on Chinese FIEs using nationally representative data from a large-scale organizational survey.

INSTITUTIONAL BACKGROUND AND LABOR MANAGEMENT STRUCTURES IN CHINESE FIES Labor relations in China have changed gradually yet decisively during the reform era. Prior to 1978, labor relations as a core component of the socialist social contract were characterized by strong state control over the workplace, lifetime employment, and a cradle-to-grave welfare system

168


(Tang & Parish, 1999). As the reform progressed, it became clear that such features were incongruent with the state’s agenda to institute a market economy. Consequently, the Chinese government implemented a series of reform policies to decentralize labor governance by promoting managerial authority, contract-based employment, and open labor markets. One milestone in the labor reform was the promulgation of China’s first unified Labor Law, which took effect on January 1, 1995. This legislation was particularly important in establishing the legal foundation for marketoriented labor management. As the government withdrew direct control and relied increasingly on legal means for labor regulation (Lee, 2002), ‘‘Chinese labor practices have shifted overwhelmingly toward favoring firm autonomy, flexibility, and managerial control of worker organizations’’ (Gallagher, 2005, p. 63). Meanwhile, China’s labor reform policies also inherited a considerable pro-worker socialist legacy. For example, the 1995 Labor Law clearly stated that it was ‘‘enacted in order to protect the legal interests of labor, regulate labor relations, establish and maintain a labor system compatible with a social[ist] market economy’’ (Warner, 1995, p. 179) and a number of its articles were indeed aimed at establishing workers’ rights and preventing labor abuses. The decentralization of labor governance during the 1980–1990s had coincided with the influx of FDI, creating interesting dynamics in how firms coped with practical challenges with their newly acquired freedom. Guthrie (1999) argues that the drastic withdrawal of state control and guidance and the turbulent market conditions create profound uncertainties which prompt Chinese firms to mimic ‘‘actors that they view as being the most market-savvy, namely, foreign investors’’ (p. 3). This leads to a process of rationalization where Western-style market strategies and management practices become increasingly prevalent in Chinese enterprises. In supporting this argument, Guthrie (1999, p. 54) shows that the degree of economic uncertainty and having joint ventures with foreign companies are both significant predictors of the presence of formal structures such as written organizational rules, hiring procedures, and grievance procedures. Whereas Guthrie’s work provides valuable insights into the diffusion of Western practices among China’s domestic enterprises, how foreign investors and FIEs structure their own labor management remains unclear. Anecdotal accounts suggest a wide array of possibilities ranging from sweatshop operations to much more sophisticated, commitment-oriented work systems, and Lee’s (1995) comparative study draws a sharp contrast between the labor regime in a Hong Kong-invested plant in southern China and that in its headquarters in Hong Kong. This means that labor practices


169

in FIEs are neither uniform across all firms nor are they a simple replica of those in the investors’ home countries. Moreover, although many modern management practices indeed originated from Western market societies, China has also developed several of its own for the same purpose of securing workers’ consent and alleviating the tension and conflicts between labor and management. These Chinese-style structures certainly also deserve our research attention. In this paper, we examine four representative labor management structures to understand how FIEs adapt to China’s distinct institutional environment during its market transition. Two of these structures were developed in China, whereas the other two are common in advanced capitalist economies. The two foreign practices are widely regarded as global ‘‘best practices’’ in human resource management which have also been introduced to China in recent years (Baron & Kreps, 1999; Becker & Gerhart, 1996; Guthrie, 2001; Huselid, 1995; Xiao & Bjo¨rkman, 2006).

Trade Union As a classic feature of socialist organizations, trade unions in China are fundamentally different from labor unions in Western democracies (Clarke, 2005, p. 4). Instead of acting as independent advocacies for the workers, all trade unions in China are affiliates of the All-China Federation of Trade Unions (ACFTU), a national organization of quasi-government stature. They are traditionally described as the ‘‘transmission belt’’ and are expected to help the communist party to better reach out to the workers. Most union officials hold managerial positions and routinely use the union organization to perform managerial functions. In FIEs, trade unions often play key roles in facilitating communication and cooperation among foreign investors, their Chinese partners, and the workers. Sensitized by the widespread labor strife and resurging nationalist ideologies, trade unions now also serve as an important channel for the state to exert influence over labor practices and to mitigate abuses in FIEs (Goodall & Warner, 1997).

Workers’ Congress The workers’ congress is a Chinese invention and has no equivalent in Western capitalist societies. It was reintroduced by the government in 1981 as an additional mechanism to counterbalance the rising managerial

170


authority (Levine, 1997, pp. 51–53; also see Guthrie, 1999). Members of this congress are officially elected by the workers and meet at least once a year to review and approve management plans and policies (Brown, 2006; Goodall & Warner, 1997, p. 584). With the trade union functioning as its executive body in day-to-day operation, the workers’ congress sometimes is able to operate as a somewhat more independent participative organ to help enforce labor-related regulations at the workplace (Warner, 1995, pp. 31, 34).

Suggestion System The suggestion system is part of the modern human resource management (Baron & Kreps, 1999; Becker & Gerhart, 1996). It aims to actively involve a broad range of employees by soliciting their inputs in order to promote both technical innovations and administrative improvements. Generally speaking, the suggestion system entails a formal communication channel, such as a ‘‘suggestion box’’ or some more evolved form, for the employees to submit suggestions. In many cases it further entails specialized personnel as well as formal criteria for suggestion evaluation and how meritorious suggestions are to be rewarded. By encouraging employees’ participation, the suggestion system is also expected to improve employee satisfaction and organizational culture.

Grievance Procedures Grievance procedures are formal structures that specify how employees can file complaints against the management and how these complaints should be resolved. These procedures are intended to address grievances before they turn into more disruptive and costly disputes or litigations. Such procedures have been widely adopted by organizations in the United States and other developed countries (Edelman, 1990; Sutton & Dobbin, 1996). In China, the emergence of grievance procedures during the 1990s represents a significant departure from the traditional socialist labor regime (Guthrie, 1999, p. 47) where worker–management conflicts were sugarcoated in the ideological domain and resolved by informal dialogues and mediations at the workplace, often through the trade unions. It is worth noting that although the trade union and the workers’ congress are often perceived as ineffective and lacking independence in domestic enterprises, the role they play in FIEs in protecting workers’ rights is


171

considerably more salient, due in part to the growing support of the state. Meanwhile, the suggestion system and formal grievance procedures are not only institutionalized in Western economies but have also been diffusing globally as ‘‘best practices’’ and as key components of the high-performance human resource management system (Bae et al., 2003; Guthrie, 2001; Huselid, 1995). Together, these four structures provide an informative set of indicators for gauging the influences of global capitalism vis-a`-vis China’s domestic institutions, which themselves are undergoing market-oriented transformations.

THEORETICAL ISSUES AND HYPOTHESES According to the neoinstitutional perspective, organizations are embedded in an institutional environment composed of ‘‘cognitive, normative, and regulative structures and activities’’ (Scott, 1995, p. xiii). The institutional environment exerts profound impacts on organizational structures, most notably through isomorphic processes where organizations incorporate taken-for-granted ‘‘institutional myths’’ to claim legitimacy (DiMaggio & Powell, 1983; Meyer & Rowan, 1977). While previous research following this approach has focused on organizations in a single nation, one unique aspect of our study lies in the fact that FIEs’ identities and operations span across multiple institutional environments demarcated by national boundaries. On the one hand, FIEs bear the norms and traditions of the investors’ home countries. Among them, subsidiaries of multinational corporations are particularly influenced by policies and practices of their parent companies as well as by global norms and standards. On the other hand, FIEs are also prone to the influences of the host country’s local institutional environment. Such ‘‘institutional duality’’ implies the coexistence of competing institutional logics and alternative normative scripts (Kostova & Roth, 2002). Under this circumstance, the taken-forgrantedness of ‘‘institutional myths’’ as highlighted by canonical statements of the neoinstitutional theory can no longer be presumed. According to the recently developed translation model in the institutional literature, when ideas and practices travel globally, they are likely to be reinterpreted by local actors and reshaped by local settings (Czarniawska & Joerges, 1996; Zilber, 2002). From this perspective, the global transfer of organizational practices often encounters cultural and institutional barriers because these practices embody different institutional logics and organizing principles than the local ones. Practices such as the suggestion system and

172


formal grievance procedures may have become highly institutionalized and taken for granted in Western countries, yet their meanings and appeals to local actors in other parts of the world cannot be presumed, neither can the effects of their adoptions. For example, Frenkel’s study (2005) showed that the transfer of Scientific Management and the Human Relations Model to Israel after 1950s was heavily shaped by the institutional power structure at the state level, and that the adoptions of these new models of management were characterized by significant adaptations to the local ideology and by the organization’s self-interest. In developing countries and transitional economies, foreign investors often command significant power, based on their control over financial capital and advanced technology, and thus are able to resist certain pressures from the host countries. This may allow them to either replicate structures common in their home countries or to adopt ‘‘best practices’’ which are increasingly regarded as the gold standard in human resource management in the global community (Bae & Lawler, 2000; Martin & Beaumont, 1998). Nevertheless, we stress that there are strong reasons for FIEs to adapt to the host country’s institutional environment by localizing their labor practices, and this may be particularly true in the Chinese context. First, FIEs need to conform to the host country’s laws, norms, and regulations to become legitimate players in the local economy. In contrast to the ‘‘shock therapy’’ approach adopted by Russia and Eastern European countries, China’s economic reform represents a rather lengthy process of incremental change, a process that is constantly monitored and tinkered with by the state. As a result, the Chinese institutional environment is characterized by not only a distinct cultural heritage but also a strong and persistent socialist legacy. Such an environment entails significant institutional and cultural barriers for the transfer of Western practices. Although FIEs are granted greater autonomy than other Chinese companies, they are still subject to significant state regulation. These regulatory constraints are especially strong in the area of labor practices and personnel policies (Child, 1994; Gallagher, 2005) and exert considerable pressure on FIEs to conform to local standards. Second, FIEs need to accommodate the views of local managers and workers in order to secure their consent and cooperation. According to the translation model, ‘‘structures and practices are perceived and understood, both consciously and unconsciously, by the members and the subgroups of organizations’’ (Zilber, 2002, p. 236). Accustomed to the socialist labor regime and practices, Chinese employees may identify more strongly with preexisting, Chinese-style structures and have difficulties digesting imported foreign practices. Third, since FIEs more or less rely on


173

indigenous organizations in their operation, conforming to local norms and established practices helps to strengthen their interorganizational ties and facilitates transactions with local business partners.1

Host-Country and Home-Country Effects It is well documented that organizations are often compelled by the state to adopt certain structures and practices. This can occur even if the state’s administrative power is weak or the legislations mandating change are somewhat ambiguous (Edelman, 1990; Kelly & Dobbin, 1999; Sutton & Dobbin, 1996). In China’s market socialism, despite recent decentralization of labor management, the state remains a formidable institutional actor except that it now relies more and more on its legislative power to cope with emerging labor issues (Lee, 2002; Gallagher, 2005). Although China’s labor policies and legislations are sometimes vague and not always well enforced, they are strong and comprehensive enough to impose significant regulatory constraints and normative pressures on industrial enterprises (cf. Edelman, 1990). In regulating labor relations in the foreign investment sector, the chief emphasis of the state has been on the establishment of trade unions. Enacted in 1992, the Trade Union Law required that trade unions be established in all firms with 25 or more employees, including FIEs (Article 12, Trade Union Law). Despite spotty enforcement, this legal requirement did prompt local trade unions and government officials to press FIEs to set up more union branches (Chan & Senser, 1997). As to the workers’ congress, although it is not required by the state (Hong & Warner, 1998, pp. 97–98), adopting this structure can still help FIEs to achieve greater legitimacy and broader worker consent. In comparison, the institutional pressure on FIEs to adopt Westernstyle human resource management structures is considerably weaker. Even though ‘‘getting on track’’ with Western practices and international standards has become a constant theme in the official rhetoric and management circle (Guthrie, 1999), there are no formal mandates from the state. Some of these structures remain largely foreign to Chinese officials, managers, and employees and are not associated with the same level of normalcy or legitimacy as they are in advanced capitalist societies. Thus, we expect that, Hypothesis 1. Trade unions and workers’ congresses are more widely adopted than suggestion systems and formal grievance procedures in China’s foreign investment sector.

174


Although we emphasize the institutional pressures on FIEs stemming from the Chinese environment, labor management structures in FIEs may also reflect the diverse institutional traditions of foreign investors’ home countries. Compared to Asia’s newly industrialized economies (NIEs) such as Hong Kong, Taiwan, Korea, and Singapore, advanced capitalist economies feature more sophisticated employment practices. The suggestion system and the grievance procedures are integral components of the high-commitment or high-performance work system (e.g., Baron & Kreps, 1999; Becker & Gerhart, 1996; Huselid, 1995) and have become highly institutionalized in Japan and Western countries (Edelman, 1990; Guthrie, 2001; Hull, Azumi, & Wharton, 1988). While these structures have been diffusing globally as part of the ‘‘best practices’’ in modern human resource management (Bae & Lawler, 2000; Bae et al., 2003), we expect them to be more favored by investors from advanced capitalist countries than by those from NIEs. Hypothesis 2. Compared to FIEs with investments from newly industrialized economies, FIEs from advanced capitalist economies are more likely to adopt suggestion systems and formal grievance procedures.

Institutional Links The new institutional perspective suggests that the adoption of formal structures is contingent on the organization’s formal and historical linkages to local institutions and it is also mediated by organizational characteristics (Edelman, 1990, 1992; Guthrie, 1999). For Chinese FIEs, one crucial form of such institutional link relates to the FIE’s ownership composition. In general, FIEs with higher Chinese equity shares have stronger connections with domestic institutions. In these firms, the Chinese parent companies tend to exercise stronger ownership-based control. Because the Chinese parent companies themselves are deeply entrenched in local regulatory, cognitive, and normative structures, their control over FIE operations can serve as an effective means for translating external institutional pressures into structural adjustments. By contrast, FIEs with high or exclusive foreign ownership are subject to stronger foreign influences and face more limited incentives to establish Chinese-style labor management structures. Hypothesis 3. FIEs with larger Chinese equity shares are more likely to establish trade unions and workers’ congresses but are less likely to adopt suggestion systems and formal grievance procedures.


175

A second aspect of FIE’s institutional link pertains to the type of ownership of the Chinese parent company. Prior to the reform, state-owned enterprises constituted the core of the socialist economy and embodied classic features of traditional socialist labor relations. During the reform period, state-owned enterprises remain the closest to the center of political power and bear a stronger socialist legacy than collective and private-owned companies. Such an institutional legacy has been found to play a significant role in shaping firm behavior in postsocialist countries (Burawoy & Krotov, 1992; Spenner, Suhomlinova, Thore, Land, & Jones, 1998; Zhou et al., 2003). Since foreign investors in China often rely on their Chinese partners and Chinese managers to handle issues in labor management (Taylor, 1999), we expect that, compared to joint ventures with other types of Chinese partners, joint ventures with state-owned enterprises should be more likely to establish and maintain Chinese-style labor management structures. Hypothesis 4. FIEs whose Chinese parent companies are state owned (as opposed to collective or private owned) are more likely to have trade unions and workers’ congresses.

Organizational Size and Age Classic organizational theories argue that larger organizations face more daunting tasks of control and coordination and tend to develop more administrative components in their formal structures (Blau, 1970). From an institutional perspective, the legal environment theory holds that larger organizations experience more intense pressure to conform to laws and regulations due to their higher visibility (Edelman, 1990). In China, organizational size has strong institutional implications. The Chinese government classifies industrial organizations into large-, medium-, and small-sized enterprises and sees large enterprises as strategic anchors for accomplishing its various political and social agendas (cf. Gallagher 2005, p. 140). Concerned with rising labor conflicts in the foreign investment sector and their impacts on social/political stability, which are further sensitized by resurging nationalist ideologies, Chinese governmental agencies have paid especially close attention to labor relations in large FIEs. As a result, large FIEs are under constant pressures to serve as exemplary modern enterprises. On the one hand, larger FIEs tend to follow the governmental regulations and adopt Chinese-style labor structures to maintain harmonious employer– worker relations. On the other hand, they are also subject to considerable

176


pressures to play a leading role in introducing sophisticated human resource management techniques from abroad, as many Chinese domestic enterprises have done (Guthrie, 1999; Xiao & Bjo¨rkman, 2006). These considerations lead to the following hypothesis: Hypothesis 5. Larger FIEs are more likely to adopt all four labor management structures. The organizational literature suggests that age may be another factor affecting the presence of different labor management structures. Organizations often retain features acquired at the time of founding for an extended period of time (Stinchcombe, 1965). Such imprints, along with other factors, contribute to strong organizational inertia (Hannan & Freeman, 1989), which impedes the adoption of new structural elements. According to this logic, older FIEs in China were established in an environment more dominated by traditional socialist institutions and therefore faced stronger pressures to incorporate traditional labor structures, particularly the trade union. Studies of multinational corporation affiliates also show that older affiliates are more likely to have adapted their human resource practices to the local culture and institutions as a result of their continuous exposure to the local environment (Rosenzweig & Nohria, 1994). In contrast, given the growing influence of global capitalism in China, the organizational imprints in FIEs established in more recent years should reflect a stronger global orientation. These FIEs also face weaker inertial forces inside and should be more open to foreign managerial models and techniques. Hypothesis 6. Older FIEs are more likely to adopt trade unions and workers’ congresses and are less likely to adopt the suggestion systems and grievance procedures.

Strategic Governance: The Roles of Industrial Context and Human Assets Although organizations are ‘‘institutionally controlled, they are not determined’’ (Friedland & Alford, 1991, p. 256). Rather, modern business organizations routinely strive to make rational decisions under institutional constraints (Brinton & Nee, 1998). This tendency is likely to be even more pronounced in FIEs. As a strategic move, the investor’s decision to enter a foreign country is explicitly aimed at enhancing profitability and global competitiveness, and the drive for efficiency should be a highly salient feature underlying FIE operation.


177

Transaction cost economics contends that the main purpose of adopting structures such as the internal labor market and formal grievance procedures is to stabilize the workforce and to better manage human assets (Williamson 1981, 1985, p. 256). Extending this perspective to Chinese FIEs, we expect their labor management structures to be contingent on the industrial context and the corresponding business strategy (cf. Sutton & Dobbin, 1996). Foreign investors come to China with different incentives and objectives: whereas some with abundant financial capital and advanced technology (e.g., some blue-chip multinational corporations) are attracted to China by its vast market, others primarily look for cheap manufacturing (Cooke, 2004) fueled by an almost endless supply of low-skilled migrant workers from the countryside (Lee, 1999, p. 44). The former tend to have long-term investment plans, operate in capital-intensive industries, and face strong incentives to adopt formal labor structures to maintain a high-quality workforce. These firms are more motivated to adopt both global ‘‘best practices’’ and Chinese-style labor structures such as the trade union and the workers’ congress which are also recognized as useful tools for facilitating labor management (Hong & Warner, 1998, p. 121). In contrast, seeking cheap manufacturing opportunities, the latter group of FIEs tend to operate in labor-intensive industries where minimum training is required and keeping labor cost low is a must. These FIEs often prefer a control-oriented production regime and are less willing to invest resources to establish elaborated organizational structures. Meanwhile, even though the central government has been striving to protect workers’ rights through legislative channels, local governments often tolerate labor abuses and other forms of violations in FIEs in order to speed up local economic development (Cooke, 2004). Given the divergent strategic orientations among FIEs and the permissive institutional environment at the local level, we expect that, Hypothesis 7. Compared to those in labor-intensive industries, FIEs in capital-intensive industries are more likely to adopt all four labor management structures. Even in the same industry, FIE’s production regime and its valuation of human assets may still vary across enterprises. At the organizational level, these characteristics can be partly captured by the worker’s wage (Sutton & Dobbin, 1996), as high wage typically signals high human capital. To the extent that employees with higher qualifications are more difficult to replace, their employers face stronger incentives to establish formal labor structures to promote employee commitment and to reduce turnovers.

178


Hypothesis 8. The higher the worker’s wage level, the more likely the FIE establishes all four labor management structures.

Labor Management Structures and the Managerial Outcome Adopting a particular labor management structure is not merely a matter of formality but rather has important ramifications for managerial outcomes (Kalev, Dobbin, & Kelly, 2006). ‘‘Best practices’’ originated in Western countries are intended to facilitate labor management and to promote organizational performance, and some studies have shown that, when introduced to several non-Western economies, they exert a similar, positive effect on the company (Bae et al., 2003; Guthrie, 2001). However, others contend that whether ‘‘best practices’’ can fulfill their intended purposes is contingent on the host country’s specific institutional context (Baron & Kreps, 1999; Robert, Probst, Martocchio, Drasgow, & Lawler, 2000). For instance, Newman and Nollen (1996) find that practices that emphasize individual responsibilities enhance organizational performance in countries cherishing individualism but not in those cherishing collectivism. Echoing these arguments and findings, the translation model suggests that, when introduced into a new environment, a specific global practice may be reinterpreted by local actors and transformed by the local setting, and as a result it may have unintended consequences and perhaps even cause detriments to the organization (Doorewaard & van Bijsterveld, 2001). In the 1990s, newly imported Western-style structures such as the suggestion system and formal grievance procedures were less well understood and less well accepted by Chinese managers and workers. Although the traditional labor management under the Chinese socialist economy also encouraged employees to raise suggestions, this practice rarely moved beyond the rhetoric and had not been formalized to the same degree as the suggestion system in Western countries. Meanwhile, as noted by Guthrie (1999), formal grievance procedures symbolize a significant departure from China’s traditional labor practices and socialist values which emphasized the fundamental unity and harmony between the workers, the management, and the work unit. Therefore, the adoptions of these newly imported practices might not be able to fulfill their intended function of alleviating the tension and conflict between labor and management. In contrast, Chinese-style labor structures such as the trade union and the workers’ congress convey strong symbolic meanings rooted deeply in the pro-worker socialist legacy,


179

thereby helping the company to construct a favorable image in the eyes of both external parties and its own employees. These structures can also serve as valuable communication channels and facilitate cross-cultural understanding (Gallagher, 2005, p. 84). Therefore, we hypothesize that FIEs benefit more from adopting the two Chinese-style labor management structures than from adopting those of Western origins. Hypothesis 9. Compared to suggestion systems and formal grievance procedures, trade unions and workers’ congresses are more effective in reducing the tension and conflict between labor and management in Chinese FIEs.

DATA AND VARIABLES Our empirical analysis employs data from the Three-Capital Enterprise Investment Experience Survey (TCEIES hereafter). Conducted by a team of US- and Taiwan-based researchers, the TCEIES collected information on two samples of FIEs in manufacturing industries during the mid-1990s. The first sample consisted of 1,066 FIEs randomly selected from a national database of 22,744 foreign-invested manufacturers registered in China as of 1991. The second sample, which contained 489 cases, was drawn from FIEs established during the period of 1991–1992. General managers of the first sample were interviewed annually from 1993 to 1997,2 and those of the second sample were interviewed from 1994 to 1998. These 1,555 FIEs were located in 184 cities in 26 of China’s 30 provinces. Compared to those used by previous studies of Chinese organizations, this dataset stands out by virtue of its large size, national representation, and exclusive focus on FIEs. Although both samples had been surveyed for 5 consecutive years, information on labor management structures was only available for the 1996 panel. This limits our research to a cross-sectional design. After pooling the two samples together and data cleaning, we obtain an effective sample of 841 FIEs.3 Descriptive statistics for all variables are reported in Table 1. Our research hypotheses entail two sets of dependent variables. The first set consists of four dummy variables indicating the presence of the four aforementioned structures. The second set pertains to the effects of these structures on labor–management relations. Here, we use two variables to gauge the tension and conflict between labor and management: (1) whether the company experienced any labor dispute and (2) the difficulty of labor management as perceived by the company’s general manager. The first is a dummy variable and reflects the workers’ perspective. As shown in Table 1,

180


Table 1.

Descriptive Statistics (N ¼ 841). Mean/Percentage

Labor management structures Trade union Workers’ congress Suggestion system Grievance procedures

60% 33% 10% 8%

Labor–management relation If had any labor dispute in 1996 Perceived difficulty of labor management (scale from 1 to 5)

6% 2.94

Investor origin NIE (Hong Kong, Taiwan, Macao, Korea, Singapore, etc.) Western (USA, Europe, Canada, and Oceania) Japan

83% 11% 6%

Chinese equity share (in proportion)

0.50

Ownership of Chinese parent company State Collective and TVE Private and others None (wholly foreign-owned FIEs)

36% 46% 7% 12%

Firm size (number of employees)

181.57

Firm age (years)

6.35

Industry Labor intensive Capital intensive Others

49% 48% 3%

Worker’s wage, ratio to city average

1.43

Control variables City-level contextual variables GDP per capita (in 1,000 Yuan) Ratio, SOE industrial output to city GDP FDI per capita (in US$) Region: Southeast coast (Guangdong, Fujian, and Hainan) East coast (Shanghai, Jiangsu, and Zhejiang) North coast (Tianjin, Shandong, Hebei, and Liaoning) Inland (all inland provinces) Beijing

19.66 0.50 511.17 33% 24% 21% 10% 12%

S.D.

0.53

0.25

327.10 2.19

0.47

14.89 0.32 536.25


181

6% of the FIEs in our sample experienced labor disputes in 1996. The second – the perceived difficulty of labor management – is a composite measure. It is calculated as the arithmetic average of five items based on the general manager’s evaluations of how difficult it was to manage employee motivation, turnover, work ethics, organizational commitment, and work quality/efficiency respectively. Each item is measured on a five-point scale and is recoded such that a higher value indicates greater perceived difficulty. According to Table 1, on average FIE managers in 1996 reported a moderate level of difficulty of labor management (mean ¼ 2.94), although there were considerable variations across companies (S.D. ¼ 0.53). In terms of investor origin, FIEs funded by Hong Kong and Taiwan investments were the most dominant, combining for a total of 75% of the sample. About 2% of the investors were from Macao, 2% from South Korea, 2% from Singapore, and 1% from other Southeast Asian countries. Together these NIEs account for 83% of the sample.4 For the remaining 17%, 6% were funded by Japanese investment and 11% were from Western countries, mostly the United States and Europe. Overall, the composition of FIE’s country origin in our sample is comparable to statistics reported by both previous studies (Chan, 1998, pp. 124–125) and the Chinese government (see Table A1 in the appendix). FIEs in China are commonly known as ‘‘Three-Capital Enterprises’’ because they belong to three official categories: wholly foreign-owned enterprises, equity joint ventures, and cooperative joint ventures.5 The average Chinese equity share in our sample is 0.50. The standard deviation of this variable is fairly large (0.25), due in part to the fact that 12% of the FIEs are wholly foreign-owned enterprises (i.e., 0% Chinese equity share). As for joint ventures, their Chinese parent enterprises can be state-owned enterprises (the reference category), collective or town-village enterprises, or private or other types of firms. The average FIE’s size, defined as the total number of employees, is 182. The average firm age was 6.35 years, consistent with the fact that all FIEs in China were established during the post-1978 reform era. We use FIE’s industrial context as a proxy for its production regime. All industries are classified into three categories: labor intensive, which includes mainly food, beverage, textile, garments, leather, furniture, wood products, and paper and paper products; capital intensive, which includes mainly chemical, mining, petroleum, machinery, and electronics; and a residual category for all others. FIEs in labor-intensive industries employ large numbers of workers with relatively lower skill requirements, whereas those in capital-intensive industries tend to place greater values on human assets and demand a more stable workforce. Based on this classification, 49% of the

182


sample falls into the labor-intensive category, 48% belongs to capital-intensive industries, and the remaining 3% falls into the residual category. To capture firm-level variations of employee human capital, we calculate the ratio between the average wage level for ordinary workers in the FIE and the average worker’s wage for the entire city.6 The higher this ratio, the higher the skill requirement relative to the local labor force, and the more difficult it is to find and train replacements following turnovers. The mean for this variable is 1.43, indicating that on average FIEs paid 43% higher than the local average. This wage gap confirms the popular perception in China that FIE employees are compensated at a level far above the domestic standard. In the statistical analyses, we also control for possible contextual effects at the city and regional levels. At the city level, we use GDP per capita to measure the economic development, the ratio of state sector industrial output to city GDP to gauge the persistence of the socialist economy, and FDI per capita to capture the city’s openness to the global economy. At the region level, we identify four distinctive regions based on previous studies (Nee, 1996; Vogel, 1989). The southeast coastal region includes three provinces – Guangdong, Fujian, and Hainan. These three provinces were among the first to implement the reform and open-door policies. They are also the leading recipients of FDI and popular destinations of migrant workers from rural China. The east coastal region consists of Shanghai, Jiangsu, and Zhejiang; the north coastal region includes Tianjin, Shandong, Hebei, and Liaoning; and all inland provinces are grouped together. We create a separate category for Beijing due to its many unique features as the nation’s capital. Table 2 presents the correlation matrix for all variables.

ANALYSIS AND RESULTS We first examine the differences in adoption rates between Chinese-style and Western-style labor management structures. The adoption rate of the trade union was 60%, substantially higher than those of the suggestion system (10%) and formal grievance procedures (8%). Although the workers’ congress was not as common as the trade union, it is adopted by 33% of the companies, still surpassing the suggestion system and grievance procedures by large margins. McNemar’s significance tests show that all these gaps are significant at the 0.001 level. These results strongly confirm our Hypothesis 1 that trade unions and workers’ congresses are more widely adopted than

Y1 Trade union Y2 Workers’ congress Y3 Suggestion system Y4 Grievance procedures Y5 Labor mgmt. difficulty Y6 Labor dispute X1 Investor origin: Western X2 Investor origin: Japan X3 China equity share X4 China parent: collective X5 China parent: nonpublic X6 China parent: none X7 Firm size X8 Firm age X9 Capital intensive X10 Other industries X11 Workercity wage ratio X12 City GDP per capita X13 City SOE output X14 City FDI per capita X15 East coastal region X16 North coastal region X17 Inland region X18 Beijing

1.00 0.18 0.02 0.06 0.14 0.06 0.06 0.05 0.12 0.03 0.03 0.09 0.12 0.12 0.12 0.04 0.06 0.04 0.06 0.04 0.06 0.05 0.02 0.02

Y1

1.00 0.04 0.01 0.10 0.13 0.05 0.01 0.23 0.10 0.03 0.18 0.04 0.04 0.09 0.01 0.04 0.13 0.12 0.10 0.04 0.14 0.03 0.07

Y2

1.00 0.15 0.03 0.01 0.05 0.14 0.00 0.06 0.09 0.01 0.08 0.01 0.11 0.01 0.10 0.02 0.02 0.02 0.02 0.00 0.00 0.02

Y3

1.00 0.02 0.12 0.09 0.10 0.06 0.11 0.07 0.02 0.06 0.02 0.09 0.02 0.05 0.03 0.06 0.02 0.04 0.05 0.01 0.02

Y4

1.00 0.11 0.11 0.07 0.07 0.02 0.01 0.05 0.01 0.00 0.05 0.01 0.10 0.02 0.07 0.07 0.09 0.07 0.01 0.03

Y5

1.00 0.01 0.02 0.03 0.04 0.05 0.03 0.07 0.01 0.01 0.02 0.04 0.10 0.02 0.08 0.02 0.03 0.03 0.01

Y6

X1

1.00 0.01 0.02 0.07 0.01 0.04 0.07 0.09 0.01 0.11 0.03 0.08 0.03 0.04 0.02 0.02 0.08

X2

1.00 0.23 0.09 0.74 0.02 0.07 0.08 0.01 0.03 0.09 0.19 0.15 0.11 0.14 0.07 0.04

X3

1.00 0.25 0.33 0.02 0.11 0.05 0.01 0.06 0.12 0.02 0.06 0.19 0.06 0.14 0.02

X4

1.00 0.10 0.04 0.01 0.02 0.02 0.03 0.03 0.06 0.01 0.02 0.06 0.05 0.08

X5

1.00 0.01 0.08 0.05 0.00 0.01 0.09 0.13 0.13 0.05 0.04 0.03 0.09

X6

1.00 0.14 0.01 0.01 0.07 0.05 0.06 0.05 0.01 0.03 0.05 0.04

X7

X9

X10

X11

X12

X13

X14

X15

X16

X17

1.00 0.07 1.00 0.05 0.17 1.00 0.03 0.01 0.02 1.00 0.26 0.13 0.01 0.27 1.00 0.23 0.07 0.07 0.07 0.36 1.00 0.19 0.08 0.03 0.24 0.60 0.38 1.00 0.16 0.09 0.04 0.00 0.03 0.11 0.20 1.00 0.11 0.04 0.05 0.16 0.19 0.11 0.15 0.29 1.00 0.14 0.01 0.04 0.17 0.16 0.20 0.27 0.19 0.17 1.00 0.03 0.02 0.04 0.19 0.19 0.45 0.30 0.21 0.19 0.12

X8

Correlation Matrix for All Variables.

1.00 0.09 0.05 0.07 0.04 0.05 0.05 0.02 0.01 0.04 0.05 0.02 0.12 0.03 0.01 0.10 0.01 0.09

Table 2.

184


suggestion systems and grievance procedures. Meanwhile, it is worth noting that even though only about 10% of our sample had fewer than 25 employees, those without a trade union reached 40%. In other words, as of 1996, four years after the enactment of the Trade Union Law, the legal requirement that firms with 25 or more employees must have a trade union was still poorly enforced. The adoption of the trade union thus cannot be considered as a simple matter of legal necessity but instead requires careful analysis. Determinants of Labor Management Structures We now turn to the determinants of the four structures. Instead of estimating four separate models, we construct one multivariate probit model that takes into account the interdependence among the four binary outcomes. This multivariate model consists of four regression equations, each corresponding to one dependent variable. As in univariate probit, each dependent variable Yj ( j ¼ 1, 2, 3, or 4) is assumed to be a manifestation of a continuous latent variable yj such that Yj is equal to 1 when yj is positive and 0 otherwise. The latent variable yj is a linear combination of the independent variables plus an error term ej, which follows a standard normal distribution. Unlike the univariate approach which models each outcome independently, the multivariate probit model assumes that the disturbance terms in the four equations are correlated and jointly follow a multivariate normal distribution.7 This may result from possible reciprocal effects among these structures themselves as well as omitted variables that affect the adoptions of multiple structures simultaneously. Formally, the multivariate probit model postulates that: 1. Y j 1ð yj 40Þ, where j 2 ð1; 2; 3; 4Þ; 2. yj ¼ Xbj þ j , where j Nð0; 1Þ; and 3. ð1 ; 2 ; 3 ; 4 Þ0 MVNð0; SÞ, and the given as 1 r12 r12 1 S ¼ r13 r23 r14 r24

variance–covariance matrix is r13 r23 1 r34

r14 r24 r34 1

where rjk is the correlation between the disturbance terms in the jth and kth equations.


185

Table 3 presents multivariate probit regression coefficients obtained with maximum likelihood estimation. Several independent variables – firm size, age, city GDP per capita, and city FDI per capita – are transformed into their natural logarithms. Because wholly foreign-owned enterprises had neither Chinese ownership (0% Chinese equity share) nor Chinese parent company, these two variables are highly correlated. To deal with this problem, we first estimate Model 1 without Chinese parent company ownership variables to better test the effect of Chinese equity share and then include both sets of variables in the full model (Model 2).8 The findings only render partial support to the prediction on the home country’s effects: FIEs from advanced capitalist economies are more likely to adopt the suggestion system and grievance procedures (Hypothesis 2). Indeed, Japanese-invested companies are significantly more likely to adopt suggestion systems than other FIEs. Our calculation based on the univariate normal distribution shows that a Japanese-invested company has a 21% chance of adopting the suggestion system, compared to 8% for an otherwise identical firm9 funded by NIE investment. This is not a surprise. Originated over a hundred years ago, the suggestion system has been embraced and utilized to a much higher degree by Japanese organizations (Hull et al., 1988). Japanese-invested companies also have the highest adoption rate of grievance procedures, although this coefficient is only marginally significant in Model 1 and becomes insignificant in Model 2. Surprisingly, Western-invested companies are by far the least likely to adopt grievance procedures. The predicted probability of adoption for this group is only 0.77%, compared to 8.91% for the reference category. In fact, among the 92 Western FIEs in the sample, only 1 reported having grievance procedures. This finding directly contradicts Hypothesis 2. It appears that rather than dogmatically adhering to established practices in their home countries, Western FIEs often make drastic adjustments based on local conditions. For example, in order to prevent thefts an American-owned company in Tianjin once required all workers to be body-searched upon leaving the premises (Gallagher, 2005, p. 89), a practice that is utterly unacceptable according to the America’s domestic standard. Field researchers’ observations further suggest that in contrast to the cultural proximity enjoyed by Japanese and other Asian investors, the tremendous cultural and political gaps between China and Western countries might have been partially responsible for Western FIEs’ complete abandonment of established practices in the investors’ home countries (Chan, 1998, pp. 126–127). This may explain the opposite results on the home country’s effects across Western- and Japanese-invested enterprises.

0.22

0.44

0.32 0.59

Firm size (log)

Firm age (log)

Industry (vs. labor intensive) Capital intensive Others

0.21

– – –

0.32 0.38

0.18

(vs. state) – – –

1.21

0.32

Chinese equity share

Ownership of Chinese parent company Collective and TVE Private and other None (wholly foreign owned)

0.14 0.03

0.51

Workers’ congress

0.20 0.15

2.72

Trade union

Local structures

0.40 0.08

0.10

0.11

– – –

0.07

0.28 0.60

2.13

Suggestion system

0.27 0.14

0.00

0.15

– – –

0.45

1.08 0.38

3.56

Grievance procedures

Imported structures

Model 1

0.32 0.59

0.40

0.24

0.26 0.39 0.29

0.18

0.18 0.16

2.40

Trade union

0.33 0.39

0.16

0.21

0.12 0.17 0.24

0.95

0.15 0.01

0.52

Workers’ congress

Local structures

0.40 0.10

0.10

0.12

0.13 0.43 0.07

0.00

0.31 0.55

2.00

Suggestion system

0.22 0.18

0.07

0.15

0.44 0.18 0.41

0.72

1.18 0.35

2.90

Grievance procedures

Imported structures

Model 2

Multivariate Probit Model of Four Labor Management Structures in Chinese FIEs (N ¼ 841).

Investor origin (vs. NIEs) Western Japan

Intercept

Table 3.

186 YANG CAO AND WEI ZHAO

0.26 0.28 0.28 0.09

Region (vs. southeast coast) East coast North coast Inland Beijing

Note: po0.10, po0.05, po0.01.

– – –

0.13 0.27 0.00

Control variables City-level contextual variables GDP per capita (log) SOE industrial output/GDP FDI per capita (log)

Correlations among disturbances (rs) Trade union Workers’ congress Suggestion system

0.11

Wage ratio to city average

0.22 – –

0.17 0.02 0.34 0.76

0.48 0.47 0.01

0.11

0.14 0.04 –

0.10 0.19 0.33 0.26

0.04 0.28 0.02

0.31

0.21 0.06 0.30

0.08 0.35 0.24 0.03

0.24 0.82 0.02

0.31

– – –

0.33 0.32 0.27 0.13

0.14 0.23 0.01

0.08

0.23 – –

0.19 0.03 0.30 0.77

0.48 0.49 0.01

0.10

0.12 0.04 –

0.07 0.16 0.39 0.28

0.08 0.24 0.03

0.30

0.21 0.04 0.28

0.02 0.29 0.31 0.06

0.22 0.75 0.00

0.26

Localization in the Age of Globalization 187

188


Another noteworthy finding in Model 1 is that investor origin has no effect on either the adoption of the trade union or that of the workers’ congress. This may be due to the relative uniformity of the regulatory constraints imposed on these FIEs by the Chinese institutional environment. Hypothesis 3 predicts that Chinese ownership serves as an institutional link that promotes the adoption of Chinese-style labor structures while reducing those of Western-style ones. The results in Model 1 provide considerable support to this hypothesis. Specifically, we find that the higher the FIE’s Chinese equity share, the more likely that it adopts the trade union and the workers’ congress, and the less likely that it adopts grievance procedures. The strongest effect is in the case of the workers’ congress. In the case of the suggestion system, the coefficient of the Chinese equity share does not reach marginal significance. Overall, these results suggest that the Chinese equity share in FIEs provides an effective institutional link that promotes the localization of their labor practices. Hypothesis 4 predicts that FIEs whose Chinese parent companies are state owned are more likely to have trade unions and workers’ congresses. The results in Model 2 are mixed. Joint ventures with state-owned parent companies are significantly more likely than those with collective and nonpublic ownership partners to adopt trade unions. Wholly foreign-owned enterprises also have a lower adoption rate, although the difference is insignificant. In contrast, state ownership of the Chinese parent company has no effect on the likelihood of establishing a workers’ congress. We suspect this is due to the relative newness of the workers’ congress. Because this structure was reintroduced during the reform period, differences across domestic firms of state, collective, and nonpublic ownerships might not have been large enough to produce differential legacies when they form joint ventures with foreign investors. A strong determinant of labor management structures in Chinese FIEs is firm size. Larger FIEs are significantly more likely to have all four formal structures. In China, large FIEs, along with other large indigenous enterprises, are under high institutional pressures to establish various formal structures to enhance legitimacy. These results lend strong support to Hypothesis 5. Anecdotal evidence also corroborates these findings. A highprofile example is Wal-Mart in China, which employed more than 30,000 Chinese workers and operated in 30 Chinese provinces. Despite its parent company’s notorious anti-union stand in the United States, Wal-Mart China was singled out and directly pressed by the All-China Federation of Trade Unions to allow the establishment of a trade union.10


189

Age exerts a significant and positive effect only on the adoption of the trade union, but not on that of the worker’s congress. FIEs established during earlier years of China’s economic reform may have been subject to stronger socialist institutional pressures and are more likely to adopt the trade union, a hallmark feature of China’s state socialism. This is not true for the workers’ congress, which is a newer development and has not become highly institutionalized in China. We find no significant age effect on establishing the two imported labor structures either. These findings render partial support to Hypothesis 6. Table 3 shows that organizational interest and efficiency consideration also mediate the adoptions of these labor structures. Consistent with Hypothesis 7, FIEs in capital-intensive industries are more likely than those in laborintensive industries to adopt all four structures. All four coefficients are positive and three of them highly significant. The only exception is the industry’s effect on formal grievance procedures, which is marginally significant in Model 1 but insignificant in Model 2. These findings reflect important cross-industry differences in labor management. As FIEs in capital-intensive industries strive for long-term employment and stable workforce, they resort to elaborated formal structures to foster commitment. In contrast, in labor-intensive industries where employee training is not as important and reducing labor cost is a must, FIEs tend to avoid those elaborated labor management structures which may put them at financial and strategic disadvantages in a highly competitive market. Clearly, situated in their specific industrial contexts, FIEs strategically select labor structures in accordance with their interests and practical needs. Worker’s wage level (relative to the city average), an organizational level indicator of production regime and human asset, plays a limited role in the adoptions of labor management structures. According to Model 2, worker’s wage level has no significant effect on the adoption of either the trade union or the workers’ congress. However, FIEs with higher wage levels are indeed more likely to adopt suggestion systems. For a typical FIE, a one standard deviation increase in the firm–city wage ratio leads to a 2.41% increase in the probability of adopting a suggestion system. The effect of wage level on the adoption of grievance procedures is positive, but it is only marginally significant in Model 1 and insignificant in Model 2. Overall, these findings appear to suggest that FIEs with valuable human assets are particularly eager to fully utilize them through the suggestion system, a structure often aimed at improving production efficiency. These results render partial support to Hypothesis 8. As for control variables, FIEs located in more developed cities (i.e., with higher GDP per capita) are less likely to adopt the workers’ congress, whereas

190


those in cities where the state sector is more dominant are more likely to have both the workers’ congress and formal grievance procedures. The city’s reception of FDI has no significant effect. At the regional level, we find the lowest trade union adoption rate in the southeast coastal region (the reference category). This may be attributable to the region’s longer reform history and its greater openness to the global economy. Both factors help to diminish the influence of the socialist legacy. Finally, FIEs in the nation’s capital – Beijing – have the lowest adoption rate of the workers’ congress. An important advantage of the multivariate model is that it allows us to gauge the interdependence among the four structures after controlling for all independent and control variables. Such interdependence can be seen in the correlation coefficients for the disturbance terms in the four equations (bottom panel of Table 3). Three of the six correlation coefficients are significant at the 0.05 level or higher, and a fourth one – that between the trade union and the suggestion system – approaches marginal significance. Substantively, these coefficients imply that net of all other effects incorporated in the model, the two Chinese-style structures – the trade union and the workers’ congress – are more likely to coexist; similarly, the two Western-style structures – the suggestion system and formal grievance procedures – are more likely to appear together; and the presence of a trade union is a negative predictor of both the suggestion system and grievance procedures. Together, these results reveal a remarkable degree of mutual exclusiveness between the Chinese-style and the Western-style labor structures. Whereas most FIEs substantially localize their labor practices, others tend to adopt multiple structures imported from industrialized economies.

Impacts of Labor Structures on Labor–Management Relations The final part of our analysis examines the effects of these four structures on labor–management relations. Table 4 presents two models: a binary probit model predicting the occurrence of labor dispute in 1996 and an OLS regression predicting the overall difficulty of labor management perceived by the company’s general manager. As expected, the two structures sanctioned by Chinese institutions – the trade union and the workers’ congress – both reduce the risk of labor dispute. These findings lend support to our arguments and confirm the observation that the workers’ congress can be especially effective in facilitating labor management communications and preventing conflicts (Levine, 1997, pp. 51–53). In contrast, the suggestion system has no significant effect, and

191


Table 4.

Effects of Four Labor Management Structures in Chinese FIEs. Binary Probit Model Predicting Labor Disputes in 1996

OLS Regression of Perceived Difficulty of Labor Management

Intercept

4.15

Formal labor management structures Trade union Workers’ congress Suggestion system Grievance procedures

0.33 0.91 0.16 0.64

0.10 0.07 0.08 0.03

0.00

0.14

Investor origin (vs. NIEs) Western (USA, Canada, Europe and Oceania) Japan Chinese equity share (in proportion)

0.22 0.23

Chinese parent company ownership (vs. state) Collective and TVE 0.15 Private and other 0.83 None (wholly foreign owned) 0.31

3.45

0.14 0.01 0.01 0.02 0.01

Firm size (log)

0.27

0.00

Firm age (log)

0.01

0.01

Industry (vs. labor intensive) Capital intensive Others

0.05 0.11

0.02 0.04

Worker’s wage, ratio to city average

0.19

0.10

City-level contextual variables GDP per capita (log) SOE industrial output/GDP FDI per capita (log)

0.54 0.28 0.01

0.13 0.06 0.03

Region (vs. southeast coast) East coast North coast Inland Beijing

0.34 0.78 0.77 0.30

0.14 0.12 0.04 0.01

2 Log-likelihood R2 N Note: po0.10, po0.05, po0.01.

322.46 – 841

– 0.07 810

192


interestingly, FIEs with formal grievance procedures are more likely to experience labor disputes. One alternative explanation is that FIEs facing greater risks of labor disputes might be more likely to establish grievance procedures in the first place. Although we are unable to rule out this possibility in a conclusive manner, additional analysis not reported here suggests this is unlikely to be the case because FIEs that had experienced labor disputes in 1994 and 1995 were no more likely to report having grievance procedures (or any of the other three structures) in 1996. We suspect that in the mid-1990s, formal grievance procedures, which were imported to China just recently, could undermine the Chinese tradition of harmonizing labor– management relations through informal dialogues and mediations. The very presence of this formal structure might symbolize a different (more formalistic and ‘‘tougher’’) management approach and might not be interpreted positively by local employees. In addition, formal grievance structures might heighten the workers’ awareness of their legal rights and increase their propensity to express grievances through open confrontations.11 Table 4 also shows that in FIEs with trade unions and workers’ congresses, company managers tend to perceive less difficulty in labor management. In contrast, the coefficients for the suggestion system and grievance procedures are positive but statistically insignificant. Overall, results in Table 4 lend considerable support to Hypothesis 9 by showing that the two Chinese-style labor governance structures help stabilize labor– management relations and reduce the tension and conflict between the two sides, whereas the two Western-style structures associated with the modern human resource management do not fulfill their intended functions in the context of China.

DISCUSSION AND CONCLUSION One of our main findings is that FIEs have significantly adapted to China’s distinct institutional environment. In contrast to the sporadic effects of investor origin, Chinese institutions have left clear imprints on FIEs’ labor practices. The majority of the FIEs in our sample reported having trade unions; perhaps more remarkably, the workers’ congress – a structural feature not required by the state – was found in a third of the cases. The high adoption rates of these Chinese-style structures stand in sharp contrast to the rarity of the two imported structures. Clearly, while FIEs have influenced domestic enterprises (Guthrie, 1999), their own practices have also been profoundly shaped by China’s distinct institutional environment.


193

These findings enrich and extend the scope of neoinstitutional analyses (Meyer & Rowan, 1977; DiMaggio & Powell, 1983). Whereas most neoinstitutional analyses in sociology examine organizations in a single institutional environment and stress their conformity to clear-cut institutional principles within a national boundary, we emphasize that FIEs face dual institutional pressures and are capable of considerable adaptations to transnational institutional environments. As suggested by the translation model (Czarniawska & Joerges, 1996; Zilber, 2002), in the Chinese context characterized by potent and distinct indigenous institutions, foreign investors make significant adjustments to fit into the host country’s local environment, rather than dogmatically transferring standard practices from their home countries. In addition to the general tendency of localization, we find that the capability and extent of FIE adaptation vary across organizations. These variations are associated with several institutional factors. A high Chinese equity share and a partnership with a state-owned enterprise provide important institutional links that facilitate FIE adaptation to the Chinese institutional environment. Our analysis shows that FIEs with higher Chinese equity shares are more likely to adopt both the trade union and the workers’ congress but are less likely to establish grievance procedures. The trade union is most commonly found in joint ventures with state-owned Chinese parent companies. Moreover, we find that larger FIEs are more likely to adopt all four structures, due in part to the fact that they are more visible and subject to more intense regulatory pressures from the Chinese government. Finally, older FIEs are more likely to establish the trade union, a hallmark of China’s traditional labor management. Although institutional factors are crucial driving forces of organizational adaptations, organizational interest and strategic considerations also play important roles in mediating the adoption of labor management structures. Indeed, on many occasions, FIEs’ adaptations represent highly purposive actions driven by organizational interests. One illustrative finding is that despite being a common organizational feature in their home countries, formal grievance procedures are almost completely absent in Westerninvested enterprises. Moreover, as FIEs operating in capital-intensive industries adopt formal structures to secure their long-term interests, those in labor-intensive industries try to avoid these structures altogether to lower the administrative cost and to maximize the exploitation of unskilled workers. Finally, FIEs employing workers with higher human capital appear particularly motivated to utilize these human assets through suggestion systems.

194


Our study further demonstrates that adopting a particular labor management structure is not simply a matter of symbolic decoration, but rather has substantive consequences. Our analysis of labor–management relations shows that both the trade union and the workers’ congress reduce the risk of labor dispute and the difficulty of labor management perceived by the company manager. In stark contrast is the complete ineffectiveness of the two Western-style structures, which seem to have largely lost their originally positive meanings and symbolic significance when they are received by Chinese employees. Through the lens of FIEs’ labor practices, this study also deepens our understanding of China’s postsocialist economy. Similar to the pattern in the state sector during the market transition (Keister, 1998; Stark, 1996), what emerged in China’s foreign investment sector is a distinct economic order characterized by mixed organizational practices. On the one hand, under the popular rhetoric of ‘‘getting on track with the international community,’’ Western-style structures have been imported and applied to China (Guthrie, 1999). On the other hand, the Chinese government has carried on a considerable socialist legacy and sought to maintain sufficient control over labor practices. Highly institutionalized Chinese-style labor management structures thus continue to prevail in FIEs. Under the dynamic interplay of the socialist legacy, emerging markets, and global capitalism, the new economic order and labor relations in the foreign investment sector are neither purely socialist nor purely capitalist (Gallagher, 2005, p. 158). Meanwhile, this distinct economic order in China’s ‘‘market socialism’’ characterized by strong state regulations and the looming managerial domination clearly differ from the ‘‘merchant capitalism,’’ which emerged in East European countries following the ‘‘shock therapy’’ (Burawoy & Krotov, 1992; Spenner & Jones, 1998). Our study of FIEs’ labor management practices in China also has broad implications for understanding globalization. Some scholars have stressed the global diffusion of sophisticated human resource management techniques (Bae & Lawler, 2000; Guthrie, 2001). Our study suggests that even between parent companies in industrialized economies and their affiliates overseas, the transfer of labor practices is neither automatic nor uniform. Instead, it is contingent on how FIEs cope with practical challenges stemming from the institutional differences across national boundaries. In China’s particular institutional context, strong state regulations and normative forces compel transnational organizations to localize their labor practices to a significant degree. It is a rich avenue for sociologists to further investigate how FIEs and other transnational organizations adapt to different local institutional


195

environments and how such adaptations (or their lack) produce diverging consequences in the age of globalization.

NOTES 1. Guthrie (1999) emphasizes that the practices of Chinese domestic enterprises are shaped by their ties with foreign companies. For example, he shows that having joint ventures with foreign investors is a significant predictor of adopting Westernstyle management practices. Applying the same logic but from a different angle, we observe that FIEs are inevitably embedded in networks of Chinese firms which should have significant influences on their orientations and practices. Chinese domestic enterprises’ mimicry of foreign investors, according to Guthrie, stems from their unfamiliarity with markets. Conversely, foreign investors may find it necessary to mimic Chinese firms due to their unfamiliarity with the Chinese institutional and cultural environment. 2. Actual interviews were conducted by personnel from the Chinese State Statistical Bureau. See Chi and Kao (1995) for details on sampling and data collection. 3. By 1996, 442 of the 1,555 FIEs sampled at the beginning of the survey had dropped out due to closures, relocations, mergers, or unspecified reasons. The average annual attrition rate was 11.76%. Among the remaining 1,113 cases, 11 were excluded due to unidentified or multiple national origin, and 261 more were discarded because of missing values. We examined the frequency distributions of a number of variables for the 1996 panel and those at the beginning of the survey and found them to be highly comparable (see Table A1 in the appendix). 4. Preliminary analysis finds no significant differences between FIEs from Hong Kong, Taiwan, and Macao and those from other NIEs. 5. We find no significant differences between equity joint ventures and cooperative joint venture. The differences between wholly foreign-owned enterprises and joint ventures are mostly explained by their differences in Chinese equity share. This is why we do not incorporate this official classification into the statistical analysis. 6. Information on FIE’s worker wage was obtained from the company’s general manager during the survey. Average wage for the city, along with other city-level contextual variables, was obtained from China’s State Statistical Bureau and merged into the TCEIES dataset by Chi and his associates (Chi & Kao, 1995). 7. We also estimated four separate univariate probit models, and the findings are virtually the same. However, the likelihood ratio test shows that the multivariate model significantly improves the model fit ( po0.01). 8. We also tested a third model that includes only Chinese parent company ownership variables but not Chinese equity share. The results are basically the same as in Model 2 except that, as expected, the coefficients for ‘‘no Chinese parent company’’ (i.e., wholly foreign-owned FIEs) in the first two equations (i.e., those for the trade union and the workers’ congress) both become larger in size and

196


statistically significant. Although this model avoids the colinearity problem, its drawback is that it does not control for Chinese equity share when examining the effect of the ownership for all other types of Chinese parent company. For these reasons, this model is not reported here. 9. All other firm characteristics and contextual variables are assumed to be either in the reference category or at the sample average. 10. See http://news.sohu.com/20060809/n244708258.shtml 11. These might even change the workers’ perceptions and definitions of ‘‘labor disputes.’’ It is possible that those firms with formal grievance procedures have a different reporting system and tend to report more labor disputes, whereas those firms with indigenous trade union and the worker’s congress tend to mediate most disputes and report only significant events. Although this scenario may raise the concern on the data collection bias, it still reflects the different approaches to dealing with labor disputes and is consistent with our general argument here.

ACKNOWLEDGMENTS This work was supported, in part, by funds provided by The University of North Carolina at Charlotte. The authors are especially grateful to Dr. Peter S. K. Chi, Professor Emeritus of Cornell University, for generously providing us with the data. We also thank Charles Brody, Lisa Keister, Xueguang Zhou, and an anonymous reviewer for their helpful comments.

REFERENCES Bae, J., Chen, S. J., Wan, D., Wai, T., Lawler, J. J., & Walumbwa, F. O. (2003). Human resource strategy and firm performance in Pacific rim countries. International Journal of Human Resource Management, 14(8), 1308–1332. Bae, J., & Lawler, J. (2000). Organizational and HRM strategies in Korea: Impact on firm performance in an emerging economy. Academy of Management Journal, 43(3), 502–517. Baron, J. N., & Kreps, D. M. (1999). Strategic human resources: Framework for general managers. New York, NY: Wiley. Becker, B., & Gerhart, B. (1996). The impact of human resource management on organizational performance: Progress and prospects. Academy of Management Journal, 39(4), 779–801. Blau, P. (1970). A formal theory of differentiation in organizations. American Sociological Review, 35(2), 201–218. Brinton, M., & Nee, V. (Eds). (1998). The new institutionalism in sociology. New York, NY: Russell Sage. Brown, R. (2006). China’s collective contract provisions: Can collective negotiations embody collective bargaining? Duke Journal of Comparative & International Law, 16(1), 31–77.


197

Burawoy, M., & Krotov, P. (1992). The Soviet transition from socialism to capitalism: Worker control and economic bargaining in the wood industry. American Sociological Review, 57(1), 16–38. Chan, A. (1998). Labor relations in foreign-funded ventures, Chinese trade unions, and the prospect for collective bargaining. In: G. O’Leary (Ed.), Adjusting to Capitalism: Chinese Workers and the State (pp. 122–149). New York, NY: M.E. Sharpe. Chan, A., & Senser, R. A. (1997). China’s troubled workers. Foreign Affairs, 76(2), 104–117. Chi, P. S. K., & Kao, C. (1995). Foreign investment in China: A new data set. China Economic Review, 6(1), 149–155. Child, J. (1994). Management in China during the Age of Reform. Cambridge, England: Cambridge University Press. Chinese Commerce Yearbook. (2005). Chinese Commerce Yearbook. Beijing, China: China Commerce and Trade Press. Chinese Statistical Yearbook. (1996 & 2006). Chinese Statistical Yearbook. Beijing, China: Chinese Statistical Press. Clarke, S. (2005). Post-socialist trade unions: China and Russia. Industrial Relations Journal, 36(1), 2–18. Cooke, F. L. (2004). Foreign firms in China: Modeling HRM in a toy manufacturing corporation. Human Resource Management Journal, 14(1), 31–52. Czarniawska, B., & Joerges, B. (1996). Travels of ideas. In: B. Czarniawska & G. Sevoń (Eds), Translating organizational change (pp. 13–48). New York, NY: Walter de Gruyter. DiMaggio, P., & Powell, W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48(140), 147–160. Doorewaard, H., & van Bijsterveld, M. (2001). The osmosis of ideas: An analysis of the integrated approach to IT management from a translation theory perspective. Organization, 8(1), 55–76. Edelman, L. (1990). Legal environments and organizational governance: The expansion of due process in the American workplace. American Journal of Sociology, 95(6), 1401–1440. Edelman, L. (1992). Legal ambiguity and symbolic structures: Organizational mediation of civil rights law. American Journal of Sociology, 97(6), 1531–1576. Frenkel, M. (2005). The politics of translation: How state-level political relations affect the cross-national travel of management ideas. Organization, 12(2), 275–301. Friedland, R., & Alford, R. (1991). Bringing society back in: Symbols, practices, and institutional contradictions. In: W. Powell & P. DiMaggio (Eds), The new institutionalism in organizational analysis (pp. 232–263). Chicago: University of Chicago Press. Gallagher, M. E. (2005). Contagious capitalism: Globalization and the politics of labor in China. Princeton, NJ: Princeton University Press. Goodall, K., & Warner, M. (1997). Human resources in Sino-foreign joint ventures: Selected case studies in Shanghai, compared with Beijing. International Journal of Human Resource Management, 8(5), 569–594. Guthrie, D. (1997). Between markets and politics: Organizational responses to reform in China. American Journal of Sociology, 102(5), 1258–1304. Guthrie, D. (1999). Dragon in a three-piece suit: The emergence of capitalism in China. Princeton, NJ: Princeton University Press. Guthrie, J. (2001). High-involvement work practices, turnover, and productivity: Evidence from New Zealand. Academy of Management Journal, 44(1), 180–190.

198


Hannan, M., & Freeman, J. (1989). Organizational ecology. Cambridge, MA: Harvard University Press. Hong, N. S., & Warner, M. (1998). China’s trade unions and management. New York, NY: St. Martin’s Press. Hull, F., Azumi, K., & Wharton, R. (1988). Suggestion rates and sociotechnical systems in Japanese versus American factories. IEEE Transactions on Engineering Management, 35(1), 11–24. Huselid, M. A. (1995). The impact of human resource management practices on turnover, productivity, and corporate financial performance. Academy of Management Journal, 38(3), 635–672. Kalev, A., Dobbin, F., & Kelly, E. (2006). Best practices or best guesses? Diversity management and the remediation of inequality. American Sociological Review, 71(4), 589–617. Keister, L. A. (1998). Engineering growth: Business group structure and firm performance in China’s transitional economy. American Journal of Sociology, 104(2), 404–440. Keister, L. A. (2001). Exchange structures in transition: Lending and trading relations in Chinese business groups. American Sociological Review, 66(3), 336–360. Kelly, E., & Dobbin, F. (1999). Civil rights law at work: Sex discrimination and the rise of maternity leave politics. American Journal of Sociology, 105(2), 455–492. Kostova, T., & Roth, K. (2002). Adoption of an organizational practice by subsidies of multinational corporations: Institutional and relational effects. Academy of Management Journal, 45(1), 215–233. Lee, C. K. (1995). Engendering the worlds of labor: Women workers, labor markets, and production politics in the south China economic miracle. American Sociological Review, 60(3), 378–397. Lee, C. K. (1999). From organized dependence to disorganized despotism: Changing labour regimes in Chinese factories. The China Quarterly, 157(1), 44–71. Lee, C. K. (2002). From the specter of Mao to the spirit of the law: Labor insurgency in China. Theory and Society, 31(2), 189–228. Levine, M. J. (1997). Worker rights and labor standards in Asia’s four new tigers. New York, NY: Plenum Press. Martin, G., & Beaumont, P. (1998). Diffusing ‘best practices’ in multinational firms. Prospects, practice and contestation. International Journal of Human Resource Management, 9(4), 671–695. Meyer, J., & Rowan, B. (1977). Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology, 83(2), 340–363. Nee, V. (1989). A theory of market transition: From redistribution to markets in state socialism. American Sociological Review, 54(5), 663–681. Nee, V. (1996). The emergence of a market society: Changing mechanisms of stratification in China. American Journal of Sociology, 101(4), 908–949. Newman, K., & Nollen, S. (1996). Culture and congruence: The fit between management practices and national culture. Journal of International Business Studies, 27(4), 753–779. Robert, C., Probst, T. M., Martocchio, J. J., Drasgow, F., & Lawler, J. J. (2000). Empowerment and continuous improvement in the United States, Mexico, Poland, and India: Predicting fit on the basis of the dimensions of power distance and individualism. Journal of Applied Psychology, 5(5), 643–658. Rosenzweig, P. M., & Nohria, N. (1994). Influences on human resource management practices in multinational corporations. Journal of International Business Studies, 25(2), 229–251.


199

Scott, W. R. (1995). Introduction: Institutional theory and organizations. In: W. R. Scott & S. Christensen (Eds), The institutional construction of organizations (pp. xi–xxiii). Thousand Oaks, CA: Sage Publications. Spenner, K., & Jones, D. (1998). Social economic transformation in Bulgaria: An empirical assessment of the merchant capitalism thesis. Social Forces, 76(3), 937–965. Spenner, K. I., Suhomlinova, O. O., Thore, S. A., Land, K. C., & Jones, D. C. (1998). Strong legacies and weak markets: Bulgarian state-owned enterprises during early transition. American Sociological Review, 63(4), 599–617. Stark, D. (1996). Recombinant property in East European capitalism. American Journal of Sociology, 101(4), 993–1027. Stinchcombe, A. (1965). Social structure and organization. In: J. March (Ed.), Handbook in organizations (pp. 142–193). Chicago: Rand McNally & Co. Sutton, J., & Dobbin, F. (1996). The two faces of governance: Response to legal uncertainty to US firms, 1955 to 1985. American Sociological Review, 61(5), 794–811. Tang, W., & Parish, W. (1999). Chinese urban life under reform. Cambridge, UK: Cambridge University Press. Taylor, B. (1999). Patterns of control within Japanese manufacturing plants in China: Doubts about Japanization in Asia. Journal of Management Studies, 36(6), 853–873. Vogel, E. (1989). One step ahead in China: Guangdong under reform. Cambridge, MA: Harvard University Press. Walder, A. (2002). Market and income inequality in rural China: Political advantage in an expanding economy. American Sociological Review, 67(2), 231–253. Warner, M. (1995). The management of human resources in Chinese industry. New York, NY: St. Martin’s Press. Williamson, O. E. (1981). The economics of organizations: The transaction cost approach. American Journal of Sociology, 87(3), 548–577. Williamson, O. E. (1985). The economic institutions of capitalism. New York, NY: Free Press. Xiao, Z., & Bjo¨rkman, I. (2006). High commitment work systems in Chinese organizations: A preliminary measure. Management and Organization Review, 2(3), 403–422. Zhou, X. (2000). Economic transformation and income inequality in urban China: Evidence from panel data. American Journal of Sociology, 105(4), 1135–1174. Zhou, X., Zhao, W., Li, Q., & Cai, H. (2003). Embeddedness and contractual relationships in China’s transitional economy. American Sociological Review, 68(1), 75–102. Zilber, T. B. (2002). Institutionalization as an interplay between actions, meanings, and actors: The case of a rape crisis center in Israel. Academy of Management Journal, 45(1), 234–254.

200


APPENDIX Table A1. Comparison of 1996 Panel with Sample at the Beginning of the Survey (1993–1994) and National Statistics Based on Chinese Statistical Yearbook. TCEIES, TCEIES, 1996 Panel 1993–1994

Investor origin: Newly industrialized economies: Hong Kong Taiwan Macao Korea Singapore Other southeast Asian countries Western countries: USA Canada Europe Oceania Japan

Yearbook 1996, FDI in 1994

Yearbook 1996, FIE in 1994

83%

86%

77%

–

55% 20% 2% 2% 2% 1%

54% 21% 2% 1% 2% 1%

58% 10% 2% 2% 3% 2%

– – – – – –

11% 7% 1% 2% 1% 6%

10% 6% 1% 2% 1% 4%

13% 7% 1% 5% 1% 6%

– – – – – –

50%

48%

–

–

Ownership of Chinese parent company: State 36% 37% Collective and TVE 46% 43% Private and others 7% 7% None (wholly foreign 12% 13% owned)

– – – –

– – – –

Firm size

Chinese equity share

181.57

187.96

–

–

Industry: Labor intensive Capital intensive Others

49% 48% 3%

51% 49% 0%

– – –

– – –

Age (years)

6.35

4.33

–

–

201


Table A1.

(Continued )

TCEIES, TCEIES, 1996 Panel 1993–1994 Region: Southeast coast East coast North coast Inland Beijing

33% 24% 21% 10% 12%

38% 21% 19% 10% 12%

Yearbook 1996, FDI in 1994

Yearbook 1996, FIE in 1994

42% 22% 17% 17% 4%

37% 20% 18% 20% 4%

A MICRO–MACRO LINK DURING INSTITUTIONAL TRANSITIONS Mike W. Peng and J. Martina Quan ABSTRACT One of the leading themes emerging out of recent management and organization research on China is the work on the micro–macro link – specifically, the connection between micro, interpersonal connections, ties, and networks on the one hand, and macro, interorganizational relationships, firm strategies, and performance on the other hand. This chapter provides an overview of the literature on the micro–macro link during China’s institutional transitions. Based on a systematic search of the literature, we review 22 papers in nine leading journals that have empirically investigated the micro–macro link, with a focus on the antecedents, contingencies, and outcomes of managerial ties and interlocking directorates. We also propose how the network structure of managerial ties will evolve from cohesion to structural holes in different phases of China’s institutional transitions. We conclude with a brief overview of the influence of China studies on research in other contexts and with a call for future research deepening our understanding of the crucial micro–macro link during institutional transitions.

Management and organization research on Asia is a relatively recent phenomenon (Bruton & Lau, 2008; Peng, 2007).1 Research on management Work and Organizations in China after Thirty Years of Transition Research in the Sociology of Work, Volume 19, 203–224 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1108/S0277-2833(2009)0000019010

203

204

MIKE W. PENG AND J. MARTINA QUAN

and organization in China2 is more recent. Yet, research on China has taken off since the late 1990s (Peng, Au, & Wang, 2001a; Peng, Lu, Shenkar, & Wang, 2001b; Quer, Claver, & Rienda, 2007). One of the leading themes emerging out of this research is what Peng and Luo (2000) label the ‘‘micro– macro link’’ – specifically, the connection between micro, interpersonal connections, ties, and networks on the one hand, and macro, interorganizational relationships, strategies, and performance on the other hand. While research on the elusive Chinese guanxi had always existed before (Yang, 1994), such research was largely relegated to the domain of area studies, published in China-specific journals such as China Quarterly (Guthrie, 1998; Walder, 1989), and set apart from mainstream, disciplinebased research. While acknowledging that such earlier scholarship is ‘‘superb and worthy of respect on its own terms,’’ Walder (2002, p. 4) criticizes such research of being ‘‘nondisciplinary.’’ It is the more theoretically oriented and empirically rigorous new research, rallied around the micro–macro link, that has propelled a new generation of China research into core disciplinary journals in management (such as the Academy of Management Journal (AMJ; Peng & Luo, 2000) and Academy of Management Review (AMR; Peng & Heath, 1996)) and sociology (such as the American Journal of Sociology (AJS; Keister, 1998) and American Sociological Review (ASR; Keister, 2001; Nee, 1989; Walder, 2003)). Such research has not only placed China on the center stage of social science inquiry (Walder, 2002), but also directly contributed to theory building in these disciplines (Li & Peng, 2008; Peng, 2003; Peng & Heath, 1996; Peng, Wang, & Jiang, 2008). As a literature rapidly develops and accumulates, it is helpful to take stock of our accomplishments, discuss limitations, and point out future directions. Therefore, in this chapter, we focus on a key question: How has the management and organization literature on the micro–macro link in China during its institutional transitions evolved?

LINKING THE MICRO AND THE MACRO IN INSTITUTIONAL TRANSITIONS A widely used, non-technical definition of institutions is that they are ‘‘rules of the game’’ (North, 1990). Consequently, institutional transitions are ‘‘fundamental and comprehensive changes introduced to the formal and informal rules of the game that affect organizations as players’’ (Peng, 2003, p. 275). The scale and scope of China’s institutional transitions are the

A Micro–Macro Link during Institutional Transitions

205

subject of thousands of articles3 and hundreds of books, and are clearly beyond the scope of this chapter. The more focused literature that we work with here deals with a crucial question: ‘‘How do firms play the new game when the new rules are not completely known?’’ (Peng, 2003, p. 283). The answer typically boils down to various forms of informal relationships centered on interpersonal connections, ties, and relationships that managers cultivate. Given that formal market-supporting laws, rules, and regulations are underdeveloped but firms need to grow during the ‘‘sink-orswim’’ reform era, firms are often forced to rely on such informal micro ties to cultivate interorganizational relationships through alliances and networks in order to attain better firm performance (Peng & Heath, 1996). According to social network theory, managers with better interpersonal connections tend to earn more income, get more frequent promotions, and have better careers – in short, there is a clear micro–micro link between ties and rewards. This micro–micro link has been widely documented in China (Li & Walder, 2001) and elsewhere (Burt, 1997; Granovetter, 1985; Walder & Nguyen, 2008). What is interesting, but also challenging, is the identification and documentation of the micro–macro link, connecting micro, personal-level relationships with macro, organizational-level actions (such as firm strategy) and outcomes (such as firm performance). Scholars interested in such a micro–macro link have paid a great deal of attention to the effects of interlocking directorates in Western corporations. Despite decades of research, evidence on a micro–macro link between board interlocks and firm performance has remained largely inconclusive (Mizruchi, 1996). Frustrated scholars note, ‘‘Perhaps alternative types of interorganizational relations y have a more consistent impact on [firm] performance, but this has yet to be demonstrated empirically’’ (Mizruchi & Galaskiewicz, 1994, p. 241). It is China research that has pushed the research agenda on the micro– macro link further. Keister (1998, p. 428) first reports that ‘‘in Chinese business groups, interlocking directorates have a positive effect on firm performance and productivity.’’ While Keister’s (1998) work is a breakthrough, board interlocks represent one special type of networks, which represent formal interorganizational relationships (Ren, Au, & Birtch, 2009). Since the number of outside directors on any given company’s board is limited (Peng, 2004), there are numerous other interpersonal network ties not captured by data on board interlocks. Focusing on managerial ties (but not board interlocks), Peng and Luo (2000) use survey data to demonstrate, for the first time, that micro interpersonal ties directly impact macro organizational performance. Peng and Luo (2000) not only empirically

206


establish the ‘‘micro–macro link,’’ but also coin this term.4 Since then, a large number of studies have followed up, leading to a proliferation of work centered on the micro–macro link. Next, we trace how research on the micro–macro link during China’s institutional transitions has evolved and how this research has influenced work beyond the China context.

METHODOLOGY Journal Selection We first identify the journals to be included in the review. We choose the following 11 journals: AMJ, Administrative Science Quarterly (ASQ), AJS, ASR, Asia Pacific Journal of Management (APJM), Journal of International Business Studies (JIBS), Journal of Management (JM), Journal of Management Studies (JMS), Organization Science (OSc), Organization Studies (OSt), and Strategic Management Journal (SMJ). These are generally regarded as the leading outlets in management and organization research in which China scholars aspire to publish their work (Peng et al., 2001a, 2001b; Quer et al., 2007). Because we choose to focus on empirical research, leading journals such as the AMR that exclusively publish theoretical (nonempirical) work are not considered.

Paper Selection After identifying these journals, we then choose the papers to be reviewed. We focus on two fields: (1) managerial ties and (2) interlocking directorates. Managerial ties are defined as ‘‘executives’ boundary-spanning activities and their associated interactions with external entities’’ (Geletkanycz & Hambrick, 1997, p. 654). Following Peng and Luo (2000), we identify two types of managerial ties in China. The first type is ties with managers at other firms (referred as ‘‘business ties’’ in Tables 1 and 2), represented by cross-firm ties among executives, such as the managerial ties between suppliers and buyers, between competitors, and so on. The second type is managerial ties with government officials (referred as ‘‘government ties’’ in Tables 1 and 2). For interlocking directorates, we employed Mizruchi’s definition: ‘‘one person affiliated with one organization sits on the board of directors of another organization’’ (1996, p. 271). In addition to managerial ties and interlocking directorates, we also examine the social capital

AMJ

AJS

AMJ

AMJ

JM

SMJ

SMJ

Keister (1998)

Peng and Luo (2000)

Li and AtuaheneGima (2001)

Luo (2001)

Park and Luo (2001)

Luo (2003)

Journal

Xin and Pearce (1996)

Paper

Sample and Method

Mail survey questionnaire to executives in 364 firms in manufacturing industries in China

Mail survey of 128 firms in China

Mail survey of 168 MNE subsidiaries in the Yangtze River Delta and the Pearl River Delta

Onsite interview with top managers in 184 new technology ventures in Beijing

Interview-based panel data on China’s largest 40 business groups and the financial performance of their 535 member firms Mail survey questionnaire to 127 top managers in China

Business and government ties



Government ties


Interlocking directorates


Ties

Main Findings

The positive effect of managerial ties on firm performance is contingent on firm and industry factors Political networking does not moderate the relationship between product innovation strategy and the performance of new technology ventures Having ties with managers at other businesses and with government officials is positively correlated with an MNE subsidiary’s local responsiveness Institutional, strategic, and organizational factors determine guanxi utilization, which leads to higher firm performance Industrial conditions influence the level of executives’ managerial networking and the relationship is moderated by firm’s strategic proactiveness

Executives in private companies are more dependent on guanxi than executives in state-owned or collective-hybrid companies There is a positive relationship between the presence and predominance of interlocking directorates and firm performance

Studies on the Micro–Macro Link in Mainland China.

Structured interview with 32 managers in various ownership companies in China

Table 1.

A Micro–Macro Link during Institutional Transitions 207

APJM

APJM

SMJ

JIBS

APJM

Wu and Leung (2005)

Li and Zhang (2007)

Zhou et al. (2007)

Gao et al. (2008)

Journal

Li (2005)

Paper

Interview-based questionnaire surveys with top managers of 129 small and medium enterprises in Zhejiang Onsite survey of 174 local and foreign companies in China

Onsite interview with senior managers in 181 foreign-invested enterprises (FIE) operating in China Structured questionnaire to 177 CEOs of small- and mediumsized enterprises in rural China Onsite questionnaire interview of 184 new technology ventures in Beijing

Sample and Method

Business ties and university ties


Government ties



Ties

Table 1. (Continued )

Firms’ strategic orientations influence the formation of managerial networks, which positively influence the performance of FIEs There is a micro–macro link from a manager’s value of reciprocity, to trust and to firm performance Managers’ political networking is positively related to new venture performance. But the positive relationship is not moderated by ownership or dysfunctional competition Guanxi-related social networks mediate the relationship between inward and outward internationalization and firm performance Absorptive capacity moderates the positive effect of managerial ties on firm’s innovativeness

Main Findings

208 MIKE W. PENG AND J. MARTINA QUAN

SMJ

JIBS

APJM

APJM

Li et al. (2008)

Luk et al. (2008)

Zhang and Li (2008)

Ren et al. (2009)

Archival data of 949 publicly listed companies

Questionnaire survey of 163 firms from the textile and IT clusters in Guangdong province

Mail survey of 189 manufacturing companies in various industries located in Beijing, Shanghai, and Guangzhou

Onsite survey of 280 firms in manufacturing industries in China





Managerial ties have different impact on the performance of foreign firms and domestic firms and the relationship is contingent on industrial factors Guanxi with government officials positively affect administrative innovativeness. Guanxi with managers at other firms has a direct positive relationship with firm performance and enhances the positive effect of administrative innovativeness on firm performance, but it does not affect product-related innovativeness Managerial ties are classified into four types, among which business ties within a cluster generate strongest influence on sales growth Chinese interlocking networks share similarities and differences with their Western counterparts. Interlocking ties occurred mainly in smaller business groups and were more prevalent among industrial peers


JMS

APJM

OSt

APJM

Peng, Au et al., 2001; Peng, Lu et al., 2001

C. Chung (2005)

H. Chung (2006)

Journal

Lee and Tsang (2001)

Paper

Sample and Method

Archival data of Taiwan’s largest 100 business groups

Case studies of five major Taiwanese business groups and demographic data of 150 Taiwanese group founders

Government ties



Business ties

Ties

Main Findings The frequency and breadth of an entrepreneur’s networking is positively related to venture growth and the relationship is contingent on firm size Compared with non-MNEs, MNEs have more densely connected interlocks, occupy more central locations in the interlocks network, and appoint more military directors Personalistic network is a necessary but insufficient condition to explain the entrepreneurship of Taiwanese group founders. Personalistic networks influenced group diversification in the early stages but the influence decreased with the development of market institutions Possession of managerial ties with government determines a business group’s decision to enter a deregulated industry

Studies on the Micro–Macro Link in other Contexts.

Questionnaire survey with 168 Chinese entrepreneurs in smalland medium-sized businesses in Singapore Archival data on the top 200 publicly listed firms in Thailand

Table 2.

210 MIKE W. PENG AND J. MARTINA QUAN

SMJ

ASQ

JIBS

Acquaah (2007)

Siegel (2007)

Luk et al. (2008)

Mail survey of 203 manufacturing companies in various industries in Hong Kong.

Archival data of 665 firms listed on the Korea Stock Exchange

Questionnaire survey of 106 manufacturing and service firms operating in Ghana


Government ties

Business, government, and community ties

Social capital developed from ties with managers at other firms, government officials, and community leaders, and is positively correlated with firm’s performance. This positive relationship is contingent on firm’s strategic orientations Having ties with regime in power increased the rate at which South Korean companies formed cross-border strategic alliances, but having ties with political enemies of the regime in power decreases that rate. Ties with government remained valuable after long period of liberalization in Korea Guanxi with managers at other firms positively affects product-related innovativeness, but does not directly affect firm performance. Guanxi with government officials does not affect administrative innovativeness


212


embedded in them. The term social capital has been defined in different ways. Despite this diversity, a general definition is ‘‘y the sum of resources, actual or virtual, that accrue to an individual or a group by virtue of possessing a durable network of more or less institutionalized relationships of mutual acquaintance and recognition’’ (Bourdieu & Wacquant, 1992, p. 119). We use the Social Science Citation Index (SSCI) database as our primary search engine, since all the journals, except APJM, are included in SSCI.5 For APJM, we use the search engine provided by the journal’s website (www.springer.com/10490) to search within that journal. After searching SSCI and APJM website, we use EBSCO to double check our search results. In SSCI, we use the search field of topic to identify our target papers. Topic includes the following fields: title, abstract, author keywords, and keywords plus. To identify papers about managerial ties, we enter the following four combinations of words into topic column, respectively: (1) guanxi, (2) ties AND China, (3) network6 AND China, and (4) social capital AND China. Then, we content-analyze each paper in the list from search results and choose those that meet the following requirements: (1) The paper investigates informal personalized ties between managers in two firms or between a manager and a government official. So the ties are at the individual level. We exclude papers investigating ties at the firm or organizational level, such as contractual ties. (2) The paper investigates personal ties with managers at other firms or government officials. We exclude those examining kinship ties, friendship ties, and other non-work-related ties. (3) The ties should occur outside the focal person’s organization. We exclude those investigating ties within a firm. (4) The antecedents, contingencies, and outcomes of ties are at the firm level or above. We exclude those investigating individuallevel antecedents, contingencies, and outcomes, such as personality traits and venture capitalists’ decisions. (5) We only consider papers with an empirical content. Based on these five criteria, we identify 14 articles about managerial ties. Among them, nine articles relate to mainland China, one is a comparative study between mainland China and Hong Kong, and four relate to other contexts but include China either in abstract or in keywords. The last set of these four articles includes China as the topic word (per SSCI). Therefore, they are clearly influenced by China studies and we include them in our review. For papers about interlocking directorates, we enter interlock AND China as the topic and identify one paper. In total, there are 15 papers about managerial ties or interlocking directorates that are published in


213

eight journals. Among these journals, SMJ published five papers, AMJ published three, JIBS published two, and ASQ, AJS, JM, JMS, and OSt each published one. For APJM, we follow the same procedure and criteria to identify papers. The only difference is that the APJM website does not allow us to search by ‘‘topic’’ function. Therefore, we search within both abstracts and titles. In doing so, we have identified five empirical papers about managerial ties and two about interlocking directorates. Among the five managerial ties papers, four relate to mainland China and one relates to other contexts. Among the two interlocking papers, one relates to mainland China and the other relates to other contexts. Finally, we search in the EBSCO database by the same procedure and criteria to double check if we have omitted any paper. No new paper is added after we search by EBSCO. Overall, in total, 22 papers are included in our review, 19 about managerial ties and 3 about interlocking directorates (see Tables 1 and 2). The trend is clearly growing (Fig. 1).

25

Number of Articles

20

15

10

5

0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Number of Articles

Fig. 1.

Micro–Macro Link Articles related to China. Published in Nine Journals during 1996–2008 (Cumulative).

214


WHAT IS KNOWN ABOUT THE MICRO–MACRO LINK The relationships explored in the 22 papers are summarized according to the links depicted in Figs. 2 and 3. Extending Li’s (2005) conceptual framework, we synthesize the literature on managerial ties in Fig. 2, including the antecedents, contingencies, and outcomes of two types of managerial ties. Link 1 depicts how firm characteristics and industry environment influence the formation of managerial ties both with managers at other firms and with government officials. Firm characteristics, including ownership, location, strategic orientation, size, technological skills, management skills, and age, have different influences on the likelihood and intensity of managerial ties. Xin and Pearce (1996) find that compared with executives in state-owned or collective-hybrid companies, executives in private companies consider guanxi more important, depend more on guanxi, build more ties with government officials, give more unreciprocated gifts, and trust their ties more. Their findings show the effect of firm ownership – specifically, private versus nonprivate ownership – on ties formation. Park and Luo (2001) support the ownership effect by differentiating stateowned with non-state-owned firms and find the latter are more likely to form both business and government ties. In addition, they report that firm’s location and strategic orientation also significantly influence the formation of both types of ties. However, firm size, organizational skills, and managerial skills are only correlated with the formation of business ties, not government ties. Finally, they find firm age to have no influence on tie formation. Consistent with the findings of Park and Luo (2001), Li (2005) shows that firm’s strategic orientation – specifically market orientation, technology orientation, and entrepreneurial orientation – is correlated with the likelihood of tie formation. Technology orientation exerts an opposite effect on different type of ties while market orientation and entrepreneurial orientation generate the same effect on both ties. Luo (2003) is the first paper that investigates how industry environment influences the level of managerial networking. It finds that among the five industrial attributes, structural uncertainty, industrial regulation, competitive pressure, and production capacity utilization are all related to networking while industrial growth is not. Link 2 shows the interacting effect between firms’ characteristics and industry environment on ties formation. Luo (2003) reports that a firm’s

1 Managerial ties with government officials

Managerial ties with managers at other firms

Ties

Fig. 2.

6

5

6

Firm’s characteristics 1. Ownership 2. Size 3. Strategic orientation* 4. Absorptive capacity

4

3

5

Industry environment 1. Business sector 2. Industry growth 3. Competitive intensity 4. Structural uncertainty

Contingencies

Outcomes

1. Firm performance 2. Business group diversification* 3. MNE subsidiaries’ local responsiveness 4. Mediating the relationship between internationalization and firm performance 5. Administrative innovativeness 6. Formation of cross-border strategic alliances* 7. Decision to enter a deregulated industry*

1. Firm performance 2. Business group diversification* 3. MNE subsidiaries’ local responsiveness 4. Mediating the relationship between internationalization and firm performance 5. New venture growth* 6. Product-related innovativeness* 7. Moderating the relationship between administrative innovativeness and firm performance

Research on the Micro–Macro Link Centered on Managerial Ties.

* The relationship is based on studies outside of Mainland China

Industry environment 1. Structural uncertainty 2. Industrial regulation 3. Competitive pressure 4. Production capacity utilization

2

Firm’s characteristics 1. Ownership 2. Location 3. Strategic orientation 4. Size 5. Technological skills 6. Managerial skills

Antecedents


216


Firm ownership* Interlocking directorates

Firm performance

Business group * The relationship is based on studies outside of Mainland China

Fig. 3.

Research on the Micro–Macro Link Centered on Interlocking Directorates.

strategic proactiveness moderates the effects of industry environment on managerial networking, while Li (2005) finds that industry’s competitive intensity moderates the relationship between a firm’s strategic orientation and managerial networking. Link 3 shows that ties with managers at other firms lead to several firmlevel outcomes. The most frequently discussed is firm performance, captured by various measures (Li, 2005; Li, Poppo, & Zhou, 2008; Luk et al., 2008; Park & Luo, 2001; Peng & Luo, 2000; Wu & Leung, 2005; Zhang & Li, 2008). In addition to the effect on firm performance, managerial ties with managers at other firms also lead to new venture growth (Lee & Tsang, 2001), MNE subsidiaries’ local responsiveness (Luo, 2001), firm’s innovation (Gao, Xu, & Yang, 2008; Luk et al., 2008), and business group diversification (C. Chung, 2005). Ties with managers at other firms also mediate the relationship between internationalization and firm performance (Zhou, Wu, & Luo, 2007) and moderate the relationship between administrative innovativeness and firm performance (Luk et al., 2008). Link 4 shows that ties with government officials lead to several firm-level outcomes. Similar as the ties with managers at other firms, ties with government officials led to higher firm performance (Li, 2005; Li et al., 2008; Peng & Luo, 2000; Park & Luo, 2001; Wu & Leung, 2005; Zhang & Li, 2008), increased MNE subsidiaries’ local responsiveness (Luo, 2001), promoted business group diversification (C. Chung, 2005), and mediated the relationship between internationalization and firm performance (Zhou et al., 2007). The unique value of government ties lies in the effect on administrative innovativeness (Luk et al., 2008), the formation of cross-border strategic alliances (Siegel, 2007), and a business group’s decision to enter a deregulated industry (H. Chung, 2006). Link 5 shows how industry environment moderates the relationship between managerial ties and firm-level outcomes. The positive effect of managerial ties on firm performance is contingent on the business sector that a firm is in (Peng & Luo, 2000), the industry growth (Peng & Luo, 2000),


217

and the competitive intensity and structural uncertainty that a firm faces in its industry (Li et al., 2008). Link 6 shows how a firm’s characteristics moderate the relationship between managerial ties and firm-level outcomes. The positive effect of managerial ties on firm performance is contingent on firm ownership (Li et al., 2008; Peng & Luo, 2000), firm size (Peng & Luo, 2000), and strategic orientation (Acquaah, 2007). In addition, the positive effect of entrepreneur’s networking on venture growth is contingent on firm size (Lee & Tsang, 2001). Finally, the positive effect of managerial ties on firm’s innovativeness is contingent on firm’s absorptive capacity (Gao et al., 2008). Fig. 3 depicts the relationship between interlocking directorates and the antecedents and outcomes. Interlocking directorates are positively associated with firm performance (Keister, 1998). They occur mainly in smaller business groups and are more prevalent among industrial peers (Ren et al., 2009). Compared with non-MNEs, MNEs have more densely connected interlocks, occupy more central locations in the interlocks network, and appoint more military directors (Peng et al., 2001a, 2001b).

THEORETICAL ARGUMENTS FOR FURTHER RESEARCH Systematically reviewing the micro–macro literature not only helps us take stock in terms of what has been done, but also helps identify missing gaps that will be useful in future research. As we can see in Fig. 2, empirical research has comprehensively investigated the antecedents, contingencies, and outcomes of managerial ties. Peng (2003) argues that the importance of managerial ties will decline as market-supporting institutions are better developed. However, no empirical research has employed a longitudinal approach to examine the evolution of managerial ties (Hite & Hesterly, 2001). Peng and Zhou (2005) argue conceptually that as government control and intervention decrease and legal frameworks become more comprehensive and effective, the intensity of business-to-business ties and business-togovernment ties will change from strong to weak during institutional transitions. Except an earlier qualitative study by Guthrie (1998), no study has comprehensively tested this claim using longitudinal and quantitative methods. Extending the arguments advanced in Peng (2003) and Peng and Zhou (2005), we propose that the network structure of managerial ties will change

218


in different periods of China’s institutional transitions. As China’s economy develops from the early phase to the late phase, more and more impersonal, rule-based, and arm’s-length transactions will emerge. At the same time, transactions based on personal connections and managerial ties will still play an important role, due to the fact that economic transactions are embedded in numerous social relations (Granovetter, 1985). In the early phase of China’s institutional transitions, managerial ties tend to build into a cohesive network, in which each manager is connected with other members by strong ties. According to Coleman (1990), dense, interconnected networks can breed trust among members and thus prevent opportunistic behaviors. Strong ties can facilitate highly non-codified knowledge transfer (Hansen, 1999) and fine-grained information transfer (Uzzi, 1997). They also promote joint problem-solving, improve members’ ability to quickly utilize market opportunities, and motivate members to share risks with each other (Uzzi, 1997). These benefits not only result from the trust that network closure generates, but also come from the third-party monitoring that cohesive network builds (Granovetter, 1985). Since members are linked with common third parties, an opportunistic behavior may bring serious damage to defectors’ reputation and lead to long-lasting negative consequences (Gulati, 1995). China’s transition economy is characterized by ‘‘weak market structures, poorly specified property rights, and institutional uncertainty’’ (Nee, 1992, p. 4). Since firm behaviors are constrained by high transaction costs, underdeveloped markets, and strong government intervention (Boisot & Child, 1996; Keister, 2009), firms have to cultivate personal connections (guanxi) rather than relying on contracts and legal framework to implement transactions. Cohesive networks can protect members’ assets by specifying property rights, and provide members with valuable resources that are not accessible through the underdeveloped strategic factor market. The normative pressures generated by cohesive networks help reduce opportunistic behaviors and thus decrease transaction costs. In all, managerial ties tend to build into cohesive networks due to the high environmental uncertainty and underdeveloped market in the early phase of China’s institutional transitions. In contrast, in the late phase of institutional transitions, the structure of managerial networks will become rich in structural holes. According to Burt (1992, 1997), people who occupy structural hole positions can acquire information benefits and control benefits by bridging the disconnected members in a social network. The weak ties that bridge these disconnected members can provide brokers with diverse and novel ideas that are socially distant from them (Granovetter, 1972; Burt, 1992). In addition to better


219

access to information, advantages in negotiations and recognition of opportunities also result from structural hole positions (Burt, 1992, 1997). In the late phase of China’s institutional transitions, the improvement of legal framework, the development of property rights, and the diminishing of government intervention greatly reduce uncertainty and facilitate the maturing of a market economy, which enables managers to take advantage of structural hole benefits (Lin, Peng, Yang, & Sun, in press). First, as market develops and transaction becomes more complex and involves more transaction partners (North, 1990), managers are able to build networks with many disconnected members. The lack of connections among members in the network provides broker managers with diversified information, which helps them identify potential opportunities, monitor market dynamisms, and acquire new knowledge and skills. Second, with the development of legal framework and property rights, managers become more willing to conduct arm’s-length transactions (North, 1990). Thus, brokers are able to reduce the search costs of disconnected managers by introducing them to each other and may seek control benefits by manipulating each partner’s behaviors (Lin et al., 2009). In this late phase, cohesive networks may decrease in number, because the gradually developed formal institutions serve many functions of the former cohesive networks, such as inhibiting opportunistic behaviors, reducing transaction uncertainty, and providing necessary resources. Meanwhile, the disadvantages of cohesive networks may outweigh the benefits. Strong ties embedded in a cohesive network tend to circulate redundant information among members, thus providing no novel opportunities (Granovetter, 1972). Cohesive networks also generate normative and conformity pressures that can force members to sacrifice their own benefits (Coleman, 1988; Portes & Sensenbrenner, 1993). The high cohesiveness of a manager’s network may lower the manager’s ability to adapt to the changing requirements of a new task environment (Gargiulo & Benassi, 2000). Finally, the inertia resided in strong ties can restrain members from fully exploring other possibilities (Hansen, 1999). Overall, we argue that the structure of managerial networks will change from a cohesive network to a network rich in structural holes as China transits from planned to market economy. We believe that these theoretical arguments regarding the evolution of managerial networks warrant attention from empirical researchers in the future. Furthermore, future research can examine how the micro-level changes in the structure of managerial networks can influence the macro-level organizational outcomes, such as those discussed in the extant literature.

220


DISCUSSION AND CONCLUSION Overall, management and organization research on China, with a focus on the micro–macro link, has been flourishing since the late 1990s. From a small number of studies appearing in area-studies outlets, research on the micro–macro link has now routinely appeared in leading discipline-based journals in management and sociology. Among several streams of recent China research, the stream on the micro–macro link has become the most widely cited and thus most influential (Quer et al., 2007). Moreover, this research has asserted its influence beyond China. As we can see from Table 2, research on managerial ties and interlocking directorates in China has influenced related studies in other contexts, such as Singapore (Lee & Tsang, 2001), Thailand (Peng et al., 2001a, 2001b), Taiwan (C. Chung, 2005; H. Chung, 2006), Ghana (Acquaah, 2007), South Korea (Siegel, 2007), Hong Kong (Luk et al., 2008), and the United States (Lin et al., 2009). These findings have enriched our understanding of the antecedents of interlocking directorates and the outcomes of managerial ties around the world. Future research may need to compare the different roles that managerial ties play in a variety of contexts. To the best of our knowledge, there are only two comparative studies. Luk et al. (2008) conduct a comparative study between mainland China and Hong Kong and show that managerial ties serve different functions in these two contexts. Lin et al. (2009) compare and contrast China and the United States in the context of alliances and acquisitions. Lin et al. report interesting findings: in China, more centrally located firms are more likely to acquire member firms in an alliance network. But in the United States, more centrally located firms are less likely to engage in similar acquisitions. Clearly, the benefits enjoyed by centrally located firms in China and the United States are different. These intriguing findings warrant further exploration. In addition, future work may explore the impact of informal institutional transitions on the evolution of network ties. For example, how does the change in cognitive pillar and normative pillar (Scott, 1995) influence managers’ perceptions, values, and beliefs? Such a change may in turn affect the ties. Furthermore, the interacting effect between formal institutions and informal institutions on managerial ties may need to be addressed in future research (Peng & Chen, in press; Peng & Jiang, 2005). In conclusion, research on the micro–macro link during China’s institutional transitions has pushed the frontier of management and organization research, and generated excitement among scholars not only interested in


221

China, but also interested in numerous other countries. Despite the progress made, we are far from achieving a complete understanding of the mechanisms underpinning the micro–macro link during institutional transitions. Therefore, more sustained efforts by scholars in and out of China will be necessary in the years to come.

NOTES 1. The leading management journal focusing on Asia, the Asia Pacific Journal of Management, celebrated its 25th anniversary in 2008 (Peng, 2007). 2. In this article, China refers to the People’s Republic of China (or mainland China), excluding Hong Kong, Taiwan, and other Chinese-speaking regions. For research on Greater China other than mainland China, see Peng et al. (2001a, 2001b), and Yeung (2006). 3. See Keister (2009) for a most recent example. 4. Quer et al. (2007, p. 368) report that Peng and Luo (2000) is the most cited empirical paper on China among 180 such papers published in 12 leading journals between 2000 and 2005. 5. Recently, APJM was accepted by SSCI as of 2008. 6. Network represents any word that includes the characters of network, such as network, networks, and networking.

ACKNOWLEDGMENTS This research was supported, in part, by the National Science Foundation (CAREER SES 0552089) and the Provost’s Distinguished Professorship at the University of Texas at Dallas. We thank Lisa Keister for editorial guidance and Erin Pleggenkuhle-Miles for helpful comments.

REFERENCES Acquaah, M. (2007). Managerial social capital, strategic orientation, and organizational performance in an emerging economy. Strategic Management Journal, 28, 1235–1255. Boisot, M., & Child, J. (1996). From fiefs to clans and network capitalism: Explaining China’s emerging economic order. Administrative Science Quarterly, 41, 600–628. Bourdieu, P., & Wacquant, L. (1992). An invitation to reflexive sociology. Chicago: University of Chicago Press. Bruton, G. D., & Lau, C.-M. (2008). Asian management research: Status today and future outlook. Journal of Management Studies, 45, 636–659.

222


Burt, S. (1992). Structural holes. The social structure of competition. Cambridge: Harvard University Press. Burt, S. (1997). The contingent values of social capital. Administrative Science Quarterly, 42, 339–364. Chung, C. (2005). Beyond guanxi: Network contingencies in Taiwanese business groups. Organization Studies, 27, 461–489. Chung, H. (2006). Managerial ties, control and deregulation: An investigation of business groups entering the deregulated banking industry in Taiwan. Asian Pacific Journal of Management, 23, 505–520. Coleman, J. S. (1988). Social capital in the creation of human capital. American Journal of Sociology, 94, S95–S120. Coleman, J. S. (1990). Foundations of social theory. Cambridge: Harvard University Press. Gao, S., Xu, K., & Yang, J. (2008). Managerial ties, absorptive capacity, and innovation. Asia Pacific Journal of Management, 25, 395–412. Gargiulo, M., & Benassi, M. (2000). Trapped in your own net? Network cohesion, structural holes, and the adaptation of social capital. Organization Science, 11, 183–196. Geletkanycz, M. A., & Hambrick, D. C. (1997). The external ties of top executives: Implications for strategic choice and performance. Administrative Science Quarterly, 42, 654–681. Granovetter, M. (1972). The strength of weak ties. American Journal of Sociology, 78, 1360–1380. Granovetter, M. (1985). Economic action and social structure: A theory of embeddedness. American Journal of Sociology, 91, 481–510. Gulati, R. (1995). Social structure and alliance formation pattern: A longitudinal analysis. Administrative Science Quarterly, 40, 619–652. Guthrie, D. (1998). The declining significance of guanxi in China’s economic transition. China Quarterly, 154, 254–282. Hansen, M. (1999). The search transfer problem: The role of weak ties in sharing knowledge across organization sub-units. Administrative Science Quarterly, 44, 82–111. Hite, J. M., & Hesterly, W. S. (2001). The evolution of firm networks: From emergence to early growth of the firm. Strategic Management Journal, 22, 275–286. Keister, L. A. (1998). Engineering growth: Business group structure and firm performance in China’s transition economy. American Journal of Sociology, 104, 404–440. Keister, L. A. (2001). Exchange structures in transition: Lending and trade relations in Chinese business groups. American Sociological Review, 66, 336–360. Keister, L. A. (in press). Organizational research on market transition: A sociological approach. Asia Pacific Journal of Management, 26. Lee, D., & Tsang, E. (2001). The effects of entrepreneurial personality, background and network activities on venture growth. Journal of Management Studies, 38, 583–602. Li, J. J. (2005). The formation of managerial networks of foreign firms in China: The effects of strategic orientations. Asia Pacific Journal of Management, 22, 423–443. Li, H., & Atuahene-gima, K. (2001). Product innovation strategy and the performance of new technology ventures in China. Academy of Management Journal, 44, 1123–1134. Li, Y., & Peng, M. W. (2008). Developing theory from strategic management research in China. Asia Pacific Journal of Management, 25, 563–572. Li, J. J., Poppo, L., & Zhou, K. Z. (2008). Do managerial ties in China always produce value? Competition, uncertainty, and domestic vs. foreign firms. Strategic Management Journal, 29, 383–400.


223

Li, B., & Walder, A. G. (2001). Career advancement as party patronage: Sponsored mobility into the Chinese administrative elite, 1949–1996. American Journal of Sociology, 196, 1371–1408. Li, H., & Zhang, Y. (2007). The role of managers political networking and functional experience in new venture performance: Evidence from China’s transition economy. Strategic Management Journal, 28, 791–804. Lin, Z., Peng, M. W., Yang, H., & Sun, S. L. (in press). How do networks and learning drive M&As? An institutional comparison between China and America. Strategic Management Journal, 30. Luk, C., Yau, O., Sin, L., Tse, A., Chow, R., & Lee, J. (2008). The effects of social capital and organizational innovativeness in different institutional contexts. Journal of International Business Studies, 39, 589–612. Luo, Y. (2001). Determinants of local responsiveness: Perspectives from foreign subsidiaries in an emerging market. Journal of Management, 27, 451–477. Luo, Y. (2003). Industrial dynamics and managerial networking in an emerging market: The case of China. Strategic Management Journal, 24, 1315–1327. Mizruchi, M. S. (1996). What do interlocks do? An analysis, critique, and assessment of research on interlocking directorates. Annual Review of Sociology, 22, 271–298. Mizruchi, M. S., & Galaskiewicz, J. (1994). Networks of interorganizational relations. In: S. Wasserman & J. Galaskiewicz (Eds), Advances in social network analysis (pp. 230–253). Thousand Oaks, CA: Sage. Nee, V. (1989). A theory of market transition: From redistribution to markets in state socialism. American Sociological Review, 54, 663–681. Nee, V. (1992). Organizational dynamics of market transition: Hybrid forms, property rights, and mixed economy in China. Administrative Science Quarterly, 37, 1–27. North, D. (1990). Institutions, institutional change, and economic performance. Cambridge: Harvard University Press. Park, S. H., & Luo, Y. (2001). Guanxi and organizational dynamics: Organizational networking in Chinese firms. Strategic Management Journal, 22, 455–477. Peng, M. W. (2003). Institutional transitions and strategic choices. Academy of Management Review, 28, 275–296. Peng, M. W. (2004). Outside directors and firm performance during institutional transitions. Strategic Management Journal, 25, 453–471. Peng, M. W. (2007). Celebrating 25 years of Asia Pacific management research. Asia Pacific Journal of Management, 24, 385–393. Peng, M. W., Au, K., & Wang, D. (2001a). Interlocking directorates as corporate governance in third world multinationals: Theory and evidence from Thailand. Asia Pacific Journal of Management, 18, 161–181. Peng, M. W., & Chen, H. (in press). Strategic responses to domestic and foreign institutional pressures in the Chinese toy industry. International Studies of Management and Organization. Peng, M. W., & Heath, P. S. (1996). The growth of the firm in planned economies in transition: Institutions, organizations, and strategic choice. Academy of Management Review, 21, 492–528. Peng, M. W., & Jiang, Y. (2005). Entrepreneurial strategies during institutional transitions. In: L. A. Keister (Ed.), Research in the sociology of work: Entrepreneurship (pp. 311–325). Greenwich, CT: JAI.

224


Peng, M. W., Lu, Y., Shenkar, O., & Wang, D. (2001b). Treasures in the China house: A review of management and organizational research on Greater China. Journal of Business Research, 52, 95–110. Peng, M. W., & Luo, Y. (2000). Managerial ties and firm performance in a transition economy: The nature of a micro-macro link. Academy of Management Journal, 43, 486–501. Peng, M. W., Wang, D., & Jiang, Y. (2008). An institution-based view of international business strategy: A focus on emerging economies. Journal of International Business Studies, 39, 920–936. Peng, M. W., & Zhou, J. Q. (2005). How network strategies and institutional transitions evolve in Asia. Asia Pacific Journal of Management, 22, 321–336. Portes, A., & Sensenbrenner, J. (1993). Embeddedness and immigration: Notes on the social determinants of economic action. American Sociological Review, 66, 647–671. Quer, D., Claver, E., & Rienda, L. (2007). Business and management in China: A review of empirical research in leading international journals. Asia Pacific Journal of Management, 24, 359–384. Ren, B., Au, K., & Birtch, T. (2009). China’s business network structure during institutional transitions. Asia Pacific Journal of Management, 26, 219–240. Scott, W. R. (1995). The institutional construction of organizations: International and longitudinal studies. Thousand Oaks: Sage. Siegel, J. (2007). Contingent political capital and international alliances: Evidence from South Korea. Administrative Science Quarterly, 52, 621–666. Uzzi, B. (1997). Social structure and competition in interfirm networks: The paradox of embeddedness. Administrative Science Quarterly, 42, 35–67. Walder, A. G. (1989). Factory and manager in an era of reform. China Quarterly, 118, 242–264. Walder, A. G. (2002). The transformation of contemporary China studies, 1977–2002. University of California International and Area Studies Digital Collection, (http://repositories. cdlib.org/unciaspubs/editedvolumes/3/8 Walder, A. G. (2003). Elite opportunity in transition economies. American Sociological Review, 68, 899–916. Walder, A. G., & Nguyen, G. H. (2008). Ownership, organization, and income inequality: Market transition in rural Vietnam. American Sociological Review, 73, 251–269. Wu, W., & Leung, A. (2005). Does a micro–macro link exist between managerial value of reciprocity, social capital and firm performance? The case of SMEs in China. Asia Pacific Journal of Management, 22, 445–463. Xin, K. R., & Pearce, J. L. (1996). Guanxi: Connections as substitutes for formal institutional support. Academy of Management Journal, 39, 1641–1658. Yang, M. (1994). Gifts, favors, and banquets: The art of social relationships in China. Ithaca: Cornell University Press. Yeung, H. (2006). Change and continuity in Southeast Asian ethnic Chinese business. Asia Pacific Journal of Management, 23, 229–254. Zhang, S., & Li, X. (2008). Managerial ties, firm resources, and performance of cluster firms. Asia Pacific Journal of Management, 25, 615–633. Zhou, L., Wu, W., & Luo, X. (2007). Internationalization and the performance of born-global SMEs: The mediating role of social networks. Journal of International Business Studies, 38, 673–690.

RECRUITING AND DEPLOYING SOCIAL CAPITAL IN ORGANIZATIONS: THEORY AND EVIDENCE Nan Lin, Yanlong Zhang, Wenhong Chen, Dan Ao and Lijun Song ABSTRACT The paper advances the argument that social capital operates on both the supply and demand sides of the labor market. Organizations have significant needs for employees with social capital capacity and skills as they do with human capital. We articulate a theory on why organizations have such needs and how social capital may be differentially and strategically deployed to different positions. Specifically, three types of positions (the top positions, the edge positions, and the exchange-oriented positions) are identified with such needs. We formulated two hypotheses derived from the theoretical articulation: (1) the deploying hypothesis – organizations are expected to strategically recruit and deploy workers with social capital capacity and skills to such key internal and edge positions and (2) the institutional contingency hypothesis – organizations in the more competitive environment (e.g., the private sector) are more likely to show such differential deployment than those in the less competitive Work and Organizations in China after Thirty Years of Transition Research in the Sociology of Work, Volume 19, 225–251 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1108/S0277-2833(2009)0000019011

225

226

NAN LIN ET AL.

environment (e.g., the state sector). The hypotheses were subjected to an empirical examination with a set of firm data from China. Both hypotheses were confirmed. Further, we also found evidence for differential deployment of human capital (education and experience) and hierarchical capital (statuses of prior positions and organizations) in different sectors. We discuss the implications of the theory and findings for future research on organizations in different economic sectors beyond China and how a theory of deploying various types of capital – social capital, human capital, and hierarchical capital – in different economic sectors may be developed.

DEPLOYING SOCIAL CAPITAL IN ORGANIZATIONS Social capital, defined as resources embedded in social networks (Lin, 1982; Bourdieu, 1983/1986; Coleman, 1990; Flap, 1999; Burt, 1992), has been envisioned as a capital that returns benefits to actors in the marketplace (Lin, 2001). Several factors contribute to the effects of social capital: it provides information otherwise inaccessible (Granovetter, 1973); it affords possible influence of a third party on behalf of the actor (Lin, Ensel, & Vaughn, 1981; Burt, 1992); it provides the actor’s social credentials to the larger community where credentials are significant (Collins, 1979); and it reinforces the actor’s self-identity and social recognition (Bourdieu, 1983/1986). That is, it offers relative advantages and added values to an actor in the marketplace. Similar to human capital, it requires investment on the part of an actor. Instead of investing technical skills and knowledge as in the case of human capital, social capital requires investment in social relations where better resources may be located (Lin, 2001; Chapter 1). Substantial evidence shows that individuals gain an advantage in the labor market when better social capital returns better job outcomes (occupational prestige, authority or supervisory positions, and, indirectly through such positional advantages, earnings) (for reviews, see Lin, 1999; Burt, 2000; Marsden & Gorman, 2001). However, the theory and utility of social capital need not be restricted to individuals seeking jobs in the marketplace – the supply side of the labor market. A labor market represents a field where a multitude of actors supply or demand capital encounters, engage in negotiations, and seek the optimal matching of such capital for both parties. Agents from organizations and enterprises, representing the demand side of the labor market, also actively seek out candidates best suited to filling positions. Social capital should play a key role from the supply side of the labor market as well. We envision this role in two ways. First, organizations and their agents are expected to

Recruiting and Deploying Social Capital in Organizations

227

actively use social connections and social networks in the process of seeking out suitable candidates for positions in their organizations – the hiring process mirrors the ‘‘getting jobs’’ process (Boxman & Flap, 1990; Boxman, De Graaf, & Flap, 1991; Burt, 1997; Erickson, 1995; Erickson, 1992; Erickson, 2001; Fernandez & Weinberg, 1997; Marsden, 1996; Marsden, 2001: Marsden & Gorman, 2001). Second, organizations and their agents may be sensitive to the demand for social capital in certain positions in the organizations – positions that are required to engage in extensive social exchanges internally or externally – and seek out candidates who best occupy these positions. Matching these socially demanding positions and socially capable candidates should promote the persistence and survival of the organizations – the ‘‘deploying process.’’ Yet there is a relative scarcity of studies investigating the utility of social capital in the demand side of the labor market. Employing social networks in the recruitment process (the hiring process) has received some research attention (see reviews in Marsden, 1996, 2001; Marsden & Gorman, 2001; Erickson, 1995; and Marx & Leicht, 1992), yet there is limited research attention given to whether organizations recruit and deploy employers for their social capital capacities in organizations (the recruiting–deploying process). This essay will focus on the latter issue and defer the former (the hiring process) to other scholars and other essays. We argue that a sound theoretical development is required to support the deploying argument for social capital; empirical work can then proceed. The purposes of this essay are threefold. First, we formulate and articulate a theory of social capital for organizations. Second, we specify two research propositions that postulate the linkage between recruiting and deploying strategies to different positions in organizations (the deploying hypothesis), and how such differential deployment is contingent on the institutional context (the institutional contingency hypothesis). Data from a large-scale cross-sectional survey on Chinese enterprises will be used to examine the hypotheses. The essay concludes with a discussion on theoretical and research implications for examining how organizations enhance and employ social capital in the future.

RECRUITING AND DEPLOYING CAPITAL IN ORGANIZATIONS We begin with a general discussion on the recruiting and deploying capital of different types in organizations. Deployment of resources in

228

NAN LIN ET AL.

organizations, especially economic organizations, constitutes a significant research enterprise for sociology and management sciences. The theoretical foundation for such a strategic deployment argument is found in both sociology and economics. Davis and Moore, 1945 proposed that in a functioning stratification system, individuals endowed with certain skills or capacity to perform should be distributed to certain positions that are more important to society. This theory, while being criticized for possible tautology (reward is justified by defining functions of positions) or class reproduction (socialization and privilege feed each other), nevertheless achieved a major paradigm status over the next six decades in guiding much of subsequent stratification analysis. Most prominently, in the studies of status attainment and social mobility (for individuals seeking positions in the labor force and stratification system; Blau & Duncan, 1967; Kalleberg, 1988) and performance in organizations (for individuals holding positions), education and training conceptualized as skills and knowledge essential for performance or human capital are conceived as the primary factors in distributing individuals to positions and determining returns to positions. Further, for organizations in certain market segments, perhaps more competitive ones, differential deployment and mobility in the organizations (i.e., the internal market) become more meaningful. A parallel development is also prevalent in economics. Neoclassical economists (Johnson, 1960; Schultz, 1961; Becker, 1964/1993) have argued that individual laborers who invest in skills and knowledge that meet the needs of certain jobs will have better opportunities for finding such jobs in the labor market and receive better returns. Human capital thus conceived, and indicated largely by education and proxies such as experience and tenure and, to a much lesser extent in research, on-the-job training is seen as the critical mechanism in the match of individuals and jobs. Again, the effects of experience and on-the-job training become more prominent in larger organizations and in the more competitive market environment. This theory has since dominated much of economic analysis of the labor market. However, human capital is only one type of capital significant in the labor market (Nahapiet & Ghoshal, 1998). What is argued here is that social capital may also be an important asset for organizations and certain positions in organizations. An organization functions as an actor in a web of networked actors – the marketplace. Its persistence and performance depends to a significant extent on its capability to negotiate with other actors for capital, labor, and commodities and goods. Such capacity largely lies in agents who represent the organization in the marketplace. Their successful maneuvers in the marketplace depend not only on their technical


229

know-how, but also on their social skills and networking capacities. Those agents with better network resources should benefit from better information as to where to find goods and services and availability of alternative sources; better persuasive and influential tactics in negotiating with other agents with whom they may have developed reciprocal relationships and trust, or gain their generalized trust through third parties; and greater confidence in selfpresentations. These social skills and assets should benefit the organization they represent in achieving closer-to-optimal returns from the marketplace. Internally, certain key positions are encumbered with responsibilities to engage and exchange with other positions and their occupants. In contrast to other positions where the occupants engage primarily in non-exchange tasks (e.g., production, accounting, computer programming, and technical), the performance of occupants in these positions depends to a significant degree on their social networking skills and capacities (e.g., the top administrators, personnel managers, and department heads). These considerations suggest the need for a theory concerning the process by which social capital is captured and used in organizations. This process may entail (1) how to recruit candidates endowed with social capital (capacity and skills) that meet the demands or requirements of positions in an organization, (2) how such candidates, once recruited, are deployed to various positions in the organization, and (3) how such recruitment and deployment impact organizational survival, performance, and individual rewards. We now proceed to explicate the first two phases of this process. The third phase is deferred to other research reports.

THEORETICAL CONSIDERATIONS The principal theoretical stance here is that social capital, like human capital, is a critical capital for organizations, especially in a competitive market. In this environment an organization’s persistence requires constant and efficient communications both internally and externally (Uzzi, 1996; Geletkanycz & Hambrick, 1997; Erickson, 2001; Scott & Davis, 2007). An organization, or a formal social structure, consists of four fundamental elements: (1) a set of social units (positions) that possess different amounts of one or more types of valued resources, (2) are hierarchically related relative to authority (control of and access to resources), (3) share certain rules and procedures in the use of the resources, and (4) are entrusted to occupants (agents) who act on these rules and procedures (Lin, 2001, p. 33; also see Sewell, 1992). The survival and performance of an organization

230

NAN LIN ET AL.

substantially depend on the coordination and communication among positions and occupants, and judicial exercises of exchanges between those in authority positions and other positions. The significance of such exchanges varies across different positions in an organization. There are positions in which performances rely heavily on exchanges, and coordination would necessarily benefit for occupants with social skills and networking capability. That is, we expect that positions in an organization can be differentiated in terms of the capacity and use of social capital. We identify three types of positions that are more demanding of capacity and exercise of social capital from their occupants (Lin, 1999). First, authority positions, especially those at the top of the organization, are expected to possess and exercise social capital. Occupants of these positions invest much time and effort in exchanges with others outside the organization and with occupants inside the organization (Erickson, 2001; Geletkanycz & Hambrick, 1997). For example, a university president is required to capably exchange with parents, alumni, the community, and potential donors. While the scholarship of the occupant is a credit, his/her social capability is paramount for maintaining and acquiring resources for the organization. Likewise, a CEO for a large corporation is expected to be capable of negotiating on behalf of the enterprise with financial institutions, legal institutions, stakeholders and shareholders, and the board of directors, among others. In a more competitive market, such capability and use of social capital should gain more significance. It has become increasingly difficult to argue whether human capital or social capital requirement is more important for these occupants. For example, does a university prefer a Nobel laureate or a top-notch communicator as the president? Does a high-tech company need a top-notch hardware or software expert, or does it need someone who is a top-notch networker? A cursory check among presidents of research universities and the CEOs in top companies seems to tilt toward social capital. A second type of position in an organization that demands social capital capacity and skills is the edge position – positions that interact with the external environment on behalf of the organization. The titles of these positions may include marketing, purchasing/supplies, sales, public relations, and development (i.e., fund raising) (Burt, 1997). These positions function primarily to protect and preserve resources of the organization or to seek additional valued resources for the organization. Negotiating and networking with agents and representatives from other organizations (both vertical linkages to suppliers of resources and to consumers, and horizontal linkages with partners), the community, and other key individuals and


231

parties are the primary job descriptions for these positions in which occupants cannot survive without social capital capacity and skills. A third type of position within an organization that may also require social capital is those whose primary responsibilities include exchanges with other occupants rather than nonoccupants in the organization. Administrators, managers, and personnel directors, for example, interact and exchange with other positions and occupants on a constant basis. In contrast to positions of production or material processing (e.g., assembly line positions, accounting, programming, etc.), these positions necessarily need a certain amount of social capital to discharge their responsibilities. These are some, certainly not all, of the positions in organizations that likely require greater demand for occupant’s social capital. From the theoretical perspective, the more critical question is whether organizations are consciously aware of these differential demands for social capital across different positions. We postulate that organizations are indeed aware, especially in a more competitive marketplace. That is, strategic recruitment and deployment of social capital characterize organizations and enterprises in such a market. We offer the following hypothesis: H1. The deploying hypothesis. Organizations strategically recruit and deploy occupants with social capital capacity and skills to certain key positions. Further, the validity of this hypothesis will need to take into account the institutional environment. In particular, the extent of differentiation may be contingent on the competitiveness of the market environment. The external environment can be seen as an institutional field in which an organization is located or competes. When the institutional field tends toward the competitive end (i.e., free market), we assume there is a greater need for internal and external coordination. Timely initiation of actions and reacting to external challenges become more urgent for sustaining and acquiring resources. Therefore, it is postulated that the differentiation will be greater. When the institutional field tends toward noncompetitiveness (i.e., criteria of survival and performance are not contingent on performance in a market), such internal and external coordination become more stable and positional performances more routinized. In this case variations in occupant (agent) capability become less demanding. Thus, the second hypothesis: H2. The institutional contingency hypothesis. The more an organization finds itself in an institutional field of greater competitiveness, the more it will differentiate in its recruitment/deployment strategy for various positions.

232

NAN LIN ET AL.

Again, we expect that this hypothesis holds true for both human capital and social capital and, tentatively, that such differentiation for social capital will be more salient for edge positions and human capital for internal positions.

RESEARCH DESIGN AND DATA The conceptual scheme and hypotheses were put to the test with a set of data collected from urban Chinese organizations in 2001. Details of the research design and sampling strategies are described below. Research Site: China At the turn of the new century, China provided an ideal research site for examining the institutional contingency hypothesis (H2). At that juncture, China was being transformed from a command economy to a hybrid economy, where multiple economic sectors, representing the state-collective sector, the private sector, and the joint-venture–foreign-venture sector, coexisted. The state-collective sector continued to operate as work units under the jurisdiction of the state and local government. While the enterprises were asked to compete in the market, much of the authority in personnel and resource issues resided within the bureaucratic structure of the government. Each enterprise was still located in a ranking hierarchy and had a supervising organization. Some of the enterprises had assumed forms of typical market enterprises, with a board of directors and a separate line of executive management. Yet the party secretary is typically either embedded in the board (more often than not as the director or deputy director) or in the authority (as the CEO or deputy CEO who also headed a separate office of the party that played decisive roles in personnel appointments and other key issues). The decision to retain, modify, or discard the enterprise resided with the sponsoring government, in conjunction with but often superseding market considerations. In the private sector, the enterprises had taken on a typical form of independence and registered as private entities. Variations remained great; many private enterprises leaned strongly on government units – meaning they relied on resources and personnel relieved from government units and remained in the deep structure (personnel, party, supply, and market), aligned with government units. Other private enterprises were started by cadre children and relatives, who could count on such guanxi to gain better


233

access to resources and support from pertinent government units. Still others were started by individual entrepreneurs. Joint ventures and foreign firms constituted another economic sector. By the turn of the century, China had welcomed and encouraged foreign firms to join forces with Chinese firms so that the combination of foreign managerial skills, capital and technology, and domestic labor and local support (land, government, etc.) could benefit both parties. At the time of the survey this sector contained firms and joint ventures of widely different characters. Most of the Chinese firms in the joint ventures were in fact government units or firms that had branched out from government units; they retained much of the character of the state sector firms. Some of the ‘‘foreign’’ firms were in fact firms from Hong Kong, Macau, and Taiwan. While capitalistic in nature, these firms operated with both market and guanxi principles socialized from their original societies. Still other foreign firms were typical capitalist firms from North America and Western Europe, but were few in number at that time. Furthermore, some of these firms were owned by Chinese expatriates. By the end of 2001, foreign-invested firms from North America and Europe numbered 27,894, or 15.6% of total foreign firms in China. The amount of investments made was 9.5 billion US dollars, or 20.4% of total FDI in China (data from China External Economic Statistical Yearbook 2002 Beijing: China Statistics Press). Thus, while diverse economic sectors had emerged, their characters were blurred at times. As described below, we gathered data from each sampled firm to explore the ‘‘true’’ political-economic nature of each firm; more detailed analysis is deferred to future projects and reports. For the present purpose, we take advantage of the multi-sector economy and make the crude assumption that variations across the three sectors were sufficiently significant to allow a preliminary test of the second hypothesis: that differentiation of recruitment and deployment would be greater in the private and joint-venture sector than the state sector. If findings from this study, with the less-than-ideal prototypical institutional segmentation formulated in the hypothesis, are consistent with the expectations, it would increase our confidence in its validity.

Sampling: A Multilevel Strategy The sampling was based on multilevel stratification. The first stratification involved dividing all urban cities in China into three regions: the coastal, the interior, and the border regions. Within each region, ‘‘typical’’ cities were

234

NAN LIN ET AL.

sampled that presumably characterized the character of each region. For the coastal regions, sampled cities were Beijing, Shanghai, and Guanzhou. This region was the first to benefit from ‘‘the reform and open’’ policy, assembled investment of resources from the central and local government, and mobilized labor forces from the rural and interior regions. These cities experienced the fastest economic growth and remained the leading region in economic development. Sampled interior cities were Shenyang, Xian, Nanchang (Jiangxi), Nanchong (Sichuan), and Wendeng (Shandong). The interior region was slower in economic development and enjoyed less support from the central government, but was rich in labor resources and some heavy industries from past development. The sampled border cities were Yuqi (Yunnan), Tacheng (Xinjiang), and Dandong (Liaoning). The border region was the slowest in economic development; had a presence, to varying degrees, of minority groups; and had begun developing trading relationships with bordering foreign countries, all of which lagged in economic development themselves. The second stratification factor was organizations (work units) in different economic sectors. In each sampled city, all work units were identified by their registered identity: state/collective, non-state/collective (private), and joint venture or foreign (joint venture) sectors. Attempts were made to sample at least one or more work units in each sector for each sampled city. More organizations were sampled in larger cities. The third stratification factor was positions in each organization. Working with Chinese colleagues and experts, we identified the ‘‘typical’’ positions for most work units: marketing/sales; advertising; supplies/purchasing; public relations; personnel; administration/management; party/union/women’s committee; technical; productive; and planning. We attempted to capture the top-rank positions (party/union/women, administrative/management, and planning); edge positions (marketing, advertising, supplies, public relations); and ‘‘exchange-oriented’’ positions (personnel) as well as other positions (production, technical). It should be noted that both the union and the women’s committees in each organization were closely aligned with the party office and widely understood as part of the party apparatus (Unger & Chan, 1995). One condition for a sampled position was that the incumbent had assumed the job within the past 12 months so as to reduce possible recall errors. Very few sampled work units had all these positions, so the sampled positions varied across work units. The numbers of sampled work units and positions by city are shown in appendix. A total of 194 work units and 526 positions were sampled. For the present study, we focus on enterprises


235

and exclude government agencies and institutes. The final sample includes 413 positions, with 134 in the state sector, 162 in the private sector, and 117 in the joint-venture sector. From each sampled work unit and position, we conducted personal interviews with the occupant, his/her immediate supervisor, and an administrator who could provide information about the work unit. We also gathered data about each city/region at the time of survey (2001), as well as over the past six or seven years. Thus, the final data set consists of four levels of data: from the occupant, from the supervisor, about the work unit, and about the city/region. The present study employs data from the occupants and the supervisors only. Data from the Supervisors and Occupants The questionnaire for the occupant covered topics ranging from his/her family background, demographics, the process by which they sought or found their present position, methods of getting the job including extensive information on the use of personal connections, routine and nonroutine practices in the current position, evaluation of their own behavior and performance, evaluations of their supervisors and the work unit, and a measure of social capital (the position generator). For the supervisors, the questionnaire also ascertained his/her family background, demographics, experience in the job market and with current work unit, how they processed the selection of candidate(s) for the focal position, criteria for evaluations, their behavior and performance, evaluations of their work units and their jobs, their own social connections and social capital (the position generator), their perception and evaluation of the occupant’s behavior, satisfaction, and performance.

RECRUITING CRITERIA For the present paper the analytic plan has two parts. First, it focuses on the recruitment criteria used by the supervisors in determining whether to hire the present occupant and second, it assesses whether such criteria had any bearings on the specific positions as reported by the occupants. Thus, the analysis was designed to examine whether the recruiting criteria used by the supervisors ‘‘matched’’ the social capital (as well as human capital) requirements of certain positions to which employees were deployed as hypothesized. The analysis therefore uses the selection criteria questions

236

NAN LIN ET AL.

from the supervisor survey and the position occupied from the occupant survey. One question in the supervisor questionnaire asked: ‘‘When you were considering the appointment, which of the following individual conditions was important or not important?’’ Response categories were (5) very important, (4) relatively important, (3) average, (2) not too important, and (1) not at all important. A total of 26 conditions or criteria followed for their ratings. These conditions ranged from demographics (age, gender, nationality, marital status), party membership, human capital (education, experience, tenure, technical level), prior position and organizational rankings, health conditions, and dedication to work, to social capacities (family background, family social activity capacity, personal social activity capacity, and the identity of the recommender), residence considerations (household registration, local resident), family considerations (housing, job, and schooling needs of spouse and children), and salary requirement. One clarification regarding the ‘‘social activity’’ items is needed here. In Chinese, ) actually means social ‘‘social activity’’ (she-hui huou-dong, networking activities, reflecting the extent to which a person engages in interpersonal networking or organizational/associational activities. In order to gain a more parsimonious picture of these criteria, we subjected the 25 items (excluding the ‘‘other’’ category) to a factor analysis for each sector. The principal component factoring with orthogonal rotations (for factors with eigen values at or greater than 1.0) was employed. The resulting selection factors appear in Tables 1, 2, and 3 for the three sectors. Seven factors emerged for the state sector. For each factor, we identified the top-loading (Z0.40) items and labeled the factor in accordance with our interpretation of these items. The first factor, family, includes items such as ‘‘housing requirement,’’ ‘‘resolving spouse’s residence or job,’’ ‘‘resolving parents’ residence or jobs,’’ ‘‘resolving children’s residence or jobs,’’ ‘‘resolving children’s school attendance,’’ and ‘‘salary requirement.’’ The second factor, human capital/party, is indicated by items such as ‘‘party membership,’’ ‘‘education,’’ ‘‘work experience,’’ ‘‘experience in the organization,’’ and ‘‘technical level.’’ Party membership, it should be noted, designates both competence and moral character in the membership induction criteria (Li & Walder, 2001; Bian, Shu, & Logan, 2001). The third factor, ‘‘personal,’’ is characterized by items such as ‘‘marital status,’’ ‘‘household registration,’’ and ‘‘local residence.’’ The fourth factor, social capital, is characterized by ‘‘family background,’’ ‘‘family social activity capacity,’’ ‘‘personal social activity capacity,’’ and ‘‘recommender.’’ The fifth factor is labeled ‘‘prior status,’’ reflecting ‘‘original position’’ and

237


Table 1. Variable

Factoring of Recruiting Criteria for Supervisor: State Sector (N ¼ 134) (Rotated Factor Loadings). Factor Factor 2 Factor 3 Factor 4 1 Family

Age 0.025 Gender 0.070 Nationality 0.133 Marital status 0.180 Household registration 0.026 Party member 0.006 Education 0.021 Overall work experience 0.110 Experience in this organization 0.175 Technical level 0.084 Original position 0.132 Original rank 0.070 Health conditions 0.034 Dedication to work 0.049 Family background 0.135 Family social activity capacity 0.201 Personal social activity capacity 0.136 Recommender 0.163 Local resident 0.217 Housing requirement 0.698 Resolve spouse’s residence 0.925 or job Resolve parents’ residence 0.939 or job Resolve children’s residence or 0.952 job Resolve children’s attending 0.950 school Salary requirement 0.537

Factor 5 Factor 6 Factor 7

Human Personal capital/ party

Social capital

Prior status

Age/ gender

Local

0.147 0.038 0.044 0.044 0.000 0.615 0.644 0.787 0.774 0.728 0.233 0.192 0.093 0.409 0.125 0.043 0.207 0.090 0.004 0.052 0.044

0.132 0.215 0.852 0.755 0.820 0.155 0.085 0.017 0.071 0.014 0.082 0.080 0.170 0.267 0.316 0.084 0.202 0.027 0.407 0.104 0.103

0.200 0.041 0.048 0.117 0.046 0.107 0.140 0.049 0.114 0.104 0.027 0.087 0.011 0.028 0.750 0.855 0.643 0.593 0.177 0.019 0.109

0.105 0.050 0.063 0.095 0.079 0.176 0.046 0.264 0.317 0.326 0.896 0.890 0.161 0.300 0.078 0.103 0.133 0.125 0.135 0.056 0.065

0.746 0.788 0.104 0.241 0.022 0.008 0.216 0.065 0.004 0.054 0.017 0.003 0.527 0.262 0.042 0.060 0.030 0.059 0.153 0.058 0.012

0.076 0.082 0.105 0.070 0.281 0.142 0.251 0.045 0.028 0.046 0.005 0.043 0.581 0.521 0.082 0.012 0.298 0.160 0.689 0.453 0.080

0.046

0.074

0.128

0.052

0.031 0.003

0.035

0.057

0.083

0.063

0.012 0.006

0.066

0.010

0.075

0.026

0.018 0.009

0.127

0.083

0.036

0.168

0.293 0.067

‘‘original organizational rank.’’ It is interesting to note that these status indicators were independent of both human capital and social capital, even though intuitively they might be related to either one or both. The sixth factor concerns age and gender. The seventh factor is a combination of ‘‘health conditions,’’ ‘‘dedication to work,’’ ‘‘local resident,’’ and ‘‘housing requirement.’’ It again indicates that a person’s devotion to work is

238

Table 2. Variable

NAN LIN ET AL.

Factoring of Recruiting Criteria for Supervisor: Private Sector (N ¼ 162) (Rotated Factor Loadings). Factor Factor 2 Factor 3 Factor Factor 5 Factor 6 Factor Factor 8 1 4 7 Family Personal/ Social party capital

Age 0.077 Gender 0.108 Nationality 0.164 Marital status 0.124 Household registration 0.214 Party member 0.098 Education 0.107 Overall work experience 0.059 Experience in this 0.049 organization Technical level 0.140 Original position 0.015 Original rank 0.055 Health conditions 0.073 Dedication to work 0.094 Family background 0.246 Family social activity 0.069 capacity Personal social 0.060 activity capacity Recommender 0.157 Local resident 0.220 Housing requirement 0.750 Resolve spouse’s 0.914 Residence or job Resolve parents’ 0.957 residence or job Resolve children’s 0.965 Residence or job Resolve children’s 0.946 attending school Salary requirement 0.223

Prior Experience Dedicate/ Age/ Education/ status social gender party

0.176 0.386 0.806 0.674 0.711 0.411 0.052 0.046 0.074

0.032 0.007 0.231 0.204 0.027 0.155 0.089 0.155 0.149

0.137 0.231 0.127 0.004 0.242 0.220 0.012 0.229 0.230

0.181 0.071 0.122 0.096 0.170 0.105 0.022 0.622 0.730

0.039 0.749 0.106 0.669 0.171 0.121 0.088 0.298 0.213 0.128 0.244 0.228 0.079 0.102 0.288 0.302 0.160 0.187

0.190 0.103 0.011 0.119 0.121 0.476 0.835 0.132 0.267

0.139 0.154 0.169 0.249 0.053 0.142 0.073

0.031 0.181 0.218 0.805 0.228 0.805 0.139 0.253 0.024 0.192 0.724 0.230 0.874 0.211

0.699 0.210 0.204 0.280 0.126 0.038 0.004

0.274 0.123 0.075 0.126 0.029 0.101 0.491 0.228 0.691 0.009 0.049 0.066 0.076 0.039

0.174 0.011 0.035 0.045 0.255 0.177 0.080

0.008 0.206 0.424 0.130 0.085

0.377

0.127

0.130

0.689

0.156

0.133

0.758 0.022 0.018 0.090 0.053 0.180 0.090 0.027

0.045 0.363 0.069 0.013

0.016 0.467 0.061 0.052

0.055 0.029 0.045 0.025

0.087 0.130 0.184 0.017

0.065

0.058

0.030

0.030

0.016

0.063

0.010

0.031

0.069

0.018

0.010

0.020

0.083

0.005

0.001

0.061

0.019

0.028

0.017

0.038

0.020

0.003

0.095 0.495

0.390

0.115

0.492

0.007

seemingly dictated by one’s physical health and no distraction from residence and housing issues. In Table 2, there were eight factors for the private sector, labeled as family, personal/party, social capital, prior status, experience, dedication/ social capital, age/gender, and education/party. Certain factors were similar to those in the state sector: family, prior status, and age/gender. Human

239


Table 3.

Factoring of Recruiting Criteria for Supervisor: Joint Venture (N ¼ 117) (Rotated Factor Loadings).

Variable

Age Gender Nationality Marital status Household registration Party member Education Overall work experience Experience in this organization Technical level Original position Original rank Health conditions Dedication to work Family background Family social activity capacity Personal social activity capacity Recommender Local resident Housing requirement Resolve spouse’s residence or job Resolve parents’ residence or job Resolve children’s residence or job Resolve children’s attending school Salary requirement

Factor 1

Factor 2

Factor 3

Factor 4

Factor 5

Family/ local

Social capital

Human capital

Prior status

Personal/ dedicate

0.153 0.108 0.362 0.237 0.472 0.512 0.257 0.124 0.014 0.150 0.079 0.131 0.321 0.109 0.332 0.294 0.257 0.459 0.507 0.713 0.922

0.058 0.139 0.246 0.321 0.277 0.034 0.143 0.000 0.293 0.048 0.142 0.179 0.070 0.195 0.685 0.819 0.708 0.591 0.494 0.401 0.176

0.155 0.105 0.107 0.043 0.098 0.188 0.585 0.834 0.684 0.836 0.302 0.336 0.126 0.421 0.034 0.003 0.142 0.261 0.073 0.208 0.086

0.680 0.341 0.098 0.193 0.136 0.286 0.080 0.104 0.318 0.232 0.770 0.669 0.621 0.181 0.143 0.001 0.256 0.058 0.049 0.113 0.113

0.305 0.639 0.690 0.664 0.543 0.378 0.164 0.031 0.025 0.017 0.156 0.262 0.126 0.464 0.409 0.115 0.182 0.107 0.283 0.134 0.132

0.937

0.148

0.069

0.084

0.115

0.944

0.137

0.066

0.100

0.091

0.939

0.144

0.076

0.105

0.061

0.439

0.161

0.120

0.491

0.234

capital is now in two separate factors: experience and education, somewhat independent of each other. Likewise, social capital is captured by two factors: family capacity and dedication. Also note that party membership is part of personal traits and part of education. Thus, part of party membership still reflects human capital, but is also seen as part of a candidate’s personal traits. Five factors emerged for the joint sector in Table 3: family/local, social capital, human capital, prior status, and personal/dedication. Interestingly, social capital, human capital, and prior status remained as independent

240

NAN LIN ET AL.

factors, but other factors found in the state and private sectors were folded into two: one concerns accommodation of family, residence, and personal needs; another concerns personal characteristics and dedication to work. It should be noted that the sequence of these factors has nothing to do with their relative degrees of significance; it merely reflects how many of the items included in the questionnaire clustered together. What is important is which items cluster together and whether they form meaningful dimensions. In general, then, social capital, human capital, prior status, family, and personal requirements were salient recruiting considerations in all three sectors. Of particular interest here is the formation of clusters pertaining to social capital and human capital. It is reassuring that they did emerge, but it is also not surprising. Also of interest are the status of the previous position and the organization. They constitute a form of capital; we defer interpretation of its meaning to the discussion section.

DEPLOYMENT DIFFERENTIATION IN DIFFERENT SECTORS We test both hypotheses simultaneously. The questions are (1) for the deploying hypothesis, do recruiting factors, especially social capital capacity, affect how workers are deployed into certain key positions (i.e., the top positions, the edge positions, and exchange-oriented positions) and (2) for the institutional contingency hypothesis, is such differential deployment more evident in a more competitive environment (the private sector and the joint sector) as compared to a less competitive environment (the state sector)? Support for the first hypothesis comes from evidence that the social capital factor in recruitment consideration is associated more with certain key positions than other positions. We employed the multinomial logistic regression technique in which the dependent variables are nine of the ten sampled positions, with production as the reference (missing) position. The reasoning was that production would probably be the position that required the least social capital capacity and skills, since it was not on the edge of the organization, not near the top of the organization, and probably among those requiring the least amount of people connections. This last argument could be contested, since some production requires coordination of workers, even on an assembly line. However, we contend these coordinations are mostly routinized and individual workers are more or less interchangeable without significant loss of performance. The independent variables were the factors of the selection criteria.


241

Methodologically, we note that this analysis used the selection criteria data from the supervisor questionnaire and the position occupied by the employee from the occupant questionnaire. Each piece of information was clearly possessed by each respondent, minimizing possible measurement errors. Also, coming from different respondents and questionnaires, the items avoided possible response set effects. To test the second hypothesis, we conducted analysis separately for each sector. Confirmation of the institutional contingency hypothesis comes from evidence that the differential associations between social capital as a selective criterion and various key positions would be stronger in the private and joint-venture sectors than those in the state sector. Results of the analyses are presented in Tables 4, 5, and 6. As in Table 4, there was almost no significant differentiation for the state sector across different positions for social capital considerations. Only public relations showed a slightly significant association with social capital, as compared to production. Human capital showed significant associations with administrative positions and party positions, and a modest association with technical positions, consistent with expectations (H1: Corollary). Family considerations were significantly associated with technical positions, and had some modest associations with party positions and advertising. This suggests that technical positions were significant in this sector so that accommodation of family needs deserved the enterprise’s attention and efforts. Also interesting were significant associations of the prior status factor (prior position and organizational rankings) with both internal (party and technical positions) and external (marketing, supplies, and, to a lesser extent, advertising) key positions in the state sector. For the private sector, as can be seen in Table 5, social capital appeared in two factors. One social capital factor was significantly associated with being an occupant of personnel and supplies, and somewhat with being in personnel planning and marketing positions. The other social capital, overlapped with dedication, was strongly associated with public relations, personnel, administration, party, and planning as well as marketing, advertising, and supplies. Human capital also appeared in two factors. The experience factor was associated with advertising and public relations and somewhat with supplies. The education/party factor was significantly associated with party, technical, planning, and advertising positions and somewhat with marketing and personnel positions as well. Age/gender was positively associated with public relations and negatively associated with party position. In the questionnaire, age and gender were listed as conditions without indicating whether the preference was for older or younger personnel, or a male or female. It is thus

0.083 0.033 0.426 0.519 0.943 0.126 0.356

Marketing

Supplies

0.659 0.175 1.102 0.273 1.490 0.333 0.488

Advertising

0.841a 0.429 0.266 0.006 0.943a 0.000 0.045 0.096 0.529 0.011 0.742a 0.556 0.178 0.528

Public relations 0.255 0.608 0.598 0.003 0.685 0.492 0.165

Personnel

Position

0.092 1.197 0.118 0.134 0.432 0.290 0.306

Administration 0.741a 0.908 0.258 0.446 0.976 0.360 0.155

Party

Planning

0.012 0.066 0.134 0.481 0.430 0.202 0.475

Technical 0.915 0.634a 0.074 0.193 0.936 0.015 0.008

Multinomial Logistic Regression of Positions on Selection Factor Scores: The State Sector (N ¼ 134).

Note: ‘‘production’’ is the base category. po0.05; po0.01. a po0.1.

Family Human/party Personal Social capital Prior status Age/gender Dedicate

Factor

Table 4.

242 NAN LIN ET AL.

0.485 0.274 0.620a 0.435 0.513 1.102 0.231 0.550a

Marketing

0.501 0.720 0.594 0.074 1.115 0.928 0.534 0.701

Advertising

Public relations 0.633 0.049 0.377 0.448 0.773 1.515 1.016 0.506

Supplies 0.983 0.001 0.898 0.065 0.841a 0.920 0.343 0.223 0.379 0.256 1.705 0.837 0.646 2.687 0.584 0.980a

Personnel

Position

0.081 0.03 0.580a 0.105 0.483 1.200 0.117 0.466

Administration 0.002 0.062 0.541 0.351 0.312 1.642 1.164 1.077

Party

0.281 0.622a 0.566 0.601 0.712 0.129 0.107 0.836

Technical

Planning 0.301 0.581 0.654a 0.351 0.391 1.041 0.086 0.765

Multinomial Logistic Regression of Positions on Selection Factor Scores: The Private Sector (N ¼ 163).

Note: ‘‘production’’ is the base category. po0.05; po0.01; po0.001. a po0.1.

Family Personal/party Social capital Prior status Experience Dedicate/social Age/gender Education/party

Factor

Table 5.

Recruiting and Deploying Social Capital in Organizations 243

0.310 0.353 0.907 0.725 0.635

1.659 0.504 0.473 0.860 0.623

Note: ‘‘production’’ is the base category. po0.05; po0.01. a po0.1.

Advertising

Marketing 0.947 0.275 0.894 0.008 0.597

Supplies

0.432 0.304 0.629 0.234 0.345

Public relations 0.671 0.248 0.003 0.544 0.250

Personnel

Position

0.687a 0.191 0.059 0.766 0.056

Administration

0.608 1.251a 0.637 0.330 1.197a

Party

0.069 0.505 0.52 0.686 0.369

Technical

0.771a 0.256 0.054 0.449 0.318

Planning

Multinomial Logistic Regression of Positions on Selection Factor Scores: The Joint Ventures (N ¼ 117).

Family Social capital Human capital Prior status Personal/dedicate

Factor

Table 6.

244 NAN LIN ET AL.


245

impossible to speculate on what the negative association inferred. Further research is needed to clarify this association. Family and personal/party considerations had no substantial associations with various positions (in contrast to production positions). Interestingly, prior status, which was significantly associated with a number of positions in the state sector, showed no significance with any positions. That is, prior position and organization rankings had little influence on deploying considerations in the private sector. Finally, as shown in Table 6, data in the joint-venture sector showed relatively few significant associations between the recruiting factors and deployment to positions. Social capital had only a modest association with party positions, and human capital was significantly but negatively related only to supplies. Prior status was also negatively associated with administrative and marketing positions. Family considerations were also negatively associated with marketing, supplies, administration, and planning. The only other positive association was between the personal/ dedication factor and being in party positions. In general, there was little evidence to support the differential deployment hypothesis. A serendipitous finding worth noting is the significance of the ‘‘prior status’’ factor for certain positions in the state sector (i.e., marketing, purchasing/supplies, party, and technical positions), while it showed no significant association with any of the positions in the private sector. Further it was negatively associated with marketing and administrative positions for the joint sector. One interesting speculation may be that previous linkages with high-status organizations and positions play a more critical role in an institutional arrangement where hierarchical structure and positions rather than market mechanisms figure more importantly for organizational survival. Finding and placing employees with such linkages would make strategic sense. We will further explore this issue in the next section.

DISCUSSION To facilitate the discussion, we summarize the findings in Table 7, where for each sector we identify the significant (at r0.05 level) associations of social capital, human capital, and prior status on various positions (relative to production positions). Several clarifications deserve mentioning here. The check sign (V) denotes a significant coefficient from the relevant table (e.g., Tables 4, 5, or 6) and the negative sign denotes a negative association. For the private sector, there were two factors each related to social capital and

246

NAN LIN ET AL.

Table 7. Summary of Findings: Selection Factors and Deploying to Positions in Different Sectors (Reference position: production and significance level o ¼ 0.05). Social Capital State sector Marketing Purchasing/supplies Administration Party Technical Private sector Marketing Advertising Purchasing/supplies Public relations Personnel Administration Party Technical Planning Joint sector Marketing Purchasing/supplies Administration Party

Human Capital

Prior Status

V V V V

V V VV V VV V V V

V V

VV V

V V V V V V

Notes: V, partial regression coefficient significant at o0.05; VV, both partial regression coefficients significant at o0.05 where there were two associated factors (e.g., two social capital or human capital factors, see text and Table 5) for the private sector. The negative sign denotes a negative association.

human capital, respectively (see Table 5), so there could be two checks (VV) when both relevant factors were significant for a given position. The first conclusion is that social capital figured strongly in the differential deployment of employees to various key internal positions (personnel, administration, party, technical, and planning) in the private sector. Certain positions were given consideration of both social and human capital, such as advertising, public relations, party, and planning positions. Social capital figured significantly in deploying employers to other important internal (administration and personnel) and edge (marketing and purchasing–supplies) positions. Prior status played no significant role in the deployment process.


247

On the other hand, social capital had little significance in differential deployment in the state sector. Human capital was important in assigning employees to key internal (administrative and party) positions. Further, prior status was significantly associated with deployment to certain key internal (party and technical) and edge (marketing and purchasing/supplies) positions. The contrast between the results from the state sector and the private sector simultaneously provides strong support to both the differential deployment hypothesis (H1) and the institutional contingency hypothesis (H2). Social capital is a major consideration in the recruitment and deployment to key internal and edge positions in enterprises in the more competitive environment, the private sector, but not in the less competitive environment, the state sector. The data, therefore, strongly supports the primary theoretical argument: that organizations, in the more competitive market environment, strategically recruit and deploy employees to certain key positions where social networking capacity and activities are deemed essential. Findings from the joint-venture sector did not support the differential deployment and institutional contingency hypotheses. We expected that the enterprises in this sector would be in the more competitive market and should show effects of social capital. One possible source of this non-finding may be due to the fact that most of the Chinese enterprises in such joint ventures are in fact state or collective enterprises. We found, for example, that about four in five (81%) of these firms received no materials from private enterprises; over half (52.5%) of the firms received no capital from private enterprises; and four in five (80%) did not receive any funding from private enterprises. It is not surprising, therefore, that many of these firms retained practices and routines similar to those in the state and collective sector. Nevertheless, the present study advances a theory about the significance of social capital for organizations and provides preliminary data to support the differential recruitment and deployment hypotheses. It urges future research on recruitment, deployment, and staffing to pay attention to the key role of social capital in organizations and how it figures in the functioning and performance of organizations. We should mention that it was not true that the private sector recruited employees with better social and human capital as compared to those in the state sector. In separate analyses, we found that the state sector employees on the average had higher education, more work experience, and more social capital (as measured in a position generator) as compared to those employed in the private sector (the employees in the joint sector seemed to have the best human and social capital, but not much).

248

NAN LIN ET AL.

Rather, the key was how organizations in the private sector more effectively deployed workers with such capital to key positions. Further, the study shows that the strategic utility of social capital gains significance for organizations in the more competitive market environment. It refutes the conventional wisdom that informal social relations or guanxi would be replaced by more ‘‘rational’’ factors (e.g., human capital) in the more competitive market environment (Guthrie, 1998). What is theorized and demonstrated is that in the more competitive environment, social capital is perhaps even more critical, and of occupants with such capital should be strategically deployed to key internal and edge positions. It also suggests that studies in other capitalistic and more industrialized societies should examine differential recruitment and deployment of social capital for organizations in the more competitive markets (e.g., the manufacturing and private sectors) as compared to the less competitive markets (e.g., bureaucratic organizations; government-sponsored enterprises (GSE), such as Freddie Mac, Fannie Mae, and federal home loan banks; industries closely associated with government – especially defense – agencies; and public sectors). Marsden (2001) has shown, for example, that social networks and social capital are more important for both employees and employers in the private sector in the United States. Finally, the factor of prior status deserves further theoretical considerations. In the present study, it was significant in differential recruitment and deployment in the state sector, more so than the significance of social and human capital. One possible clue to understanding what it represents is the general knowledge in economic sociology that markets, networks, and hierarchies represent three components of social structure that dictate the functioning of economic activities (Williamson, 1975, 1985; North, 1990; Granovetter, 1985; Powell, 1990). It may well be that prior status in fact captures in part the significance of hierarchical linkages in the state sector. Transactions among bureaucratically linked units are critical to their survival and persistence. The enterprise would indeed benefit from employees who came with such hierarchical capital. It is now possible, we believe, to formulate theoretical propositions regarding the differential recruitment and deployment of the three types of capital (social capital, human capital, and hierarchical capital) in different economic sectors. These considerations lend support to our contention that strategic recruitment and deployment of social capital has its significance beyond the present-day China. Sectors or variations in institutional arrangements exist in almost all societies. Economic activities engage the three types of capital everywhere. It behooves us to explore their persistent significance and differential variations in all economies and regimes.


249

The next logical step in the present analysis of social capital in organizations in China is to explore the relative advantages of the recruitment and deployment of social capital for the deployed occupants and the organizations, especially in the more competitive markets. Returns to employees may include appropriate use of talents (e.g., more exchange opportunities in and outside the organization), job satisfaction, and socioeconomic rewards, and returns to organizations may include extensity of markets (e.g., supplies of talents, materials and capital, and reaches of markets beyond local or regional boundaries), sustainability (e.g., international investments and markets, quality and quantity of supplies), and performance (e.g., duration, market shares, and revenues and profits). We are currently undertaking some of these investigations.

ACKNOWLEDGMENTS To appear in Work and Organization in China, Volume 18, Sociology of Work, edited by Lisa Keister. We appreciated the reading and commenting on a previous version by Steve McDonald, Bonnie Erickson, and Joon-mo Son.

REFERENCES Becker, G. S. (1964/1993). Human capital. Chicago, IL: University of Chicago Press. Bian, Y., Shu, X., & Logan, J. R. (2001). Communist party membership and regime dynamics in China. Social Forces, 79(2), 805–841. Blau, P. M., & Duncan, O. D. (1967). The American occupational structure. New York: Wiley. Bourdieu, P. (1983/1986). The forms of capital. In: J. G. Richardson (Ed.), Handbook of theory and research for the sociology of education (pp. 241–258). Westport, CT: Greenwood Press. Boxman, E. A. W., De Graaf, P. M., & Flap, H. D. (1991). The impact of social and human capital on the income attainment of Dutch managers. Social Networks, 13, 51–73. Boxman, E. A. W., & Flap, H. D. (1990). Social capital and occupational chances. Presented at the International Sociological Association XII World Congress of Sociology, July, Madrid. Burt, R. S. (1992). Structural holes: The social structure of competition. Cambridge, MA: Harvard University Press. Burt, R. S. (1997). The contingent value of social capital. Administrative Science Quarterly, 42, 339–365. Burt, R. S. (2000). The network structure of social capital. Research in Organizational Behavior, 22, 345–423. Coleman, J. S. (1990). Foundations of social theory. Cambridge, MA: Harvard University Press. Collins, R. (1979). The credential society. New York: Academic Press.

250

NAN LIN ET AL.

Davis, K., & Moore, W. E. (1945). Some principles of stratification. American Sociological Review, 10(April), 242–249. Erickson, B. (1992). A structural approach to network and cultural resources. Presented in the XII International Sunbelt Social Network Conference, San Diego, CA, USA. Erickson, B. H. (1995). Networks, success, and class structure: A total view. Presented at the Sunbelt Social Networks Conference, February, Charleston, SC. Erickson, B. H. (2001). Goods networks and good jobs: The value of social capital to employers and employees. In: N. Lin, K. Cook & R. S. Burt (Eds), Social capital: Theory and research (pp. 127–158). New York: Aldine de Gruyter. Fernandez, R. M., & Weinberg, N. (1997). Sifting and sorting: Personal contacts and hiring in a retail bank. American Sociological Review, 62(December), 883–902. Flap, H. D. (1999). Creation and returns of social capital: A new research program. La Revue Tocqueville, XX(1), 5–26. Geletkanycz, M. A., & Hambrick, D. C. (1997). The external ties of top executives: Implications for strategic choice and performance. Administrative Science Quarterly, 42, 654–681. Granovetter, M. (1973). The strength of weak ties. American Journal of Sociology, 78, 1360–1380. Granovetter, M. (1985). Economic action and social structure: The problem of embeddedness. American Journal of Sociology, 91, 481–510. Guthrie, D. (1998). The declining significance of guanxi in China’s economic transition. The China Quarterly, 154, 31–62. Johnson, H. G. (1960). The political economy of opulence. Canadian Journal of Economics and Political Science, 26, 552–564. Kalleberg, A. (1988). Comparative perspectives on work structures and inequality. Annual Review of Sociology, 14, 203–225. Li, B., & Walder, A. G. (2001). Career advancement as party patronage: Sponsored mobility into the Chinese administrative elite, 1949–1996. American Journal of Sociology, 106, 1371–1408. Lin, N. (1982). Social resources and instrumental action. In: P. V. Marsden & N. Lin (Eds), Social structure and network analysis (pp. 131–145). Beverly Hills, CA: Sage. Lin, N. (1999). Social networks and status attainment. Annual Review of Sociology, 25, 467–487. Lin, N. (2001). Social capital: A theory of social structure and action. London: Cambridge University Press. Lin, N., Ensel, W. M., & Vaughn, J. C. (1981). Social resources and strength of ties: Structural factors in occupational status attainment. American Sociological Review, 46(4), 393–405. Marsden, P. V. (1996). The staffing process. In: A. L. Kalleberg, D. Knoke, P. V. Marsden & J. L. Spaeth (Eds), Organizations in America: Analyzing their structure and human resource practices (pp. 133–156). New York: Kluwer Academic. Marsden, P. V. (2001). Interpersonal ties, social capital, and employer staffing practices. In: N. Lin, K. Cook & R. S. Burt (Eds), Social capital: Theory and research (pp. 105–125). New York: Aldine de Gruyter. Marsden, P. V., & Gorman, E. H. (2001). Social networks, job changes, and recruitment. In: I. Berg & A. L. Kalleberg (Eds), Sourcebook of labor markets: Evolving structures and processes (pp. 467–502). New York, NY: Kluwer Academic/Plenum Publishers. Marx, J., & Leicht, K. T. (1992). Formality of recruitment to 229 jobs: Variations by race, sex and job characteristics. Sociology and Social Research, 76, 190–196. Nahapiet, J., & Ghoshal, S. (1998). Social capital, intellectual capital, and the organizational advantage. The Academy of Management Review, 23, 242–266.

251


North, D. C. (1990). Institutions, institutional change and economic performance. Cambridge: Cambridge University Press. Powell, W. W. (1990). Neither market nor hierarchy: Network forms of organization. Research in Organizational Behavior, 12, 295–336. Schultz, T. W. (1961). Investment in human capital. The American Economic Review, LI(1), 1–17. Scott, W. R., & Davis, G. F. (2007). Organizations and organizing: Rational, natural and open system perspectives. In: H. E. Aldrich (Ed.), Organizations and environments. Stanford, CA: Stanford University Press. Sewell, W. H., Jr. (1992). A theory of structure: Duality, agency, and transformation. American Journal of Sociology, 98(1), 1–29. Unger, J., & Chan, A. (1995). China, corporatism, and the East Asian Model. The Australian Journal of Chinese Affairs, 33(January), 29–53. Uzzi, B. (1996). The sources and consequences of embeddedness for the economic performance of organizations: The network effect. American Sociological Review, 61(4), 674–698. Williamson, O. E. (1975). Markets and hierarchies: Analysis and antitrust implications. New York: Free Press. Williamson, O. E. (1985). The economic institutions of capitalism. New York: Free Press.

APPENDIX. SAMPLED ORGANIZATIONS AND POSITIONS BY CITY City Beijing Shanghai Guanzhou Shenyang Xian Nanchang, Jiangxi Nanchong, Sichuan Wendeng, Shandong Yuqi, Yunnan Tacheng, Xinjiang Dandong, Liaoning Total

Positions

Organizations

78 68 68 74 48 48 28 28 30 28 28

23 46 29 22 15 15 15 4 9 19 4

526

194

PART II WORK

NETWORK RESOURCES AND JOB MOBILITY IN CHINA’S TRANSITIONAL ECONOMY Yanjie Bian and Xianbi Huang ABSTRACT Information and influence are distinct network resources that are embedded in and mobilized from networks of personal contacts. A five-city survey shows that Chinese job changers obtain both kinds of network resources from social ties of varying strengths. During the first 20 years of China’s market reforms, job changes were increasingly network facilitated; despite the growth of labor markets network allocation of labor had reached dominance by 1992. Job changers using information and influence networks to search for new employment were more likely to increase both job search time and job–worker matching; however, those using influence networks, not information networks, were likely to move into jobs of higher earning opportunity. These results are interpreted in a dynamic context of increasing market competition and growing allocative efficiency.

INTRODUCTION How do networks of personal contacts matter for job mobility? Sociologists have offered two alternative arguments. One argument, which is tied to Work and Organizations in China after Thirty Years of Transition Research in the Sociology of Work, Volume 19, 255–282 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1108/S0277-2833(2009)0000019012

255

256

YANJIE BIAN AND XIANBI HUANG

market economies, points to the role of weak ties of infrequent interaction and low intimacy to relay nonredundant information about job openings that are not immediately available to job seekers (Granovetter, 1973, 1974). As argued by Granovetter (1981) among others, this network-channeled information allows for a wider search, reduces uncertainties on workers and employers, and increases the efficiency of matching jobs to qualified workers. The other argument, which is linked to state socialist economies and originated from empirical studies of prereform China, contends that strong ties of trust and obligation tend to relay influence from authorities to individuals whose jobs are assigned through a hierarchy of government agencies and state-run employers (Bian, 1997). While influence accessed through strong-tie contacts helped Chinese youth break free from the rigid control of jobs by a redistributive, coercive state (Walder, 1986; Davis, 1990; Lin & Bian, 1991; Yang, 1994), its effect entailed favoritism, a source of inefficient allocation of labor that has been expected to decline in an increasingly market-oriented Chinese economy after reforms (Guthrie, 1998, 2002). Capitalizing on a multicity probability sample of Chinese urbanites whose job histories span from 1957 to 1999, in this paper we study three interrelated implications of the information and influence arguments for job mobility in the Chinese transitional economy. The first addresses the extent to which weak ties are more likely to relay information and strong ties to relay influence. While this likelihood has been assumed by previous researchers, we assess it statistically for the first time in the Chinese context. The second is how information and influence networks affect job search efficiency, as measured in our study by job search time, job–worker matching, and mobility into positions of higher earning opportunity. Finally, the third concerns the ways in which the effects of information and influence networks on job mobility have evolved across periods of varying degrees of marketization.

THEORETICAL ARGUMENTS AND RESEARCH HYPOTHESES Information and Influence in Job Mobility A market economy is imperfect because of information asymmetry problems. Job seekers and prospective employers, for example, have insufficient or inaccurate information about each other, causing unemployment and less efficient matching of jobs and workers. From the perspective

Networks and Mobility in China

257

of job seekers, Granovetter’s (1973, 1974) pioneer research has shown that networks of personal contacts are informal channels through which job seekers learn about a broader range of job openings than through formal ‘‘market’’ channels. Because weak ties of infrequent interaction and low intimacy tend to bridge people across group boundaries, weak ties, rather than strong ties, tend to provide nonredundant information about jobs, and in doing so are better able to facilitate job mobility for career advancement. Although others have modified the specifications of Granovetter’s theory, the argument about the role that networks of personal contacts play in the flow of nonredundant information remains essentially the same in the theories of social resources (Lin, 1982) and structural holes (Burt, 1992). In a state socialist economy, as in prereform China and elsewhere, information asymmetry was not a characteristic but the deliberate design of the system. There urban jobs were considered as national resources subject to state planning, and youth were required to ‘‘wait for assignments’’ from state labor bureaus (Bian, 1994). Job mobility was kept minimal and occurred mostly as the result of workers’ reassignments or organized transfers to satisfy the plan (Walder, 1986; Davis, 1990). Under these circumstances, information about jobs became unessential because those with information about certain jobs they wanted to get were given no freedom to apply for them. Some did indeed get what they wanted, but such a condition surfaced only when they were connected, directly or indirectly, to job-assigning authorities (Bian, 1997). Bian’s study revealed that these connections were mainly strong ties of kin and close friends, and that the key resource flowing through them was influence of power – certain actions had been taken behind the scenes by the authorities to manipulate job assignments in someone’s favor. There are two broad empirical implications of the literature just reviewed. The first is that information and influence are qualitatively different network resources whose flows are structurally constrained by the strengths of ties between job seekers and their personal contacts. Whereas weaker ties are more likely than stronger ties to flow job information, stronger ties are better able to flow influence of power than weaker ties. The problem is that this likelihood has only been assumed in previous studies of China and elsewhere, partly because of the lack of measures of information and especially influence. In this study, we apply this assumption to an empirical test for the first time in Chinese context: Hypothesis 1. Job changers tend to obtain qualitatively different network resources from personal contacts of varying strengths. More specifically,

258


job changers tend to obtain information about jobs from weaker ties than from stronger ties, and obtain influence of power from stronger ties than from weaker ties. Network Resources and Institutional Contexts A second broad implication is that mobilizations of network resources from personal contacts of varying strengths are constrained by institutional contexts.1 Under a market institution, networks of weak ties are expected to mobilize information about jobs and workers in order to reduce uncertainties brought about by the information asymmetries between job seekers and prospective employers. Under a redistributive institution, on the other hand, networks of strong ties are expected to mobilize influence of power from job-assigning authorities, who in turn would act in someone’s favor. Two reversal scenarios have not been considered by previous researchers, and both are also of interest to this study: network mobilization of influence under a market institution and network mobilization of information under a redistributive institution. We are interested in these reversal scenarios partly because the post-1980 Chinese economy has been under tremendous transformation away from a total redistributive system before reforms, and partly because it still remains a highly mixed system under a durable and quite powerful party-state after reforms. A brief review of China’s recent employment policies is in order. China’s ‘‘socialist transformation’’ around 1956–1958 marked the beginning of a new policy of state job assignments. Under this policy, all urban jobs were monopolized and assigned by the state. Lifelong employment was promised and institutionalized in the state sector, but at the same time labor markets were eliminated, individuals’ rights to choose and move between jobs were literally denied and subject to state plans, and ‘‘labor hoarding’’ became a systematic behavior of state managers (Walder, 1986). Around 1980, state monopoly of urban jobs began to change when household businesses, private companies, and international joint ventures hired people outside state plans (Gold, 1990). Job assignments continued in the state sector, but a labor contract system was implemented to recognize individuals’ right to choose and move between jobs; employment services opened in all cities; and state managers, not just labor bureau officials, began to play an important role in hiring decisions (Davis, 1990). A dual track system existed in labor allocation (state assignments and labor markets), as well as in pricing (plan and market prices), production (in-plan and out-plan projects), and wages (budgetary and extrabudgetary wages) (Naughton, 1995). In the


259

post-1992 period, nonstate sectors experienced full-blown expansion, massive numbers of rural migrant laborers flooded the cities, and workers of stateowned enterprises in large numbers lost jobs. Under a ‘‘grab the big and release the small’’ policy, however, major industries and giant enterprises reorganized as business groups, remained under the state, and were closely monitored by the central, provincial, or municipal levels of government (Keister, 2000). Importantly, labor allocation in all sectors was subject to the new Labor Law, published in 1994, which made labor contracts mandatory. State assignments were finally abolished, and job turnovers became the typical experience of a great majority of Chinese urbanites. On the whole, labor markets grew in their own right after 1992. In China’s emerging labor markets, the roles that networks of personal contacts play in job mobility appear to be complex, and mixed findings have contributed to a lively debate about increasing or decreasing significance of guanxi, or influence networks of favor exchanges (Yang, 1994, 2002; Bian, 1994, 2002, 2008; Guthrie, 1998, 2002; Hanser, 2002; Huang, 2008). While no one has denied that networks of personal contacts flow labor market information in the Chinese economy just as it does in the West, some researchers may not recognize this as the practice of guanxi. Much of the controversy about guanxi is also generated by an empirical approach that is sensitive to when, where, in what sectors, and on what kinds of jobs the researchers conducted their studies. Given that China is a huge country of great interregional variation in the pace of reforms and economic development, controversial findings are not surprising. Taking a theoretical approach and informed by the evolvement of China’s employment policies just reviewed, we identify two directions of change concerning relative activeness of information and influence networks in job mobility across periods of Chinese transitional economy. One dimension of change is the increasing degree of marketization with time. That is, compared to prereform periods in which state job assignments operated in a closed, controlled system, since 1980 the allocation of China’s labor has been increasingly operated in open, competitive labor markets. Labor market competition has brought pressures on both job seekers and employers. On job seekers it brought the pressure to collect as much job information as they could, and the pressure to channel as much positive information about themselves to prospective employers. On employers it brought the pressure to be increasingly efficiency oriented, thus filling job slots with qualified and reliable workers. These pressures make networks of personal contacts more and more relevant and useful for flowing job and personal information. Thus,

260


Hypothesis 2. Information networks are more frequently used in later stages than earlier stages of Chinese transitional economy. More specifically, H2a. Job changers more frequently use weak ties in later stages than earlier stages of Chinese transitional economy. H2b. Job changers more frequently obtain job information through personal contacts in later stages than earlier stages of Chinese transitional economy. The other dimension of change is a decentralized system in which hiring power is in the hands of employers. These employers, state and nonstate, work in a highly mixed and dynamic institutional context in which resourceful and powerful connections are essential to helping them fight against future uncertainties or gain a comparative edge over heightened market competitions. On an individual level, employers are socially constrained in all hiring decisions they make since they are both embedded in networks of personal contacts and under the influence of government officials who are strategic and important to their careers. Thus, Hypothesis 3. Influence networks continue to be significant across periods of Chinese transitional economy, and especially so in hierarchical search channels in which government influences are direct and strong. More specifically, H3a. Job changers frequently mobilize strong ties to help with job mobility in both earlier and later stages of Chinese transitional economy. H3b. Job changers frequently seek influence of power from network contacts to help with job mobility in both earlier and later stages of Chinese transitional economy. H3c. Job changers who use hierarchical search channels more frequently seek influence of power from their network contacts than those who use market search channels.


261

Network Resources and Search Efficiencies Efficiency is an important criterion of job mobility. In theory, a society has a high efficiency in labor allocation when it is low in unemployment, high in job–worker matching, and optimal in labor productivity and wage distribution (Javanovic, 1979). In empirical evidence, individuals use formal and informal methods to search for jobs, ‘‘it is unclear whether these contrasting systems differ dramatically in efficiency’’ (Granovetter, 1995, p. 164). Under a stylized market image, Cahuc and Fontaine (2009) develop a theoretical model on relative efficiencies of formal and informal search methods. They show that social networks are an inexpensive but inefficient job search method, and alternative methods such as using private agencies and posting advertisements in newspapers are more efficient and more costly. In this study, we ask the empirical question regarding the level of efficiency when Chinese workers move to a new job, with or without the assistance of personal contacts. We argue that networks of personal contacts are efficient when they: (1) reduce search time or increase search intensity in which one secures a wanted job within a short period of time; (2) match job changers’ qualifications to technical requirements of the landed jobs in order to maximize the utility of their human capital investments; and (3) send job changers to positions of higher earning opportunity to increase material returns on their skills. We elaborate these three search efficiencies and propose research hypotheses in turn.

Search Time Efficiency In the Western context, job searches through personal contacts are widely believed to be inexpensive since obtaining or sharing information about job openings with someone is ‘‘cheap’’ (Granovetter, 1995). Formal search methods, on the other hand, are more costly because people must pay to use them (Osberg, 1993). There is no research evidence, however, on search time efficiency of formal or informal search methods. In the Chinese context, mobilizing personal contacts for job search purposes seems more costly than in the West because it usually involves targeting specific persons of power who would in turn act on someone’s behalf to influence hiring decisions. Under the closed, controlled system of state job assignments, this is especially true when job changers must work out agreements with their current employers to release them, with prospective employers to hire them, and with labor bureau officials to generate labor

262


quotas for them; each step of this lengthy process is a difficult job to accomplish, as exemplified in a personal story told by an informant (Bian, 1994, p. 976). In China’s emerging labor markets after reforms, this process is expected to be simplified since the focus is now only on prospective employers. But in-depth interviews (Bian, 2002; Huang, 2008) indicate that to secure success a job changer may mobilize multiple contacts for many different jobs, or use multiple contacts to influence a specific employer for a targeted job. We expect: Hypothesis 4. Search time is longer for job changers who mobilize influence networks, it is shorter for job changers who mobilize information networks, and it is shortest for job changers who mobilize no networks of personal contacts. Job–Worker Matching Networks of personal contacts are understood to contain what economists call ‘‘person-specific information’’ about worker qualifications (Saloner, 1985). In the West, employers welcome and use this information to sort out workers and match a job’s required tasks to a worker’s ability to do the job (MacDonald, 1980). When this occurs, a better job–worker matching is reached (Simon & Warner, 1992). While person-specific information may be available through schools or other institutional channels (Rosenbaum, DeLuca, Miller, & Roy, 1999), the most useful information comes from employee referrals and old boy networks of ongoing relationships (Saloner, 1985). In all these analyses, person-specific information is assumed to have been delivered to prospective employers, a scenario that implies a certain degree of personal contact or communication that the job seeker or his/her agent has with the employer. In the Chinese context, we differentiate two types of networks: information networks that assume that job changers and their personal agents have no contact with prospective employers, and influence networks that do assume the existence of such contact. If person-specific information is offered through information networks it is expected to increase job– worker matching, so far as Chinese employers, like their Western counterparts, are efficiency driven and want to hire the best qualified persons into job slots. We have already discussed the tendency of Chinese employers to be increasingly efficiency driven with time, and that they are more efficient when they are outside the influence of the party-state. Influence networks are more complicated. When a go-between acts on someone’s behalf to contact a prospective employer, he/she tends to do both


263

of the following things: offer person-specific information about the job changer to impress the employer in a positive way, and try to use personal power to influence the employer in someone’s favor. While the first is expected to increase job–worker matching, the second is an exercise of favoritism and expected to decrease job–worker matching because it tends to send marginally qualified or unqualified persons into jobs of higher requirements. It remains to be seen whether favoritism exercised through influence networks has reduced as China’s market reforms entered the later stages. For the periods through the 1990s, our hypotheses are Hypothesis 5. Networks of personal contacts are expected to affect job– worker matching in the following ways: H5a. Compared to those not using networks, job changers using information networks to search for jobs tend to have a higher degree of job–worker matching. H5b. Compared to those using information networks, job changers using influence networks to search for jobs may have a higher degree of job– worker matching or, if favoritism exercises strongly, a lower degree of job–worker matching. Mobility into Positions of Higher Earning Opportunity While economists have examined wage returns to job search methods (Javanovic, 1979), sociologists have studied the prestige or status of occupations as an outcome of job mobility (see reviews by Granovetter, 1995; Lin, 1999). One assumption is that job changers are rational players who move into new jobs to advance their careers. Logically, the information and influence that are gained through networks of personal contacts are the social capital or social resources that help with career moves (Lin, 1982, 2001). For China, our approach is to assess mobility into jobs of varying earning opportunities. If networks of personal contacts are efficient and effective, then workers who use them to search for jobs are expected to move into jobs of higher rather than lower earning opportunity. While this expectation was learned repeatedly from in-depth interviews with job changers (Bian, 2002; Huang, 2008), we summarize it into the following, final hypothesis: Hypothesis 6. Job changers who search for jobs through information or influence networks are more likely to move into jobs of higher earning

264


opportunity than those not using networks of personal contacts to search for jobs.

RESEARCH DESIGN Data Data for this study comes from a 1999 five-city survey (Changchun, Tianjin, Shanghai, Guangzhou, and Xiamen), conducted by the first author. These five cities cover a large interregional variation in economic development and pace of labor market growth. Changchun is located in northeast China and is the last among the five cities to develop labor markets. Tianjin is in the north, Shanghai in the southeast, and both cities are on the Pacific coast and are studied by international scholars most frequently. Guangzhou and Xiamen are southern cities; as of 1999 economic reforms there are more far reaching and local labor markets most developed. Each city’s survey was based on the same multistage probability sampling design by taking advantage of the city’s geoadministrative structure (city district, subdistrict, and neighborhood committee) to select households. In all sampled households, an adult member aged 18 and older who ever had work experience in the civilian labor force was randomly selected to be the survey respondent. In total, the five-city survey provides a combined sample size of 4,752 respondents. Table 1 presents descriptive data about the respondents in the total and the analytic sample of job changers that we define shortly. A special challenge was to proportionally sample migrant households. In the Chinese context, migrants refer to those who came from rural areas and resided or worked in the cities, where they registered with local governments as ‘‘nonpermanent population,’’ a category that the government created to distinguish it from the ‘‘permanent population’’ of urbanites. While the distinction between permanent and nonpermanent populations is a legacy of the past planned system, China’s census in 2000 reported 130 million migrants, or 10 percent of the country’s population. The problem is that migrants are hard to locate in a household survey: they are highly mobile geographically, at best only half registered with local governments, and live in temporary shelters that are usually unnumbered. All these factors make them difficult to sample by following a standard procedure. The approach adopted in the five-city survey to sample migrant households included two steps. First, in each city the percentage of migrant labor in the city’s labor

265


Table 1. Variables and Descriptive Statistics of the Chinese Five-City Sample, 1999. Variables Total sample Respondents Changchun respondents Tianjin respondents Shanghai respondents Guangzhou respondents Xiamen respondents Respondents who started work after 1956 Respondents ever changed jobs

Mean or % (S.D.)

Valid N

100% 19.7% 21.1% 21.1% 17.0% 21.1% 96.5% 50.3%

4,752 943 1,001 1,000 808 1,000 4,588 2,309

Dependent variables Weeks to complete job searches (ln) Weeks to complete job searches Job–worker matching Job of hierarchical bridging (on a 6–24 scale) Job of market connectedness (on a 3–12 scale)

11.2 1.8 69.1 14.9 8.0

(13.9) (1.2) (46.2) (5.1) (3.5)

1,788 1,788 2,199 2,177 2,165

Explaining variables Job change through a hierarchical channel Job change through a market channel Job change through a network channel (using a contact) Contact was a kin Contact was a friend Contact was an acquaintance Contact offered information (information network) Contact offered influence (influence network) Contact offered both (dual-resource network)

29.7 39.3 69.9 25.1 38.8 3.8 13.2 20.0 31.6

(45.7) (48.8) (45.8) (43.3) (48.7) (19.2) (33.9) (40.0) (46.5)

2,309 2,309 2,309 1,952 1,952 1,952 1,971 1,971 1,971

Other variables Job changes in 1957–1979 Job changes in 1980–1992 Job changes in post-1992 Gender (male ¼ 1) Age Age2 Years of schooling Communist party membership Migrants Prior occupation as a professional Prior occupation as a manager Prior occupation as a clerical worker Prior occupation as a manual worker Prior job in state sector Prior job in a workplace with a high administrative rank

14.9 29.3 55.8 54.3 38.9 1639.5 11.4 17.1 21.0 10.6 15.5 9.4 64.6 58.4 53.8

(35.6) (45.5) (49.7) (49.8) (11.1) (916.8) (3.2) (37.6) (40.7) (30.7) (36.2) (29.1) (47.8) (49.3) (49.9)

2,199 2,199 2,199 2,199 2,199 2,199 2,199 2,199 2,199 2,199 2,199 2,199 2,199 2,199 2,199

Analytic sample: job changers

266


force was estimated by using diverse sources of aggregate statistics, and the residential distribution of migrant households was also estimated. Second, these estimates were used to draw a community-specific quota sample of migrant households. Slightly more than one-fifth of the five-city sample was migrant labor.

Periods and Channels of Job Change In the current analysis, we are less interested in regional variation than in historical change in the mode of labor allocation. For our purpose of analyzing job mobility, three periods are identified: (1) the 1957–1979 prereform period, in which state job assignments were the predominant mode of labor allocation; (2) the 1980–1992 dual track system period, in which state job assignments and labor markets coexisted; and (3) the post1992 period, in which the policy of state job assignments was abolished and labor markets became the predominant mode of labor allocation. These three periods apply to all five cities, with the second period starting earlier in the two southern cities. A job change occurs when a worker moves from one employer to another. Based on this definition, our respondents’ most recent job changes span from 1957 to 1999 and were facilitated by many methods. To satisfy our analytic purpose, these methods are classified into three distinctive search channels: (a) Hierarchy refers to various modes of job assignments and organized transfers by state or organizational authorities, and this channel essentially ignores personal freedom and choices over what jobs to change to; (b) Market is a job change channel that honors such personal freedom and personal choices, does not involve hierarchical authorities, and practically includes self-employment, direct application, and the use of employment services, job advertisements, and media methods; and (3) Network refers to the use of personal contacts to obtain information or more concrete assistance through which to connect job changers to their prospective employers. A network channel can be used independently or in conjunction with a hierarchical or market channel.

Network Measures Tie strength between the respondents and their personal contacts is measured by a scheme of three role relations, namely kin (strong), friend

267


(less strong), and acquaintance (weak). A proxy to tie strength, the scheme allows for comparison with previous studies. Network resource, one of the innovative measures from the five-city survey, refers to information and influence that the respondents obtained from their personal contacts during job searches. It is regarded as an informational assistance if a personal contact ‘‘gave me information about a job’’ or ‘‘assisted with preparing my application,’’ the second of which implies a good deal of information being offered about the jobs or employers to which one applied. It is regarded as an influential assistance when a personal contact ‘‘delivered my application the employer,’’ ‘‘contacted someone of importance on my behalf,’’ or ‘‘solved a concrete problem for the employer’’ (see appendix for coding details). A large number of respondents contacted more than one person for help, with an average of 2.3 contacts, and nearly half obtained both information and influence. For simplicity we will call the resulting three categories of networks ‘‘information network,’’ ‘‘influence network,’’ and ‘‘dual-resource network’’ (for similar terms in an American study, see Corcoran, Datcher, & Duncan, 1980).

Search Efficiencies Our hypotheses call for three measures of search efficiency. Search time efficiency is measured by the number of weeks spent from the start to the completion of a search; a shorter time is considered more efficient.2 Job– worker matching is a dichotomous variable; it is coded ‘‘1’’ when a worker matches the employer’s announced requirement for education, job skill, or prior work experience, and coded ‘‘0’’ otherwise.3 Finally, positions of varying earning opportunity are measured in two ways (Bian & Logan, 1996). First, jobs are paid a higher salary and wage when they have a higher degree of ‘‘hierarchical bridging,’’ or frequent contact with different positions within the organization. Second, jobs can be paid a higher salary and wage when they have a higher degree of ‘‘market connectedness,’’ or frequent contact with market positions outside the organization (see coding details in appendix).

Control Variables To assess independent effects of network measures on three search efficiencies, we include in our regression models (to be presented shortly)

268


a set of control variables, including the respondent’s gender, age, age squared, years of schooling, membership in the Communist party, migrant/ nonmigrant status, and prior job’s occupational category, economic sector, and workplace ranking (this recognizes the level of government jurisdiction over the employer). Previous studies show that all these variables are significant predictors of job mobility processes and outcomes.

RESULTS Changing Channels of Job Change across Periods Table 2 is self-explanatory and our description is brief. Of the 4,588 respondents who started their first urban civilian employment after 1956, slightly more than 50 percent, or 2,309 respondents, had ever changed jobs; these job changers compose our analytic sample. While migrants account for 21.0 percent of this sample (see Table 1), their representation increases to 33.1 percent in the post-1992 period when we observe a dramatic growth of migrants in all Chinese cities. Migrants are believed to be more mobile between jobs than nonmigrants. Our interest is not in the rate of job turnover but relative frequencies of job changes that are facilitated through distinctive search channels across Table 2.

Job Change by Period, Chinese Five-City Survey, 1999.

Variables

1957–1979

1980–1992

Post-1992

Number of respondents Number of job changers Number of migrant job changers Migrant job changers

1,255 352 4 1.1%

1,416 671 59 8.8%

1,917 1,286 426 33.1%

Job change through Hierarchical channels Market channels Network channels

71.3% 11.6% 44.6%

46.3% 25.6% 63.9%

9.6% 54.0% 80.0%

Job change through Only hierarchy Only market Only network Network in market Network in hierarchy

57.2% 4.5% 17.5% 5.1% 15.7%

31.2% 8.5% 30.1% 15.5% 14.6%

4.6% 14.9% 36.3% 40.2% 4.1%


269

periods. The second panel of Table 2 shows that across-period relative frequencies contrast sharply between hierarchy and market, dramatic declining and fast increasing, respectively, but network channels also increase by an extremely large margin of 35.4 percent. Since job changers may combine networks with hierarchical or market channels, the percentages add up to more than 100 percent within each period. The third panel of Table 2 adjusts this complication by using a scheme of five exclusive search channels. As can be seen, network channels were sometimes used independently but were more frequently coupled with hierarchy and, especially after 1992, market channels. In this most recent period, the combination of ‘‘only network’’ (36.3 percent) and ‘‘network in market’’ (40.2 percent) facilitated more than three-fourths of job changes.4 Clearly, networks of personal contacts became both active and dominant when labor markets grew in their own right after 1992.

Network Ties and Network Resources Results in Table 3 were obtained to test Hypotheses 1–3 concerning crossperiod patterns in which job changers obtained information and influence resources from their personal contacts of varying strengths during job searches. Under the assumption that no measures of information or influence are available, Panel 1 replicates what previous researchers have produced about social ties. In each period, personal contacts leaned heavily on kin and friends, a finding that lends support to H3a. Acquaintances, on the other hand, account for less than 16 percent at best, but decreasing over time. This goes against H2a. Kin and acquaintance ties declined somewhat in the post-1992 period, when migrants (58.1 percent) joined nonmigrants (58.5 percent) to more frequently use friendship ties to change jobs. It appears that in China friends take the role, at least quantitatively, that acquaintances in the West do in labor markets. Panel 2 includes measures of information, influence, and dual-resource networks. It shows that overtime information networks did not necessarily increase (against H2b) and influence networks did not necessarily decrease (consistent with H3b). The two kinds of networks may be complementary to each other after reforms, as implied in a significant though not dramatic increase in the frequencies of dual-resource networks in the two postreform periods. An insignificant chi-square test (p ¼ .067) signals a likely rejection of H2b and likely support of H3b.

270


Table 3.

Social Ties, Network Resources, and Their Associations.

1. Ties by period Kin (strong) Friends (likely strong) Acquaintances (weak)

1957–1979 (144) 40.3% 43.8% 15.9%

1980–1992 (362) 40.6% 46.1% 13.3%

Post-1992 (959) 34.2% 58.4% 7.4%

2. Network resource by period Information network Influence network Dual-resource network

Chi-square ¼ 29.82, 1957–1979 (100) 25.0% 34.0% 41.0%

d.f. ¼ 4, po.001 1980–1992 (304) 16.5% 35.5% 48.0%

Post-1992 (873) 21.3% 28.9% 49.8%

3. Network resource by type of ties

Chi-square ¼ 8.77, d.f. ¼ 4, p ¼ .067 Kin (443) Friends (717)

Information network Influence network Dual-resource network 4. Network resource by channel of job change Information network Influence network Dual-resource network 5. Network resource by channel of job change, post-1992, nonmigrants Information network Influence network Dual-resource network 6. Network resource by channel of job change, post-1992, migrants Information network Influence network Dual-resource network

14.7% 37.4% 47.9%

21.3% 29.0% 49.7%

Chi-square ¼ 37.34, d.f. ¼ 4, po.001 Network in Network in market (538) hierarchy (172) 23.6% 22.9% 53.5%

18.6% 43.6% 37.8%


19.5% 53.7% 26.8%


33.3% 50.0% 16.7%

Acquaintances (117) 36.7% 17.1% 46.2% Network independent (524) 15.9% 34.9% 49.2% Network independent (250) 12.8% 33.6% 53.6% Network independent (120) 20.0% 36.7% 43.3%

Chi-square ¼ 13.32, d.f. ¼ 4, p ¼ .010

Panel 3 presents a test of Hypothesis 1, a widely held assumption that different kinds of network resources flow disproportionately through network ties of varying strengths. From kin to friends to acquaintances, information networks increase from 14.7 percent to 21.3 percent to 36.7


271

percent, and influence networks decrease from 37.4 percent to 29.0 percent to 17.1 percent. These contrasting tendencies, though confirming Hypothesis 1 with a significant chi-square test, must be cautiously interpreted, since a great majority use kin and friends, many acquaintance ties are used to mobilize influence as well as information, and overall nearly half of job changers mobilize both kinds of network resources in each of the three periods. Informed by these findings, we tentatively conclude that tie strength is not an ideal proxy to network resources in the Chinese context. We will drop tie strength from further analysis and focus our attention on three types of networks: information, influence, and dualresource networks. When these three networks are cross-tabulated with three channels of job change, in Panel 4 one observes a strong and statistically significant association (po.001). This association is strongest in the post-1992 period, as shown in Panels 5 and 6. For nonmigrants (Panel 5), under ‘‘network in market channel,’’ information networks are slightly more active than influence networks (22.7 percent vs. 21.5 percent, with the remainder in dual-resource networks); under ‘‘network in hierarchy channel,’’ information networks somewhat decrease but influence sharply increase (19.5 percent vs. 53.7 percent); and under ‘‘network-independent channel,’’ information networks are substantially lower than influence networks (12.8 percent vs. 33.6 percent). These patterns, with small and insignificant variations, are repeated by migrants (Panel 6). Overall, findings from Panels 4–6 support H3a, H3b, and H3c.

Search Time Efficiency On average, it takes 11.2 weeks to complete a job search, and the average deviation is 13.9 weeks as shown in Table 1. Table 4 presents two OLS regression models estimated to predict the log-transformed number of weeks spent to complete a job search. Results in Model 1 partially support Hypothesis 4: compared to those not using personal networks to complete a search, those using information networks increased the search time by 53 percent (e.4241), those using influence networks by 42 percent (e.3491), and those using dual-resource networks by 95 percent (e.6661). Going against Hypothesis 4 is a narrow but reversal gap in search time between those using information networks (higher) and influence networks (lower). Nevertheless, the large coefficient for dual-resource networks makes it

272


Table 4.

OLS Regression for the Effects of Network Resources on Search Time Efficiency.

Predictors

(Ln) Number of Weeks to Complete Job Searches Model 1

Model 2

Network resource Information network Influence network Dual-resource network

.424 .349 .666

.430 .365 .649

Year of job mobility (reference ¼ 1957–1979) 1980–1992 Post-1992

.198 .301

.215 .315

Network–period interactions Information 1980–1992 Influence 1980–1992 Dual-resource 1980–1992 Information Post-1992 Influence Post-1992 Dual-resource Post-1992

– – – – – –

Control variables Gender (male ¼ 1) Age Age2 Years of schooling Party membership Migrant Prior job Professional (reference ¼ manual) Managerial Clerical Economic sector (state ¼ 1) Rank of workplace (high ¼ 1) Cities (reference ¼ Changchun) Tianjin Shanghai Guangzhou Xiamen Constant R2 Number of cases

.098 .364 .367 .021 .253 .273

.059 .056 .001 .021x .005 .260

.056 .058 .001 .021 x .004 .264

.090 .146 .008 .013 .078

.087 .140 .010 .014 .075

.380 .429 .480 .045 .199 .094 1,788

.372 .425 .473 .053 .163 .097 1,788

Notes: Two-tailed significance: xpo.10; po.05; po.01; po.001.Tie users who failed to report network resources are included in the analysis as a control (not reporting ¼ 1, otherwise ¼ 0).


273

explicit that the more resources one mobilizes from one’s networks, the more effort it requires, and the longer search duration will be. Model 2 adds network–period interactions to explore if network effects are altered across periods. They are not, but there is a period effect on labor market competition: search time increases by 13 percent from the early- to later-reform period (e.315e.215). Moreover, jobs seem more available in larger cities, such as Shanghai, Tianjin, and Guangzhou, where search time is substantially shorter (all negative coefficients) than in the two smaller cities, Xiamen and Changchun (omitted). On an individual level, younger workers and migrants take a much shorter time to complete job searches than older workers and nonmigrants, respectively, and search time increases with years of education. Search time is basically equal among people with different attributes in gender, party membership, occupation, economic sector, and workplace ranking.

Job–Worker Matching Confined to our measure, 69.1 percent of job changers in the analytic sample passed the threshold of job–working matching (see Table 1). Table 5 presents binary logistic regression models estimated to test H5a and H5b concerning network effects on job–worker matching. Model 1 shows that compared to those not using networks to complete a search, those using information networks increase the probability of job–worker matching by 40.1 percent (odds ratio 1.401–1), those using influence networks by 53.9 percent, and those using dual-resource networks by 50.8 percent. These large percentage margins lend strong support to H5a. The results also clarify one thing: influence networks are not just about favoritism; rather, they may relay person-specific information that in turn increases job–worker matching. This finding lends support to H5b. Results in Model 2 are confirmative. The positive network effects on job– worker matching all survive significance tests after network–period interactions are incorporated; these interactions show that network effects persist across the periods. But job–worker matching was substantially increased in the post-1992 period by a huge margin of 106 percent; it also increased with years of schooling, for professionals, and in cities in which economic reforms started earlier and had been more far reaching by 1999. These are significant signs that China’s emerging labor markets were becoming more and more efficient after 1992.

274

Table 5.


Logistic Regression for the Effects of Network Resources on Job–Worker Matching.

Predictors

Matched to Unmatched Odds Ratios Model 1

Model 2

Network resource Information network Influence network Dual-resource network

1.401 1.539 1.508

1.388x 1.488 1.486

Year of job mobility (reference ¼ 1957–1979) 1980–1992 Post-1992

1.390 2.238

1.276 2.060

Network–period interactions Information 1980–1992 Influence 1980–1992 Dual-resource 1980–1992 Information Post-1992 Influence Post-1992 Dual-resource Post-1992 Control variables Gender (male ¼ 1) Age Age2 Years of schooling Party membership Migrant Prior job Professional (reference ¼ manual) Managerial Clerical Economic sector (state ¼ 1) Rank of workplace (high ¼ 1) Cities (reference ¼ Changchun) Tianjin Shanghai Guangzhou Xiamen Constant Nagelkerke R2 Number of cases

– – – – – –

.837 1.170 .632 .487 .562 .471

.959 .980 1.000 1.105 .920 1.402

.969 .978 1.000 1.106 .915 1.430

2.455 1.182 1.115 1.187 1.296

2.484 1.175 1.103 1.188 1.285

1.367x 2.016 1.909 1.678 .181 .112 2,199

1.315 1.950 1.919 1.626 .210 .117 2,199



275

Mobility into Positions of Higher Earning Opportunity Table 6 presents our final set of regression results. On mobility into jobs of higher hierarchical bridging (Models 1 and 2), influence and dual-resource networks but not information network show significant and persistent effects across periods. Information network becomes important only when it is coupled with influence network, as shown in a positive and statistically significant coefficient for dual-resource network in both models. Take Model 2. We subtract .533 from .820 to get .287; this is the difference that information makes in a dual-resource network. In other words, without this informational component, a dual-resource network would reduce to an influence network whose effect is substantially lower, to the level of .533. While this finding lends partial support to Hypothesis 6, one must understand the theoretical meaning of the coupling effect of information and influence: as far as mobility into hierarchical bridging jobs is concerned, Chinese guanxi networks are fundamentally networks of influence because the information that is transmitted through one’s personal contacts would not matter by itself. This tentative conclusion is not altered in Models 3 and 4, designed to estimate network effects on mobility into jobs of market connectedness. In Model 3, none of the network variables is statistically significant. In Model 4, while the network variables generate no main effects, influence–period interactions show significant results: in two postreform periods influence network helped a job changer to move into a job of higher market connectedness than he/she would in the prereform period. Once again, it is influence of power that makes the difference on job mobility, even if it occurs after reforms. Nonnetwork variables in Models 2 and 4 reveal who is likely to move into hierarchical bridging jobs and market connected jobs, independent of network effects. From the prereform to the later-reform period, there were increasing opportunities for obtaining market-connected jobs but not for hierarchical bridging jobs. For getting hierarchical jobs younger persons, party members, and nonmanual workers moving from the state sector have an advantage, but for getting market-connected jobs men, manual and clerical workers, and managers have more opportunities. A higher level of education is a significant source of advantage on mobility into both kinds of jobs.

DISCUSSION AND CONCLUSION The 1999 five-city survey of China has given us an opportunity to examine the extent to which networks of personal contacts matter for job mobility.

276

Table 6.


OLS Regression for the Effects of Network Resources on Job Positions.

Predictors

Hierarchical Bridging Positions Model 1

Network resource Information network Influence network Dual-resource network Year of job mobility (reference ¼ 1957–1979) 1980–1992 Post-1992 Network–period interactions Information 1980–1992 Influence 1980–1992 Dual-resource 1980–1992 Information Post-1992 Influence Post-1992 Dual-resource Post-1992 Control variables Gender (male ¼ 1) Age Age2 Years of schooling Party membership Migrant Prior job Professional (reference ¼ manual) Managerial Clerical Economic sector (state ¼ 1) Rank of workplace (high ¼ 1) Cities (reference ¼ Changchun) Tianjin Shanghai Guangzhou Xiamen Constant R2 Number of cases

.329 .555x .801

Model 2

.268 .533x .820

Market Connectedness Positions Model 3

Model 4

.039 .174 .021

.006 .099 .024

.758 1.161

.990 1.354

.381 .165

.194 .002

– – – – – –

.852 .926 .681 1.354 1.173 .613

– – – – – –

.028 .153 .002 .376 1.221 .310

.020 .153 .002 .375 1.231 .336

.720 .003 .000 .102 .243 .349

1.673

1.665

.569

.591

3.591 3.415 .801

3.588 3.430 .805

.734 .936 1.192

.735 .931 1.180

1.032

1.026

.105 .316 .341 .470x 6.759 .158 2,165

.056 .260 .359 .425x 6.621 .161 2,165

.355

.600x .400 .387 .014 11.621 .290 2,177

.363

.527 .336 .380 .067 11.421 .291 2,177

1.416 1.406x .785 .692 1.327x .354 .722 .000 .000 .100 .236 .336



277

From the prereform to the early- and to the later-reform periods, hierarchical channels of job change declined from dominance to unimportance. Meanwhile, market channels grew significantly, but so did network channels. In the post-1992 period ‘‘only network’’ and ‘‘network in market’’ combined to facilitate more than three-fourths of job changes. This means that by the end of the last century, network allocation of labor had reached a point of dominance in China’s urban labor markets. How exactly did networks of personal contacts facilitate job changes? We have found that Chinese job changers predominantly use kin and friends and rarely use acquaintances; this tendency is associated with kinds of network resources that are mobilized from social ties of varying strengths. While Westerners tend to obtain information about job openings from acquaintances and other contacts (Granovetter, 1995; Lin, 1999), the Chinese are interested in more than information; they are strongly oriented toward obtaining more substantive assistance that is unlikely to be provided by acquaintances. Such assistance includes delivering someone’s application to the employer, contacting relevant authorities on someone’s behalf, arranging an informal interview at a banquet, and helping solve ‘‘concrete’’ issues for the hiring organization. These are a variety of processes through which one can not only relay information about job seekers and their prospective employers but, more importantly, can also influence hiring decisions in someone’s favor. We thus make a conceptual distinction between ‘‘information network’’ and ‘‘influence network.’’ When both information and influence are embedded in and mobilized from one’s networks of personal contacts, we call this ‘‘dualresource network.’’ These conceptual distinctions are key to a longstanding controversy about the increasing or decreasing significance of guanxi, or influence networks of favor exchanges. Under the assumption that market competition rationalizes economic decisions, Guthrie’s (1998, 2002) hypothesis on the declining significance of guanxi predicts that favor exchanges will decrease in job searches and worker hiring after reforms. This hypothesis does not deny the possibility that the role of networks in information flow may increase. Under a relational image of labor markets that recognizes the centrality of social actors and their relationships, Bian’s (2002) hypothesis on the persistent significance of guanxi predicts that job searches and hiring decisions can be affected by networks of favor exchanges as well as networks of information exchanges when ‘‘market’’ institutions alone fail to coordinate employment processes. Our analysis opposes Guthrie but goes beyond Bian to suggest that guanxi’s significance in job mobility processes increased through the end of the 1990s.

278


This conclusion is supported by our survey data in several ways. First, by 1992 relative frequency of information networks had decreased, while that of influence networks and dual-resource networks had increased substantially. Second, when labor markets grew in their own right after 1992, influence networks and dual-resource networks accounted for four-fifths of job changes among all network users of both migrant and nonmigrant groups. Third, influence networks, not information networks, are found to increase one’s opportunities of moving into jobs of higher market connectedness after reforms. Finally, influence networks and dual-resource networks, not information networks, increased one’s opportunities to move into jobs of higher hierarchical bridging, and this pattern was observed across all three periods. This last set of findings about network effects on job mobility into hierarchical bridging jobs is of particular importance to the conclusion regarding the increasing significance of guanxi. Since a dual-resource network contains both information and influence, its effect can be interpreted as a result of equal importance of information and influence as far as information networks and influence networks can each make an independent effect. However, this is not true, since information networks do not have an independent effect on job mobility. In other words, the information that is transmitted through one’s personal contacts would not matter by itself. Thus, Chinese guanxi networks, as in the case of job mobility, are primarily networks of influence of power in which information resources are a byproduct. The increasing significance of guanxi must be carefully interpreted in a dynamic Chinese economy of growing market competition and allocative efficiency. In the post-1992 period, for example, comparable workers spent more time in completing a successful job search than before, a sign of increased labor market competition after 1992; job–worker matching also significantly increased and especially in cities whose market reforms were more far reaching; and education and professionalism rather than gender, age, or party membership were the leading variables of individual attributes to match workers to jobs. It is in this dynamic context that we also found that influence networks and dual-resource networks, along with information networks, increased job–worker matching. Although past researchers have assumed that influence networks are a relational structure for exercising favoritism or nepotism, in an increasingly marketized Chinese economy they may contain and flow person-specific information and other positive resources that are important for allocative efficiencies in the labor markets. This implication awaits further theoretical justification and empirical tests.

279


NOTES 1. Institutions can be defined in cognitive, cultural, and legal-regulative perspectives (Scott, 2003). We are fully aware of a strong guanxi culture in China where things are normatively done through personal networks of favor exchanges. While a cross-country study would help address the effect of cultures on guanxi practices, this study is within China and we address implications of the institutional distinction between market and redistributive economies. 2. An alternative measure of search intensity being measured by the average number of jobs applied per week was considered in initial analysis and produced similar results. This measure was dropped from the subsequent analysis. 3. Employers may not announce all of these requirements, but many stressed at least one at the time of advertising job openings. In initial analysis, we tried three single-criterion measures by taking the matching of education, job skill, and prior work experience one at a time, and a continuous measure that scales from none to three-way matches. None of these produced results that ran contrary to those from the chosen dichotomous measure. 4. Migrants made their contribution to this tendency, with 34.5 percent (138) using ‘‘only network’’ and 48.3 percent (193) using ‘‘network in market’’ channels to change jobs.

ACKNOWLEDGMENT The authors thank Lisa Keister and an anonymous reviewer for helpful comments on an earlier draft. A grant (HKUST6052/98H) from Hong Kong’s Research Grants Committee to the first author financed the survey reported in the paper, and a postdoctoral fellowship from the University of Queensland, Australia, supported the second author during the revision of this paper.

REFERENCES Bian, Y. (1994). Guanxi and the allocation of urban jobs in China. The China Quarterly, 140(December), 971–999. Bian, Y. (1997). Bringing strong ties back in: Indirect ties, network bridges, and job searches in China. American Sociological Review, 62, 266–285. Bian, Y. (2002). Institutional holes and job mobility process: Guanxi mechanisms in China’s emerging labor markets. In: T. Gold, D. Guthrie & D. Wank (Eds), Social connections in China: Institutions, culture, and the changing nature of Guanxi (pp. 117–136). New York: Cambridge University Press. Bian, Y. (2008). Urban occupational mobility and employment institutions: Hierarchy, market, and networks in a mixed system. In: D. Davis & F. Wang (Eds), Creating wealth and poverty in China (pp. 165–183). Palo Alto, CA: Stanford University Press.

280


Bian, Y., & Logan, J. (1996). Market transition and the persistence of power: The changing stratification system in China. American Sociological Review, 61, 739–758. Burt, R. S. (1992). Structural holes: The social structure of competition. Cambridge, MA: Harvard University Press. Cahuc, P., & Fontaine, F. (2009). On the efficiency of job search with social networks. Journal of Public Economic Theory, 11, 411–439. Corcoran, M., Datcher, L., & Duncan, G. (1980). Information and influence networks in labor markets. In: G. J. Duncan & J. N. Morgan (Eds), Five thousand American families: Patterns of economic progress (Vol. 8, pp. 1–37). Ann Arbor, MI: Institute for Social Research. Davis, D. S. (1990). Urban job mobility. In: D. Davis & E. F. Vogel (Eds), Chinese society on the eve of Tiananmen (pp. 85–108). Cambridge, MA: Harvard University Press. Gold, T. (1990). Urban private business and social change. In: D. Davis & E. F. Vogel (Eds), Chinese society on the eve of Tiananmen (pp. 157–180). Cambridge, MA: Harvard University Press. Granovetter, M. S. (1973). The strength of weak ties. American Journal of Sociology, 78, 1360–1380. Granovetter, M. S. (1974). Getting a job: A study of contacts and careers. Cambridge, MA: Harvard University Press. Granovetter, M. S. (1981). Toward a sociological theory of income differences. In: I. Berg (Ed.), Sociological perspectives on labor markets (pp. 11–47). New York: Academic Press. Granovetter, M. S. (1995). Afterword 1994: Reconsiderations and a new agenda. In: Getting a job: A study of contacts and careers (2nd ed., pp. 139–182). Chicago: The University of Chicago Press. Guthrie, D. (1998). The declining significance of Guanxi in China’s economic transition. The China Quarterly, 154, 254–282. Guthrie, D. (2002). Information asymmetries and the problem of perception: The significance of structural position in accessing the importance of Guanxi in China. In: T. Gold, D. Guthrie & D. Wank (Eds), Social connections in China: Institutions, culture, and the changing nature of Guanxi (pp. 37–56). New York: Cambridge University Press. Hanser, A. (2002). Youth job searches in urban China: The use of social connections in a changing labor market. In: T. Gold, D. Guthrie & D. Wank (Eds), Social connections in China: Institutions, culture, and the changing nature of Guanxi (pp. 137–161). New York: Cambridge University Press. Huang, X. (2008). Guanxi networks and job searches in China’s emerging labour market: A qualitative investigation. Work, Employment and Society, 22, 467–484. Javanovic, B. (1979). Job matching and the theory of turnover. The Journal of Political Economy, 87, 972–990. Keister, L. A. (2000). Chinese business groups: The structure and impact of interfirm relations during economic development. Oxford: Oxford University Press. Lin, N. (1982). Social resources and instrumental action. In: P. V. Marsden & N. Lin (Eds), Social structure and network analysis (pp. 131–145). Beverly Hills, CA: Sage. Lin, N. (1999). Social networks and status attainment. Annual Review of Sociology, 25, 467–488. Lin, N. (2001). Social capital: A theory of social structure and action. New York: Cambridge University Press. Lin, N., & Bian, Y. (1991). Getting ahead in urban China. American Journal of Sociology, 97, 657–688.


281

MacDonald, G. M. (1980). Person-specific information in the labor market. The Journal of Political Economy, 88, 578–597. Naughton, B. (1995). Growing out of the plan. Cambridge: Cambridge University Press. Osberg, L. (1993). Fishing in different pools: Job-search strategies and job-finding success in Canada in the early 1980’s. Journal of Labor Economics, 11, 348–385. Rosenbaum, J. E., DeLuca, S., Miller, S. R., & Roy, K. (1999). Pathways into work: Shortand long-term effects of personal and institutional ties. Sociology of Education, 72, 179–196. Saloner, G. (1985). Old boy networks as screening mechanisms. Journal of Labor Economics, 3, 255–267. Scott, R. W. (2003). Organizations: rational, natural, and open systems (5th ed.). Upper Saddle River, NJ: Prentice-Hall. Simon, C. J., & Warner, J. T. (1992). Matchmaker, matchmaker: The effect of old boy networks on job match quality, earnings, and tenure. Journal of Labor Economics, 10, 306–330. Walder, A. G. (1986). Communist neo-traditionalism: Work and authority in Chinese industry. Berkeley, CA: University of California Press. Yang, M. (1994). The art in Guanxixue: Ethics, tactics, and etiquette. In: Gifts, favors, and banquets: The art of social relationships in China. Ithaca, NY: Cornell University Press. Yang, M. (2002). The resilience of Guanxi and its new deployments: A critique of some new Guanxi scholarship. The China Quarterly, 170(June), 459–476.

APPENDIX. RESPONSE CATEGORIES AND CODING SCHEMES FOR SELECTED VARIABLES Informational assistance/information networks: My helper Gave me information about a job opening Gave me information about the workplace where there are job openings Gave me advice about how to apply for ‘‘that job’’ Assisted me in preparing my job application Prepared my application Influential assistance/influence networks: My helper Delivered my application to the hiring organization (helper is an employee) Contacted ‘‘a relevant person’’ of the hiring organization (phone call or ‘‘social eating’’) Did something concrete to boost my chances of being hired Arranged an informal meeting with a VIP of the hiring organization (at a banquet) Escorted me to visit the home of a VIP of the hiring organization (gift was most likely) Solved a problem leading to my hire at the organization Helped the organization resolve a concrete program (scarce resource provided) Promised to help with a proposal

282


APPENDIX (Continued ) Jobs of market connectedness: ‘‘How frequently did you contact people in the following positions outside the organization when you performed your job on a daily basis – would you say 1 ¼ never, 2 ¼ rarely, 3 ¼ sometimes, or 4 ¼ often?’’ A. Customers/service receivers B. Clients/suppliers C. All kinds of visitors Jobs of hierarchical bridging: ‘‘How frequently did you contact people in the following positions/units within the organization when you performed your job on a daily basis – would you say 1 ¼ never, 2 ¼ rarely, 3 ¼ sometimes, or 4 ¼ often?’’ A. Upper-level leaders B. Lower-level colleagues C. Colleagues of equal ranks D. Supervisory units E. Subordinary units F. Other work units

DANWEI AND SOCIAL INEQUALITY IN CONTEMPORARY URBAN CHINA Yu Xie, Qing Lai and Xiaogang Wu ABSTRACT Prior research showed that danwei, the work unit, was very important in determining workers’ social, economic, and political lives in pre-reform urban China. In this chapter, we argue that danwei continues to be an agent of social stratification in contemporary urban China. Using data from a 1999 survey that we conducted in three large Chinese cities, Wuhan, Shanghai, and Xi’an, we assess the extent to which workers’ socioeconomic well-being depends on the financial conditions of their danwei. Results show that the financial situation of danwei remains one of the most important determinants of earnings and benefits. However, the explanatory power of danwei’s financial situation is much greater for earnings than for benefits.

INTRODUCTION Walder’s (1986) pioneering work on pre-reform China demonstrated well that the role of danwei prior to the economic reform was all-encompassing for


283

284

YU XIE ET AL.

urban citizens: danwei defined one’s social, economic, and political life. Individuals depended on danwei for almost everything, a phenomenon referred as ‘‘organized dependency.’’ The post-1978 economic reform has changed, but not diminished, the role of danwei. Certain functions of danwei in pre-reform China have been displaced or at least substantially weakened by the market, including housing provision, food rationing, entertainment provision, and political rights conferment. Non-state economic entities, such as private firms and joint ventures with foreign firms, have appeared and gained more prominence in the economy. Even state-owned or statetransformed employers, such as large public-traded firms, universities, and hospitals, have now lost their previous characters as all-encompassing danwei. Technically, they are no longer part of the state’s command economy, which was characterized by ‘‘soft-budget’’ constraints (Kornai, 1986). It is thus an open question whether danwei continue to play an important role in determining social stratification in reform-era urban China. Before we attempt to answer this question, we observe that the government, especially local government, continues to play a strong role in the Chinese economy. At least at the local level, business interests are allied closely with government interests. In the words of Oi (1995, p. 1137), ‘‘local governments in China [are] fully fledged economic actors, not just administrative-service providers as they are in other countries.’’ Oi (1992) has termed the merger of state and economy at the local level ‘‘local state corporatism.’’ One important reason for this merger is that the local government has a vested interest in promoting local economic interests so that it can provide welfare and public services to its population. While Oi’s work focused on rural China, her characterization can be readily extended to urban China. A concrete manifestation of the danwei system is that an employer in contemporary China may not wish to maximize profits by reducing labor costs as low as the labor market equilibrium would allow (Xie & Wu, 2008). Rather, it may be interested in protecting the economic interests of current employees by paying them at levels substantially higher than may be justified by market competition. Thus, Xie and Wu (2008) show that workers’ earnings are strongly affected by their danwei’s financial conditions. In this chapter, we argue – and demonstrate with survey data – that danwei, the work unit, continues to be a main agent of social stratification in contemporary urban China. Using data from a 1999 survey that we conducted in three large Chinese cities, Wuhan, Shanghai, and Xi’an, we assess the extent to which workers’ earnings and benefits depend on their danwei’s financial conditions. We draw heavily from our earlier work on the relationship between danwei’s financial conditions (referred as ‘‘danwei profitability’’ then)

Danwei and Social Inequality in China

285

and earnings inequality (Xie & Wu, 2008) but provide additional results on the role of danwei’s financial conditions in determining three other indicators of well-being, that is, benefits that were provided by danwei in pre-reform China: medical benefits, housing conditions, and pensions for retirees. Taken as a whole, our results reveal that danwei financial conditions continue to be one of the most important determinants of social inequality in urban China as of the late 1990s, although the explanatory power of danwei is much greater in explaining earnings than in explaining benefits.

THEORETICAL ISSUES Danwei in Pre-Reform China The dominant view among Western scholars on contemporary China traces the origin of the Chinese danwei system either to the Communist movement in China or to the corporate culture of the Republic period (Lu¨ & Perry, 1997). However, some scholars in China believe that danwei can also be seen as a modern incarnation of the bureaucratic system that played a central role in the very long history of imperial China. Liu (2000), author of Danwei China, expresses this view forcefully. After agreeing with Weber’s (1951) observation that cities in ancient China were set up less as commercial centers, as in Europe, than as political and military centers representing the emperor, Liu remarks (p. 101): As a result, the establishment of the danwei system does not contradict the cultural and social logic of Chinese society. Instead, it is a structural manifestation of this logic. From ancient times, Chinese cities possessed a variety of the political, economic, and moral elements of danwei; these cultural properties have now been transplanted to the contemporary urban danwei. Thus, to understand the properties of the contemporary and urban version of the danwei system, one cannot overlook the historical tradition.

From this perspective, danwei now serve the mediation role between the central court and the general population – a crucial role played by assigned bureaucratic officials in imperial China (not the noble class as in Europe) (Sun, 2004). In the governance system of imperial China, local officials were not judged on how closely they followed orders from above, but on how well they provided for the well-being of the populations under their governance. The ideal official was someone considered to be a true ‘‘father and mother’’ of the people (Nylan, 1996).

286

YU XIE ET AL.

Sociologists have long studied danwei and its importance in contemporary China (Bian, 1994; Walder, 1986; Whyte & Parish, 1984). Prior to the economic reform in China, practically all urban workers were organized as part of a danwei, be it a factory, a store, a school, or a government office. The danwei organizations had multiple social, political, and economic functions, providing employees with a permanent ‘‘membership’’ of life employment (Naughton, 1997; Hussain, 2000). Workers and their families were totally dependent on their affiliated work units for material resources and career chances (Walder, 1986, 1992). The Chinese urban society was organized as a hierarchy, in which each work organization functioned as a social ‘‘unit’’ in a system dominated by the state. Indeed, membership in a danwei was an important sign of social status and an important vehicle for status attainment and social mobility (Lin & Bian, 1991). Danwei differed in their capability of providing for workers (Bian, 1994; Liu, 2000). Government and party agencies were de facto redistributors. State-owned work units had priority in acquiring resources from the government so as to provide housing and subsidized coupons to their employees. In general, the closer a danwei was to the state administrative hierarchy, the more power it had in bargaining with central planners, and thus the more material resources it had at its disposal for dispensing goods and services to workers (Bian, 1994; Li & Li, 2000; Walder, 1986, 1992). In pre-reform urban China, salaries were to a large extent fixed by the government, based on occupational status, seniority, and administrative position (You, 1998; Korzec & Whyte, 1981; Walder, 1987; Yuan & Fang, 1998, pp. 42–64). There was not much leeway left for a work unit to pay its employees higher cash salaries than another work unit. However, social inequality across danwei was large, albeit more in the form of benefits than in salary (Bian, 1994; Lin & Bian, 1991; Liu, 2000; Walder, 1986; Whyte & Parish, 1984; Yang & Zhou, 1999).

Danwei during the Economic Reform In 1984, the Chinese redistribution system already described was fundamentally reshaped by the market-oriented reform, transforming social inequalities during this era of change. A key policy aspect of this transformation was the loosening of state control over individual work units. Once these danwei had more decision-making power, they were able to keep a more significant share of the surpluses they generated than previously and to distribute these funds to workers as they chose, either by paying them out in bonuses or by


287

spending them on shared living space (Walder, 1987). In the meantime, a growing new market sector began offering alternative resources and life opportunities outside the redistributive sector, such that coordination from above became less essential than previously. During the reform era, scholars continued to observe that work units played a major role in the stratification of urban China (Wang, 2008; Wu, 2002; Xie & Wu, 2008). There is a simple institutional explanation for this: cadres and employees in the same danwei had a common interest in trying to keep the state from extracting resources generated by the danwei, as had also been the case in pre-reform China. A common strategy was to set up new subsidiary firms using assets that were essentially state-owned, staying, however, within the confines of the law in justifying this practice (Guthrie, 1999). For example, factories were allowed to rent out empty offices to generate income and start their own shops, restaurants, or hotels. These danwei-owned businesses are commonly called ‘‘tertiary industries’’ or san chan. The revenues they generated could often be excluded from the accounts of supervising agencies and were mostly retained at the discretion of the work unit (You, 1998, pp. 129–130). The private sector, which consists of private enterprises, foreign companies, joint ventures, and the self-employed, has grown rapidly in the Chinese economy during the economic reform. Despite the private ownership, the Chinese people habitually refer them as danwei. Interestingly, this is more than a matter of inertia in language use. Evidence shows that danwei as an institution is reproduced even in the private sector, as danwei-like institutional features could be observed even in newly emerging sectors such as privately owned high-tech firms in urban China (Francis, 1996), or even industrialized villages in rural China (Lin, 1995). Thus, danwei are likely to continue and to play an important role no matter what economic system takes shape, including the private sector. However, the form the danwei take may vary from one type of organization to another. For example, in the newer private firms, income redistribution typically takes the form of salary or wage increases, while state-owned and state-controlled danwei redistribution is likely to be hidden in bonuses and various forms of welfare benefits.

Danwei and Social Inequality Why do danwei, after being released from tight state control, actively pursue revenue generation and then convert some of the resulting revenue into extra

288

YU XIE ET AL.

earnings and tangible benefits for their employees? The answer is that danwei are not necessarily interested in maximizing profit for danwei property owners, which in most cases are still the state. This may be less true for private firms. However, given the institutional legacy of danwei, they are far from free in pursuing profit. If profit maximization were the sole objective of danwei, highly profitable danwei would be interested in keeping the labor cost low by not redistributing the profit among employees, as they could easily replace their workers at low, market levels. Rather, the internal structure within work units is such that management has an interest in providing additional resources to workers. Despite policies in the early years of the economic reform targeted at breaking up the ‘‘big pot of rice’’ (da guo fan), danwei as a social institution have remained. Thus, distributions of bonuses and other welfare benefits have become a major part of the internal politics within danwei (Dittmer & Lu¨, 1996). As discussed earlier, danwei managers served as intermediates between the state on the one hand and the workers on the other. Walder’s (1986) work on pre-reform danwei clearly documents a dependency relationship between managers and workers. While managers sought cooperation from workers, workers needed managers to provide means and services for their livelihoods. This patron–client relationship enabled workers to extract resources from their managers. During the economic reform, factory managers have increasingly taken on dual roles: they are supposed to make workers more productive while at the same time taking responsibility for their overall well-being. To meet the challenges of this role conflict, managers often retain extra funds and use them to upgrade workers’ living standards. According to a national survey of 2,765 factory managers in 1995, to the question ‘‘What is the most important thing in your management?’’ 48% answered that maximizing workers’ incomes was their priority (You, 1998, p. 140). This expectation of a danwei manager is reminiscent of the ‘‘father and mother’’ image of an ideal local official in imperial China. Our thesis that danwei remains an important agent of social stratification can be interpreted to mean that redistribution within danwei is more equal than the market would tend to justify. Wang’s (2008, pp. 114–117, 150–168) recent book cites a large number of existing studies and provides additional results showing relative income equality within danwei. One of his interesting findings is that in two recent surveys, respondents overwhelmingly complained that the degree of inequality was too high for China as a nation but not too high for their own danwei (p. 167). As a result of this within-danwei egalitarian culture, workers’ earnings fluctuate directly as a function of danwei’s financial conditions (Xie & Wu, 2008).


289

Danwei and Benefits Recall that in pre-reform urban China, the role of danwei was allencompassing, affecting almost every aspect of Chinese life, particularly earnings and benefits. Our earlier work (Xie & Wu, 2008) demonstrated the continued relevance of danwei for earnings. In this chapter, we would like to extend our earlier work to understanding whatever effects danwei may still have on benefits, specifically medical benefit, housing, and pensions. Before we formulate our research hypotheses, let us review the history of changes in the provision of benefits during the economic reform. The ties between benefits and danwei were very close in pre-reform China (Yang & Zhou, 1999). This was true in large part because the state regulated salaries so formulaically that, when possessing extra resources, danwei could only redistribute them to workers through benefits. After the economic reform, the Chinese government wished to create an efficient labor market and found the close ties between danwei and benefits to be a major impediment to labor mobility. Thus, one of the major changes in the reform period has been the breakup of benefits from danwei, through various means. For housing, this has been a story of privatization (Zang, 1999). For medical benefit and pension, it has involved social security, as we will explain below. Prior to the economic reform, housing was a benefit offered by danwei to their employees. As a result, a worker’s housing size and amenities were subject to his/her danwei’s resources in providing the benefit (Bian, Logan, Lu, & Wang, 1997; Zhou & Logan, 1996). As early as 1978, ideas about privatizing housing were first proposed by Chinese leaders. In 1980, the National Urban Housing and Residence Meeting formalized its housing commercialization plans: ‘‘Workers are permitted to build privately owned housing units, purchase publicly owned housing units, and own property. The state would make plans to construct housing units and sell them to the public.’’ In 1988, the National Housing Reform Meeting held by the State Council approved a scheme encouraging workers to buy the existing public housing stock. Finally, a 1998 directive of the State Council declared that the state and the work units stop constructing publicly owned housing units but channel this source of funds toward subsidizing workers in purchasing their own housing units (Pan, 2000). This last measure effectively brought the welfare-oriented housing allocation regime to an end and prompted rapid growth of the real-estate industry in the ensuing years. Our review of recent changes to health insurance and pension policies is heavily drawn from Song (2001) and Song, Zhang, and Zheng (1998). China’s urban health insurance has consisted of two components: Labor

290

YU XIE ET AL.

Insurance and Government Employee Insurance. Labor Insurance began in the 1950s as a danwei-based self-insurance system that covers medical expenses for workers and often their dependents. Government Employee Insurance provides health coverage to employees of state institutions (such as government agencies, schools, and hospitals), funded by general revenues but administered through danwei. In the late 1980s, reform efforts were initiated to reduce the dependency of Labor Insurance on danwei, culminating in the government’s major policy decision in 1998 to establish a social insurance program for urban workers, with contributions jointly from the state, the enterprises, and individual workers. Following trials in Zhenjiang City and Jiujiang City in 1994 and 1995, the Medical Benefit Reform was carried out on a full scale. The new Labor Insurance has two key features: (1) specified contributions from both employers and workers toward health insurance funds managed at the local government level and (2) pre-defined coverage of medical services, including co-pays (around 10– 20%), at the local government level. While thus far, the local government for most places is still at the prefectural/city level, the aim of reformers is to shift responsibility for Labor Insurance to the provincial government. In the meantime, Government Employee Insurance has been modified so as to shift a higher portion of expenses than before to individual employees in the form of co-pays (at 10–20%). Since both Labor Insurance and Government Employee Insurance only provide limited coverage of medical services, a small number of highly profitable danwei may choose to provide premium medical insurances purchased in the open market. The pension insurance system was first established in the early 1950s. It covered the government and public sector, state-owned enterprises (SOE), as well as certain collectively owned enterprises (COE) in urban China, with funds from both danwei and the state. The situation has remained unchanged for the government and public sectors. In 1966, however, state financing retreated from the SOEs’ and COEs’ pension program, leaving danwei solely responsible for retirement benefit provision. During the economic reform, such a heavy financial burden seriously handicapped many enterprises. In the 1983 National Insurance and Welfare Meeting, it was decided to shift the SOEs’ pension burden to the ‘‘society’’ (meaning public) level, and trials followed in a few cities. In 1987, the National Institutional Reform Committee and the Ministry of Labor and Human Resources urged cities and counties to complete the transition of pension provision from danwei to society-based programs within two years. By 1991, the State Council required that pension insurance be jointly contributed by the state, enterprises, and individual workers, and be managed at the


291

province level. In 1995, the No. 6 Document by the State Council extended the coverage of pension insurance to include employees in the private sector. In 1997, the State Council Decisions on establishing the Unified Basic Pension Insurance System for Enterprise Employees specified the new pension insurance program, requiring both workers and their employers to make financial contributions. While the goal is to manage the funds at the provincial level, some of this responsibility still remains at the prefectural/ city level. Hypotheses From the existing literature (e.g., Wang, 2008) and our earlier work (Xie & Wu, 2008), we know that danwei continue to affect workers’ earnings. When a danwei is profitable, some of the profit is directly rechanneled as compensation to its workers. However, compensation does not have to take the form of earnings. In pre-reform China, indeed, such compensation primarily took the form of benefits. In this chapter, we are interested in the potential role of danwei in influencing workers’ benefits, as well as their earnings. We consider three forms of benefits: housing, medical benefits, and retirement pensions. As stated in the preceding session, provision of these three types of benefits has changed radically during the economic reform. Change has mainly taken the form of detaching benefits from danwei so as to facilitate the labor market and encourage job mobility, involving privatization in the case of housing, and public welfare and marketization in the case of medical benefits and retirement pensions. However, it is plausible that danwei have continued to affect benefits in urban China through at least three causal mechanisms. First, a danwei in a good financial situation may continue to provide additional resources to workers, in the form of housing subsidies and/or premium health/ pension benefits purchased in the private market. Second, danwei in a poor financial condition may not comply with government regulations concerning medical benefits and retirement pensions. We suspect that this mechanism was particularly likely to be in operation in the late 1990s, the period under our study. Finally, a component of the health insurance system and the pension system consists of individual accounts, the amounts of which are proportional to individual’s earnings and thus are indirectly affected by danwei’s financial situation. Hence, we propose: Hypothesis 1. Danwei still influences benefits, as well as earnings, in contemporary urban China.

292

YU XIE ET AL.

We recognize, however, that the influence of danwei on benefits in postreform urban China should be much smaller than that on earnings, for several reasons. First, the economic reform has strategically diminished the ties of these benefits to danwei as impediments to a true labor market. Second, unlike earnings, benefits are entitlements, subject to membership boundary restrictions but otherwise given on a formulaic basis. That is, they are supposed to be distributed equally. Third, while earnings can be easily fluctuated according to a danwei’s financial condition, perhaps on a monthly basis, benefits are usually set on relatively long terms and cannot be easily changed. For example, no family can easily change a housing situation on a short-term basis. Wang (2008, p. 81) shows that, using per-capita living space as a measure, inequality in housing has been relatively stable between 1986 and 2000, a period that witnessed a rapid increase in earnings inequality. For these reasons, we propose: Hypothesis 2. Danwei’s effects on benefits, if still present, are smaller than those on earnings. Besides these two key hypotheses that motivate the study, we also consider the possibility that the role of danwei may vary by region or sector. It may vary by region because the economic reform in China has been regionally uneven, with some regions far ahead of others (Xie & Hannum, 1996). It may vary by sector because danwei were an institutional legacy from a socialist planned economy, and its role should be more pronounced there than in the newly emerged private sector. Thus, we propose: Hypothesis 3. Danwei’s effects are smaller in more marketized regions or sectors than in less marketized regions or sectors.

RESEARCH DESIGN While Xie and Wu’s (2008) study has confirmed the important role of danwei in determining earnings inequality, a question remains as to whether danwei affect other indicators of well-being that are associated with workers’ benefits. A study of the influence of danwei on benefits is important because, prior to the economic reform, the strong influence of danwei on workers’ well-being was most clearly manifested through benefits rather than through direct earnings (Bian, 1994; Lin & Bian, 1991; Liu, 2000; Walder, 1986; Whyte & Parish, 1984). As discussed earlier, following the economic reform, danwei have been able to use discretionary funds to reward workers directly.


293

It is thus possible that the economic reform has changed the mechanism by which danwei affect workers from being benefits-based to being earningsbased. This conjecture is sensible given that many benefits previously available only through danwei (such as housing and medical insurance) can now be purchased through the open market. This chapter represents our effort to study the potential roles of danwei in affecting worker’s benefits as well as their earnings. We ask whether or not danwei continue to serve as a major agent of social stratification in affecting not only earnings, but also benefits in urban China. We attempt to answer this question in the following analysis of data from a survey, ‘‘Study of Family Life in Urban China,’’ which we conducted in the summer of 1999 in three Chinese cities: Shanghai, Wuhan, and Xi’an. We also refer to the study as the ‘‘Three-City Survey.’’ In the Three-City Survey, we designed a few items that specifically allowed us to address our research question. At each of the research sites, the Three-City Study initially targeted a probability sample of 1,000 households, with a two-stage probability sampling method. At the first stage, 50 neighborhood communities were randomly chosen in proportion to size. Within each selected neighborhood community, 20 households were randomly chosen. A Kish table was used to select an adult respondent (18 years or older) within each selected household. If the person being interviewed was younger than 60 years, we first interviewed the person with Questionnaire A. We then interviewed one of his/her parents with Questionnaire Aþ, which was specifically tailored to the elderly. If the person initially selected was 60 years or older, we interviewed the person with Questionnaire B, which is similar to Questionnaire Aþ for elderly respondents. We then randomly selected one of his/her children for an interview with Questionnaire Bþ, which is very similar in content to Questionnaire A for adult respondents. There were four outcome variables for the four separate analyses in our study. For the first analysis, we measured earnings by the total annual earnings in 1998. The earnings were composed of three parts: monthly salary from regular job, monthly bonus, and year-end bonus. This variable was analyzed in Xie and Wu (2008). The analytical sample for the analysis of earnings consisted of respondents across the forms between ages 20 and 60 years who earned at least 1,000 RMB yuan in 1998 regular salary and had valid responses to other independent variables. After the restriction, we were left with a total of 1,771 cases. For the second analysis, we measured medical benefit from responses to the question ‘‘If you were hospitalized and your medical costs were 1,000 Yuan, how much would the state or your work unit

294

YU XIE ET AL.

pay towards this expenditure, either by paying directly or by reimbursing you later?’’ For the study of medical benefit, we further restricted the sample to observations for which responses to this outcome variable were valid, resulting in 1,753 cases. For the third analysis, we measured housing size by the area (in square meters) of apartment divided by the total number of persons living in it. Again, for the analysis of housing size, we deleted observations which had missing values for the outcome variable, reducing the number of cases to 1,745. For the fourth analysis, we measured pension amount by the annual amount of total pension that was received by a retiree. Here, we utilized a particular strength of the data and focused on retired persons receiving pensions among the elderly population in the study (i.e., those responding to Form Aþ and Form B). After deleting missing values, we obtained an analytical sample with 1,449 cases. For all the four analyses, we took the natural logarithm of the four outcome variables – earnings, medical benefit, housing size, and pension amount – to form the dependent variables in regression analyses. Besides correcting for skewness in distribution for these variables, dependent variables after the logarithm transformation rendered the regression coefficients easily interpretable, as their exponentials represented multiplicative effects. Thus, regression coefficients had comparable interpretations across regressions with the four different dependent variables, as relative effects. The independent variables in this study included the following: city in which the respondent resided and was interviewed, the highest education level attended, years of work experience, gender, cadre status for the job worked at the survey time in 1999, sector of the job worked at the survey time in 1999, and danwei’s financial situation for the job worked at the survey time in 1999. For the third analysis on housing, we added the logged total annual household income in 1998 as a control variable. For the fourth analysis on pension, we included three additional independent variables: (1) a dummy indicating lixiu status as those who belonged to this category held higher official positions and were given retirement privileges; (2) number of years since retirement; and (3) a crude measure of pre-retirement income in 1989, which was imputed from responses to interval categories retrospectively ascertained in 1989. We presented the descriptive statistics of the variables used in the study in the appendix by analyses. The cities were Shanghai (reference category), Wuhan, and Xian. There were six categories of the highest education level attended: no schooling, primary school, junior high school, senior high school or technical senior high school, junior college, and four-year college or graduate school. This variable was obtained from a sequence of questions of respondents’ education history.


295

Work experience was measured in years and was calculated by subtracting the year beginning the first job from the year ending the last job or from the survey year, 1999, for those who were still working. If the year ending the last job was missing, the year retired was used. For those who never worked before, year of work experience was coded 0. Gender is a dichotomous variable, with male as the reference group. Cadre status is a dichotomous variable and was coded 1 if one’s job rank at the survey time was a cadre at the rank of department level (ke ji) or above and 0 if the rank was below ke ji or not a cadre. Sector is a categorical variable with four groups: government or public sector (reference category), SOE, COE, and privately owned enterprise, which includes Chinese private firms, foreign firms, joint ventures, and the self-employed. There were three cases of peasant work or doing odd jobs without a work unit, and these were excluded from this study. The information for work experience, cadre status, and sector was obtained from questions about respondent’s job history. The key explanatory variable in this study pertains to a measure of danwei’s financial situation. In the questionnaire, we asked the respondent to answer the following question: ‘‘Is the economic situation of your work unit much better, better, about the same, worse, or much worse than that of other work units of this city?’’ The respondent was given five options: ‘‘Much better,’’ ‘‘Better,’’ ‘‘About the same,’’ ‘‘Worse,’’ and ‘‘Much Worse.’’ Note that the respondent was instructed to interpret danwei broadly, as the work unit that issued ‘‘worker’s card’’ (gongzuo zheng). It thus covered all types of work organizations, including government agencies, institutions, state firms, collective firms, and private firms, for which there was no single objective measure of extra revenue generated. Obviously, the measure employed here was a very crude way to assess a danwei’s financial situation, as it provided no more than a subjective assessment on the part of the respondent. To make things worse, it was possible that the respondent’s subjective assessment of the danwei’s financial situation could be influenced by the amount of earnings he/she received, our dependent variable in this study. We reverse coded the variable so that a higher number/category represents better financial conditions. We used the measures in two specifications, as a set of dummy variables (with ‘‘Much Worse’’ as the reference category) and as an interval variable from 1 to 5. The methodological limitation of this measure was discussed in Xie and Wu (2008). The statistical analyses proceeded in three stages. In the first stage, we examined the amount of variance in earnings that could be explained by the danwei financial condition measure, relative to amounts of variance explained by other variables. We did this both separately (i.e., variable by

296

YU XIE ET AL.

variable) and jointly in a regression analysis. Second, we examined the net impact of the danwei financial condition on the three benefit outcome variables, as well as earnings, in a regression framework controlling for other covariates. Third, we tested the interaction effects of the danwei financial condition with city and sector of employment. These tests revealed whether danwei’s financial situation mattered more in a more marketized region/sector than in a less marketized region/sector. Throughout the three analytical stages, we compared the results across the four analyses with four different outcome variables.

RESULTS In Table 1, we present results that compare the explanatory power of different determinants of the four outcome variables. The table is divided into four panels, with results for each outcome variable presented in a separate panel. The first column, labeled ‘‘DF,’’ is the degrees of freedom for a particular determinant. For danwei financial condition, we use both a linear function (1 DF) and a set of dummies (4 DF). The second column, labeled ‘‘R2,’’ is the proportion of the variance in logged earnings that is explained by a determinant, without control of any other variables. In the third column, ‘‘DR2(1),’’ we give the net proportion of the variance explained by a determinant after the inclusion of danwei financial condition in its linear form. Finally, the fourth column, ‘‘DR2(2),’’ represents the net proportion of the variance explained by a determinant after the inclusion of danwei financial condition in its linear form and all the other predictors considered in the study. Several notable findings emerge from Table 1. First, danwei financial condition is a statistically significant predictor of all the four outcomes. Second, the explanatory power of danwei financial condition varies, its effect on earnings being much more pronounced than its effects on the three benefit outcomes. Specifically, while danwei financial condition explains between 9 and 13% of the variation in logged earnings, depending on specification, it explains around 3% of the variation in medical benefit and 1% or lower of the variations in housing size and pension amount. Third, while the variation in the three benefit outcomes across cities is statistically significant, these regional variations are much smaller than those for earnings. Now let us compare the role of the determinants across the four outcome variables. Results pertaining to earnings were discussed in Xie and Wu (2008). We find that the determinants as a whole play a much stronger role


Table 1.

297

Percent Variance explained in Earnings, Medical Benefit, Housing Size, and Pension Amount. DF

R2

DR2(1)

DR2(2)

Earnings (N ¼ 1,771) City Education Level Experienceþexperience2 Gender Cadre status Sector Danwei financial condition (linear) Danwei financial condition (dummies)

2 5 1 1 1 3 1 4

17.47 7.82 0.23 4.78 3.08 3.54 12.52 12.89

18.11 5.49 0.17 4.84 2.27 2.18

19.12 4.46 0.05 3.05 0.63 1.80 9.30

Medical benefit (N ¼ 1,753) City Education level Experienceþexperience2 Gender Cadre status Sector Danwei financial condition (linear) Danwei financial condition (dummies)

2 5 1 1 1 3 1 4

8.33 2.54 2.41 0.26 1.69 13.79 3.09 3.76

13.52 2.08 3.20 0.27 1.39 13.63

8.08 1.10 0.87 0.01 0.14 11.23 2.92

Housing size (N ¼ 1,745) City Education level Experienceþexperience2 Gender Cadre status Sector Danwei financial condition (linear) Danwei financial condition (dummies) 1998 household income

2 5 1 1 1 3 1 4 1

4.13 3.51 3.40 0.04 2.17 1.86 1.02 1.09 0.33

4.00 3.17 3.15 0.04 1.98 1.73

3.86 0.80 1.77 0.04 0.47 1.00 0.30

0.12

0.42

Pension amount (N ¼ 1,449) City Education level Experienceþexperience2 Gender Cadre status Sector Danwei financial condition (linear) Danwei financial condition (dummies) Lixiu (privileged retirement) Years since retirement 1989 salary

2 5 1 1 1 3 1 4 1 1 1

1.45 0.64 2.15 1.74 0.90 2.01 1.18 1.58 1.11 1.55 1.47

1.37 0.59 1.81 1.49 0.69 1.47

1.53 0.47 0.61 0.08 0.11 0.87 0.30

0.94 1.71 1.06

0.21 2.45 0.50

Variables

Note: DF refers to degrees of freedom. DR2(1) refers to the incremental R2 after the inclusion of danwei’s financial situation (linear). DR2(2) refers to the incremental R2 after the inclusion of all the other variables. Source: 1999 Three-City Survey. po0.05, po0.01, po0.001, based on F-tests.

298

YU XIE ET AL.

for earnings than for benefits. Across the three measures of benefits, they are more important in determining medical benefits and least important in determining pension amount, with housing size in between. These results make sense, because by definition benefits are distributed as entitlements and are thus more subject to equal distribution than earnings are. For example, while a large portion of earnings for many workers consists of bonuses and subsidies that are flexible depending on many factors such as assigned tasks and business cycle, pension amounts are basically set formulaically by the local government based on such hard factors as position prior to retirement, pre-retirement salary, work experience, and lixiu status. For medical benefit, the most significant explanatory factor is sector, contributing 11–14% of variation. This is because, while universally provided by all state employers, medical benefits were still at the discretion of employers in the private and collective sectors in 1999. Thus, workers in the private and collective sectors were at a severe disadvantage compared to workers in the state sector. In addition to the sector differences, danwei financial condition still plays a large role – indeed larger than individuallevel attributes such as education and work experience. There is also a relatively large regional variation. For both housing and pension, danwei financial condition claims little explanatory power (ranked 6th and 7th, respectively). Sector of employment is more important. Between these two outcomes, individual characteristics play a more important role in determining housing outcome. The estimated effects of education on pension are not even statistically significant. Therefore, in terms of housing size and pension amount, danwei are no less important as individuals’ personal attributes. To evaluate the influence of various factors on the four outcomes more directly, we present the estimated coefficients from four multivariate regression models in Table 2. The regression models we choose to present are simple additive models. They differ slightly across the four outcome variables, as additional control variables are included for the analysis of housing size and pension amount. As we observed before, there are regional variations in all four outcomes across the cities. However, the directions of the variations are different. For earnings and medical benefit, we see the expected order of Shanghai– Wuhan–Xi’an, as would be predicted by their relative levels of economic development. Given the log scale of the dependent variable, we estimate that the average earning in Wuhan is at the 58% (exp(0.539)) level compared with that in Shanghai, and that the average earning in Xi’an is at


Table 2.

299

Regression Coefficients for Four Outcome Variables (Logged).

Variables

Earnings (b)

Intercept City (Shanghai ¼ excluded) Wuhan Xi’an

8.237 0.539 0.658

Education (none ¼ excluded) Primary Junior high Senior high Junior college College

0.414 0.447 0.592 0.778 0.923

Medical Benefit (b) 2.540 1.222 2.035 0.799 1.490 1.606 1.886 2.515

Experience Experience (1,000) Experience2 (1,000)

2.421 0.017

78.816 1.240

Gender (male ¼ excluded) Female

0.225

0.056

0.185

0.395

Cadre status (non-cadre ¼ excluded) Cadre

Housing Size (b)

Pension Amount (b)

1.700 0.288 0.285

5.310 0.534 0.566

0.155 0.209 0.217 0.107 0.027 24.838 0.728

33.558 0.207

0.026

0.134

0.151

Sector (government þ institution ¼ excluded) State-owned 0.043 Collectively owned 0.224 Privately owned 0.114

0.223 1.118 2.723

0.057 0.153 0.048

0.227

0.572

0.040 0.076

Danwei financial condition (linear) 1998 household income (logged) Lixiu (privileged retirement) Years since retirement

0.213 0.060 0.546 0.577 0.107 0.406 0.052 0.192

1989 salary (logged) R2 (%) Sample size (N)

0.244 0.156 0.049 0.154 0.480

43.92 1,771

28.05 1,753

12.17 1,745

9.98 1,449

Note: Results are derived from multivariate regressions with different dependent variables. Source: 1999 Three-City Survey. po0.05, po0.01, po0.001, based on two-side t-tests.

the 52% (exp(0.658)) level compared with that in Shanghai. For housing size, residents in Shanghai are disadvantaged relative to those in Wuhan and Xi’an, reflecting crowding and a very high housing cost in Shanghai. Retirees in Wuhan and Xi’an have lower pensions (at about 40% less) than those in Shanghai.

300

YU XIE ET AL.

While education has very strong monotonic effects on earnings, it has no influence on benefits, except that workers with college educations enjoy significantly more medical benefits. This is not surprising, given our earlier discussion that benefits are usually given equally to all workers within danwei. The estimated relationship between experience and housing size is U-shaped in that those in the middle years of work experience (around 15– 25 years) are most disadvantaged. There are two reasons for this U-shaped curve. The first is compositional, as the outcome measure is per-capita living space. Both old and young residents may not have children living with them at home, boosting per-capita living space. The second is rooted in the history of the housing reform. While older residents were able to purchase apartments that were assigned to them by their danwei at deeply discounted rates, some middle-class young Chinese may have purchased apartments in the open housing markets in recent years. The advantage of a cadre for earnings and housing is apparent, at a premium, respectively, at 20 and 15%. Further, sector of employment appears to affect different outcomes differently. It is interesting that workers in the private sector have highest earnings but receive the least amount of medical benefit, as if they trade for their less generous medical benefit with higher earnings. Workers in the collective sector are disadvantaged relative to workers in the public sector, except for housing, for which they have an advantage. Of particular interest to our study are the coefficients of danwei financial condition. Here, we observe large effects on earnings and medical benefit but small effects on housing and pension. The coefficients mean that one unit change in danwei financial condition is associated with a 25% increase in earnings and a 77% increase in medical benefit. For housing size and pension amount, the corresponding change is only 4 and 11%, respectively. Still, for all three benefit outcomes, the effect of danwei financial condition is statistically significant at the 5% level. This result supports our first hypothesis that danwei still plays a role in influencing benefits in contemporary urban China. Results in Tables 1 and 2 also shed light on our second hypothesis. By bivariate R2 and incremental R2 (reported in columns 2–4 in Table 1), we know that danwei financial condition is a very important determinant of earnings, second only to region, explaining between 9 and 12% of the total variation. In contrast, danwei financial condition explains a very small amount of variation in benefit outcomes (between 0 and 3%). In terms of estimated effect sizes, we commented earlier that danwei financial condition has the highest effect on medical benefit, followed by its effect on earnings. The effects on housing and pension are small. Taken as a whole, the results


301

confirm the second hypothesis that danwei affects earnings more strongly than benefits. In the final step of the data analysis, we add to the additive models presented in Table 2 the interaction effects between danwei financial condition and two structural measures that approximate marketization: city and sector. We know that Shanghai is more marketized than Wuhan, and Wuhan is more marketized than Xi’an (Xie & Wu, 2008). If marketization weakens the importance of danwei in earnings determination, we would expect a significant difference in the effect of danwei’s financial condition among the three cities. Similarly, we expect the following order from more marketization to less marketization among the sectors: privately owned, collectively owned, state-owned, and government/public. Again, if marketization weakens the role of danwei, we would expect significant interaction terms between danwei financial condition and sector. To test these interactions, we ran a total of eight new models. Out of the total of 20 interaction parameters in the 8 models, only 1 coefficient is statistically significant: the interaction effect between Xi’an and danwei financial condition on medical benefit (estimated coefficient ¼ 0.612, with standard error of 0.168). This interaction effect means that danwei financial condition is much more important for medical benefit in Xi’an than in Shanghai. By and large, however, the results of the interaction analysis reveal that the effect of danwei financial condition does not vary much by city or sector. This finding means that danwei continues to play a strong role in determining earnings among Chinese workers, irrespective of city and sector. In particular, we reject our third hypothesis that danwei is much less important in a more developed city than in a less developed city, or less important in a more marketized sector than in a less marketized sector.

CONCLUSION In this chapter, we examine the role of danwei in social stratification in contemporary urban China, extending the focus from mere earnings to benefits plus earnings. Our analysis of survey data collected in 1999 in three major Chinese cities (Shanghai, Wuhan, and Xi’an) has yielded three main findings. First, danwei financial condition still exerts positive effects on worker’s benefits, as well as earnings. Second, danwei financial condition affects earnings more strongly than it affects benefits. Third, danwei’s effects on benefits and earnings do not vary by region or sector.

302

YU XIE ET AL.

The first finding shows that danwei continues to play a very important role in determining the economic well-being of workers in urban China. This finding corroborates well with the observation that danwei serves an important function as the mediating agent between the individuals on the one hand and the very large society of China as a whole on the other hand (Liu, 2000; Sun, 2004; Xie & Wu, 2008). It is also used effectively as a boundary to protect group interests (Wang, 2008). While the economic reform has dramatically changed many facets of the economic and social structure, this mediating function of danwei is still relevant today. The second finding suggests that the concrete mechanism through which danwei affect workers has changed in post-reform China. In pre-reform China, danwei were largely responsible for generating inequality in benefits, but not in earnings. During the economic reform, the central role of danwei in determining benefits has been displaced either by the private market (in the case of housing) or by social welfare (in the case of medical benefit and pension). As a result, danwei financial condition now holds much smaller explanatory power for benefits than for earnings. The third finding indicates that danwei remains an important agent of social stratification, irrespective of region and sector. We conjecture that this surprising finding confirms the structural, mediating role of danwei in China. As we commented earlier, one possible difference between private firms and state-owned or state-controlled work units is that income redistribution in the former is directly reflected in salaries and wages rather than hidden in bonuses and other forms of welfare benefits, as in the latter. For danwei to remain an important stratification agent, labor mobility needs to be low. This was the case until recent years (Naughton, 1997; Wang, 2008). Not until the late 1990s did the Chinese government begin to change social conditions that would facilitate labor mobility. While competitive labor markets have existed in China in recent years, particularly for the two extremes of very low-skilled workers and very highly skilled workers, mobility has been relatively low for ordinary Chinese middle-class workers, for whom a job with a good employer still means lifetime employment. As reviewed by Wang (2008, p. 177), ‘‘a 1999 national survey of urban Chinese workers found that as many as 78 percent of respondents had only held one job, and only 6 percent had held three or more jobs.’’ In this low-mobility regime, securing a job with a good employer is the surest way of securing one’s economic well-being (Bian, 1994). While an individual’s personal attributes associated with productivity, such as education, have become more and more important in China (Hauser & Xie, 2005), they are confounded and constrained by danwei as a social institution. Hence, we


303

conclude by quoting Xie and Wu (2008, p. 580): ‘‘one cannot truly understand social stratification in China without properly understanding the important role played by danwei.’’

ACKNOWLEDGMENTS The research is supported by a research grant from the National Science Foundation, a Guggenheim Fellowship from the John Simon Guggenheim Memorial Foundation, and a Fogarty grant for international research from the National Institute of Child Health and Development, and a grant from the Hong Kong Research Grants Council (GRF 644208). The authors are grateful to Xi Song and Jia Yu for research assistance.

REFERENCES Bian, Y. (1994). Work and authority in urban China. Albany: State University of New York Press. Bian, Y., Logan, J. R., Lu, H., & Wang, Y. (1997). Work units and housing reform in two Chinese cities. In: X. Lu & E. J. Perry (Eds), Danwei: The changing Chinese workplace in historical and comparative perspective (pp. 223–250). Armonk, NY: M.E. Sharpe. Dittmer, L., & Lu¨, X. (1996). Personal politics in the Chinese danwei under reform. Asian Survey, 36, 246–267. Francis, C.-B. (1996). Reproduction of danwei institutional features in the context of China’s market economy: The case of Haidian District’s high-tech sector. The China Quarterly, 147, 839–859. Guthrie, D. (1999). Dragon in a three-piece suit. Princeton, NJ: Princeton University Press. Hauser, S., & Xie, Y. (2005). Temporal and regional variation in earnings inequality: Urban China in transition between 1988 and 1995. Social Science Research, 34, 44–79. Hussain, A. (2000). The social role of the Chinese state enterprises. In: M. Warner (Ed.), Changing Workplace Relations in the Chinese Economy (pp. 57–76). London: Macmillan. Kornai, J. (1986). Contradictions and dilemmas: Studies on the socialist economy and society. Cambridge, MA: The MIT Press. Korzec, M., & Whyte, M. K. (1981). Reading notes: The Chinese wage system. The China Quarterly, 86, 248–273. Li, L., & Li, H. (2000). China’s work unit organization (in Chinese). Hangzhou: Zhejiang Renmin Press (Zhejiang People’s Press). Lin, N. (1995). Local market socialism: Local corporatism in action in rural China. Theory and Society, 24, 301–354. Lin, N., & Bian, Y. (1991). Getting ahead in urban China. American Journal of Sociology, 97, 657–688.

304

YU XIE ET AL.

Liu, J. (2000). Danwei China (in Chinese). Tianjin, China: Tianjin People’s Press. Lu¨, X., & Perry, E. (1997). Introduction: The changing Chinese workplace in historical and comparative perspective. In: X. Lu¨ & E. Perry (Eds), Danwei: The changing Chinese workplace in historical and comparative perspective (pp. 3–17). Armonk, NY: M.E. Sharpe. Naughton, B. (1997). Danwei: The economic foundations of a unique institution. In: X. Lu¨ & E. Perry (Eds), Danwei: The changing Chinese workplace in historical and comparative perspective (pp. 169–194). Armonk, NY: M.E. Sharpe. Nylan, M. (1996). Confucian piety and individualism in Han China. Journal of the American Oriental Society, 116.1, 1–16. Oi, J. C. (1992). Fiscal reform and the economic foundations of local state corporatism in China. World Politics, 45(1), 99–126. Oi, J. C. (1995). The role of the local state in China’s transitional economy. The China Quarterly, 144, 1132–1149. Pan, G. (2000). The process of Chinese housing reform. Chinese Reform No. 11 (pp. 15–19, in Chinese). Song, X. (2001). Chinese social welfare institutional reform and development report (in Chinese). Beijing: Zhongguo Renmin Daxue Chubanshe (China Renmin University Press). Song, X., Zhang, Z., & Zheng, D. (1998). 20 years of Chinese social welfare institution (in Chinese). Zhengzhou: Zhongzhou Guji Chubanshe (Zhongzhou Ancient Literature Press). Sun, L. (2004). Transition and breakup (in Chinese). Beijing: Qinghua University Press. Walder, A. G. (1986). Communist neo-traditionalism: Work and authority in Chinese industry. University of California Press. Walder, A. G. (1987). Wage reform and the web of factory interests. The China Quarterly, 109, 22–41. Walder, A. G. (1992). Property rights and stratification in socialist redistributive economies. American Sociological Review, 57, 524–539. Wang, F. (2008). Boundaries and categories: Rising inequality in post-socialist China. Stanford, CA: Stanford University Press. Weber, M. (1951). The religion of China: Confucianism and Taoism. Glencoe, IL: Free Press. Whyte, M. K., & Parish, W. (1984). Urban life in contemporary China. Chicago: University of Chicago Press. Wu, X. (2002). Work units and income inequality: The effect of market transition in urban China. Social Forces, 80(3), 1069–1099. Xie, Y., & Hannum, E. (1996). Regional variation in earnings’ inequality in reform-era urban China. American Journal of Sociology, 102, 950–992. Xie, Y., & Wu, X. (2008). Danwei profitability and earnings inequality in urban China. The China Quarterly, 195, 558–581. Yang, X., & Zhou, Y. (1999). China’s danwei system (in Chinese). Beijing: Chinese Economy Press. You, J. (1998). China’s enterprise reform: Changing state/society relationship after Mao. London: Routledge. Yuan, Z., & Fang, Y. (1998). The change of employment system in China, 1978–1998 (in Chinese). Taiyuan: Shanxi Jingji Press (Shanxi Economy Press). Zang, X. (1999). Urban housing reforms in China. In: X. Zang (Ed.), China in the reform era (pp. 49–79). Commack, NY: Nova Science Publishers. Zhou, M., & Logan, J. (1996). Market transition and the commodification of housing in urban China. International Journal of Urban and Regional Research, 20, 400–421.


305

APPENDIX. DESCRIPTIVE STATISTICS Variable

Analysis 1: Earnings Mean

S.D.

Analysis 2: Medical Benefit Mean

S.D.

Analysis 3: Housing Size Mean

S.D.

Analysis 4: Pension Amount Mean

S.D.

Dependent variable

9.027

0.624

5.073

2.819

2.429

0.594

8.028

1.967

Independent variables City Shanghai Wuhan Xi’an

0.312 0.353 0.335

0.463 0.478 0.472

0.309 0.353 0.338

0.462 0.478 0.473

0.312 0.351 0.337

0.464 0.477 0.473

0.289 0.376 0.335

0.454 0.485 0.472

0.005 0.030 0.298

0.071 0.170 0.457

0.005 0.030 0.297

0.071 0.170 0.457

0.005 0.030 0.299

0.072 0.172 0.458

0.239 0.260 0.246

0.426 0.439 0.431

0.439

0.496

0.440

0.497

0.437

0.496

0.167

0.373

0.158

0.365

0.156

0.363

0.158

0.365

0.041

0.199

0.071

0.256

0.071

0.257

0.070

0.256

0.046

0.210

Education No schooling Primary Junior high school Senior high school Junior college College Experience Experience Experience2 Gender Male Female Cadre status Cadre (ke ji or above) Sector Government and public institution State-owned Collectively owned Privately owned Danwei financial condition (linear)

20.107 8.973 20.110 8.965 20.201 8.923 31.335 8.816 484.752 375.857 484.712 376.124 487.667 374.893 1059.581 542.861 0.577 0.423

0.494 0.494

0.578 0.422

0.494 0.494

0.575 0.425

0.494 0.494

0.411 0.589

0.492 0.492

0.094

0.292

0.095

0.293

0.095

0.293

0.155

0.362

0.219

0.414

0.218

0.413

0.219

0.414

0.174

0.379

0.516 0.094

0.500 0.292

0.517 0.095

0.500 0.293

0.517 0.092

0.500 0.289

0.635 0.160

0.482 0.367

0.172

0.377

0.170

0.376

0.171

0.377

0.031

0.174

2.954

0.862

2.952

0.863

2.955

0.860

2.605

1.052

306

YU XIE ET AL.

APPENDIX. (Continued ) Variable

Analysis 1: Earnings Mean

S.D.

Danwei financial condition (dummies) Very poor 0.076 0.265 Relatively 0.147 0.355 poor Average 0.545 0.498 Fairly good 0.211 0.408 Very good 0.021 0.143

Analysis 2: Medical Benefit Mean

S.D.

Analysis 3: Housing Size Mean

S.D.

0.075 0.147

0.264 0.354

0.190 0.236

0.393 0.425

0.546 0.209 0.021

0.498 0.407 0.144

0.547 0.211 0.021

0.498 0.408 0.142

0.373 0.179 0.021

0.484 0.383 0.145

9.728

0.604 0.060

0.238

10.773

7.163

7.678

0.801

Years since retirement 1989 salary (logged)

Source: 1999 Three-City Survey.

S.D.

0.266 0.354

Lixiu (privileged retirement)

1,771

Mean

0.076 0.147

1998 household income

N

Analysis 4: Pension Amount

1,753

1,745

1,449

THE PHASE-OUT OF THE UNFIT: KEEPING THE UNWORTHY OUT OF WORK Dorothy J. Solinger Ever since the old danwei [or ‘‘work unit’’]1 – whose celebrated ‘‘iron rice bowl’’ with its inviolable pledges of stability and security – bit the dust in the course of the relentless march of the market and its technologies into China, not just a protected job, but the Office, the Shop and the Factory writ large have become effectively off-bounds to hordes of ordinary people in contemporary urban China. By now it is well known that many millions of the masses who once possessed gongzuo, or ‘‘work,’’ in these places of business saw their livelihoods lost in and after the late 1990s.2 Indeed, where ‘‘work’’ in Chinese cities once meant nothing more nor less than permanent formal-sector employment in a state-run entity, (Henderson & Entwisle, 2000; Henderson et al., 2000) that sort of activity is itself slipping away: During the 9th Plan (1995–2000), employees in the traditionally formal sectors (state and collectively owned firms) decreased substantially; according to a report done in 2000, 44.3 million employees had been thrown out by then, and the ratio of this type of labor to total employment had dropped from the 73.5 percent as of 1995 to 45.1 percent (Hu, n.p., n.d.). This is occurring just as enormous numbers of those who partook of state work in the past are now more likely to be found in their homes (often sickly, almost always listless) or out on the streets (whether selling, sweeping, doing service work, or sounding off) than Work and Organizations in China after Thirty Years of Transition Research in the Sociology of Work, Volume 19, 307–336 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1108/S0277-2833(2009)0000019014

307

308

DOROTHY J. SOLINGER

engaged, either indoors or out, in any kind of salaried assignment. Surely related to these discharges, somewhere between 20 and 50 million urbanites,3 probably some 70–80 percent of whom come from the ranks of the newly jobless,4 have grown to constitute a novel grouping in the municipalities, a collectivity known as the ‘‘tekunhu,’’ meaning households in special difficulty, one special segment among the newly constituted ‘‘ruoshi qunti,’’ weak masses.5 These are the people seen as too old, too weak, or too ignorant to fit into what is felt in China to be ‘‘modern society.’’6 It would quite likely be accurate to claim that practically none of these people has reentered into the formal workforce since being expelled from it over the past decade. Quite true, I’m convinced, no leaders in Beijing – nor even functionaries at the grass roots – would openly assert that these individuals should be prevented from participating in labor; my research – whether documentary or in the field – uncovers nothing to that effect. Nor, I imagine, is that outcome even anyone’s hidden agenda. Still, from my observations it appears that the story of their lack of placement could be read to demonstrate that both official regulations and the actions of local administrators, both bent on making China more and more modern, operate to achieve that end, that is, to keep them out of the workforce. This is the case I set out to make. I first provide some background as to how these two groups – first those who lost their work posts, and then those who became destitute – came into being; I then explain what is being done by the state to attenuate their difficulties. I go on to argue that in the nation’s march toward affluence and state-of-the-art metropolises – despite its politicians’ persistent pleas to bring down the rate of unemployment – people of this sort are at least subliminally handled as if or – one might say, effectually treated as if – best kept out-of-sight, even off the work rolls. I explore how a set of disincentives built into employment policy, along with the externalities (what I call, in sync with contemporary jargon, ‘‘collateral damage’’) of other, apparently unrelated policies, have come together to discourage job holding by the new middle-aged workless and indigent in the cities since the 1990s.

BACKGROUND7 The cities, for the most part, appeared up until the middle of the 1990s to be islands within the larger Chinese political economy in which job-secure workers could be certain that their livelihood, health, education, and living

The Phase-Out of the Unfit: Keeping the Unworthy Out of Work

309

abodes would evermore undergird their and their children’s sustenance. At least until the late 1980s, urbanites who stuck with the state sector even considered good treatment on the job a kind of birthright, an entitlement that was sure to be enforced. In the cities, true, there had always been the disadvantaged after 1949 – those without offspring or spouses, the disabled, and people unable to support themselves. But this relatively tiny batch of individuals generally survived in the shadows and out of sight, subsisting – but just barely – as members of the ‘‘three withouts’’ on a mere pittance, in the form of meager ‘‘social relief’’ disbursed by civil affairs departments.8 Not only did cities seem immune from the perils of hardship up until just over a decade ago, but also there was even reason to hope that the rise in living standards that followed China’s opening up after 1980 would continue for everyone in them. For the dawn of the switch to the market economy in the early 1980s was accompanied by the credo that the wealth being generated – first by the fertilization lent by the inrush of foreign capital along the coast and later by rapidly shooting sprouts of the private sector – would in time shed seeds that fostered prosperity much more widely. Optimism, especially in macro terms, was fueled by statistics such as the following: during the Ninth Five-Year Plan period (1996–2001), per household disposable income increased at an average annual per person rate of 5.7 percent in real terms in the cities, while the urban wage for staff and workers’ experienced an average per person annual increase of 15.9 percent. Permanent residents’ average disposable income had climbed up to 6,860 per year, an improvement of 19-fold over the year 1978 when the post-Mao economic reforms began.9 But the drama of the displacement of these now disqualified citizens unfolded precisely in the midst of their state’s hell-bent drive to ‘‘develop,’’ a critical part of its leadership’s push for modernity, and, in fact, inseparable from that push. At first the effects were slow to appear. Throughout the late 1980s, there had been scattered reports of job losses for ‘‘redundant workers’’ (U.S. Foreign Broadcast Information Service, 1988; Walder, 1991, p. 477). In the main, however, managers were still constrained then from dismissing employees openly (Walder, 1991, p. 473). In the early 1990s, though, the state-owned firms began to succumb to competition from imports, as well as from the nonstate and foreign-invested sectors (Naughton, 1992). For enterprises in these other portions of the economy lacked the responsibility that the state had long bestowed on the firms it owned and ran to provide welfare and other benefits for their staff and labor forces. Clearly, this discrepancy enhanced the nonstate firms’ profitability

310

DOROTHY J. SOLINGER

and competitiveness at the expense of their state-owned rivals, and, as we will see, was to have a major impact on workers’ jobs. Other direct causes of the plight of the official enterprises that came to visit adversity upon the old work force included the obsolescence of much of these firms’ equipment in the wake of burgeoning technological imports from the developed world. Added to this was the growing and serious mismatch between, on the one hand, the largely unschooled nature of a huge segment of the workforce (owing to its coming of age during the years of the Cultural Revolution, when schools were shut and the only education on offer entailed ‘‘learning from the workers, peasants and soldiers’’), and the type of demand issuing from the labor market as the economy underwent its marketization, on the other hand. But it was finally with the several-year-long austerity program introduced by then-Vice Premier Zhu Rongji in mid-1993 and the significantly heightened market pressures it induced with its temporary but stiff curtailment of the customarily lavish and guaranteed credit that state firms had regularly counted on – and the consequent rise in firm losses – that a sudden outbreak of unemployment began (Naughton, 1995, 1996; Woo, 1997). In 1994, when a new Labor Law was written granting firms’ management freedom to fire if near bankruptcy or in serious difficulty,10 the phenomenon called xiagang (according to which employees in name retained their tie to their danwei but were in fact without any work to do) began to gather speed (Niu, 1998). By spring 1996, urban unemployment, once strictly anathema to the Communist Party, was being termed ‘‘inevitable in a market economy,’’ which in China by that time unquestionably obtained.11 Subsequently, a surge of job loss gathered new and ever-escalating momentum after the Communist Party’s 1997 Fifteenth Party Congress. There, then-Party General Secretary Jiang Zemin put forward two critical chores: first, to ‘‘adjust and improve the ownership structure’’; and second, to ‘‘accelerate the reform of state-owned enterprises.’’12 Neither of these objectives can be divorced from the subsequent flood of layoffs that followed in the wake of the convention. At the end of the year, the Ministry of Labor’s National Work Conference announced, apparently with much chagrin, ‘‘Dismissing and laying off workers is a move against our will taken when we have no way to turn for help, but also the only way to extricate ourselves from predicament.’’13 The upshot was that, for a significant section of the old proletariat in the factories of China’s metropolises, mounting losses among state companies came to spell involuntary unemployment.14 According to China’s own


311

National Statistical Bureau, ‘‘the number of workers employed in the stateowned sector fell from 113 million to 67 million, a decline of 40 percent over the five years 1996–2001.’’15 And so for the first time in the cities of the People’s Republic, there were widespread instances of people with work ability and a desire to work who were unable to land ‘‘jobs,’’ or, more loosely, to find employment of any sort.16 These moves, executed with reluctance as they may have been, resulted in the generation of a whole new sub-sector of city society, a segment apparently set to stay. The poverty stricken (i.e., the newly poor, the portion of the ruoshi qunti who were made poor in the course of economic reforms, a set of people later to go under the label of dibaohu, or households receiving the dibao17) then emerged as a subset of those who had been discharged from their firms. These were the people most disadvantaged, whether by their age, their poor health or disability, or their total lack of any credentials, unable to find a way to sustain their existence on their own. The presence of this part of the population stranded amidst much plenty is also the outcome of the as yet far from completely fulfilled need for the Chinese polity to undergird the livelihood of the unemployed masses in the cities with a workable social welfare system (Duckett, 2003; Solinger, 2005).18

PROGRAMS AIMED AT ASSISTANCE AND ALLEVIATION For the Newly Let Go The state has not been passive in the face of the hardship its policies engendered. Toward the unemployed or, so-called ‘‘xiagang’’ [laid off], as early as 1994, an experimental program, the ‘‘Reemployment Project’’ (REP), was piloted in 30 cities, and then extended nationwide the following year. It comprised a monumental effort directed at a sadly overambitious aim: to somehow arrange for the settlement of all the laid-off workers (Ru, Lu, & Dan, 1998). The problem was the weighty and critical limitations on the entire endeavor from the start – a scarcity of funds; a widespread connection of the unemployed with firms that had either gone bankrupt or were suffering serious losses and deeply in debt;19 incalculable levels of corruption among local cadres and firm managers, who intervened between policymakers and intended recipients, taking substantial cuts, to put it

312

DOROTHY J. SOLINGER

kindly; and, perhaps most serious, a vast insufficiency in the supply of potential work posts in the economy.20 The REP was designed as a set of transitional measures to insure the laidoffs’ basic sustenance through the granting of ‘‘basic livelihood allowances,’’ and to provide them with opportunities for work during the time when the country’s nascent labor market was admittedly yet imperfect and the nation’s social insurance system seriously incomplete (Wang, 2000; Lei, 1998; Wang, 1997, 1998; Shen & Ma, 1999; Xinhua, 1998). ‘‘Reemployment service centers’’ [zaijiuye fuwu zhongxin] were formed to care take or act as ‘‘trustee’’ [tuoguan] for furloughed workers for a three-year period from the layoff date for each. The activities of these ‘‘centers’’ entailed disbursing the livelihood allowances [jiben shenghuofei]; paying into funds for their staff and workers’ medical insurance and pensions; retraining them; and finding them new employment.21 Such centers proliferated nationwide, generally at the firm level.22 Those eligible to be termed xiagang according to the official definition (Guo, 1999) were to be (but, in practice, by no means always were) sheltered by the measures these centers were to enforce. These properly ‘‘laid-off’’ individuals – defined officially as those who met all three of the following conditions: (1) s/he began working before the contract system was instituted in 1986 and had a formal, permanent job in the state sector (plus those contract laborers whose contract term is not yet concluded); (2) because of his/ her firm’s problems in business and operations, has been let go, but has not yet cut off relations with the original firm; and (3) has not yet found other work in society – also were to be (but again, often were not) the target of a set of active labor market policies, taking the form of so-called ‘‘preferential policies’’ once they produced their laid-off certificate [xiagangzheng].23 Around the end of 2002, about four and a half years after the program was launched in earnest in May 1998, documents and speeches appeared urging its termination. The original promise had been for just three years of succor; by that point those who had been dismissed as late as the end of 1999 and who were counted as qualified xiagang workers had supposedly already received their intended benefits, and reemployment centers were urged to push their occupants ‘‘out to society.’’ Funds that had been targeted at keeping the ‘‘laid-off’’ going were to be merged with unemployment insurance, and the two tracks of livelihood subsidies and unemployment insurance were to be merged.24 It is probably impossible to assess the extent to which the program managed to attain any or even some of the results it was meant to accomplish, though it has been speculated that its termination was related to its general ineffectiveness.


313

For the Impoverished For the poverty stricken, a new program was also crafted, during the same years. Entitled the zuidi shenghuo baozhang zhidu [minimum livelihood guarantee system] (in Chinese, colloquially known as the dibao), like the Reemployment Project, it also was initiated in Shanghai and spread nationally. The charge of the dibao was to provide for urban residents whose household income failed to reach a locally determined minimal threshold; the method was to supplement that income to the extent necessary to bring the family’s monthly wherewithal up to the level deemed requisite for basic survival in that region (Hussain et al., 2002). The measures became formalized and relatively standardized nationally in a State Council circular in August 1997, and in the next month all localities at the urban and county levels across the country were called upon to set up the system. By the end of September 1999, though some 500 cities had put it in place, the recipient population amounted to just 2.82 million people. Still, progress of a sort had been made: of those being served, just about a fifth were traditional targets of civil affairs relief, with the remainder being the newly poor. Monies allotted had risen substantially as well: in the first nine months of that year, 1.5 billion yuan was allocated for this program, a tenfold increase over the total of funds that had been spent on relief just seven years before. In September 1999 the State Council issued its ‘‘Regulations for the Minimum Livelihood Guarantee for Urban Residents,’’ to go into effect in October that year.25 With the exception of Beijing, Shanghai, Shandong, Jiangsu, Zhejiang, Fujian, and Guangdong, all the other provincial-level units received financial subsidies in varying amounts from the central government for the dibao (Tang, 2002a). But by the end of the year 2000, when the figure served had risen to 3.2 million people, the Ministry of Civil Affairs calculated that somewhat under one quarter of the urban poor as of that time was being helped (Tang, 2001). During the year 2001, however, there was a massive upswing in the monies that the central government devoted to the program, with the result that the numbers of recipients shot up quickly.26 As of late 2003, though a claim was published that over 30 million people had been assisted by the project, up to the end of 2007 the numbers served at any given time had never surpassed 22.8 million;27 the amounts of funds invested never went above .11 percent of gross domestic product, nor over .6 percent of national government expenditures (see Tables 1 and 2). What is meant to occur is that community28 workers are to endeavor to uncover poor families’ total income. The managers then compare that

314

DOROTHY J. SOLINGER

Table 1. Year 1999 2000 2001 2002 2002 2003 2004 2005 2006 2006 2007

Number of Participants Served by the Dibao, 1999–2007. No. of Participants (Unit ¼ Million)

(late) (3rd qtr.) (end) (July) (end) (end) (end) (end) (Oct.) (end) (end)

2.800 3.237 11.700 19.307 20.647 22.468 22.050 22.342 22.2937 22.401 22.7090

Sources: 1999, Tang Jun, ‘‘The Report,’’ pp. 15–16; for 2000, ibid., p. 18; 2001 and 2002 (July), Hong Zhaohui, ‘‘Lun shehui zuanli de ‘pinkun’ – zhongguo chengshi pinkun wenti de genyuan yu zhili lujing’’ [Poverty of social rights: Dilemmas of urban poverty in China], Xiandai zhongguo yanjiu [Modern China studies], No. 79 (4, 2002), pp. 9–10; for 2006 (Oct.), ‘‘2006 nian 10 yuefen quanguo xian yishang dibao qingkuang’’ [National dibao situation for county and above, October 2006], http://www.mca.gov.cn/news/content/public/20061120150856.htm, accessed 8/17/2007; for 2007, ‘‘National urban and rural residents, the minimum livelihood guarantee system for equal coverage,’’ http://64.233.179.104/translate_c?hl ¼ en&sl ¼ zh-CN&u ¼ http://jys.ndrc.gov.cn/xin/hich/af45/dbch/af13/loch/f45xi/t20080y, accessed March 18, 2008. For 2002, 2003, 2004, 2005, 2006 (end of year figures), Zhonghua renmin gongheguo guojia tongjiju, Zhongguo tongji nianjian-2007 [Chinese people’s republic statistical bureau, Chinese statistical yearbook, 2007], p. 899.

income, on a per-capita basis, with their city’s officially, locally set, per person ‘‘minimum livelihood standard’’ [zuidi shenghuo biaozhun]. Each household whose total income falls below the local livelihood standard on a per-capita basis is entitled to a supplement equal to the difference between the total of its income, on one hand, and the sum of the city’s livelihood standard times the number of household members living together in that city, on the other. But that this procedure is far from always honored as it ought to be is sometimes the source of difficulties for the recipients. Nonetheless, strictly observed or not, the policy is miserly. The idea behind it from the start – and into the present – amounted (and amounts) to supplying the individuals with funds that were ‘‘just enough to keep body and soul together,’’ in the words of its leading scholar within China, Tang Jun.29 Indeed, what is most striking in evaluating the success of this program is that the nourishment, educational, and health conditions among the individual dibaohu have remained remarkably unchanged and essentially

315


Table 2. Dibao, Government Expenditures, Dibao as Percent of Government Expenditures, GDP, and Dibao as Percent of GDP, 1999–2007. Year

Dibao

Government Expenditures

Dibao as % of Expenditures

GDP

Dibao as % of GDP

1999 2000 2001 2002 2003 2004 2005 2006 2007

1.50 3.00 4.20 10.53 15.00 n.a. 19.50 20.33 27.796

1318.767 1588.650 1890.258 2205.315 2464.995 2848.689 3393.028 4042.273 4956.540

0.113 0.188 0.220 0.477 0.608 n.a. 0.570 0.503 0.561

8967.7 99214.6 10965.5 12033.3 13582.3 15987.8 18386.8 21180.8 24660.0

0.01600 0.0300 0.0380 0.0875 0.1100 n.a. 0.1060 0.0960 0.1127

Note: Unit ¼ billion yuan. Sources: For the dibao, the figures are either taken from or estimated from the following sources: Tang Jun, ‘‘Jianli zonghe de zuidi shenghuo baozhang zhidu’’ [Establish a comprehensive minimum livelihood guarantee system], accessed on March 18, 2008 at http:// thjp.vip.sina.com/M.htm; Xinhuanet (Beijing), July 19, 2002; Tang Jun, ‘‘Jiasu zuidi shenghuo’’; Tan Jun, ‘‘Tiaozhengzhong de chengxiang’’; and from ‘‘2006 nian shi yuefen’’. For government expenditures (1999–2006), Zhonghua renmin gongheguo guojia tongjiju bian [Chinese people’s republic national statistical bureau, ed.], 2007 Zhongguo tongji nianjian [China Statistical Yearbook] (Beijing: Zhongguo tongji chubanshe) [China Statistics Press], p. 279. For GDP (1999–2006), Zhonghua renmin, op. cit., p. 57. For 2007, Wen Jiabao, ‘‘Report on the Work of the Government,’’ Delivered at the First Session of the Eleventh National People’s Congress on March 5, 2008 (http://www.chinadaily.com.cn/china/2008npc/2008-03/19/content_6549177.htm, accessed May 26, 2008); and Ministry of Finance, ‘‘Report on the Implementation of the Central and Local Budgets for 2007 and on the Draft Central and Local Budgets for 2008’’(http://www.china.org.cn/government/NPC_CPPCC_sessions2008/ 2008-03/18/content_12y accessed May 26, 2008).

abysmal over the years of its operation, as a comparison of the late 1998 field notes of Tang Jun’s research team with my own interview material from 2007 and 2008 documents.30

UNFIT FOR WORK: THE MODALITIES OF DISCOURAGEMENT In the stance of the state toward the great majority of these impoverished and out-of-work city-dwellers a consistent subtext can be read. I contend

316

DOROTHY J. SOLINGER

that what in the realm of employment appears on the surface to be simply an acquiescence to the playing out of market forces is a phenomenon undergirded by an implicit impetus – both within the larger society and at the level of the policy elite – to eliminate the style of employment at use in the socialist era and to create in its place a purely modern labor market, one staffed by fully up-to-date wage-earners and by them alone. This ideal marketplace for labor is, therefore, to be one that has been swept clean of what is popularly deemed to be the dross spewed up from the old planned economy – those over age 35 as of the late 1990s, the undereducated, and the unskilled urban workers whose machine-based, assembly-line driven jobs had been allocated to them in the bygone age of the planning mechanism.31 I do not mean to claim that the business of barring such people from the shop floor and the shop is executed consciously or by specific design. I choose instead to underline the several ways in which this cohort of oncecomrades has been in essence discouraged – disincentivized, or, because of other policies, effectually discarded – with the outcome of being excluded from professional places of business. I categorize the two principal modalities through which this outcome has occurred as disincentives and collateral damage, respectively. I illustrate my remarks with comments from interviews I have conducted or arranged over the decade 1998–2008, in Wuhan, a city in central China. Over the years 1999–2002 I interviewed nearly 100 laid-off workers; in 2007 and 2008 I arranged and/or participated in another 65 talks with recipients of the dibao. These interviews were not the product of random samples but, from all I’ve read, appear to have been representative cases. That many of the subjects were female reflects the fact that we more often found women at home to speak with us than we did men; nonetheless, often women talked about problems that the couple as a unit faces (Table 3).32

Disincentives The most apparent obstacle erected against the employment of the xiagang and the dibao beneficiaries are the want-advertisements in the local newspapers: the typical ad calls for a college graduate, preferably holding a Master’s degree, under the age of 30 or 35, with two or more years of work experience. As a news article admitted: ‘‘This won’t work for those who couldn’t get higher education because of ‘historical reasons’ [a veiled reference to the Cultural Revolution, as mentioned above] and those middle-aged workers who have lost their jobs.’’33 Nor would an electronic

317


Table 3. Interviewees’ Number of Household Members, Stated Dibao Amounts, Stated Sources, and Amounts of Other Income, Stated Total Income, and Stated Per Capita Income, by Month, Wuhan, August 2007. Case No.

No. of Household Members

Dibao

Other Income/From What

Total Incomea

Income/ Capitaa

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

3 2 4 3 2 1 3 4 3 2 3 3 3 1 3 3 4

200 415 460 n.a. 300þ n.a. 200 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

700 633 1,600 n.a. 300þ n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

233 317 400 n.a. 150þ n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a.

18 19

3 3

100 80

500 n.a.

n.a. n.a.

20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

4 2 3 3 1 3 3 4 3 3 2 3 3 3 3 3 2

200þ 200 n.a. 210 120 290 210 200 n.a. 300 460 420 150 160 300 180 319

500/n.a. 218/pension 900/pension; 200–300/odd jobs n.a. None n.a. Odd jobs, darning clothes 400þ/migrant odd jobs No work No work n.a. n.a. 150 170 210 150 Pension/severance pay (here an elsewhere, severance pay is one-time) 400/severance pay; odd jobs Severance pay; odd jobs; departmental store job 700/pension n.a. n.a. 400/odd jobs 0 380þ/nusemaid 460 Odd jobs 800/odd jobs; relatives; work unit 300/work unit; brother 0 Severance pay Small stall 500/wages 500/odd jobs; relatives 470/wages; work unit 120/housing subsidy

900þ n.a. n.a. 610 0 670þ 670 n.a. n.a. 600 460 420 n.a. 660 800 650 439

225 n.a. n.a. 203 120 n.a. 223 n.a. n.a. 200 230 n.a. n.a. 220 266 217 220

318

DOROTHY J. SOLINGER

Table 3. (Continued ) Case No.

37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53

No. of Household Members

Dibao

4 1 3 3 3 4 2 3 3 2 4 4 3 3 3 3 2

240 160 300 260 n.a. n.a. n.a. n.a. n.a. n.a. n.a. Won’t say n.a. 136 234 n.a. 200

Other Income/From What

600/odd jobs 80/work unit 400/odd jobs 400/wages 600/odd jobs; pension Both adults do odd jobs n.a. Sanitation work Unstable construction work n.a. Odd jobs 400–500 temporary work n.a. 300/work unit Odd jobs Stall; odd jobs 195/work unit

Total Incomea

Income/ Capitaa

840 240 700 660 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 436 n.a. n.a. 395

210 240 233 220 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. 136 n.a. n.a. 197.5

Note: Unit ¼ billion yuan. The information is not available for one of three reasons, and since I participated in only seven of the interviews (because it seemed that the subjects were more apt to voice complaints when I was not present), I am unable to state which of these is correct: (1) the interviewer did not record the reply; (2) the interviewer failed to ask for the information; or (3) the subject refused to provide it.

a

screen at Wuhan City’s central labor market that posted job openings in 1999 suit their needs. For virtually all the jobs described were for people under 30, some even for those under 22 years old.34 Many an informant in her mid-30 s or beyond bemoaned to me that her advanced age was the reason for her joblessness,35 as in the pervasive refrain that, ‘‘Only people under 30 can find work.’’ One woman, echoing many others, who would have loved to become a saleswoman, at 37 could not hope to be hired because she was ‘‘too old.’’36 As a 48-year-old male expressed his predicament: When no one introduces me, I go to the labor market that the government manages to search for work. But the competition there is rather fierce. Everyone without work goes there looking, while a lot of rural young people also go, so that people like me of an older age with my physical capability wanting, often cannot contend. Sometimes work is very hard to find, if it’s too tiring or heavy, after working a few days I can’t go on.37


319

Insufficient education is a related issue, since ex-workers of a certain age were the same ones who had missed out on schooling in the years between the mid-1960s and the late 1970s (Hung & Chiu, 2003). Indeed, official statistics for 1998 show that more than half (53.5 percent) of those released from their posts had only a junior high education or even less.38 A 38-yearold man began ‘‘going out’’ to Guangdong for odd jobs not long after he was laid off, prompted mainly by plans to provide for his son’s schooling (in the hope shared by all such folk that the next generation will do better). ‘‘But after going out, his circumstances were poor,’’ revealed his wife, ‘‘because without any wenhua [used in this instance to refer to education] and not having mastered any specialized craft, the work he engaged in wasn’t secure, the situation was sometimes good and sometimes bad. Sometimes the money he made was just enough to maintain his livelihood while he was out.’’39 Because calls for job applicants specify so precisely age limits and educational attainments that eliminate the laid off, the ads themselves serve as a disincentive to job searches. Not just the middle-aged who missed their chance at schooling will fail to qualify for work from their lack of wenhua. The problem is apt often to be reproduced down the generations. One informant’s daughter, aged 19, is doing well in senior high, and, the mother related, ‘‘would like to attend a vocational school so she can go out to work sooner and lighten the household’s burden. But our family basically can’t pay the tuition,’’ sighed the woman. ‘‘We just hope we can borrow some money from the bank.’’40 Meanwhile, the daughter of this family’s neighbor was already attending a public-health vocational school, planning to become a nurse. For her first two years, that mother used her work unit’s severance pay to cover the fees, but for the third year, she fretted, ‘‘In the family we already don’t have the money to continue to pay, just now worrying about this.’’41 The dibao program itself contains a range of measures that directly discriminate against the poor’s daring to take up work. Most blatant is the stipulation that a dibaohu with a household member earning some wages stands to sees its allocation cut back substantially. In fact, even a recipient’s acquisition of a tiny increment in income through occasional labor could result in drastic reduction in his/her household’s dibao disbursement, so some (in my one sample, though, there was just one, out of 53 informants) did feel altogether disinclined to seek employment at all. Several interviewees in Wuhan did find their families’ dibao funds cut back or cut off when a member took on some wage-earning work. In one case a wife’s street sweeping led to deductions that left four people to survive

320

DOROTHY J. SOLINGER

on some 500-plus yuan per month.42 In another case, the wife in a family of three bravely reflected that: We can still go on, use the dibao money to raise our son (then age 12), each month we get a subsidy of 234 yuan. Though it’s not much, some is always better than nothing, the family has one person working, so the subsidy was lowered a lot. We’re not thinking of arguing about it, we all are very submissive people, so we wouldn’t bicker over money. If you give us 200-plus yuan it still can be of use.43

A third woman, aged 34, lamented that, ‘‘People like us are at the age for working, but we have no skill or culture, basically can’t find any good job.’’ The questioner pointed out that the woman’s husband was out of the city doing odd jobs [dagong], and that she was managing a stall, and inquired whether their monthly quota was therefore decreased. ‘‘Yes,’’ she replied, and continued: It’s a no way affair [mei banfa de shiqing]. In my stall in one month I can earn only so much money, his work also isn’t stable, but now our work is calculated into our income, then they have to cut the subsidy. But this income fluctuates, sometimes we have it and sometimes we don’t, only relying on the dibao, that little money means that basically there’s no way to live.44

Rigorous procedures used to judge whether or not a household is deserving serve the goal of ensuring that the allowances are calibrated to what is officially deemed to be the income necessary for mere survival – and nothing more. Applicants’ journey toward becoming recipients begins with a written entreaty, accompanied by documentary proof of their penury, to be submitted to the community [shequ] office in charge.45 After filing the request, community officials who manage the dibao have a certain amount of time (set locally, usually from 5 to 10 days) to assess the candidate’s needs and to attempt to ascertain the veracity of the paperwork s/he has presented. This certification of a claimant’s qualification can be a particularly invasive process. The steps begin with a thorough physical search of the household in question, along with close inquiry of its members. What follows is a particularly intrusive, sometimes even insidious, procedure, calling for interviewing neighbors and visits to the candidate’s place of work, if any, to make sure that the applicant has spoken truthfully. Most embarrassing of all, the results of all the scrutiny are to be posted upon a public board [the gongshilan], in order to solicit the views not just of immediate neighbors but of everyone in the community acquainted with the applicant family’s true state of eligibility and of everyone in a position to see the targeted family members’ daily comings and goings.46


321

Indeed, communities managing the system as they are ordered to do exhibit such a notice board proclaiming how many members live in every payee household; how much money each one is receiving; what special subsidies it is being granted; and how much voluntary work (such as neighborhood sanitation, public security, guarding, or gardening) its relevant members had performed in a given week, such activity being a necessary condition of enjoying the allowance.47 Once the community officers have made their tentative appraisal of a case, the file goes up to the street level, where another week or so is spent reviewing the materials. Street officials’ deliberations are also posted publicly on the community’s board. After the same length of time has passed, the records are delivered to the district level, where managers do a reexamination. The judgments about those who so far have seemed to meet the necessary conditions must once again be subjected to public view. If and only if there are no objections from other residents, finally the City Civil Affairs Bureau gives its stamp of approval and the applying family then becomes a full-fledged ‘‘dibaohu.’’ Subsequent, regular inspections (sometimes as frequently as every three months, in other cases just every six, mainly to discover whether or not the recipients have found work48) are meant to certify that the family remains qualified to enjoy the subsidy. When its situation or income undergoes changes (perhaps because of a retirement, a death in the family, a new odd job, or alterations in health), the household head is to notify the office in its community that takes care of examination and approvals, to arrange for stopping, reducing, or increasing their outlays.49 Such check-ups tend to classify those with wages as cheats if they do not disclose every cent they earn; in this way the home visits – often most unpleasant – serve as a form of surveillance that incline people not to work. As expressed by one couple, ‘‘The dibao’s examinations are too strict, each time they come to check it’s very upsetting, causes a lot of chaos, the people coming to investigate have no manners.’’50 Such scrutiny and surveillance must lend to informal labor a flavor of the illegitimate, a sense of sneakiness that some would prefer to eschew. In fact, many of the jobless, if earning any money, attempt as much as possible to conceal what they are doing, with the result that such secret work has earned the label ‘‘hidden employment.’’ As early as the late 1990s, when millions were first thrown off the job, the term began to surface; it has remained in use up through the present. Much writing on the laid-off censured this activity. One study of 1,000 dismissed workers in eight enterprises reported that over 70 percent of them belonged to this category,

322

DOROTHY J. SOLINGER

as it criticized such people for collecting some ‘‘basic living allowance’’ from their former firm while obtaining wages from a job elsewhere.51 But material on the income and mundane miseries of these people evinces the genuine necessity of this activity for those thrust into the vacuum between the state security of the past and the very inchoate, totally insufficient social security system of the present. The requirement of doing the oxymoronic ‘‘voluntary obligatory work’’ [ yiwu laodong] required of all able-bodied dibao ‘‘targets’’ [dibao duixiang] is yet another form of deterring these charity subjects from moving to the regular, external labor market. As a condition of receiving their statedispensed subsidy, each household must supply one day of unpaid labor per week to his or her community; failure to do so jeopardizes the family’s take from the program.52 The need to be on hand for that duty in order to accept the more certain, reliable money from the dibao program could, obviously, interfere with accepting paid work. Still another example of rules militating against job holding is the prohibition in some cities (of which Wuhan is one), against performing money-earning activities on the sidewalks, a rule that keeps the streets from having an appearance of scruffiness or disorder – traits with which the petty businesses of the poor would surely mark them. The upshot is that the droves of the destitute who could have scraped out a measly income in the absence of this regulation have had to put their trade tools out of the way. As the section chief of the dibao office in the Gansu provincial civil affairs department explained to me, ‘‘Letting them earn money is a way of cutting down their numbers. If their skill level is low, their only means of livelihood can be the street-side stalls they set up themselves.’’53 Thus, a talented but hard-up woman in Wuhan complained that the fees for advertising her artwork on the streets had escalated substantially over time, so that she was forced to abandon the effort to try to make sales.54 A woman who cannot afford to rent a shop is able only to repair shoes for friends and relatives on the sly in an unheated shed attached to her apartment building, rather than to make her skill known on the road as she would have preferred to do.55 Her previous workshop was demolished by the city’s ‘‘urban management bureau’’ [the chengguan] – the police in charge of maintaining order in public spaces – in 2003, likely at least in part because the structure looked too dingy and unattractive, though the official explanation was that the building was unregistered. That it was, but the shoemaker did not have the funds to register it. Yet one more instance were the pedicabs. From the late 1980s up until 2002 when a reported 40,000 of them, steered by the furloughed, graced the


323

avenues of Wuhan, these vehicles provided a tiny source of income for those who had lost their original jobs and who were strong enough to pedal them. But their old-fashioned unsightliness, as much as their chaotic swerving about, induced city authorities to ban them that year, throwing their operators out of business for a second time. And sellers of small tools and implements who lined the streets in the early layoff days of the late 1990s and in the first years of the 2000s were shooed into rent-charging stalls inside or – in the case of those unable to pay the charges – were made to carry their wares into their homes, where the items became available only to those already aware of their existence. All of these efforts to clear the streets occurred in and after 2002; this jibes exactly with a statement from one of Kellee Tsai’s informants, who noted that around 2000 the city government of Wuhan was more active in assisting laid-off workers than it was thereafter (Tsai, 2007). Two further types of measures clearly yield perverse outcomes in denying funding to truly needy people. The first of them also dissuades such people from behaviors that could put them into the job market. That disincentive consists of rules that exclude engagement in a range of actions that could help the actors to better themselves. These include an August 2007 order from the city of Ji’nan mandating that anyone who had purchased a computer or who often used a cell phone could not enjoy the dibao.56 Beijing’s regulations for assisting the poor preclude persons from getting the dibao who had bought cell phones or who had arranged for their children to attend schools of their own choice or private schools. In Liaoning, using a household phone more than 15 percent more than the local dibao standard or even having received gifts whose value was above the poverty line disqualified potential partakers.57 Elsewhere, some places banned people from becoming recipients if they had a family business, regardless of its profits or losses – firms losing money and incapable of supporting the family’s livelihood could be known to spark quarrels between civil affairs departments and an applicant.58 In Wuhan, the following circumstances could deprive the destitute of succor: having a motorized vehicle (unless it was required because of disability); having electrical fees surpassing 15 yuan per month, except in the high heat months of July, August, and September; running up phone fees beyond 40 yuan per month; using a cell phone or other hand-held communication device (even if having obtained it as a gift or a loan) or going on the web on a computer.59 Also forbidden was arranging for one’s children to select their own school or to enroll in special classes for study or training; doing odd jobs for which the wages are hard to verify; or for a child

324

DOROTHY J. SOLINGER

in the family to be studying with a foreigner. At least some grantees took these guidelines seriously, as evinced in this quotation from a mother of a 16-year-old boy: This year his grades could qualify him to transfer to the Number 3 Senior High School, a provincial-level keypoint institution. But I don’t have the money and secondly, if it’s discovered that there’s a child in the family who has transferred to a keypoint high school, our dibao qualification would be eliminated. We can’t take this risk. He really wants to study in that school, but he knows the family’s conditions, so he doesn’t demand it of me; I feel I have really let my son down.60

Whereas the exclusions of households whose so-labeled ‘‘extras’’ leave them out, the ‘‘as if’’ ostracizations, while also reducing a locality’s financial responsibility, treat workless people as they would treat those holding jobs. This behavior is justified thus: ‘‘since household income is very difficult to determine, hidden employment is pervasive, and hidden income and assets [are known to exist], flexible standards are adopted everywhere.’’58 According to this logic, families may be rejected simply because they are perceived as having the ability to work even if they have not found any waged labor, thereby considering them as having received the wages they would have earned had they been on a job. Such reckoning ‘‘regards as income’’ salary or benefits that properly speaking ought to have been – but were not – paid to a person, using their area’s minimum wage or unemployment insurance subsidy to assess the amount of the supposedly received income or benefit and then taking that sum to be the person’s actual income.61 While this rule does not dissuade applicants from working, it does penalize a person living in poverty for his or her failure to find a place willing to hire him or her. Most telling of all among my examples of an official bias against putting the poor to work was the justification of the program as a ‘‘tranquilizer’’ [a dingxinwan, literally, a ‘‘pill to stabilize the heart’’] that was offered by one of its early proponents (Ding, 1999). The reasoning here seems to have been that providing the indigent with a pittance would serve to push them back into their domiciles, where, in gratitude (and, perhaps as much as sedated), they would refrain from inciting disturbances against the economic reforms that had destroyed their old way of life and livelihood. At the same time, given the restrictions laid out above, the piddling funds they were granted did act to restrain them from venturing out to seek employment. This amalgam of limitations, linked together, works to leave those seen as incompetent, unworthy, unfit, in a word, lowly, outside the pale of the laboring public. While this outcome may not always be intended, the influences from these various factors often achieve that effect.


325

Collateral Damage Alongside these implicit disincentives, situated squarely within the directives on the dibao, exist a host of regulations whose externalities, or what could also be called collateral damage, in practice act to obstruct the entry of the dismissed and the dibao-holders into the work force. These include the privatization of medical insurance and the termination of reimbursing the ill for their health care costs; the decentralization of finances to localities and their management; and the Cultural Revolutionary era’s educational practices, whose impact on the past-40 populace has already been noted above; plus various and sundry state macroeconomic policies, including under-investment in labor-intensive industry, a bias toward lending just to the larger firms where risk is lower, and the ongoing inadequacy of the state’s active labor market policies (such as sponsoring job training not calibrated to match the nature of the market). Privatization of medical care – the end of free service for workers and half-price charges for family members with the disappearance of the danwei [workplace] system, the rise in costs for treatment and medicine, the creation of insurance packages that the jobless cannot possibly buy into, and the introduction of user fees – has meant that the very poor must languish in illness with no hope of recovery. Indeed, in all of my interviews where someone was in poor health the patient stayed at home, lay on a bed nearly all the time, was unable to work, and contrived to subsist, if barely, by swallowing a minimal amount of medicine, visiting a hospital only in times of dire emergency. Here are several particularly poignant examples: First the husband speaks: ‘‘My wife [aged 44] got uremia in 2002 [urine poison illness]; she’s from the countryside and has never worked, for her medical funds she’s completely dependent on me. Before, when she wasn’t sick, she could do household chores, now she can only lie on the bed, can’t do anything. The medical fees are very high, she sometimes gets dialysis. We basically despise this illness, everyday she stays home, takes a little medicine, and in this way drags on.’’

Soon the wife chimes in: ‘‘The doctors in the hospital would let you stay in the hospital for treatment, but we haven’t so much money, basically we can’t afford it. Each day I can take some medicine to control the illness, and that’s very good, I can’t hope to cure the illness, can just live a day and write it off [huo yitian, suan yitian], sometimes I think if I can only lie on the bed all day like this, unable to do anything, it’s the family’s burden, it’s not as good as dying earlier.’’ As she speaks, there’s a tear in her eye and her daughter quietly goes away.62

326

DOROTHY J. SOLINGER

In another home, an old mother is prostrate, paralyzed on the bed, as she has been for half a year. ‘‘Now she’s very old,’’ explains her daughter-in-law, ‘‘her health situation is very poor, her pension is all used up in seeing doctors and buying medicine.’’63 Yet one more futile scenario features a wife, aged 47, again confined to her bed. She contracted a thyroid disease nine years before. ‘‘At first, it wasn’t serious and we didn’t pay much attention to it,’’ she recounted. ‘‘Afterward, it slowly got severe, and I took a lot of hormone-type medicine. Now you can see I got fat, it’s really a side effect of the medication. Each month, must take about 100 yuan worth of it to control the illness. The doctor can examine me every month and check the condition of my relevant body signs, but a general check-up costs 300–400 yuan and we just can’t afford it. Ordinarily I’m at home, and keep track myself, I can do what I can to control it, but I don’t lightly go to the hospital.’’64 In one more home the wife is also laid up in bed, having claimed to be ill with erythema lupus (a connective tissue disorder); in two others it is rheumatism, while elsewhere a husband down with cancer makes do with a petty 70 yuan from his former work unit.65 Perhaps the most pitiful tale is one related by a young wife in tears: when she married her husband she was not aware that he was plagued by schizophrenia. Having been hospitalized four times since their 2003 wedding, he takes a medication that renders him dull and stiff, his reactions slowed down, so he is routinely cheated each time he goes out to work.66 A vicious cycle spins relentlessly in such cases: poverty – with the poor nourishment, shabby living accommodations, and unsanitary environs – brings on ill health; next, the inability to afford sufficient and appropriate medical care prolongs the problems indefinitely; and next, ill health precludes participation in labor, leading onwards to deeper and incurable penury. Decentralization of finances has left some localities (not just the inhabitants living in them) without the wherewithal to support business activity.67 Just as fundamental as an individual’s lack of start-up capital in establishing a small business is the fact that in the areas where impecunious people reside there simply is no market. As one man explained, ‘‘We tried in the past to sell breakfast in the community. We made some hot noodles, things like that. But because of the area where we live, the masses everywhere have low incomes, the level of consumption is really low, everyone just eats breakfast at home, so it basically didn’t work out.’’68 Another head of household, asked whether he had considered setting up a small shop, had a nearly identical reply: ‘‘Where we live it’s rather out of the


327

way, many residents are the nearby state enterprise’s laid-off staff and workers. Those who found other good work have basically all moved away, so setting up a retail food and fruit stall is useless – there would be few who would buy. If they have a little money they buy some ordinary vegetables and that’s it.’’69 A web-based analysis frames the problem within a more analytical framework: In poor communities the economy creates social segregation, which causes economic units’ incomes to fall. This makes the units leave the area, so capital migrates away, which only deepens the poverty. Then there’s no way to supply jobs, or to create tax income for the community. Purchasing power is low, and the return on investment is negligible, so outside businesses won’t enter and there’s no way internally to generate economic entities. Even if there are those with economic initiative, they’ll leave, so the community will lose their role as a model.58

In short, there are simply no spare funds in such locales, neither for forming nor for patronizing, commercial ventures. And finally, a major issue is the nature of state macroeconomic policies.70 Since the late 1990s, these have included priority treatment for large corporations through state investment, preferential taxation rates, and easy credit, while refraining from serious and substantial efforts to promote labor-intensive employment of the sort that could offer work to these people (the exception is infrastructural development, though it is not known how jobs on large state-funded projects are allocated71). The problems of lack of capital for tiny firms is fully evident in Kellee Tsai’s book, Back-Alley Banking, (Tsai, 2002) which describes the alternate sources of credit to which small entrepreneurs are compelled to resort. Scores of laid-off interviewees over the years have complained to me about the inappropriateness of the training available as against the nature of the labor market. In one case, in my presence, a friend suggested to a 38-year-old laid-off former thread factory employee that, ‘‘A lot of departments have organs specially to take care of middle-aged workers – train them to do household labor, beauty salon work, cooking, tailoring; they teach how to drive a cab.’’ But speaking of her hopeless feeling in the face of her low chance to make such training work out, the dismissed woman despondently mused that, I’d like to drive a cab, but have no money to buy the cab. Even working for someone else is also a problem – it requires giving 70 yuan even if you don’t earn any money. And you still have to pay for gas; anyway, there’s so many taxis now.72

328

DOROTHY J. SOLINGER

In conversations with the dibaohu in the past year or two, I heard comments such as this one: The residents’ committee organized some professional training, like computer training classes, and one on studying flower arranging, all of them were free. I really wanted to be employed, but the work still must be found by oneself, the residents’ committee can’t solve big problems, it’s still [a matter of] mainly relying on oneself.’’73

So in these various ways choices on other issues taken by top policy elites – with regard to health care, financial decentralization, macroeconomic policy, and state investments, and, in the distant past, education – skew the employment opportunities and possibilities for the poverty stricken and the out-of-work.

CONCLUSION From the year when China’s policymakers began to promote ‘‘reform and opening up’’ in 1978 up to the time of the present writing in late 2008, 30 full years have gone by. During the first half of this three-decades-long period, most of China’s populace – especially those who reside in the cities – saw clear material improvements in their standard of living. Wages climbed steadily upward; consumption choices became much more numerous; new and more spacious housing was built; more varieties of, and more nutritious, foods were available at affordable prices; advanced electronics products were suddenly within reach for the millions; automobile ownership even began to become a possibility for the upper incomed; and for the first time since the 1950s clothing was cheap, colorful, and classy. But soon on the heels of the early 1990s workers without suitable schooling or skills, people moving into middle-age, and those in questionable health all came to be viewed as out of tune with the times, or generally as not fit either for the style or for the scientific turn or the slick business world of the moment – all of these changes parts of a set of accelerating trends that left out and behind such unfortunates. Before long these out-ofdate people were pressured from their posts, millions thereby suddenly impoverished both relatively and absolutely. And there, on the outskirts of the hum of life some millions among them seem destined to remain. This paper shines its beam on this segment of the populace, illuminating how its members’ enforced worklessness has not been purely or only market-driven. Nor is their absence from the labor market a situation wholly of their own making. Instead, stuck away in the heart of what is


329

otherwise hectic urban life these people simply endure, discouraged by writ from taking a job. Indeed, I have argued here, even as the state designs programs to plant a floor for sustenance under their feet, at the same time it trips them up as they struggle to stand up upon it.

NOTES 1. A relatively recent volume contains essays on the changing fate of the danwei as the Chinese economy was removed from the plan (Lu & Perry, 1997). 2. We will never know just how many millions these have been. See Solinger (2001). 3. A 2001 report by the Chinese Communist Party’s Organization Department disclosed that an investigation done by the National Statistical Bureau, the State Council Research office, and other units, discovered that, nationwide, 20–30 million staff and workers had fallen into poverty in the prior few years. With their family members it was judged that altogether these people amounted to 40–50 million (Zhonggong zhongyang zuzhibu ketizu [Chinese central organization department research group], 2000–2001 Zhongguo diaocha baogao – xin xingshixia renmin neibu maodun yanjiu [2000–2001 Chinese investigation report – research on internal contradictions within the people under the new situation], (Beijing: Zhongyang bianyi chubanshe [Central Complation & Translation Press], 2001), pp. 170–171), or almost 13 percent of what was considered to be the urban population as of that time. 4. For instance, as of late 1999, when 2,306 cities and towns had installed a new program to cater to these individuals, 79 percent of the nearly three million residents entered into the system were individuals who had just lately become poor. See Tang (2002b). Portions of the report were later translated and published as Solinger (2004). 5. There has not been much research on these people. For some background, see Yang and Huang (2003), Zhang (2002), Lu (2002), and Sun (2002). Among Western scholars, two rare studies of the emerging urban poverty are Cook (2002) and Hussain et al. (2002). There is a cottage industry of economic studies of the poverty rate, but these often display no awareness of the substantive dimensions of the issue or of the repercussions for people’s lives. 6. Not only were such individuals likely to lose their jobs in the massive shakedown of the late 1990s and early 2000s, but lacking skills, they could rarely hope to find new work. Tang Jun, ‘‘Selections,’’ Chapter 3 makes this point repeatedly. According to an investigation of the dibao targets reported in a 2006 piece (‘‘Zhongguo chengshi,’’ op. cit.), it was found that among adult targets, those with primary education and below represented 24.1, and 46.5 percent had been to junior high school, together amounting to 70.6 percent without any senior high school training. A mere 27.6 percent of these people boasted of having some sort of professional or handicraft skill, while just 2.9 percent claimed to have some work. As for their health, the Ministry of Civil Affairs announced that in a national study of 10,000 dibao households, 33.7 percent have disabled people, and 64.9 percent had one or more members with a chronic illness or serious illness.

330

DOROTHY J. SOLINGER

7. The first few paragraphs below draw on my introduction to Solinger (2004). 8. The term ‘‘three withouts’’ refers to those unable to work, those without means of livelihood and those without family support. See Chow (1988) and Wong (1998). 9. Reported in China News Digest, September 28, 2002, taken from Agence France Press. 10. Translated in FBIS, July 19, 1994, pp. 18–26, from XH, July 5, 1994. 11. FBIS, June 14, 1996, p. 52, from Jinrong shibao (Beijing) [Financial Times], April 15, 1996, p. 1. 12. Jiang’s report to the congress is in SWB FE/3023 (September 13, 1997), pp. S1/1–S1/10. 13. Ming Pao [Bright Daily], December 19, 1997, in SWB FE/3107 (December 20, 1997), p. G/7. 14. Indeed, by 1996 a sharp increase had occurred in the number of state-owned firms throughout the country that were losing money, such that state industry for the first time experienced an overall loss (Rawski, 1999). 15. Cited in a recent paper by Park and Giles (2003). This paper, based on data from the China Urban Labor Survey conducted in five large Chinese cities at the end of 2001, found that unemployment reached double digits in all sample cities between the years 1996 and 2001. The Chinese source cited is National Statistical Bureau (2002). Later, a revised version of this paper was published under the same title, that time with the authors listed as Giles, Park, and Zhang (2006). It should be pointed out, however, that one reason for the huge drop in numbers in the state-owned firms (SOEs) is the reclassification of many former SOEs as joint-stock companies and as entities with other similar new names. 16. Earlier periods of urban unemployment existed in the PRC, as in the early 1950s, the mid-1960s, and the late 1970s. But in each case the government was able to devise programs – sometimes distasteful ones, as in the 1960s’ rustication movement – that to a large extent disposed of the problem. Besides, in these earlier eras, it was for the most part the never-employed who searched for jobs; in the present era it is a case of massive dismissals of the labor-age, already employed, population. 17. As will be discussed soon, this term is the short name for the ‘‘minimum livelihood guarantee’’ program. 18. Though the situation is gradually improving, the remarks in these pieces still hold largely true, as of late 2008. 19. According to N.a., ‘‘1998 nian qiye xiagang zhigong jiben qingkuang’’ [The basic situation of the laid-off enterprise staff and workers in 1998] Laodong baozhang tungxun [Labor insurance bulletin] (LBT), No. 1 (1999), p. 10, laid-off workers let go by enterprises losing money represented 67 percent of all laid-off workers as of the end of 1998. 20. According to Zhang (1998), ‘‘to solve the problem of reemployment for the unemployed, the laid-off, and surplus labor, we lack at least 30 million jobs.’’ 21. Yang (1999) calls the center a product of ‘‘a special historial stage, a transitional measure which can solve its special contradictions.’’ 22. Originally – for instance, in Wuhan in 1998 – entire sectors, such as the textile trade, entrusted their cutback staff and workers (in this case, numbering 10,000 such persons) to one service center. The work of that center is described in a document handed to me privately in summer 1998 in Wuhan, ‘‘Wuhan Shi fangzhi zaijiuye


331

fuwu zhongxin yuncuo qingkuang huibao’’ [A summary report on the operations situation of the Wuhan City Textile Reemployment Service Center], prepared by the center, Wuhan, March 18, 1998. Later it seems that each firm able to afford to do so ran its own center. 23. Those who meet the criteria for xiagang are qualified to obtain such a certificate. 24. N.a., Laodong baozhangbu buzhang Zeng Silin zai quanguo laodong baozhang tingjuzhang zuotanhuishang de jianghua (Minister of Labor and Social Security Zeng Silin’s speech in the national labor and social security department heads’ forum), ‘‘Jiakuai tuijin zaijiuye gongzuo, quanmian wancheng jinnian gexiang laodong baozhang gongzuo renwu’’ [Speed up the work of promoting reemployment work, fully complete this year’s various tasks in labor and social security work] LBT, No. 9 (2003), pp. 8–14; Zhong fa [2002] 12 hao [Central Document No. 12, 2002]: ‘‘Zhonggong zhongyang Guowuyuan guanyu jinyibu zuohao xiagang shiye renyuan zaijiuye gongzuo de tongzhi’’ [Central Committee and State Council Circular on progressively doing well the reemployment work of laid-off and unemployed personnel], dated September 30, 2002 LBT, No. 2 (2003), pp. 4–8. 25. For the 1999 Regulations officially announcing the program, see ‘‘Chengshi jumin zuidi shenghuo baozhang tiaoli’’ [Regulations on the urban residents’ minimum livelihood guarantee] (hereafter ‘‘Chengshi jumin’’) Zhongguo minzheng [Chinese civil affairs] (hereafter ZGMZ), No. 11 (1999), pp. 16–17. 26. ‘‘Needy urban residents in China get assistance,’’ accessed at www.xinhuanet. com/english/2002-10/24/content_607365.htm, October 24, 2002. In the following year the numbers served continued to increase. 27. ‘‘China’s subsistence allowance system benefits urban, rural poor equally,’’ accessed January 24, 2008, at http://english.people.com.con/90001/90776/ 6344770.html 28. ‘‘Community’’ or, in Chinese, shequ, is a formal term, explained below. See note 45. 29. Tang (2002b), p. 4. 30. Tang Jun’s notes are in Solinger (2004); mine are available upon request. 31. In this paper I ignore the rural migrants who are still in demand to perform such work. I discuss just the city-registered people who once held formal-sector jobs in the urban economy. 32. I refer to the latter group by their number in my footnotes; the information on these discussions can be obtained upon request (see Table 3). 33. Jingji cankaobao [Economic reference news], March 24, 1998, reprinted in Gongyun cankao ziliao [Workers’ movement reference materials] April 1998, p. 16. 34. Visit to the market, September 4, 1999. 35. This problem seems to have been particularly severe for women; men over 40 had the same difficulty. 36. Interview, September 1, 1999. 37. Interview 4, 2007. 38. N.a., ‘‘1998 nian qiye xiagang zhigong jiben qingkuang’’ [The basic situation of the laid-off enterprise staff and workers in 1998] LBT, No. 1 (1999), p. 10. 39. Interview 7, 2007. 40. Interview 1, 2007.

332

DOROTHY J. SOLINGER

41. Interview 2, 2007. 42. Interview 8, 2007. At the time, this amounted to about $75 US. 43. Interview 11, 2007. 44. Interview 12, 2007. 45. A shequ is a residential unit organized in cities nationwide after 2000, usually comprising two or three of the former residents’ committees. The residents’ committee was a standard urban entity (though not formally part of the official governmental hierarchy) created in the early 1950s to monitor, assist, and regulate the affairs of people who lived in adjacent compounds. Such a ‘‘community’’ can contain anywhere from several hundred to several thousand residents. 46. Interview with officers at community W, an area with about 1,600 residents, of whom only about 1 percent are dibaohu, on August 30, 2007. 47. Interviews at community Y containing over 4,000 people, and community Z, both on August 29, 2007. 48. Interview with the director of the dibao office at the Gansu provincial Civil Affairs Department, September 5, 2007. 49. Wang Zhikun, op. cit., 19. Interview with dibao workers at community X, where there are 1,099 households, of which 7.9 percent are dibaohu, August 27, 2007. 50. Interview 13, 2007. 51. Xu (1999); Li and Xie (1999); Gu (1998); and Yang (1999), which says 60 percent of the registered unemployed were working, as ‘‘hidden employed.’’ 52. This point is made in Hammond (2008). 53. Interview, September 5, 2007, Lanzhou. Lanzhou is the capital of Gansu province. 54. Interview 2, August 26, 2007. 55. Personal friend. 56. ‘‘Jinan guiding maidiannao jingchang yongshoujizhe buneng xiangshou dibao’’ [Jinan regulates that those who bought a computer or often use a cell phone can’t enjoy the dibao], Zhongguowang, October 9, 2006, www.china.com.cn, accessed August 17, 2007. 57. ‘‘Zhongguo chengshi jumin zuidi shenghuo baozhang biaojun de xiangguan fenxi, jingji qita xiangguan lunwen’’ [Chinese urban residents’ dibao norm’s relevant analysis economic and other related treatises], http://www.ynexam.cn/html/jingjixue/ jingjixiangguan/2006/1105/zhonggochengshijiminy, accessed August 18, 2007. 58. ‘‘Zhongguo chengshi.’’ 59. Interview at community X, August 27, 2007. 60. Interview 6, 2007. 61. ‘‘Chengshi dibao: tashang xin zhengcheng’’ [The urban dibao: step onto a new journey], ZGMZ 1 (2000), pp. 24–25. 62. Interview 1, 2007. 63. Interview 3, 2007. 64. Interview 4, 2007. 65. Interviews 6, 10, 23, and 33, all in 2007. 66. Interview 50, 2007. 67. As explained in Tao, Liu, and Wang (2008), ‘‘upper levels of government attempt to externalize their fiscal burdens to lower levels of government. The central government initiated a series of reforms since 1994 to centralize revenue yet leaves


333

expenditure decentralized.’’ Also, Guan, (2008) notes that, ‘‘As the result of decentralization of social expenditure, local government should take responsibility to pay for most of the social welfare projects. Because of the limitation in financial capacity, and their strong motivation in promoting local economic development, most of the local governments tended to control their social expenditure [sic.].’’ A study showing how this move affected health care within localities is Duckett (2002). 68. Interview 4, 2007. 69. Interview 9, 2007. 70. In 1998, government advisor Hu Angang noted the process of ‘‘capital deepening’’ that was favoring capital-intensive industry while it was weakening the economy’s ability to absorb labor; accordingly, he proposed putting more emphasis on labor-intensive industries, including the service sectors, as well as developing new economic sectors that would enable the creation of new nonstate jobs and loans for self-employment (see Hu, 1998). He continued to make such recommendations three or four years later, calling in particular for governmentally created jobs, for the promotion of employment and labor-intensive industry, and for the development of informal, private sectoral, and service market jobs in Hu (n.p., n.d.). 71. Infrastructural investment in the three years 1998–2001 amounted to 3.7586 trillion yuan, with 398.5 billion yuan of it coming from the fiscal budget, according to Hu Angang, ‘‘China’s Present Economic Situation,’’ p. 7. 72. Interview, September 6, 1999. 73. Interview 2, 2007.

REFERENCES Chow, N. W. S. (1988). The administration and financing of social security in China. Hong Kong: Centre of Asian Studies. Cook, S. (2002). From rice bowl to safety net: Insecurity and social protection during China’s transition. Development Policy Review, 20(5), 615–635. Ding, L. (1999). Cong danwei fuli dao shehui baozhang – ji zhongguo chengshi jumin zuidi shenghuo baozhang zhidu de dansheng [From unit welfare to social security – recording the emergence of Chinese urban residents’ minimum livelihood guarantee system]. Zhongguo minzheng [Chinese civil affairs] (ZGMZ), 11, 7. Duckett, J. (2002). State self-earned income and welfare provision in China. Provincial China, 7(1), 1–19. Duckett, J. (2003). China’s Social Security Reforms and the Comparative Politics of Market Transition. Journal of Transition Politics and Post-Communist Studies, March. Giles, J., Park, A. F., & Zhang, J. W. (2006). How has economic restructuring affected China’s urban workers? The China Quarterly (CQ), 185, 61–95. Gu, Y. (1998). Xiagang xhigong yinxing jiuye wenti chuyi [Preliminary opinions on hidden employment among the laid-off staff and workers]. ZGJY, 6, 27. Guan, X. (2008). The relationship between central and local governments in social policy actions in China: From an example of the MLS’ Development. Paper presented at Provincial China Workshop, Nankai University, Tianjin, October 27–30, pp. 6–7.

334

DOROTHY J. SOLINGER

Guo, J. (1999). Guoyou qiye xiagang yu fenliu you he butong? [What’s the difference between laid-off and diverted workers in the state firms?]. Zhongguo gongyun [Chinese workers’ movement], 3, 32. Hammond, D. (2008). Local variation and failure to transfer nationally in urban social assistance policy: The Community Public Service Agency in Dalian City. Paper presented at the Provincial China Workshop 2008, Tianjin, October 27–30, p. 4. Henderson, G. E., & Entwisle, B. (2000). Introduction. In: B. Entwisle & G. Henderson (Eds), Re-drawing boundaries: Work, households, and gender in China (p. 5). Berkeley, CA: University of California Press. Henderson, G. E., Entwisle, B., Ying, L., Mingliang, Y., Siyuan, X., & Fengying, Z. (2000). Re-drawing the boundaries of work: Views on the meaning of work (Gongzuo) (pp. 33–50). Berkeley, CA: University of California Press. Hu, A. (n.p., n.d.). Cong jihua tizhi zhuanxiang shichang jizhi: dui zhongguo jiuye zhengce di pingjia (1949–2001) [The transition from the planned system to the market mechanism: An assessment of Chinese employment policy], pp. 14, 21–26. Hu, A. (n.p., n.d.). Creative destruction of restructuring: China’s urban unemployment and social security (1993–2000), p. 26. Hu, A. (1998). Employment and Development: China’s Employment Problem and Employment Strategy, No. 6 National Conditions Report (Beijing: National Conditions Analysis and Study Group of the Chinese Academy of Sciences, April 30, 1998), pp. 17, 19, 22, and 28. Hung, E., & Chiu, S. W. K. (2003). The lost generation: Life course dynamics and Xiagang in China. Modern China, 29(2), 204–236. Hussain, A., et al. (2002). Urban Poverty in the PRC. (Asian Development Bank Project No. TAR: PRC 33448). Lei, P. (1998). Zhigong peixun yu jiuye cujin – chengshi fupin de zongyao [Staff and workers’ training and the promotion of reemployment – the important path in subsidizing urban poverty]. LDNC, 11, 30–31. Li, B., & Xie, Y. (1999). Yinxing shiye yu ‘yinxing jiuye [Hidden unemployment and ‘hidden employment’]. Zhongguo laodong [Chinese labor], 4, 46. Lu, X., & Perry, E. J. (Eds). (1997). Danwei: The changing Chinese workplace in historical and comparative perspectives. Armonk, NY: M.E. Sharpe. Lu, X. (2002). Dangdai zhongguo shehui jieceng yanjiu baogao [A research report on China’s current social structure] (Beijing: shehui kexue wenxian chubanshe). National Statistical Bureau. (2002). China Statistical Yearbook. Beijing: China Statistical Press. Naughton, B. (1992). Implications of the state monopoly over industry and its relaxation. Modern China, 18(1), 14–41. Naughton, B. (1995). Growing out of the plan: Chinese Economic Reform 1978–1993. New York: Cambridge University Press. Naughton, B. (1996). China’s emergence and prospects as a trading nation. Brookings Papers on Economic Activity, 2, 294. Niu, R. (1998). Xiagang zhigong chulu sikao. [Thoughts on the way out for the laid-off staff and workers]. Lingdao canyue [Leadership consultations], 1, 8. Park, A., & Giles, J. (2003). How has economic restructuring affected China’s urban workers? (Ms., October 2003), p. 1. Rawski, T. G. (1999). Reforming China’s economy: What have we learned? The China Journal, 41(January), 144.


335

Ru, X., Lu, X., & Dan, T., (Eds.), (1998). nian: Zhongguo shehui xingshi fenxi yu yuce [1998: Analysis and prediction of China’s social situation] (Beijing: shehui kexue wenxian chubanshe [Social science documents publishers], 1998), p. 86. Shen, W., & Ma, R. (1999). Zaijiuyezhong de zhengfu xingwei [The government’s behavior in reemployment]. ZGLD, 2, 19. Solinger, D. J. (2001). Why we cannot count the unemployed. CQ, 167(September), 671–688. Solinger, D. J. (2004). Guest editor: Tang Jun, Selections from report on poverty and antipoverty in urban China. Chinese Sociology & Anthropology, Winter 2003–4/Spring 2004. Solinger, D. J. (2005). Path dependency reexamined: Chinese Welfare Policy in the transition to unemployment. Comparative Politics, 38(1), 83–101. Sun, L. (2002). 90 niandai zhongqi zylai zhongguo shehui jiegou yanbian di xin qushi. [New trends in the evolution of Chinese social structure since the mid-1990s]. Dangdai zhongguo yanjiu [Modern China Studies], 9, 3. Tang, J. (2001). Dibao zhiduzhong de shehui paichi. [Social discrimination in the minimum living guarantee system]. Paper presented at the Conference on Social Exclusion and Marginality in Chinese Societies, sponsored by the Centre for Social Policy Studies of the Department of Applied Social Sciences, the Hong Kong Polytechnic University and the Social Policy research Centre, Institute of Sociology, the Chinese Academy of Social Sciences, Hong Kong, November 16–17, p. 9. Tang, J. (2002a). The New Situation of Poverty and Antipoverty. Report published as a chapter of a book edited by Ru Xin, Lu Xueyi, Li Peilin, et al., 2002 nian: Zhongguo shehui xingshi yu yuce (shehui lanpishu) [Year 2002: Analysis and Forecast of China’s Social Situation (Blue Book on Chinese Society)], January 1, 2002 [FBIS Translated Text]. Tang, J., (2002b). The Report of Poverty and Anti-Poverty in Urban China – The Poverty Problems in Urban China and the Program of Minimum Living Standard, ms., pp. 15–17. Tao, R., Liu, M., & Wang, J. (2008). Making the best of transfers: Central policies and local responses: Evidence from a state-designated poor county in Gansu. Paper presented at the Provincial China Workshop 2008, Nankai University, Tianjin, October 27–30, p. 21. Tsai, K. S. (2002). Back-alley banking: Private entrepreneurs in China. Ithaca, NY: Cornell University Press. Tsai, K. S. (2007). Capitalism without democracy: The private sector in contemporary China (p. 78). Ithaca, NY: Cornell University Press. U.S. Foreign Broadcast Information Service. (1988). China Daily Report (FBIS), September 14, 1988, p. 36, and September 28, 1988, pp. 52–53. Wang, T. (1997). Kunjing yu chulou. [Difficult straits and the way out]. Shehuixue yanjiu [Sociology Research], p. 6. Wong, L. (1998). Marginalization and social welfare in China. London: Routledge. Wang, T. (1998). Xinhua wengao [New China documents], 3, p. 25. Wang, D. (2000). Jianchi zhengque fangxiang, fahui xuehui gongneng wei jianshe you zhongguo tese di laodong he shehui baozhang shiye fuwu [Persist in an accurate direction, foster competence in studying, to serve in developing a labor and social security enterprise with Chinese characteristics]. Zhongguo laodong [Chinese labor] (ZGLD), 4, 5. Walder, A. G. (1991). Workers, managers and the state: The reform era and the political crisis of 1989. CQ, 127pp. 473, 477.

336

DOROTHY J. SOLINGER

Woo, W.T. (1997). Crises and institutional evolution in China’s industrial sector. Joint Economic Committee, Congress of the United States, (Ed.), China’s economic future: Challenges to U.S. Policy. Armonk, NY: M.E. Sharpe, pp. 164–165. Xinhua, (1998). New China News Agency, October 14, 1998, in Summary of World Broadcasts (SWB) FE/3358 (October 15, 1998), p. G/3. Xu, J. (1999). Yinxing jiuye de xianzhuang ji jiejue duice [The present conditions of hidden employment and measures to solve it]. Zhongguo jiuye [Chinese employment] (ZGJY), 3, 34. Yang, Y. (1999). Ruhe kandai dangqian di shiye wenti [How to consider the present issue of unemployment]. Neibu canyue [Internal consultations], 447, 18. Yang, S. (1999). Zaijiuye yao zou xiang shichanghua [In reemployment we must go toward marketization]. Zhongguo jiuye [Chinese employment] (ZGJY), 3, 19. Yang, Y., & Huang, Y. (2003). Zhongguo jumin shouru fenpeo xin geju [The new pattern in income distribution among Chinese urbanites]. In: R. Xin, L. Xueyhi & L. Peilin (Eds), Shehui lanpishu: 2003 nian: zhongguo shehui xingshi fenxi yu yuce [Social blue book: 2003 analysis and predictions of China’s social situation] (pp. 226–234). Beijing: shehui kexue wenxian chubanshe [Social science documents company]. Zhang, H. (1998). Dangqian zaijiuye gongcheng de qi da wuqu. [Seven big misunderstandings in the present reemployment project]. Lingdao neican [Leadership internal reference] (LDNC), 7, 27. Zhang, W. (2002). Twenty years of research on stratified social structure in contemporary China. Social Sciences in China, 23(Spring), 1.

GENDER INEQUALITY IN THE CHINESE LEGAL PROFESSION Ethan Michelson ABSTRACT In China’s urban context of labor retrenchment, women are faring poorly relative to their male counterparts. Is the same true in China’s incipient, dynamic, and expanding legal profession? Findings from four sources of quantitative data suggest that gender inequality in China’s private and highly market-driven legal profession is a microcosm of larger patterns of female disadvantage in China’s evolving urban labor market. Although employment opportunities for women lawyers have greatly expanded quantitatively, their careers are qualitatively less successful than those of their male counterparts in terms of both income and partnership status. In the Chinese bar, women’s significantly shorter career trajectories are perhaps the most important cause of their lower incomes and slimmer chances of becoming a law firm partner. Future research must identify the causes of this significant career longevity gap between men and women in the Chinese legal profession.

As market institutions replace bureaucratic planning, and as state-owned enterprises restructure, privatize, and shut down in urban China, workers who formerly enjoyed virtually guaranteed employment security have experienced increasing difficulty in finding and keeping work (Cai, Park, & Zhao, 2008; Gold, Hurst, Won, & Li, 2009; Hurst, 2009; Rosemont, 2000). Work and Organizations in China after Thirty Years of Transition Research in the Sociology of Work, Volume 19, 337–376 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1108/S0277-2833(2009)0000019015

337

338

ETHAN MICHELSON

Contraction and retrenchment are the story of work in China’s urban public sector. China’s total urban labor force has expanded despite rising urban unemployment largely because of a continuous influx of migrant workers, the swelling ranks of self-employed, and employment growth in unregistered private enterprises (Knight & Li, 2006, p. 105; Knight & Song, 2005, pp. 30–38). While existing scholarship on the fate of women in China’s reform era documents persistent and even widening gender gaps in labor market outcomes, we know little about patterns of gender inequality in China’s emerging professions. This chapter aims to help fill this gap by exploring various dimensions of gender inequality in China’s legal profession. This chapter is concerned exclusively with ‘‘urban’’ lawyers (lu¨shi). Generally speaking, legal practitioners in rural China go by the name ‘‘law workers’’ (falu¨ gongzuozhe) and fall under a different licensing and regulatory regime (Michelson, 2006, p. 11, n. 9). Moreover, the scope of Chinese lawyers also excludes judges, prosecutors, in-house counsel, members of the legal academy, and other actors commonly included in definitions of the ‘‘legal profession’’ elsewhere in the world (Liu, 2009). Drawing on multiple sources of quantitative data, my analyses center on two set of questions. First, how has the gender composition of the Chinese bar changed over time? If public-sector retrenchment in urban China has raised barriers to stable, high-pay employment for women, has the almost total privatization of the Chinese bar had a similarly negative impact on women’s opportunities to practice law? Second, after examining changes in the quantity of jobs for women lawyers, I will analyze the gender gap in the quality of jobs in the Chinese bar. Relative to their male counterparts, how have women fared in China’s incipient legal profession in terms of income, chances of making partner, and career longevity? Despite remarkable feminization, available evidence suggests that the Chinese legal profession remains hostile to women after their entry. My research findings show that women’s significantly shorter career trajectories are perhaps the most important cause of their lower incomes and slimmer chances of becoming a law firm partner. As we will see, differences in career trajectories help explain both the income gap and the partnership gap.

WOMEN AND WORK IN URBAN CHINA Social scientists have amassed a considerable and growing body of evidence that women’s work opportunities and rewards have suffered relative to

Gender Inequality in the Chinese Legal Profession

339

men’s in post-Mao urban China. No different from Soviet Bloc countries through the mid-1980s (Kornai, 1992, pp. 206–207), urban female labor force participation was near-total in China’s pre-reform era of high socialism (Parish & Busse, 2000). In the wake of China’s economic reforms, however, women in urban China have borne the brunt of unemployment. While labor force participation rates have declined for both men and women, women have suffered greater setbacks than men (Li & Gustafsson, 2008, p. 244). In urban China, women are not only more likely than men to be unemployed (Zhang, Hannum, & Wang, 2008), but when they are employed they are disproportionately represented in low-status, low-pay jobs and are less likely than men to get promoted to better positions (Cao, 2001; Cao & Hu, 2007; Du & Dong, 2009; Shu, 2005). Among those with jobs in urban China, the gender gap in income has remained fairly stable over time with women’s incomes in the vicinity of 80–85% of men’s incomes (Bishop, Luo, & Wang, 2005; De´murger, Fournier, & Chen, 2007; Parish & Busse, 2000; Shu & Bian, 2003). Such accounts of a stable income gap, however, have generally ignored the selection of women out of employment. If people who have lost their jobs or who are unable to find jobs are also considered in analyses of income inequality, the male–female gap exhibits a clear yawning over time (Li & Gustafsson, 2008). In some accounts, human capital differences are key to explaining gender gaps (e.g., Cao, 2001; Bishop et al., 2005; Shu & Bian, 2003). In other accounts, by contrast, human capital is of limited or no utility in explaining gender gaps, either in China (Cao & Hu, 2007; Zhang et al., 2008) or in Russia (Gerber & Mayarova, 2006). Although scholars disagree about the extent to which human capital explains female disadvantages in China’s evolving urban labor market, in all accounts, market institutions are the root culprit. In a process known as ‘‘unhooking and privatizing’’ (tougou gaizhi or tougou zhuanzhi), the Chinese bar’s transformation beginning in 1992 from an almost fully public profession to an almost fully private profession (Michelson, 2007) mirrors China’s larger workplace transformation. Do general patterns of urban gender inequality extend to China’s legal profession? Or are different dynamics at play in the Chinese bar? We will see in this chapter that the answer to both questions is yes. The dramatic expansion of the Chinese bar – which in the 20 years spanning 1988 and 2007 more than quadrupled nationwide and ballooned by more than a factor of 10 in Beijing (ZFN; ZTN) – is associated with feminization. However, despite this narrowing job quantity gap, significant job quality gaps persist. In these regards, gender inequality exhibits remarkably similar features in the Chinese bar as it does in bars elsewhere in the world.

340

ETHAN MICHELSON

WOMEN AND WORK IN LEGAL PROFESSIONS OUTSIDE CHINA The expansion of legal professions elsewhere in the world is associated with lower entry barriers for women and greater opportunities for would-be women lawyers. At the same time, however, gender inequality among lawyers persists in terms of income, upwards mobility, and career longevity. Feminization: Job Quantity Gaps Narrow Between 1980 and 2000, the population of American lawyers doubled from .54 to 1.07 million, while the share of women in the profession more than tripled from 8% to 27% (Chiu & Leicht, 1999, p. 566; Carson, 2004, p. 27). In the United States, ‘‘The massive influx of women into the bar occurred simultaneously with the expansion of the legal profession’’ (Mather, 2003, p. 42). Elsewhere in the world, the influx of women has been similarly dramatic and similarly associated with bar expansion. Between 1980 and 2005, the population of English solicitors almost doubled, while the share of women more than quintupled from 8% to over 40% (Bolton & Muzio, 2007, p. 48). Similar patterns have emerged in Australia (Hunter, 2003), Canada (Kay & Brockman, 2003), Germany (Schultz, 2003a, p. liii, 2003b), Poland (Fuszara, 2003), and South Korea (Kim, 2007, pp. 52–53). A state-level analysis of American lawyers at three time points spanning 1980 and 2000 further supports the empirical association between the feminization and expansion of the legal profession. Fig. 1 contains scatter plots of women as a proportion of all lawyers by lawyer population density. Following scholarly convention (e.g., Abel, 1989; Carson, 2004; Schultz, 2003a), I measure lawyer population density as the population/lawyer ratio (i.e., the number of people per lawyer) rather than as lawyers per capita (but cf. Halliday, 1986). As the American bar doubled between 1980 and 2000, the population/lawyer ratio shrank from 424:1 to 298:1 (Fig. 1A). Because, ceteris paribus, population/lawyer ratios shrank as lawyers grew in number, the scatter plots reveal negative correlations that are somewhat counterintuitively interpreted. Consistent with Halliday (1986, p. 62) and Sander and Williams (1989, p. 444), we can see in Fig. 1A that prior to 2000 the female share of lawyers was considerably higher in census data than in data supplied by Martindale-Hubbell and published in the American Bar Foundation’s Lawyer Statistical Report (e.g., Carson, 2004; Chiu & Leicht, 1999, p. 566). While Fig. 1A illustrates the association between bar

.10

.15

.20

Population per Lawyer

.05

50 150 250 350 450 550 650 750 850 950

IN80

.10

.15

MD00 NM00 NY00 MA00 CA00NH00 ND00 MA90 WV00 DC80 VA00MN00 WA00 KY00 DE00 ME90 AK90 NM90 CT00 TN00 IN00 TX00 NC00 PA00 CT90 HI00 CA90 ME00 CO00 ND90 NJ00 IL00 MI00 NY90 FL00 RI00 GA00 AZ00 KS00 NC90 PA90 MS00 CO90 VA90 NJ90 LA00 MO00 NH90 WI00 MD90 VT80 AR90 AR00 KY90 MN90 AZ90 AK00 OH00 NV00 TX90 TN90 OK00 OR90 VT00 DE90 AL00 OR00 WA90 IL90 LA90 SC00 IN90 WI90 NE00 SC90 OK90 MO90 FL90 MI90 IA00 ID80 NE90 IA90 GA90 OH90 HI90 ID00 WV90 UT00 KS90 NV90 HI80WY00 CA80 RI90 MT00 MA80 VT90 OR80 AZ80 AL90 NE80 GA80 ID90 NH80 UT90 CO80 MN80 PA80 SC80 WY90 CT80 AK80 NY80 SD00 NC80 MT80 MD80 NV80 VA80IL80 MO80 UT80 KY80 WV80 OK80 NM80 TX80 IA80 MS90 LA80 RI80 FL80 KS80 WA80 MI80 OH80 WI80 AL80 ME80 MT90 NJ80 TN80 WY80 IN80 AR80 SD90 ND80 DE80 R = –.63 MS80 SD80

DC90 DC00

Population per Lawyer

USA80 Pennsylvania, R = –1.00

PA80

.20

.25

.30

.35

.40

B. All States

200 250 300 350 400 450 500 550 600 650 700

MA80

.35 MA00 Massachusetts, R = –1.00 USA, R = –.97 MA90 USA00 .30 IN00 PA00 Indiana, R = –.97 PA90 .25 USA90 IN90 Proportion of Lawyers Female

Fig. 1. Gender Composition of Lawyers by Population/Lawyer Ratio, United States, 1980–2000. Source: Data on lawyers from the US Census Bureau’s 5% Public-Use Microdata Samples (PUMS). Population data are from US Census Bureau tabulations. Lawyer data for all years from 5% PUMS: http://www2.census.gov/. 1980 and 1990 population data: http:// www.census.gov/population/www/censusdata/files/table-16.pdf; 2000 population data: http://www.census.gov/population/ www/cen2000/maps/files/tab02.pdf. Notes: N ¼ 153 in Fig. 1B (50 states plus District of Columbia for each of 3 years). Place labels (with two-digit year) correspond to US Postal Service state abbreviations. 1990 and 2000 counts of lawyers are multiplied by ‘‘person weights’’ included in the PUMS data. Because weights are not included in the 1980 PUMS data, counts of lawyers are multiplied by 20. ‘‘Lawyer’’ counts are limited to the detailed occupation code with this label (‘‘178’’ in 1980 and 1990; ‘‘210’’ in 2000) and thus do not include judges and other members of the American bar using other occupational codes. Census figures on lawyers are not always consistent with Martindale-Hubbell data published in the American Bar Foundation’s Lawyer Statistical Report (e.g., Carson 2004).

Proportion of Lawyers Female

A. USA, Massachusetts, Pennsylvania, and Indiana

Gender Inequality in the Chinese Legal Profession 341

342

ETHAN MICHELSON

expansion and feminization with selected examples, Fig. 1B includes all states plus the District of Columbia. I hasten to add that this relationship persists net of state-level per capita income (in constant dollars) (details not presented). Although the expansion of the American bar is closely correlated with rising levels of income and wealth (Pashigian, 1978; Sander & Williams, 1989), state-level variation in income does not explain away the relationship in Fig. 1. (The raw data used to construct Fig. 1 as well as income data are available from the author upon request.) Sander and Williams (1989) convincingly rule out the possibility of reverse causality – that the influx of women caused the expansion of the American legal profession. More plausibly, bar expansion serves to expand opportunities for women in the legal profession. Despite the strong association between expansion and feminization, however, I have not established a causal link between the two. Indeed, perhaps the feminization of legal professions would have happened anyway – even in the absence of the bar’s expansion – owing to cultural shifts accompanying the rise and diffusion of civil rights in the 1960s (Halliday, 1986). Nonetheless, there is some theoretical support for a causal link between expansion and feminization: Although they do not empirically test it, Chiu and Leicht (1999), drawing on Reskin and Roos’ (1990) ‘‘queuing theory,’’ posit that growing demand for lawyers puts better qualified women ahead of less qualified men in the ‘‘job queue.’’ While a deeper exploration of causal mechanisms is beyond the scope of this chapter, I have established in the foregoing a dual process of expansion and feminization in legal professions around the world.

Gender Inequality after Entry: Job and Career Quality Gaps Persist Although women have made great inroads numerically, gender inequality in other forms stubbornly endures in legal professions around the world. Research on gender inequality among lawyers has consistently found that women remain overrepresented in lower-pay, lower-status segments of the profession (Bolton & Muzio, 2007; Hull & Nelson, 2000; Rhode, 2001; Wilder, 2007), that women are less likely than men to get promoted to partner (Dau-Schmidt, Galanter, Mukhopadhaya, & Hull, 2009; Hagan & Kay, 1995; Kay, 1997; Noonan & Corcoran, 2004; Noonan, Corcoran, & Courant, 2008; Reichman & Sterling, 2002; Sterling & Reichman, 2004), that women’s incomes are lower than men’s (Dau-Schmidt et al., 2009; Dinovitzer, Reichman, & Sterling, 2009; Dixon & Seron, 1995; Hagan, 1990;


343

Heinz, Nelson, Sandefur, & Laumann, 2005; Noonan & Corcoran, 2004; Noonan et al., 2005, 2008; Sterling & Reichman, 2004; Wilder, 2007), and that women leave law firms and legal practice altogether at higher rates than their male counterparts (Kay, 1997; Noonan & Corcoran, 2004; Noonan et al., 2008; Reichman & Sterling, 2002; Sterling & Reichman, 2004; Wilder, 2007). Despite scholarly consensus of persistent job and career quality gaps among lawyers, there is disagreement on reasons for these gaps. Different studies provide empirical support for different theoretical explanations, such as gender gaps in human capital, gender-specific effects of family events (e.g., marital status, parenthood, and parental leaves), and gender discrimination (see Dinovitzer et al., 2009; Kay & Gorman, 2008 for reviews; also see Dau-Schmidt et al., 2009 on the effects of childcare). Rather than shedding new light on the issue of causal explanations, my goal in this chapter is far more modest: to establish an empirical baseline for feminization trends and gender gaps in the Chinese bar.

DATA AND METHODS Studies of gender inequality in legal professions elsewhere in the world have considered the gender composition of lawyers in different practice settings, such as private firm practice, solo practice, government counsel, corporate inhouse counsel, judgeships, prosecutors’ offices, and the legal academy. Because the scope of Chinese ‘‘lawyers’’ is limited almost entirely to law firm practice, the data I analyze in this chapter are limited to lawyers in this single practice setting: law firms. I draw on four sources of quantitative data: (1) official yearbook data on lawyers; (2) data from a 2000 survey of lawyers in 25 cities; (3) data from a 2007 survey of lawyers in 3 cities and 5 provinces; (4) and event history data from the population of registered lawyers in the city of Beijing between 1995 and 2004. In this section, I will first describe these four sources of data before discussing my methods and key measures.

Government Yearbook Data Although they contain annual counts of lawyers, neither the national China Statistical Yearbook (ZTN) nor the national China Law Yearbook (ZFN) disaggregate lawyer data by province or by gender. While it may be possible

344

ETHAN MICHELSON

to compile regionally disaggregated lawyer data from municipal and provincial yearbooks, data on lawyer gender is more difficult to find. To the best of my knowledge, the China Lawyer Yearbook (ZLN) is the most comprehensive source of data on Chinese lawyers. Less complete information on the gender composition of the Chinese bar has been published in Zhu (2007, p. 345) and Jia (2003). Because the China Lawyer Yearbook (ZLN) was published only between 2000 and 2005, official gender data for subsequent years are not publicly available.

2000 Survey Data In the summer of 2000, I carried out two surveys of lawyers: one in Beijing (N ¼ 462) and one in 24 small and mid-sized cities in 16 provinces outside Beijing (N ¼ 518).1 For convenience, I call this survey the ‘‘2000 survey’’ or the ‘‘25-city survey.’’ All questionnaires were self-administered. In Beijing, questionnaires were distributed and collected by 33 local undergraduate student research assistants, all female. Using a 1999 roster of Beijing law firms published in the Zhongguo Lu¨shi Bao (China Lawyer News), I randomly assigned Beijing law firms to the student research assistants. In Beijing, the firm-level refusal rate (e.g., receptionists’ turning away interviewers and firm directors’ refusing to permit lawyers’ participation) was 23%. In total, I collected data from lawyers in 131 identifiable firms, representing 38% of all Beijing law firms in 2000. The 24 survey sites outside Beijing were not selected randomly, but purposively. Outside Beijing, survey questionnaires were distributed and collected by 26 research assistants hired in Beijing, 19 of whom were female. These assistants originated from the cities they surveyed and were traveling home for the summer holiday anyway. Whereas the Hohhot and Nanning surveys were administered by two assistants, the remaining surveys were each administered by one assistant. Because law firm directories were available for only 10 of the 24 cities outside Beijing, in only a fraction of the survey sites outside Beijing could law firms be randomly sampled. Information on firm-level refusals in the surveys outside Beijing was not recorded. Moreover, the proportion of all firms accounted for by the 185 identifiable firms in the surveys outside Beijing is impossible to estimate given the absence of a comprehensive national law firm directory. However, in the 10 cities with available local law firm directories, I surveyed an average of 34% of all firms. Outside Beijing, the average number of firms surveyed per city was 7.5, and the maximum was 19 (Nanning).


345

I trained all assistants working on both surveys to select lawyers randomly within firms. Although information on individual-level refusals was recorded in neither sample, all indications are that these are reasonably representative samples (see Michelson, 2007, pp. 378–379, for details).

2007 Survey Data During February–April 2007, the All-China Lawyers Association administered a mail and email survey of lawyers (N ¼ 1,337) in three major cities (Beijing, Shanghai, and Guangzhou) and five provinces (Jilin, Zhejiang, Qinghai, Jiangsu, and Shandong). For convenience, I call this survey the ‘‘2007 survey’’ or the ‘‘8-province survey.’’ The survey instrument was designed by Shanghai Jiaotong University Koguan Law School’s Professor, JI Weidong, who generously shared the data. Because it was administered through official channels, the response rate – in the 80–85% range – was unusually high for a survey of its kind. With only a few exceptions, one lawyer in each sampled law firm was selected to participate in the survey. I have no additional information about how law firms were selected or about how individuals were selected within firms. However, judging by the proportion of partners in the two surveys (61% in the 2007 survey compared to 20% in the 2000 survey), the 2007 survey appears to overrepresent law firm leaders. I also have no additional information about the geographical distribution of law firms within provinces.

1995–2004 Beijing Lawyer Population Data In order to estimate Chinese lawyer attrition rates, I compiled a database of the population of lawyers who ever registered in Beijing between 1995 and 2004. Early rosters of law firms and their members were published in the Zhongguo Lu¨shi Bao (Chinese Lawyer News). This newspaper was discontinued and replaced by the Baixing Xin Bao (News for the Common People) in 1999 or 2000, and subsequent lawyer rosters were published by the Zhongguo Lu¨shi (Chinese Lawyer) magazine. The Baixing Xin Bao was later replaced by the Falu¨ Fuwu Shibao (Legal Service Times). All incarnations of the defunct Zhongguo Lu¨shi Bao have been published by the Fazhi Ribao (Legal Daily). The first lawyer roster in Beijing was published in 1993. Owing to the incompleteness of the early rosters, however, I limit my analysis to the 10 years spanning 1995 and 2004.

346

ETHAN MICHELSON

Lawyer rosters contain the following information supplied by law firms on mandatory annual registration (nianjian zhuce) forms: the names of law firms, the names of lawyers within law firms, law firm addresses (including postal code), and law firm telephone numbers. Since rosters published prior to 2001 are publicly available in hardcopy form only, research assistants computerized the information manually. Rosters published between 2001 and 2004 were published in electronic form on the website of the All-China Lawyers Association. I merged each roster together by matching lawyer names. Name redundancy – a problem of particular salience in China given the popularity of certain first names and a very limited supply of surnames – problematized this name-matching process. The assumptions I developed to make coding decisions about identical names observed in different firms in different years conservatively err on the side of career retention. In other words, whenever possible, in order to minimize the possibility of producing inflated attrition estimates, I coded the appearance of identical names across years and law firms as lateral moves between law firms and not as exits out of the bar. However, the effect of these conservative coding decisions on attrition estimates is probably offset by a countervailing source of bias: Since my estimated exit rates include lawyers who in actuality did not leave the bar but merely experienced a temporary absence (e.g., a maternity leave or illness) or who moved to another city, my estimates probably exaggerate the true attrition rate. Given these two countervail sources of bias, my attrition estimates are probably somewhere close to the truth. My consolidated event history database contains 51,653 observations (personyears) consisting of 11,500 unique lawyers (with 10,784 unique names and 6,840 unique first names) in 893 unique law firms.

Plan of Analysis I use the above-described sources of data to analyze four types of gender gaps among Chinese lawyers: (1) the numerical representation gap (a.k.a. feminization) and its changes over time; (2) the income gap; (3) the partnership gap; and (4) the career longevity gap. My first analysis – of feminization trends – is primarily descriptive and relies on government yearbook data. In my second analysis – of income inequality – I apply quantile regression techniques to the 2000 and 2007 survey data. Previous studies of lawyer income inequality elsewhere in the world, without exception, apply classical (ordinary least squares) linear regression techniques to income data. Most


347

studies, in an effort to conform to the assumption that values are normally distributed, model the logarithm of income in order to rein in outliers (e.g., Dinovitzer et al., 2009; Heinz et al., 2005; Hersch, 2003; Noonan & Corcoran, 2004; Noonan et al., 2005). Despite skewed income distributions caused by outliers, some studies model raw income values (Hagan, 1990; Hagan & Kay, 1995). Even if the normality assumption is satisfied, however, income data rarely satisfy an additional assumption of homoscedasticity: the assumption that the level of certainty about the effect of an independent variable on income is constant across in the income distribution (Hao & Naiman, 2007). Because it greatly relaxes these and other assumptions, quantile regression, also known as median regression, has become increasingly popular in analyses of income inequality (see Bishop et al., 2005; Knight & Song, 2005 for examples of quantile regression analysis of Chinese urban income inequality). Because it models median values, quantile regression can accommodate extreme outliers that are part of the real world. An additional advantage of quantile regression is the ease by which it permits assessments of the extent to which effects vary across the income distribution. Previous research on income inequality in legal professions has not examined whether the effect of an independent variable varies between lower-income and higher-income lawyers. With respect to my third analysis, in the North American partnership law firm context of ‘‘up or out’’ in which lawyers who do not ‘‘make partner’’ are typically forced out of the firm and sometimes out of legal practice altogether, cross-sectional data (from a single time point) introduce selection bias by underrepresenting certain groups of candidates more likely to drop out before the time of observation. As Hagan and Kay (1995, p. 84) put it, crosssectional data ‘‘ignore many lawyers who set out to become partners but have since left firms or the practice of law entirely.’’ In the Chinese context, by contrast, partnership is less an up-or-out promotion system and more strictly a system of law firm ownership. The Chinese legal profession generally lacks an analog to the North American ‘‘associate’’ lawyer, who, unless promoted to partner within a limited probationary time period, will exit the law firm or the bar as a whole. Generally speaking there is no such time pressure in China. Because lawyers in China are generally not selected into partnership according to a process that pushes other lawyers out of the bar, cross-sectional data are appropriate for analyzing partnership gaps in the Chinese bar. However, cross-sectional data do introduce real selection bias into the analysis of career longevity gaps – my fourth and final analysis. If women leave the bar at significantly higher rates than men, then the career durations of women who survive to the point of observation will be biased upward.

348

ETHAN MICHELSON

In an effort to overcome this source of selection bias, I analyze event history data from the Beijing lawyer rosters.

Key Measures I use data from the two surveys to estimate income gaps, partnership gaps, and career longevity gaps. The 2000 25-city survey measured income (gross annual pre-tax income from legal practice over the previous 12 months) as a categorical variable. Given limited numbers of income values (19 in Beijing and 17 in the remaining city samples), the income data are poorly equipped to support quantile regression analysis of the determinants of median income. For this reason, I model the mean of the categorical midpoints of income and billings. Not only are income and billings highly correlated (R ¼ .84), but combining the two categorical variables in this way produces a continuous variable with 115 unique values. But because billings are considerably greater than income (double or triple the value of income for the typical lawyer), this combined variable should not be interpreted as income. The 2007 8-province survey measured income (total gross income from legal practice in 2006) as a continuous variable in units of f10,000 (i.e., respondents were given space to enter a five-digit number and were instructed to use this space to enter income in wan, a standard unit in China for large values). I suspect, however, that a handful of lawyers who reported anomalously high income values may have reported income in f1 rather than in f10,000. For example, someone with an income of f2,000 (extremely low but nonetheless not impossible) may have entered f2,000 instead of f0.2 wan. For this reason, I dropped from the income analysis 22 observations with income values higher than f4 million.2 I hasten to add that the results of my income analysis are highly robust. The inclusion of these 22 observations in no way changes my substantive conclusions. In the time between the 2000 and 2007 surveys, partnership law firms eclipsed all other forms of law firm ownership in China (Michelson, 2007). Whereas in 2000, almost 60% of all law firms in China were nonpartnerships (e.g., state-owned and cooperative law firms) (Zhu, 2007, p. 331); by 2007, the vast majority of Chinese law firms were privately owned partnerships. For this reason, I limit the 2000 analysis of partnership status to lawyers in partnership law firms. However, in the 2007 survey data, many of the lawyers in the small minority of nonpartnership law firms nonetheless called themselves ‘‘partners.’’ I therefore include all lawyers in the 2007 analysis of partnership status.


349

With respect to career longevity, years of practice was recorded as a continuous variable in the 2000 survey and as a categorical variable in the 2007 survey. In order to preserve comparability between the two surveys, in this chapter I analyze years of practice as a categorical variable. I include a variety of control variables in the multivariate models that could potentially help explain gender gaps. However, I do not include hours of work and marital status in any of my models because they were not measured in the 2007 survey. Likewise, owing to excessive amounts of missing educational data in the 2007 survey, I also exclude this measure of human capital from my models. Although hours of work and certain educational credentials are significant predictors of both income and partnership status in models using data from the 2000 survey, their effects (details not reported) neither explain away other effects I report nor call into question any of my substantive conclusions. At the same time, marital status has no clear effect in the 2000 data (details not reported). Finally, I do not include age in any models because it is so closely correlated with years of practice and thus introduces multicollinearity problems. Although in previous research, the effect of parenthood on partnership status is positive for men and negative for women (Hagan & Kay, 1995; Hersch, 2003), neither of the Chinese surveys contains information about children. Descriptive statistics for all variables included in my analyses are presented in the appendix Tables A2 and A3. After presenting findings from my analysis of survey data, I use the Beijing lawyer roster data to estimate lawyer attrition rates. I added the following additional individual-level variables to my consolidated database of lawyers ever registered in Beijing between 1995 and 2004: year of entry into the Beijing bar (if the first observation point is after 1995, owing to left censoring); year of exit from the Beijing bar (if the last observation point is prior to 2004, owing to right censoring); and years of lateral moves between firms. The rosters contain no information about partnership status – either at the firm level (law firms that are partnerships) or at the individual level (lawyers who are law firm partners). Because the rosters also do not contain gender information, two Chinese research assistants guessed the gender of each name by assigning one of the following mutually exclusive categories: (1) 60–80% certainty of female gender; (2) W80% certainty of female gender; (3) 60–80% certainty of male gender; and (4) W80% certainty of male gender. In this chapter, I define gender separately for each coder (‘‘Coder 1’’ and ‘‘Coder 2’’) according to the lower threshold of certainty. According to this definition of gender, Coder 1 and Coder 2 agreed 89% of the time. Moreover, as we will see

350

ETHAN MICHELSON

shortly, their guesses are highly consistent with official yearbook data. As a source of external validation, I used a ‘‘gender guesser’’ computer program to code the gender of each name according to the composition of Chinese characters in the name (see http://www.chinese-tools.com/tools/genderguesser.html for the PHP web interface). Coder 1 and Coder 2 each agreed with the computer program’s brute-force codes 85% of the time. Because the human coders had a higher rate of agreement, I used the gender data they generated for the analyses in this chapter. However, I successfully replicated all analyses using gender data generated by the ‘‘gender guesser’’ computer program (details not presented). In this chapter, I do not estimate lateral moves between firms. Such a high degree of intercoder reliability – both between the two coders and between the coders and the computer program’s output – lends confidence to the findings I report from my event history analysis of lawyer career longevity gaps.

FINDINGS I will begin with descriptive findings of feminization trends over time before presenting the results of my analyses of income gaps, partnership gaps, and career longevity gaps between men and women in the Chinese bar. Feminization Fig. 2 displays feminization trends among lawyers in both Beijing and China as a whole. Fig. 3 displays changing numbers of lawyers over time – both in absolute numbers and as ratios of population to lawyers. Five patterns jump out from these two figures. First, female representation among lawyers has grown consistently over time (Fig. 2). Second, gender composition estimates based on the ‘‘guesses’’ of Coder 1 and Coder 2 are consistent with official yearbook data (Fig. 2). Indeed, according to official government data, in the 5 years spanning 2000 and 2004 the proportion of women in the Beijing bar averaged 28%, precisely the average value of the two sets of estimates generated by Coder 1 and Coder 2 for the same time period. Third, gender composition estimates from both the 2000 25-city survey and the 2007 8-province survey are consistent with official yearbook data. These consistencies enhance our confidence both in the representativeness of the survey data and in the accuracy of the coders’ gender guesses. Fourth, the population of Chinese lawyers has exploded over the past two decades from a very low base (Fig. 3). Finally, the declining ratio of population to lawyers

351


Gender Inequality in the Chinese Legal Profession 30% 29% 28% 27% 26% 25% 24% 23% 22% 21% 20% 19% 18% 17% 16% 15%

Beijing Coder 2 Coder 1 Of f icial Yearbooks Coders (Beijing) 2000 Survey (Beijing only) 2007 Survey (Beijing only) 2007 Survey (all 8 provinces) China

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Fig. 2. Lawyer Gender Composition, China and Beijing, 1995–2007. Source: Official yearbooks: ZLN. Note: Lawyer populations are limited to full-time, parttime, and specially appointed lawyers (i.e., they exclude ‘‘administrative assistants’’ [xingzheng zhuli]).

reflects the fact that the growth of lawyers has been far faster than general population growth (Fig. 3). In order to assess the link between bar expansion and feminization, Fig. 4 contains alternative representations of the same data contained in Figs. 2 and 3, plus data from a few additional province-level units (viz. Shanghai, Tianjin, and Zhejiang) contained in the appendix, Table A1. This figure shows that the analysis of American census data presented in Fig. 1A is perfectly replicated with Chinese data. The relationship between bar expansion and feminization is not limited to North America and Europe, but clearly extends to China. We can also see in Fig. 4 that, despite its explosive growth, the Chinese legal profession remains relatively small. Even if we were able to add nonlawyers who do legal work (e.g., corporate in-house counsel, prosecutors, etc.) to the population of Chinese lawyers, China’s population/lawyer ratio would still be many times greater than Germany’s ratio of about 800:1 (Schultz, 2003a, p. liii) and the United States’ extremely low ratio of less than 300:1. However, China’s most economically developed and globally integrated cities are clearly catching up. According to the data in Fig. 3, by the year 2007 the population/lawyer ratio in Beijing had already dropped to less than 800:1.

352

ETHAN MICHELSON 17

16,000

16 15

14,000 12,000

Lawyers

10,000

13

Population per Lawyer (right axis)

12 11

Lawyers (÷10) (left axis)

10 9

8,000

8 7

6,000

Lawyers (left axis)

4,000

China

6

Beijing

5 4

Population (1,000s) per Lawyer

14

3 2,000

2 Population per Lawyer (right axis)

1 0

0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Fig. 3. Lawyers and Ratio of Population to Lawyers, China and Beijing, 1995–2007. Source: ZTN, BTN. Notes: Lawyer populations are limited to full-time, part-time, and specially appointed lawyers (i.e., they exclude ‘‘administrative assistants’’ [xingzheng zhuli]). Population per lawyer in Beijing is limited to the officially registered population.

While Fig. 4 illustrates the association between bar expansion and feminization with selected examples from China, Fig. 5B includes all province-level units. Fig. 5A shows that, no different from the American case discussed earlier, the expansion of the Chinese legal profession is highly associated with economic development. At the same time, also no different from the American case, the data show that feminization is not reducible to economic development. OLS regression models (embedded in Fig. 5B) show not only that the effect of economic development on feminization is explained away entirely by the population density of lawyers but also that the effect of lawyer density on feminization persists net of economic development. Finally, Fig. 5B also reveals enormous regional variation. In some of the poorer parts of China, including Guangxi, Guizhou, and Jiangxi, population/lawyer ratios were in excess of 15,000:1 and female representation was below 15% in 2004 and 2005.

Fig. 4.


Tianjin, R = –.99

Zhejiang,

2002

2004 2005 2003

China, R = –.91

2000 R = –.94

2005 2004 2002 2003 2001

2000

2002 2001

2004 2003

2005

2000

2001

Proportion of Lawyers Female .25

.26

.27

.28

.29

.30

1

2002 2004 2001 2003

2001 2004 2003

2

3

1998 1999 1997 1996

2000

2000

4

5

1995

Source of Gender Data: ZLN Coder 1


2002

R = –.90

2005

B. Beijing

Selected Illustrations of the Relationship between the Population Density of Lawyers and the Gender Composition of Lawyers, 1995–2005. Source: ZTN; BTN; ZLN; see appendix, Table A1.


1 2 3 4 5 6 7 8 9 10 11 12 13 14

Shanghai, R = –.93

.16

.15

2000

2002 2003 2001

2004

2005

.17

.18

.19

.20

.21

.22

.23

.24

A. China, Shanghai, Tianjin, and Zhejiang


2

4

6

8

10

12

14

16

18

20

22

24

26

28

5,000

10,000 20,000 30,000 50,000

Per Capita GDP (yuan/person, log scale)

3,000

JX04 GX01 JX03 GX00 GX03 HeN00 GX02 JX01 JX05 GS01 AH00 JX02 HeN01 AH01 GS00 GX04 HeB00 JX00 GX05 HeN02 HeB01 GS04 GS02 SC01 GS05 GS03 HeN04 AH03 SC00 HeB03 HeN05 SaaX00 AH05 YN00 HuN01 AH04 YN01 HuN00 HB01 HeN03 HB00 HeB02 HuN03 AH02 YN02 HB02 HuN02 HeB04 HuN04 HB03 JL05 HeB05 SC02 HB04 HuN05 QH01 SaaX01 YN03 HB05 QH00 HLJ01 SC03 SC04 SaX04 QH02 SD00 NM02 SD01 SaX00 QH04 NM01 JL04 NM00 YN04 SC05 QH03 HLJ00 SaX01 JS00 JL00 SD02 FJ01 SaaX05 NM05 SaaX02 JL01 JL02 NM04 FJ00 JS01 YN05 NM03 HLJ04 JL03 SD04 SaX03 HaN03 HaN02 SD03 QH05 HLJ05 SaaX04 HaN04 FJ02 SaaX03 XJ01 SaX05 XJ02 CC00 SD05 HLJ02 HaN00 HaN05 SaX02 HLJ03 JS02 FJ03 FJ04 ZJ00 JS03 NX04 JS05 FJ05 XJ05 LN01 XJ04 NX00 NX02 NX05 JS04 XJ00 LN02 NX01 LN00 XJ03 CC03 CC02 HaN01 ZJ01 GD00 LN03 NX03 CC04 CC01 CC05 LN04 GD01 ZJ02 LN05 ZJ03 ZJ04 GD02 ZJ05 GD03 GD04 GD05 TJ00 TJ01 TJ02 TJ03 TJ04 TJ05 R = –.79 SH00 SH01 SH02 SH03 BJ00 SH04 BJ01 SH05 BJ04 BJ02 BJ03 BJ05

GZ04 GZ03 GZ02 GZ05

GZ00 GZ01


.05

.10

.15

.20

.25

.30

.35

GZ03 GZ04 GZ05

GZ02


2 4 6 8 10 12 14 16 18 20 22 24 26 28

R = –.60

LN05 HLJ05 HLJ04 SaaX04 HLJ01 LN03 LN02 QH01 LN04 BJ00 NM04 NM05 TJ00 LN01 HaN05 BJ02 TJ05 XJ05 NM02 SH05 SaX03 TJ04 HLJ00 XJ04 NM00 JL05 TJ03 LN00 NX04 HaN04 JL03 HeB03 XJ02 CC02 JL04 SaX04 GD04 SH04TJ02 XJ03 XJ01 HeB05 QH05 NX03 NX02 TJ01 NM01 NX01 NX05 JL02 NM03 HeB04 JL01 XJ00 QH03 QH04 SH02 GD05 QH02 SH03 GD00 SaX05 HaN03 SaaX05 SH01 GD03 SaX00 GD01 HaN01 SaX01 HaN02 HeB02 JS04 SH00 ZJ05 NX00 SD04 SD03 YN03HeB01 ZJ04 ZJ02 ZJ03 GS05 ZJ01 SC02 JX01 SD05 CC01 SaaX02 CC03 SC01 SC04 JS05 HaN00 GS03 HeN04 CC04 SD02 SC03 YN02 SC00 CC05 SaaX01 JX02 SaX02 JS01 GS04 ZJ00 JL00 SC05 HeN05 GD02 QH00 SaaX00 JS02 FJ01 HB02 HeN00 YN01 YN04 YN00 HeN02 CC00 HB01 FJ02 FJ04 JX00 YN05 GX02 FJ03 FJ05 HB05 AH05 GS00 HeB00 GX04 HuN00 JX03 HB04 GS02 JX04 AH04 JX05 FJ00 SD00 HB00 HeN03 HuN04 HeN01 AH03 AH02 HuN03 GX03 SD01 HuN01 JS03JS00 GX00 HuN05 AH01 GS01 GX05 HuN02AH00 GX01 HuB03

BJ01 BJ05 BJ04 BJ03

HLJ02 SaaX03

HLJ03

B. Gender Composition of Lawyers

GZ00

GZ01

Fig. 5. Overall Regional Variation in the Population Density of Lawyers and the Gender Composition of Lawyers, 2000–2005. Source: ZLN, http://chinadataonline.org; see appendix, Table A1. Notes: N ¼ 180 (30 province-level administrative units for each of 6 years between 2000 and 2005). Place labels (with two-digit year) are as follows: AH, Anhui; BJ, Beijing; CC, Chongqing; FJ, Fujian; GS, Gansu; GD, Guangdong; GX, Guangxi; GZ, Guizhou; HaN, Hainan; HeB, Hebei; HLJ, Heilongjiang; HeN, Henan; HuB, Hubei; HuN, Hunan; NM, Inner Mongolia; JS, Jiangsu; JX, Jiangxi; JL, Jilin; LN, Liaoning; NX, Ningxia; QH, Qinghai; SaaX, Shaanxi; SD, Shandong; SH, Shanghai; SaX, Shanxi; SC, Sichuan; TJ, Tianjin; XJ, Xinjiang; YN, Yunnan; ZJ, Zhejiang.


A. Population/Lawyer Ratio

354 ETHAN MICHELSON


355

Notwithstanding great inroads women have made into the Chinese legal profession, limited evidence may reflect the presence of durable obstacles blocking female entry. Evidence from two prominent Beijing law schools suggests that men are overrepresented among newly recruited lawyers in law firms. Women represented almost two-thirds of incoming undergraduate law majors at Peking University in 2000 (He, 2002). At the Renmin University of China Law School, women as a proportion of undergraduate students hovered in the 49–58% range between 1999 and 2005 (Zhu, 2007, p. 540). Likewise, Szto (2008) reports that women accounted for two-thirds of the entering class of 2004 at a law school in Fujian Province. Meanwhile, according to data from the 25-city survey and the 8-province survey, in 2000 and 2007, respectively, women accounted for 43% and 48% of the youngest age cohort of lawyers in Beijing (20–29 years of age). More conclusive analyses of entry barriers to law firms must await better data, including a source of comprehensive data on law school matriculation and graduation.3

Income Gaps Beginning with this section, we will see that, even if women lawyers are not disadvantaged in their prospects of entering law firm practice, they are greatly disadvantaged after entry. The benefits of quantile regression discussed above are immediately apparent. In both 2000 and 2007, as lawyer incomes increased, both income dispersion and the gender gap likewise increased. Indeed, income dispersion in the top two deciles was so great in both the 2000 and 2007 data that I limit all analyses to the bottom eight deciles. Let us first consider income in 2000. In models without controls (Fig. 6A), women’s median incomes in 2000 were between f5,000 and f8,000 lower than men’s in the bottom two deciles. In the top half of the income distribution, by contrast, the gap was between f20,000 and f30,000. Fig. 6 also shows that much of the gender gap is explained away by years of practice. Controlling for years of practice reduces the gender gap in median income from about f14,000 (Fig. 6A) to about f7,500 (Fig. 6B) across all displayed income quantiles. Indeed, years of practice alone shrinks the income gap (in Fig. 6B) roughly to the same extent as a battery of alternative control variables (in Fig. 6C). In the fully saturated model (Fig. 6D), the gender gap disappears completely in all but three income deciles. Even where the gender gap persists, however, it nonetheless shrinks considerably net of controls. The benefits of quantile regression are even more apparent in models calculated from the 2007 data. Changes in the size and character of the gender

356

ETHAN MICHELSON

25,000

A. Baseline Model

15,000

15,000

5,000 0 -5,000

5,000 0 -5,000

Female

Female

25,000

-15,000

B. Baseline Model Plus Years of Practice

-15,000

-25,000

-25,000

-35,000

-35,000

-45,000

-45,000 -55,000

-55,000 .1

.2

.3

.4

.5

.6

.7

.8

.1

.2

Income Quantile

25,000

.5

.6

.7

.8

D. Fully Saturated Model

15,000

5,000 0 -5,000

Female

Female

15,000

.4

Income Quantile

C. Saturated Model Minus Years of Practice

25,000

.3

-15,000

5,000 0 -5,000 -15,000

-25,000

-25,000

-35,000

-35,000

-45,000

-45,000

-55,000

-55,000 .1

.2

.3

.4

.5

Income Quantile

.6

.7

.8

.1

.2

.3

.4

.5

.6

.7

.8

Income Quantile

Fig. 6. Female Dummy Variable Coefficients (with 90% Confidence Intervals) from Quantile Regression Models of Factors Associated with Lawyers’ Income, 25-City Survey, 2000. Notes: N ¼ 870. A female (yes ¼ 1) dummy variable is the only independent variable in the baseline model. The fully saturated model includes nine additional independent variables. Black lines indicate changes in median income with each unit increase of a given independent variable for a given income quantile.

gap across the income distribution are far greater in the 2007 data than in the 2000 data. Whereas the gender gap widened in favor of men in the 2000 data, it widened in favor of women in the 2007 data as lawyers climbed the income distribution. In 2007, whereas in the bottom half of the income distribution median incomes were lower for women than for men, in the top half of the income distribution median incomes were higher for women than for men. In 2007, the effect of being female on median income was large and positive at the high end of the income distribution. Because at the top of the income distribution women’s reported incomes were higher than men’s, they pulled

357

390,000 360,000 330,000 300,000 270,000 240,000 210,000 180,000 150,000 120,000 90,000 60,000 30,000 0 -30,000

A. Baseline Model

Female

Female


390,000 360,000 330,000 300,000 270,000 240,000 210,000 180,000 150,000 120,000 90,000 60,000 30,000 0 -30,000

.2

.3 .4 .5 .6 Income Quantile

.7

.8

C. Saturated Model Minus Years of Practice

.1

.2

.3

.4 .5 .6 Income Quantile

.7

.8

B. Baseline Model Plus Years of Practice

.1

Female

Female

.1

390,000 360,000 330,000 300,000 270,000 240,000 210,000 180,000 150,000 120,000 90,000 60,000 30,000 0 -30,000

390,000 360,000 330,000 300,000 270,000 240,000 210,000 180,000 150,000 120,000 90,000 60,000 30,000 0 -30,000

.2

.3


.7

.8

D. Fully Saturated Model

.1

.2

.3


.7

.8

Fig. 7. Female Dummy Variable Coefficients (with 90% Confidence Intervals) from Quantile Regression Models of Factors Associated with Lawyers’ Income, 8-Province Survey, 2007. Note: N ¼ 950. See note given for Fig. 6.

up the average income for all women. Indeed, overall mean income was considerably higher for all women than for all men (f410,292 vs. f345,945) even though overall median income was lower for all women than for all men (f100,000 vs. f120,000). For this reason, even though women enjoyed no overall advantage in median income and enjoyed an advantage only at the top of the income distribution, in conventional linear regression models the female coefficient is nonetheless – and highly misleadingly – positive. Much of the gender gap in the 2007 data – no different from the 2000 survey data – is explained away by years of practice. In the bottom half of the income distribution, controlling for years of practice reduces the gender gap in median income from about f12,000 (Fig. 7A) to f0 (Fig. 7B). Indeed, in the bottom

358

ETHAN MICHELSON

half of the income distribution, years of practice alone is about as effective (in Fig. 7B) as a battery of alternative control variables (in Fig. 7C) at shrinking the income gap. In the fully saturated model (Fig. 7D), the gender gap disappears completely in the bottom half of the income distribution. Figs. 8 and 9 contain all remaining coefficients from the fully saturated models (i.e., all coefficients except those for gender, which are presented in Figs. 6D and 7D). The results are remarkably similar across the two surveys. Both in 2000 and 2007, the effects on median income of being a partner, of years of practice, of billings/cases from corporate clients as a proportion of all billings/cases, of prior work experience in the court system, and from local economic development (per capita GDP) are positive and increase as lawyers climb the income distribution. Only the effect of trial work (cases in court as a proportion of all cases) differs between the two surveys. In terms of magnitude, the largest effects both in 2000 and in 2007 were exerted by partnership and years of practice. Toward the top of the income distribution, the effect of being a partner on median income was over f100,000 in 2000 and over f200,000 in 2007. Also toward the top of the income distribution, lawyers with over 20 years of work experience enjoyed a median income advantage of over f50,000 in 2000 and of over f100,000 in 2007 (compared to lawyers with fewer than 4 years of work experience). We now turn to a closer look at the determinants of both partnership status and years of practice.

Partnership Gaps Two key patterns emerge from the results in Table 1. First, both in 2000 and 2007, the probability of being a law firm partner was considerably smaller for women than for men. Transforming the logit coefficients in Table 1 into odds ratios shows that, absent controls, women were 38% less likely than men to be partners in 2000 (Model 1, 1e.476 ¼ .379) and 43% less likely than men to be partners in 2007 (Model 5, 1e.566 ¼ .432). Second, in both surveys, years of practice completely wipe out this partnership gap. Controlling for years of practice, the female coefficients remain negative but lose statistical significance in both surveys (Models 2 and 6). In both surveys, the gender gap in partnership status shrinks far more when the models control for years of practice than when they control for the remaining five control variables (Models 3 and 7). Indeed, in the 2007 data, the remaining five control variables slightly widen the gender gap (Model 7). In the saturated models (Models 4 and 8), the gender gap persists but is far below any reasonable threshold of statistical significance. As in my analysis

City-Level Per Capita GDP (log), 2000

Years of Practice 4-10 Years

0

200

400

600

800

1,000

1,200

1,400

0

5,000

10,000

15,000

20,000

25,000

30,000

35,000

10,000

25,000

40,000

55,000

70,000

.1

.1

.1

.2

.2

.2

.3 .4 .5 .6 .7 Income Quantile



.8

.8

.8

40,000 35,000 30,000 25,000 20,000 15,000 10,000 5,000 0 -5,000 -10,000

0

10,000

20,000

30,000

40,000

50,000

60,000

70,000

80,000

90,000

0

25,000

75,000 50,000

100,000

125,000

150,000

175,000

200,000

225,000

.1

.1

.1

.2

.2

.2




.8

.8

.8

-1,200

-1,000

-800

-600

-400

-200

0

20,000 0 -20,000

60,000

100,000

140,000

180,000

220,000

260,000

30,000 27,000 24,000 21,000 18,000 15,000 12,000 9,000 6,000 3,000 0 -3,000

.1

.1

.1

.2

.2

.2




.8

.8

.8

Fig. 8. Coefficients (with 90% Confidence Intervals) for Remaining Variables in Fully Saturated Quantile Regression Model of Lawyer Income, 25-City Survey, 2000. Note: N ¼ 870. Black lines indicate changes in median income with each unit increase of a given independent variable for a given income quantile.

% Billings from Corporate Clients

Partner Years of Practice 11-20 Years Prior Work in Court

Firm Size (log) Years of Practice>20 Years % Cases in Court

85,000


Fig. 9.

Provincial Per Capita GDP (log), 2006

Years of Practice 4-10 Years

.1

.1

.1

.2

.2

.2




.8

.8

.8

0

25,000

50,000

75,000

100,000

125,000

150,000

175,000

200,000

10,000

30,000

50,000

70,000

90,000

110,000

130,000

150,000

170,000

0

50,000

100,000

150,000

200,000

250,000

300,000

350,000

400,000

450,000

.1

.1

.1

.2

.2

.2




.8

.8

.8

1,000 900 800 700 600 500 400 300 200 100 0 -100 -200

0

50,000

100,000

150,000

200,000

250,000

300,000

350,000

70,000 65,000 60,000 55,000 50,000 45,000 40,000 35,000 30,000 25,000 20,000 15,000 10,000 5,000 0 -5,000

.1

.1

.1

.2

.2

.2




.8

.8

.8

Coefficients (with 90% Confidence Intervals) for Remaining Variables in Fully Saturated Quantile Regression Model of Lawyer Income, 8-Province Survey, 2007. Note: N ¼ 950. See note given for Fig. 8.

5,000 4,500 4,000 3,500 3,000 2,500 2,000 1,500 1,000 500 0

100,000 90,000 80,000 70,000 60,000 50,000 40,000 30,000 20,000 10,000 0 -10,000

% Cases from Corporate Clients

Partner Years of Practice 11-20 Years Prior Work in Court

Firm Size (log) Years of Practice>20 Years % Cases in Court

175,000 160,000 145,000 130,000 115,000 100,000 85,000 70,000 55,000 40,000 25,000 10,000

360 ETHAN MICHELSON

.893 (.211) .007 599

2.649 (.545) .150 599

1.734 (.437) 3.046 (.537) 3.389 (.642)

.137 (.161)

Model 2

.464 (.148) .536 (.205) .018 (.004) .002 (.005) .819 (.235) 7.365 (2.406) .057 599

.312 (.218)

Model 3

.363 (.182) 1.017 (.355) .017 (.003) .002 (.006) .685 (.263) 4.022 (2.748) .188 599

1.749 (.405) 3.037 (.558) 3.471 (.538)

.033 (.167)

Model 4

.540 (.525) .009 1,253

.566 (.204)

Model 5

1.422 (.502) .181 1,253

1.674 (.479) 3.196 (.406) 4.208 (.716) 1.0544 (.619)

.212 (.282)

Model 6

.223 (.105) .254 (.343) .013 (.004) .008 (.007) 1.3934 (.778) 14.9014 (8.208) .125 1,253

.586 (.163)

Model 7

8-Province Survey, 2007

.101 (.093) .401 (.368) .011 (.003) .007 (.007) 1.202 (.801) 10.737 (8.810) .253 1,253

1.720 (.497) 3.018 (.466) 3.947 (.906) 1.413 (.503)

.275 (.237)

Model 8

Notes: 4pr.10, pr.05, pr.01, pr.001, two-tailed tests. Robust standard errors that correct for nonindependence within cities/ provinces are in parentheses. ‘‘% Corporate clients’’ refers to billings in the 2000 survey and to cases in the 2007 survey. Models calculated from the 2000 25-city survey are limited to lawyers in partnership law firms.

Pseudo R2 N

Intercept

Per capita GDP (log)

% Cases in court

% Corporate clients

Prior work in court

o4 years (omitted) Firm size (log)

Missing

W20 years

11–20 years

(.206)

.476

Model 1

25-City Survey, 2000

Unstandardized Coefficients from Logistic Regression Models of Factors Associated with Lawyer Partnership Status, China, 2000 and 2007.

Years of practice 4–10 years

Female

Table 1. Gender Inequality in the Chinese Legal Profession 361

362

ETHAN MICHELSON

of income gaps, the results of my analysis of partnership gaps are remarkably consistent across the two surveys. In both surveys, and consistent with findings elsewhere in the world (Kay & Brockman, 2003), firm size reduces the likelihood of being a partner, although in 2007 this effect is explained away by years of practice. Likewise, in both surveys, prior work experience in the court system increases the likelihood of being a partner, although only in 2000 this effect is statistically significant. Finally, in both surveys the effect on partnership chances of representing corporate clients is large and statistically significant.

Career Longevity So far I have established that years of practice, more than anything else, account both for the income gap and for the partnership gap between male and female lawyers in China. But what accounts for years of practice? Results from my analysis of years of practice in Table 2 reveal two important patterns. First, the gap between men and women in years of practice is huge. In both 2000 and 2007, the chances of being one point higher on the four-point measure of years of practice was 38% lower for all women in 2000 (Model 1, 1e.482 ¼ .382), 49% lower for women in partnership law firms in 2000 (Model 3, 1e.669 ¼ .488), and 49% lower for women in all firms in 2007 (Model 5, 1e.664 ¼ .485). Second, these career longevity gaps persist net of all controls in saturated models (Models 2, 4, and 6). In short, no observed information can account for the wide career longevity gap between men and women in the Chinese bar. Survival analysis of my 1995–2004 event history data collected from rosters of Beijing lawyers confirms the existence and scale of such a notable career longevity gap. Table 3 contains average annual exit rates and Table 4 contains total exit rates over the full span of time (9 years owing to censoring). Both sets of analyses show statistically significantly higher rates of exit for women than for men. Results in Table 3 show that average annual exit rates were 20–30% higher for women than for men (5.9% vs. 4.6% according to Coder 1 and 5.7% vs. 4.7% according to Coder 2). Results in Table 4 show that total exit rates were 20–25% higher for women than for men (41.1% vs. 32.7% according to Coder 1 and 39.3% vs. 33.2% according to Coder 2). A similar pattern emerges with respect to estimated career life expectancies. Both restricted survival times (estimated after excluding right-censored data) and extended survival times (estimated with the

363


Table 2. Unstandardized Coefficients from Ordered Logistic Regression Models of Factors Associated with Lawyers’ Years of Practice, China, 2000 and 2007. 25-City Survey, 2000 All law firms

Model 1

Model 2

8-Province Survey, 2007

Partnership law firms only Model 3

Model 4

Model 5

Model 6

.664 .738 .482 .446 .669 .625 (.113) (.124) (.134) (.132) (.097) (.178) .373 .335 Firm size (log) .339 (.092) (.096) (.099) .001 Prior work in court .329 .636 (.262) (.299) (.181) .006 % Corporate clients .009 .010 (.006) (.005) (.003) % Cases in court .001 .0005 .004 (.003) (.003) (.003) .515 .835 Per capita GDP (log) .409 (.108) (.149) (.159) Cut 1 1.015 5.495 .960 6.547 1.818 11.081 (.129) (.910) (.163) (1.380) (.203) (1.561) .3864 8.740 Cut 2 1.171 3.250 1.358 4.144 (.170) (.894) (.214) (1.365) (.222) (1.499) 5.163 .308 3.050 5.967 Cut 3 4.317 .078 (.214) (.882) (.358) (1.455) (.325) (1.188) .006 .021 .011 .033 .008 .043 Pseudo R2 N 920 920 599 599 1,230 1,230 Female

Notes: 4pr.10, pr.05, pr.01, pr.001. Robust standard errors that correct for nonindependence within cities/provinces are in parentheses. The dependent variable (years of practice) has four values: (1) o4 years, (2) 4–10 years, (3) 10–20 years, and (4) W20 years. ‘‘% Corporate clients’’ refers to billings in the 2000 survey and to cases in the 2007 survey. Cut points are analogous to intercepts in single-equation regression models and are used to calculate predicted probabilities.

inclusion of right-censored data) show statistically significantly shorter careers for women than for men. Table 5 shows that the gender gap in restricted survival time was 4–5% shorter for women than for men (7.2 vs. 7.5 years according to both coders) and that the gender gap in extended survival time was 20–30% shorter for women than for men (21 vs. 27 years according to Coder 1 and 22 vs. 26 years according to Coder 2).

364

Table 3.

ETHAN MICHELSON

Estimated Annual Lawyer Attrition Rates, Beijing, 1995–2004. Average Annual Exit Rate (%)

95% Confidence Interval Lower bound

Upper bound

Years at Risk

N (Lawyers)

Coder 1 Female Male Total

5.9 4.6 5.0

5.4 4.4 4.7

6.4 4.9 5.2

9,103 24,227 33,330

2,583 6,565 9,148


5.7 4.7 5.0

5.3 4.4 4.7

6.2 5.0 5.2

9,598 23,720 33,318

2,716 6,429 9,145

Table 4.

Estimated Total Lawyer Attrition Rates, Beijing, 1995–2004. Failure Function, Ninth Year (%)


Upper bound

N (Lawyers)


41.1 32.7 35.0

36.8 31.1 32.9

45.9 35.3 37.2

2,583 6,565 9,148


39.3 33.2 35.0

35.3 30.8 32.8

43.9 35.8 37.2

2,716 6,429 9,145

DISCUSSION AND CONCLUSIONS I began this chapter by asking whether China’s incipient, dynamic, and expanding legal profession is afflicted by or immune to the more general forces producing negative labor market outcomes for women in urban China. In China’s larger urban context of labor retrenchment, women have fared poorly relative to their male counterparts in terms of employment opportunities, income, and upwards mobility. I posed the following question: Is the same true in China’s market for legal services? And I tried to answer this question by measuring the same gender gaps in China’s legal profession. My findings suggest that gender inequality in China’s private

365


Table 5.

Estimated Longevity of Lawyer Careers, Beijing, 1995–2004. Restricted Mean (Years)


Upper bound

Extended Mean (Years)

N (Lawyers)


7.17 7.52 7.42

7.04 7.44 7.35

7.30 7.60 7.49

21.1 26.8 25.0

2,583 6,565 9,148


7.22 7.51 7.42

7.09 7.43 7.36

7.35 7.58 7.49

22.0 26.4 25.0

2,716 6,429 9,145

Notes: Restricted means are underestimated because they are limited to right-censored data (i.e., they exclude unobserved future events). Extended means are designed for right-censored data and are calculated by exponentially extending the survival curve to zero.

and highly market-driven legal profession is a microcosm of larger patterns of female disadvantage in China’s evolving urban labor market. At the same time, China’s legal profession also mirrors patterns observed in market-driven legal professions elsewhere in the world. No different from the American case, for example, the expansion of the Chinese legal profession is associated with broader economic expansion. Meanwhile, also consistent with a more general pattern, feminization has accompanied expansion. The global feminization of legal professions, however, has not leveled the playing field for men and women lawyers. Although employment opportunities for women lawyers around the world have greatly expanded quantitatively, their careers remain qualitatively less successful than those of their male counterparts. China is no exception to this general pattern of persistent gender inequality. Despite numerical growth in jobs for women in the Chinese bar, women do not hold their jobs as lawyers for as long as men do. In China, years of practice appear to be the lynchpin either holding together or unhinging the careers of women lawyers. Women’s shorter survival times are perhaps the greatest source of their plight. If women stayed in the bar for longer periods of time, they would accumulate more work experience and would reap greater professional rewards. For women, greater stocks of this essential source of human capital would enhance their chances of professional success.

366

ETHAN MICHELSON

Years of practice, more than anything else, explain both the income gap and the partnership gap between men and women in the Chinese bar. The story of gender gaps in the Chinese bar is overwhelmingly about the gender gap in work experience. Women’s reported incomes were significantly lower than men’s in 2000 (and in 2007 at the bottom half of the income distribution) in large part because women’s careers were shorter than men’s. For precisely the same reason, women’s chances of being a law firm partner were significantly lower than men’s. In the American context, Hersch (2003), too, found that the gender gap in lawyers’ career longevity explains most of the gender gap in lawyers’ income.4 On the basis of this finding, Hersch (2003) concludes that the gender gap in lawyers’ income will spontaneously shrink if women continue to enter the bar and if cohort replacement increases female representation among lawyers with the most work experience – but only barring the persistence of other causes of their relatively short careers, such as temporary and permanent exits due to parenthood and overt gender discrimination. Although I have identified a significant career longevity gap between men and women as a formidable obstacle preventing women from reaping professional rewards, viz. income and upwards mobility, I have not identified causes of this career longevity gap. One possibility is that gender discrimination is a more important cause than human capital differences. Consistent with this possibility, regression results show that, although work experience – an important source of human capital – explains much of the income gap and the entire partnership gap in 2000 and 2007, the gender gap in work experience persists net of controls. In other words, even women who were otherwise seemingly identical to men – with otherwise seemingly identical clients doing otherwise seemingly identical legal work in otherwise seemingly identical contexts – had significantly less work experience than men. My failure to explain away this gap could be an indication of gender discrimination pushing women out of the bar. But it could also be an indication that women are leaving law firm practice to attend to family priorities, which the surveys did not measure. Or it could reflect other unobserved effects. Because I cannot identify the root causes of women’s lower levels of work experience, I cannot conclusively assess the relative contributions of human capital and gender discrimination to income and partnership gaps. In addition to the possibility of parenthood and gender discrimination, barriers to longer careers in law firm practice could also include contextually specific business practices. Based on their interview research in the central city of Wuhan, Wang and Gao (2000, p. 13) paint a picture of women


367

avoiding trial work in general and criminal defense work in particular because the essential process of building relationships with judicial insiders – which often involves young female escorts and heavy drinking at karaoke clubs – is particularly difficult for women. Reports of sexual harassment and judges’ demands for sexual favors are common enough that, in the course of my fieldwork in Beijing, I repeatedly heard lawyers remark that women in the profession were either unable to get married or unable to stay married. Szto (2008) reports that this phenomenon is not limited to trial work and criminal defense, but is an integral and more general part of law firm efforts to cultivate and manage clients of all kinds. Another factor both Wang and Gao (2001, p. 13) and Szto (2008) identify as an obstacle to the integration of women into the Chinese legal profession is the frequency of out-of-town business trips, which apparently are more difficult, less desirable, and less culturally appropriate for women than for men. At the same time, however, if differences between the 2000 and 2007 survey data reflect real changes over time, gender gaps and gender discrimination may indeed have begun to shrink in the Chinese bar. First, income gaps in the 2000 data never entirely closed, but rather persisted net of controls across much of the income distribution in saturated models. In the survey data collected 7 years later, however, gender gaps (that favored men) completely disappeared after the introduction of control variables into the models. Insofar as unexplained gender gaps are indications of gender discrimination, these differences between the 2000 and 2007 data suggest the possibility of waning gender discrimination. Second, at no place on the 2000 income distribution did women enjoy an income advantage over men. In the 2007 data, by contrast, a gender gap in lawyers’ income favored women toward the top of the income distribution. If the data are believable, they could be an indication that women are joining elite members of the corporate bar in growing numbers. At the same time, however, the consistently wide partnership gap in both the 2000 data and the 2007 data suggest precisely the opposite: that women’s chances of joining the elite ranks of China’s lawyers remain limited. This chapter represents a first step toward a more complete understanding of gender inequality in China’s emerging legal profession. While this chapter helps establish an empirical baseline, much work remains to be done. Foremost in future research agendas must be efforts to identify root causes underlying the significant career longevity gap between male and female lawyers in China. How often do women exit by choice, and how often are they pushed out? Under what specific circumstances do they exit law firm practice? These and other questions that await future research.

368

ETHAN MICHELSON

NOTES 1. The 24 cities include 22 prefecture-level cities (in Hebei: Tangshan, Qinhuangdao, and Baoding; in Shanxi: Changzhi; in Inner Mongolia: Hohhot; in Liaoning: Dandong and Liaoyang; in Heilongjiang: Shuangyashan and Mudanjiang; in Zhejiang: Wenzhou; in Fujian: Quanzhou; in Jiangxi: Nanchang; in Henan: Anyang and Xinxiang; in Hubei: Yichang; in Hunan: Zhuzhou; in Guangxi: Guilin, Nanning, and Liuzhou; in Hainan: Haikou; in Qinghai: Xining; and in Ningxia: Yinchuan) and 2 county-level cities (Guangdong’s Nanhai and Xinjiang’s Changji). At the time of the survey, the average urban population in these cities was 800,000, a small fraction of Beijing’s 11 million. 2. Currency exchange rates in Yuan per US dollar at the time of the 2000 and 2007 surveys were about f8.3 and f7.8, respectively. 3. Surprisingly, my 1995–2004 event history database of the Beijing lawyer population shows virtually no differences between the gender composition of entering lawyers and the gender composition of all lawyers. I expected more recent cohorts of lawyers entering the bar to have greater female representation than the bar as a whole. According to the gender data supplied by both Coder 1 and Coder 2, the proportion of women in entering cohorts is only about two percentage points higher than the proportion of women among all lawyers and far below the 40–50% levels of female representation among lawyers in Beijing under age 30 in the two surveys. 4. In other studies of lawyers in Canada and the United States, however, years of practice does not explain away the gender gap in income (Dinovitzer et al. 2009; Dixon & Seron, 1995; Hagan & Kay, 1995; Heinz et al., 2005; Noonan & Corcoran, 2004).

ACKNOWLEDGMENTS I am grateful to Professor JI Weidong for generously sharing his 2007 survey data collected with the financial support of the Center for Legal Dynamics of Advanced Market Societies, Graduate School of Law, Kobe University, Japan. Sida LIU helped broker this arrangement and has provided critical feedback and suggestions at various stages of this project. Wudi LIU supplied his ‘‘gender guesser’’ PHP computer program, which Jianling JIANG skillfully modified for the research behind this chapter. I would also like to thank Wei HE, Hui ZHENG, WANG Xiaobei, and Weiwei SHEN for their research assistance. Of course I am solely responsible for all defects and omissions. (Chinese surnames in this chapter are capitalized to avoid confusion created by competing norms on different sides of the Pacific Ocean.) Text on the methodological details of the 25-city


369

of Chinese lawyers I conducted in the year 2000 was previously published in Michelson (2007).

REFERENCES Abel, R. L. (1989). American lawyers. New York: Oxford University Press. Bishop, J. A., Luo, F., & Wang, F. (2005). Economic transition, gender bias, and the distribution of earnings in China. Economics of Transition, 13(2), 239–259. Bolton, S. C., & Muzio, D. (2007). Can’t live with ‘Em; Can’t live without ‘Em: Gendered segmentation in the legal profession. Sociology, 41(1), 47–64. BTN (Beijing Tongji Nianjian [Beijing Statistical Yearbook]). Various Years. Beijing: Zhongguo Tongji Chubanshe. Cai, F., Park, A., & Zhao, Y. (2008). The Chinese labor market in the reform era. In: L. Brandt & T. G. Rawski (Eds), China’s great economic transformation (pp. 167–214). Cambridge, MA: Cambridge University Press. Cao, Y. (2001). Careers inside organizations: A comparative study of promotion determination in reforming China. Social Forces, 80(2), 683–712. Cao, Y., & Hu, C.-Y. (2007). Gender and job mobility in postsocialist China: A longitudinal study of job changes in six coastal cities. Social Forces, 85(4), 1535–1560. Carson, C. N. (2004). The lawyer statistical report: The US legal profession in 2000. Chicago: The American Bar Foundation. Chiu, C., & Leicht, K. T. (1999). When does feminization increase equality? The case of lawyers. Law & society review, 33(3), 557–593. Dau-Schmidt, K. G., Galanter, M. S., Mukhopadhaya, K., & Hull, K. E. (2009). Men and women of the bar: The impact of gender on legal careers. Michigan Journal of Gender and Law, 16, forthcoming. De´murger, S., Fournier, M., & Chen, Y. (2007). The evolution of gender earnings gaps and discrimination in urban China, 1988–95. Developing Economies, 45(1), 97–121. Dinovitzer, R., Reichman, N., & Sterling, J. (2009). The differential valuation of women’s work: A new look at the gender gap in lawyers’ incomes. Social Forces, forthcoming (December). Dixon, J., & Seron, C. (1995). Stratification in the legal profession: Sex, sector, and salary. Law & Society Review, 29(3), 381–412. Du, F., & Dong, X. (2009). Why do women have longer durations of unemployment than men in post-restructuring urban China? Cambridge Journal of Economics, 33(2), 233–252. Fuszara, M. (2003). Women lawyers in Poland under the impact of post-1989 transformation. In: U. Schultz & G. Shaw (Eds), Women in the world’s legal professions (pp. 371–386). Oxford: Hart Publishing. Gerber, T. P., & Mayarova, O. (2006). Dynamic gender differences in a post-socialist labor market, 1991–1997. Social Forces, 84(4), 2047–2075. Gold, T. B., Hurst, W. J., Won, J., & Li, Q. (Eds). (2009). Laid-off workers in a workers’ state: Unemployment with Chinese characteristics. New York: Palgrave Macmillan. Hagan, J. (1990). The gender stratification of income inequality among lawyers. Social Forces, 68(3), 835–855.

370

ETHAN MICHELSON

Hagan, J., & Kay, F. (1995). Gender in practice: A study of lawyers’ lives. New York: Oxford University Press. Halliday, T. C. (1986). Six score years and ten: Demographic transitions in the American legal profession, 1850–1980. Law & Society Review, 20(1), 53–78. Hao, L., & Naiman, D. Q. (2007). Quantile regression. Thousand Oaks, CA: Sage Publications, Inc. He, W. (2002). He Weifang: Faxue Mantan. (He Weifang: An informal discussion on legal studies.) Speech to the incoming class of the Peking University School of Law, October 9, 2000. Available at http://flwh.znufe.edu.cn/article_show.asp?id ¼ 142. Retrieved on January 4, 2009, on file with author. Heinz, J. P., Nelson, R. L., Sandefur, R. L., & Laumann, E. O. (2005). Urban lawyers: The new social structure of the bar. Chicago: The University of Chicago Press. Hersch, J. (2003). The new labor market for lawyers: Will female lawyers still earn less? Cardozo Women’s Law Journal, 10(1), 1–58. Hull, K. E., & Nelson, R. L. (2000). Assimilation, choice, or constraint? Testing theories of gender differences in the careers of lawyers. Social Forces, 79(1), 229–264. Hunter, R. (2003). Women in the legal profession: The Australian profile. In: U. Schultz & G. Shaw (Eds), Women in the world’s legal professions (pp. 87–102). Oxford: Hard Publishing. Hurst, W. J. (2009). The Chinese worker after socialism. Cambridge, MA: Cambridge University Press. Jia, W. (Ed.) (2003). 2002 Niandu Zhongguo Lu¨shi Fazhan Shuju Tongji Fenxi Baogao [2002 Report on the Statistical Analysis of Data on the Development of China’s Lawyers]. Report dated August 20, prepared using data jointly supplied by the Ministry of Justice and the ACLA. Kay, F. M. (1997). Flight from law: A competing risks model of departures from law firms. Law & Society Review, 31(2), 301–335. Kay, F. M., & Brockman, J. (2003). Barriers to gender equality in the Canadian legal establishment. In: U. Schultz & G. Shaw (Eds), Women in the world’s legal professions (pp. 49–75). Oxford: Hard Publishing. Kay, F. M., & Gorman, E. (2008). Women in the legal profession. Annual Review of Law and Social Science, 4, 299–332. Kim, J. (2007). Legal profession and legal culture during Korea’s transition to democracy and a market economy. In: Raising the bar: The emerging legal profession in East Asia (pp. 47–80). Cambridge, MA: East Asian Legal Studies, Harvard Law School. Knight, J., & Li, S. (2006). Unemployment duration and earnings of re-employed workers in urban China. China Economic Review, 17(2), 103–119. Knight, J., & Song, L. (2005). Towards a labour market in China. New York: Oxford University Press. Kornai, J. (1992). The socialist system: The political economy of communism. Princeton, NJ: Princeton University Press. Li, S., & Gustafsson, B. (2008). Unemployment, earlier retirement, and changes in the gender income gap in urban China, 1995–2002. In: B. Gusafsson, L. Shi & T. Sicular (Eds), Inequality and public policy in China (pp. 243–266). Cambridge, MA: Cambridge University Press. Liu, S. (2009). The logic of fragmentation: An ecological analysis of the Chinese legal services market. Ph.D. Dissertation, Department of Sociology, University of Chicago.


371

Mather, L. (2003). Gender in context: Women in family law. In: U. Schultz & G. Shaw (Eds), Women in the world’s legal professions (pp. 33–47). Oxford: Hard Publishing. Michelson, E. (2006). The practice of law as an obstacle to justice: Chinese lawyers at work. Law & Society Review, 40(1), 1–38. Michelson, E. (2007). Lawyers, political embeddedness, and institutional continuity in China’s transition from socialism. American Journal of Sociology, 113(2), 352–414. Noonan, M. C., & Corcoran, M. E. (2004). The mommy track and partnership: Temporary delay or dead end? Annals of the American Academy of Political and Social Science, 596, 130–150. Noonan, M. C., Corcoran, M. E., & Courant, P. N. (2005). Pay differences among the highly trained: Cohort differences in the sex gap in lawyers’ earnings. Social Forces, 84(2), 853–872. Noonan, M. C., Corcoran, M. E., & Courant, P. N. (2008). Is the partnership gap closing for women? Cohort differences in the sex gap in partnership chances. Social Science Research, 37(1), 156–179. Parish, W. L., & Busse, S. (2000). Gender and work. In: W. Tang & W. L. Parish (Eds), Chinese life under reform: The changing social contract (pp. 209–231). Cambridge, MA: Cambridge University Press. Pashigian, B. P. (1978). The number and earnings of lawyers: Some recent findings. American Bar Foundation Research Journal, 3(1), 51–82. Reichman, N. J., & Sterling, J. S. (2002). Recasting the brass ring: Deconstructing and reconstructing workplace opportunities for women lawyers. Capital University Law Review, 29(4), 923–977. Reskin, B. F., & Roos, P. A. (1990). Job queues, gender queues: Explaining women’s inroads into male occupations. Philadelphia, PA: Temple University Press. Rhode, D. L. (2001). The unfinished agenda: Women and the legal profession. ABA Commission on Women in the Profession, American Bar Association. Rosemont, H., Jr.. (2000). China’s new economic reforms: Replacing iron rice bowls with plastic cups. In: T. B. Weston & L. M. Jensen (Eds), China beyond the headlines (pp. 171–192). Lanham, MD: Rowman & Littlefield. Sander, R. H., & Williams, E. D. (1989). Why are there so many lawyers? Perspectives on a turbulent market. Law & Social Inquiry, 14(3), 431–479. Schultz, U. (2003a). Introduction: Women in the world’s legal professions: Overview and synthesis. In: U. Schultz & G. Shaw (Eds), Women in the world’s legal professions (pp. xxv–lxii). Oxford: Hard Publishing. Schultz, U. (2003b). The status of women lawyers in Germany. In: U. Schultz & G. Shaw (Eds), Women in the world’s legal professions (pp. 271–293). Oxford: Hard Publishing. Shu, X. (2005). Market transition and gender segregation in urban China. Social Science Quarterly, 86(Suppl), 1299–1323. Shu, X., & Bian, Y. (2003). Market transition and gender gap in earnings in urban China. Social Forces, 81(4), 1107–1145. Sterling, J. S., & Reichman, N. (2004). Sticky floors, broken steps, and concrete ceilings in legal careers. Texas Journal of Women and the Law, 14(1), 27–76. Szto, M. (2008). Gender and the Chinese legal profession in historical perspective: From heaven and earth to rule of woman? Unpublished manuscript. Available at http://works. bepress.com/mary_szto/1. Retrieved on January 10, 2009.

372

ETHAN MICHELSON

Wang C., & Gao, Q. (2000). Lu¨shi Zhiye de Xianzhuang Diaocha: Wuhan Lu¨shi Fangtan Zongshu. [A survey on the current state of the legal profession: A summary of interviews from Wuhan.] Zhongguo Lu¨shi [Chinese Lawyer], December, pp. 5–13. Wilder, G. Z. (2007). Women in the profession: Findings from the first wave of the after the JD study. Overland Park, KS: The NALP Foundation for law Career Research and Education and The National Association for Law Placement, Inc. (NALP). ZFN (Zhongguo Falu¨ Nianjian [China Law Yearbook]). (1986–2001). Beijing: Zhongguo Falu¨ Nianjian She. Zhang, Y., Hannum, E., & Wang, M. (2008). Gender-based employment and income differences in urban China: Considering the contributions of marriage and parenthood. Social Forces, 86(4), 1529–1560. Zhu, J. (Ed.) (2007). Zhongguo Falu¨ Fazhan Baogao: Shujuku he Zhibiao Tixi. [Report on China law development: Database and indicators.] Beijing: Zhongguo Renmin Daxue Chubanshe. ZLN (Zhongguo Lu¨shi Nianjian [China Lawyer Yearbook]). (2001–2006). Beijing: Renmin Fayuan Chubanshe. ZTN (Zhongguo Tongji Nianjian [China Statistical Yearbook]). Various Years. Beijing: Zhongguo Tongji Chubanshe.

Beijing (BJ) Tianjin (TJ) Hebei (HeB) Shanxi (SaX) Inner Mongolia (NM) Liaoning (LN) Jilin (JL) Heilongjiang (HLJ) Shanghai (SH) Jiangsu (JS) Zhejiang (ZJ) Anhui (AH) Fujian (FJ) Jiangxi (JX) Shandong (SD) Henan (HeN) Hubei (HuB) Hunan (HuN) Guangdong (GD)

27.0% 18.7% 13.2% 19.1% 23.3%

23.0% 14.8% 23.4%

17.2% 12.2% 14.9% 10.6% 12.8% 13.7% 12.8% 14.4% 12.8% 13.1% 19.5%

4,280 2,015 3,035

4,671 4,820 4,000 3,210 2,514 2,559 6,200 4,266 3,032 4,205 7,362

13.22 70.69 45.01 62.78 34.10 41.64 89.75 95.27 59.36 65.15 74.99

41.35 26.27 36.98

11.08 9.12 66.71 31.96 23.01

f34,547 f11,773 f13,461 f4,867 f11,601 f4,851 f9,555 f5,444 f7,188 f5,639 f12,885

f11,226 f6,847 f8,562

f22,460 f17,993 f7,663 f5,137 f5,872

% Population Per Female (millions) Capita GDP

2000

4,447 4,907 4,387 3,188 2,495 2,097 5,753 4,556 2,936 4,123 7,744

4,141 1,978 2,665

6,036 1,575 3,619 2,407 1,806

Lawyers

19.1% 15.0% 16.0% 11.5% 14.5% 16.0% 12.0% 12.6% 14.0% 12.0% 18.7%

25.1% 20.5% 28.1%

29.7% 21.2% 17.2% 18.7% 21.3%

13.27 70.98 45.20 63.25 34.40 42.12 90.24 96.03 59.57 65.40 75.65

41.47 26.37 37.15

11.22 9.14 67.02 32.20 23.19

f37,382 f12,922 f14,655 f5,221 f12,362 f5,221 f10,465 f5,924 f7,813 f6,054 f13,730

f12,041 f7,640 f9,349

f25,523 f20,154 f8,362 f5,460 f6,463


2001

5,117 5,833 4,791 3,760 2,793 2,001 6,581 4,897 3,195 4,478 8,997

4,330 2,032 3,079

8,130 1,715 4,180 3,089 1,748

Lawyers

19.9% 14.5% 16.5% 12.3% 13.9% 15.0% 15.3% 14.2% 14.5% 10.8% 14.7%

27.9% 21.0% 34.4%

24.6% 21.5% 18.0% 15.0% 24.1%

13.34 71.27 45.36 63.69 34.66 42.63 90.69 96.83 59.78 65.64 76.49

41.55 26.49 37.32

11.36 9.19 67.42 32.44 23.35

f40,646 f14,391 f16,838 f5,817 f13,497 f5,829 f11,645 f6,436 f8,319 f6,565 f15,030

f12,986 f8,334 f10,184

f28,449 f22,380 f9,115 f6,146 f7,241


2002

Official Data on Lawyers and General Population, China, 2000–2005.

4,414 1,483 3,717 2,343 1,756

Lawyers

Table A1.

APPENDIX


Beijing (BJ) Tianjin (TJ) Hebei (HeB) Shanxi (SaX) Inner Mongolia (NM)

89,331

Total

7,916 1,729 4,017 2,678 1,972

Lawyers

2,040 575 2,591 4,978 938 2,286 1,883 1,273 382 530 1,938

Guangxi (GX) Hainan (HaN) Chongqing (CC) Sichuan (SC) Guizhou (GZ) Yunnan (YN) Shaanxi (SaaX) Gansu (GS) Qinghai (QH) Ningxia (NX) Xinjiang (XJ)

Lawyers

47.24 7.61 30.91 84.08 36.77 40.77 35.72 25.34 4.80 5.54 18.46

1,238.19 f7,858

f4,319 f6,894 f5,157 f4,784 f2,662 f4,637 f4,549 f3,838 f5,087 f4,839 f7,470

29.3% 22.7% 22.3% 23.5% 20.6%

11.49 9.26 67.82 32.68 23.50 f32,061 f26,532 f10,513 f7,435 f8,975


2003

16.2%

11.6% 15.5% 14.0% 15.2% 11.5% 14.3% 14.6% 13.2% 14.7% 17.0% 20.1%


2000

9,355 1,816 4,466 2,283 1,941

Lawyers

90,057

1,997 589 2,745 4,652 1,000 2,345 2,170 1,202 385 576 1,493

Lawyers

Table A1.

47.58 7.70 30.98 84.37 37.10 41.07 35.90 25.51 4.83 5.63 18.76 1,245.84

f8,622

f4,668 f7,135 f5,654 f5,250 f2,895 f4,866 f5,024 f4,163 f5,735 f5,340 f7,913

29.6% 23.3% 20.6% 21.9% 25.5%

11.63 9.33 68.22 33.35 23.60

f37,058 f31,550 f12,918 f9,150 f11,305


2004

17.4%

9.3% 18.7% 15.8% 15.8% 13.9% 14.4% 15.0% 11.4% 27.5% 21.0% 21.8%


2001

(Continued).

11,373 1,856 4,615 2,943 1,996

Lawyers

102,063

2,241 626 3,038 5,337 1,130 2,411 2,566 1,301 383 595 1,689

Lawyers

47.91 7.79 31.14 84.75 37.48 41.41 36.11 25.68 4.87 5.72 19.05 1,253.97

f9,398

f5,099 f7,803 f6,347 f5,766 f3,153 f5,179 f5,523 f4,493 f6,426 f5,804 f8,382

29.5% 24.5% 21.8% 19.2% 25.3%

11.81 9.39 68.65 33.55 23.52

f45,444 f35,783 f14,782 f12,495 f16,331

Per % Population Female (millions) Capita GDP

2005

17.7%

13.6% 18.5% 22.0% 16.0% 12.6% 15.3% 15.9% 13.0% 19.8% 21.3% 22.1%


2002

374 ETHAN MICHELSON

12.2% 19.2% 15.8% 15.3% 4.2% 16.9% 32.5% 15.4% 20.1% 21.5% 21.8%

2,295 621 2,857 5,783 1,153 2,639 2,768 1,377 413 554 2,018

18.6%

19.6% 11.9% 16.5% 12.4% 13.6% 13.1% 16.9% 12.7% 8.2% 12.1% 18.5%

5,529 6,318 4,989 3,686 3,058 1,880 7,219 5,583 3,687 4,251 9,808

106,491

28.0% 22.3% 36.6%

4,585 1,988 3,120

1,262.07

48.30 7.90 31.30 85.29 37.87 41.76 36.42 25.81 4.92 5.80 19.34

13.42 71.64 45.52 64.10 34.88 43.02 91.08 97.68 60.01 65.99 77.23

41.62 26.59 37.24

f10,542

f5,969 f8,316 f7,209 f6,418 f3,603 f5,662 f6,480 f5,022 f7,277 f6,691 f9,700

f46,718 f16,809 f20,147 f6,455 f14,979 f6,678 f13,661 f7,570 f9,011 f7,554 f17,213

f14,258 f9,338 f11,615

113,289

2,505 656 3,454 5,845 1,153 3,008 2,832 1,334 411 537 2,040

6,061 6,888 5,278 3,744 3,250 1,943 7,537 5,450 3,774 4,518 11,160

4,839 2,078 3,133

19.5%

13.2% 22.6% 15.3% 15.6% 4.2% 14.3% 28.1% 14.9% 20.0% 22.5% 23.3%

21.8% 17.4% 16.6% 13.0% 13.8% 13.0% 17.0% 15.4% 13.1% 12.6% 21.8%

27.2% 21.9% 28.7%

1,272.61

48.82 8.06 31.44 85.95 38.31 42.30 36.74 25.93 4.99 5.88 19.63

13.52 72.06 45.77 64.61 35.11 43.63 91.63 98.88 60.16 66.42 78.05

41.73 26.62 37.61

f12,336

f7,196 f9,450 f9,608 f8,113 f4,215 f6,733 f7,757 f5,970 f8,606 f7,880 f11,199

f55,307 f20,705 f23,942 f7,768 f17,218 f8,189 f16,925 f9,470 f10,500 f9,117 f19,707

f16,297 f10,932 f13,897

121,713

2,598 687 3,696 6,331 1,288 3,308 2,780 1,382 438 586 2,088

7,129 7,243 5,514 3,771 3,345 1,963 8,213 5,594 3,887 4,728 12,016

5,134 2,013 3,145

19.2%

10.8% 25.0% 15.1% 14.8% 4.1% 13.6% 19.1% 16.1% 21.7% 21.0% 24.9%

24.1% 15.4% 16.8% 13.2% 13.5% 12.9% 16.0% 14.7% 13.4% 11.5% 20.1%

29.2% 23.1% 29.6%

1,280.92

48.94 8.19 31.69 86.42 38.68 42.70 37.04 26.00 5.04 5.96 19.62

13.60 72.53 46.02 65.16 35.35 43.84 92.12 100.10 60.31 66.74 79.00

41.89 26.69 37.68

f14,053

f8,788 f10,871 f10,982 f9,060 f5,052 f7,835 f9,899 f7,477 f10,045 f10,239 f13,108

f51,474 f24,560 f27,703 f8,675 f18,646 f9,440 f20,096 f11,347 f11,431 f10,426 f24,435

f18,983 f13,348 f14,434

Source: ZLN, http://chinadataonline.org. Note: Discrepancies between sum of provincial figures and totals in are due to the exclusion of Tibet (i.e., totals include data from Tibet, where available). Per capita GDP is in current Yuan.

Total

Liaoning (LN) Jilin (JL) Heilongjiang (HLJ) Shanghai (SH) Jiangsu (JS) Zhejiang (ZJ) Anhui (AH) Fujian (FJ) Jiangxi (JX) Shandong (SD) Henan (HeN) Hubei (HuB) Hunan (HuN) Guangdong (GD) Guangxi (GX) Hainan (HaN) Chongqing (CC) Sichuan (SC) Guizhou (GZ) Yunnan (YN) Shaanxi (SaaX) Gansu (GS) Qinghai (QH) Ningxia (NX) Xinjiang (XJ)


376

Table A2.

ETHAN MICHELSON

Descriptive Statistics of Variables in Multivariate Analysis, 25-City Survey, 2000.

Income (Yuan) Female Firm size (log) Partner Prior work in court Years of practice: o4 years Years of practice: 4–10 years Years of practice: 11–20 years Years of practice: W20 years % Billings/cases from corporate clients % Cases in court City-level per capita GDP (log)

Table A3.

Mean

Standard Deviation

Minimum

Maximum

N

132,256 0.242 2.853 0.201 0.086 0.299 0.488 0.201 0.012 46.696

341,768 0.428 0.664 0.401 0.280 0.458 0.500 0.401 0.107 20.276

4,000 0.000 1.099 0.000 0.000 0.000 0.000 0.000 0.000 0.000

3,600,000 1.000 4.898 1.000 1.000 1.000 1.000 1.000 1.000 100.000

915 980 966 980 980 948 948 948 948 965

37.930 9.535

19.271 0.553

0.000 8.444

100.000 10.349

980 980

Descriptive Statistics of Variables in Multivariate Analysis, 8-Province Survey, 2007. Mean

Income (Yuan) 356,721 Female 0.189 Firm size (log) 2.413 Partner 0.608 Prior work in court 0.031 Years of practice: o4 years 0.149 Years of practice: 4–10 years 0.435 Years of practice: 11–20 years 0.322 Years of practice: W20 years 0.040 Years of practice missing 0.055 % Billings/cases from 35.335 corporate clients % Cases in court 59.602 City-level/provincial per 10.549 capita GDP (log)

Standard Minimum Maximum Deviation

N

568,333 0.392 0.749 0.488 0.174 0.356 0.496 0.468 0.195 0.227 29.847

3,000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

3,580,000 1.000 5.176 1.000 1.000 1.000 1.000 1.000 1.000 1.000 100.000

998 1,293 1,323 1,301 1,337 1,337 1,337 1,337 1,337 1,337 1,337

34.587 0.471

0.000 9.373

100.000 11.052

1,337 1,337

FLEXIBLE WORK, FLEXIBLE HOUSEHOLD: LABOR MIGRATION AND RURAL FAMILIES IN CHINA C. Cindy Fan ABSTRACT The assumption that the family migrates as a unit downplays migrants’ circularity. This chapter focuses on China’s rural–urban labor migrants that travel back and forth between the sites of work and home community and between places of work. I argue that migrants and their households pursue work flexibility in order to obtain the best of the urban and rural worlds, by gaining earnings from urban work and at the same time maintaining social and economic security in the countryside. Work flexibility demands flexibility in household organization, in the form of division of labor and collaboration between genders, generations, and households. Based on a study in Sichuan, I examine household biographies and narratives to identify migrants’ work and household strategies. Migrants change jobs frequently, switch from one type of work to another and one location to another readily, and often return to the home village for months or even years before pursuing migrant work again. Not only are migrants ready to split the household between the city and the countryside, but also they frequently change from one form of division of labor to another. The inside–outside model, where the wife stays in the village and the husband does migrant work, used to be the dominant Work and Organizations in China after Thirty Years of Transition Research in the Sociology of Work, Volume 19, 377–408 Copyright r 2009 by Emerald Group Publishing Limited All rights of reproduction in any form reserved ISSN: 0277-2833/doi:10.1108/S0277-2833(2009)0000019016

377

378

C. CINDY FAN

arrangement. Over time, the outside–outside model, where both the husband and wife migrate to work and leave behind other family members, is increasingly popular. This is facilitated by intergenerational and interhousehold division of labor in the form of assistance by the extended family. Intergenerational division of labor takes place when the second generation is replacing the parents in migrant work. This research’s findings support the notion that rural–urban migrants are fast becoming a hybrid segment of Chinese society, playing dual roles of farmers and urban workers and straddling the peasant and urban worlds.

INTRODUCTION The assumption that the family – customarily considered to consist of two parents and their children – migrates as a unit underlies many studies of migration. Yet, a split-household arrangement, where the migrant worker leaves behind the rest of the family at his/her place of origin, is not uncommon. Labor migration in less developed economies, international migration, and transnational migration are often characterized by migrants’ circularity between the sites of work and their home community. Rural– urban migrants in China, likewise, persistently pursue the split-household strategy. China’s floating population, estimated to be 150 million, consists mostly of rural migrants working in urban areas (National Bureau of Statistics, 2006). Most of these migrants leave the family behind, and many change jobs frequently, switch from one type of work to another and one location to another readily, and return to the home village for months or even years before pursuing migrant work again. This chapter has two objectives. First, I seek to illustrate the ways in which rural migrants in China are flexible in their work. I also offer an explanation, based on the concept of security, for such flexibility. Second, I argue that migrants’ work flexibility demands flexibility in household organization, and I identify the forms of organization that entail division of labor between genders, generations, and households. My analysis uses household biographies and narratives from a field study of a village in Sichuan. The next section reviews selected studies on the relationship between labor migration and household organization. It is followed by a section that articulates an argument that security is the key to explaining for Chinese migrants’ work flexibility. Then, I describe the field site and the biographical materials used in the analysis. Part one of the analysis focuses on the work

Labor Migration and Rural Families in China

379

flexibility of rural Chinese, and part two examines migrant households’ organization and division of labor.

LABOR MIGRATION AND HOUSEHOLD ORGANIZATION In the literature on migration, the relationship between labor migration and household organization has generated much debate and interest. The conventional approach to analyze such relationship focuses on family migration, that is, the family moving as a unit. Because family migration often aims at improving the earnings or advancing the career of the primary wage earner, most likely men, other members of the family are viewed as tied migrants. Castro and Rogers (1983), for example, examine the migration of children with their parents and of wives with their husbands. Other studies have supported the argument that family migration has detrimental effects on the employment and earnings of trailing wives (e.g., Mincer, 1978), an argument increasingly challenged by recent findings that women also benefit from family migration (e.g., Clark & Withers, 2002). Underlying these studies is the assumption that members of the family, especially husbands and wives, stay together in one physical location, regardless of who benefit more from migration and whether they decide to migrate or not. It is also this assumption that underlies the observation that families migrate less than individuals. Odland and Ellis (1988), for example, find that potential migrants may forgo the economic benefits of migration in order to keep the household intact. At the same time, three bodies of work, focusing mostly on contexts other than internal migration in Western advanced economies, address migrations that split the household into two or more locations. Studies of rural–urban migration in less developed economies examine situations where some household members undertake migrant work and others stay behind. Temporary, circular migration that results in split households has always been common in Africa, Asia, and Latin America (Chant, 1991; Deshingkar & Start, 2003; Nelson, 1976; Wilkinson, 1987). In countries experiencing rapid urbanization and industrialization, such as Vietnam, Cambodia, and China, rural–urban circular migration is rapidly increasing (Deshingkar, 2005). Research by Stark (1991, pp. 23–31) and his associates has advanced the New Economics of Migration theory, which considers remittances as part of an implicit agreement between the migrant and the left-behind that is grounded

380

C. CINDY FAN

on a plan for the migrant’s eventual return. Contrary to the expectation that a large family deters migration, Stark and Taylor (1991) argue that having more adult members in the household facilitates labor migration because the leftbehind can assume the migrant worker’s farm duty. The second body of work focuses on international labor migration. Skinner (1976) shows that laborers from southern China that worked on US mines and railroads in the 19th century were expected to eventually return, despite their long absences from home (Pan, 1998, p. 261). Prior to the mid1990s, most Mexican immigrants to the US did not settle permanently because they did not intend to stay or were compelled to return (e.g., Reyes, 2004). Women from the Philippines, Indonesia, and Thailand leave behind husbands and children for overseas domestic work and return home only after long periods and between contracts (Silvey, 2006). The above are just some of the many examples of international labor migrants splitting the family between the host country and the sending country. The above two bodies of literature concern primarily poor segments of society, many from rural areas, that have few options other than migrant work to improve their families’ livelihood (e.g., Deshingkar & Start, 2003). The third body of literature deals also with high-skilled, highly educated migrants. Studies on transnationalism focus on sustained flows across national borders, including migrants with multiple identities and ties, and familial, social, and professional relations and networks stretched across nations and facilitated by increased ease in mobility and communication (Mitchell, 1997; Ong, 1999; Saxenian, 2005). There are overlaps between studies on international labor migration and studies on transnationalism, but the latter emphasize fluidity, flexibility, shifts in mobility, and repeated activities across national borders (Willis, Yeoh, & Fakhri, 2004). Research on transnationalism focuses on not only labor migrants seeking to alleviate poverty, but also those that champion mobility and flexibility, including transplanting the family overseas, as a formula for success. Waters’ (2002) study of ‘‘astronaut wives’’ in Vancouver, for example, examines ‘‘flexible families’’ of highly educated immigrants from Taiwan and Hong Kong. Unlike earlier examples, the astronaut syndrome is one where the wives and children are ‘‘left behind’’ on foreign soil while the breadwinner husband returns to the home region to work (Skeldon, 1997). Although the above three bodies of literature examine migrations that differ in context, scale, and composition, they share the commonality of involving split households. They challenge the conventional assumption that keeping the family physically together is the norm and is of primary importance to migrants. The prevalence of split households begs the


381

question why – why do families, and husbands and wives, tolerate being apart for long periods of time, and why do they choose this arrangement? One explanation portrays migrants in a passive light, that is, migrants want to bring their family along but are unable to do so, because they are not citizens of the host country, they have only limited access to the social benefits of the host society, they cannot afford the high cost of living there, etc. Laborers barred from entering the US due to the Chinese Exclusion Act of 1882, guest-workers from Eastern and Central Europe no longer welcome by Western European countries during the economic downturn of the 1970s, and Indonesian domestic workers at the end of their contracts are some examples where migrant workers have little control over how long they can stay, not to mention bringing their family along. The second explanation is that migrants prefer circular migration to permanently leaving the home region. Hugo (2003a, 2006) argues that migrant workers do not always desire to settle in destination countries. Modern forms of transport and communication have reduced the friction of distance and allowed migrants to maintain closer and more intimate linkages with their home communities than before. By keeping some family members at the origin, migrants can maintain valued traditions and family ties via frequent visits. In contrast to the first explanation, this explanation emphasizes migrants’ agency and strategies. When a family is split into two locations, its members can tap into resources and opportunities in both locations. Migrants can obtain the best of both worlds by earning in highincome destinations and spending in low-cost origins. For rural–urban migrants, the city offers employment and higher wages and the countryside has farmland and is inexpensive. For the astronaut families, Asia offers attractive economic opportunities to breadwinners, whereas Canada and US afford quality education for their children. In reality, decisions on splitting the household to accommodate labor migration are likely based on a combination of the above two explanations. Likewise, rural–urban migrants in China can be viewed in a passive light, but it is important to also recognize the strategies they pursue to benefit from both the city and the countryside.

CHINA’S RURAL–URBAN LABOR MIGRANTS Split households are prevalent in rural China. The typical model is one where one or more adult members of the family undertake migrant work, send home remittances, and return home infrequently, while other members

382

C. CINDY FAN

stay in the village to farm and fulfill care-giving responsibilities. Although there is some evidence of increase in family migration (Zhou, 2004), split households remain the most popular from of organization among rural– urban labor migrants. In other words, most rural–urban labor migrants are physically separated from their families for extended periods of time. What explains split households in rural China? Despite the proliferation of research on migration in China, relatively little attention has been paid to migrants’ household organization. The dominant view of rural migrants is that they are passive, helpless victims. In particular, a large body of work highlights the hukou (household registration) system as a source of migrants’ plight. Hukou is a socialist institution that bifurcates China into urban and rural segments and blocks rural citizens from the rights and benefits reserved for their urban counterparts. In cities, rural migrants do not have access to the full range of jobs and services, and are exploited, segregated, and discriminated against. The impacts of hukou on migrants are a subject of many studies (e.g., Alexander & Chan, 2004; Chan & Buckingham, 2008; Fan, 2002, 2008; Solinger, 1999; Wang, 2005) and are not repeated here, but a popular view is that hukou prohibits migrants to stay in the city permanently and bring their families along, forcing them to split the household into two places. There is no question that hukou disadvantages rural migrants and prevents them from becoming full-fledged urban citizens. Yet, this explanation alone downplays the desire, agency, and strategies of migrants and their households. Below, I articulate another explanation, one that centers on the concept of security and a strategy of flexibility. Security is related to protection, safety, continuity, reliability, and a sense of future and permanence. After more than two decades of massive rural– urban labor migration, the countryside continues to be the basis of economic security for China’s rural migrants and their families. The countryside is the site of three types of economic security. First, peasants have access to farmland contracted from village authorities. Agriculture is a source of subsistence and serves as an insurance against adversity – for example, when migrant work fails – and a security for migrants’ future return. In addition, rural Chinese are not free to buy or sell farmland, and conversion of farmland into other purposes is strictly controlled by government authorities (Lin & Ho, 2005; Yusuf & Nabeshima, 2006, pp. 57–58).1 Although the Chinese government has recently signaled a willingness to allow farmers to lease or trade their land-use rights, they still do not have land-ownership rights (Lim, 2008). Thus, migrants are motivated to leave behind family members to farm, ask relatives to farm for them, or lease out the land to other villagers. Such arrangements not only


383

yield output, but also guard the farmland’s boundaries (against infringement by others) and prevent it from being fallow. Second, Chinese peasants are allocated non-farm land where they can build houses. To many rural Chinese, building a new house or renovating or expanding an existing house is of great importance. For example, it is expected that in order to find a wife, a man must have a house or a partition of his parents’ house. Only when a new house or an extension to an existing house is complete can a married son and his wife and children have their own space independent from the parents’. Indeed, housing-related expenses are a main reason for migrant work. Third, the cost of living in the countryside is low. Affordability to live in a place permanently is an important dimension of economic security. By contrast, the city does not offer rural migrants economic security. Migrant jobs are neither stable nor secure and most demand youth and manual labor. The global financial crisis since 2008 that led to the closing of many manufacturers and rendered millions of migrants unemployed, is a vivid reminder of the precarious nature of migrant work. As migrant workers age, their chance of being hired deteriorates and they must contemplate returning to the countryside. In addition, cities are expensive. Unless migrants have considerable savings and access to health care, education, housing, and other benefits, cities are less desirable places to live than the countryside. The countryside is also the basis of peasant migrants’ social security, namely a support system made up of the spouse, children, parents, siblings, relatives, and fellow villagers. In rural China, the age-old concept of jia (family or home) – which refers to not only the nuclear family, but also the extended family and even the home village – remains strong. Members of a family are related to each other by blood or marriage, and their budgets, properties, and interests are interconnected (Croll, 1994, p. 163; Woon, 1994). The institution of marriage is fundamental to the social structure and is keenly protected. Within marriage, traditional gender norms are determined by the age-old inside–outside (IO) ideology, which defines the woman’s place to be inside the family and the man’s sphere to be outside (nan zhu wai nu zhu nei) (Entwisle & Henderson, 2000, p. 298; Hershatter, 2000; Mann, 2000). The boundary between the feminine inside and the masculine outside is not fixed and has, indeed, shifted such that the inside now includes all village responsibilities, and the outside refers to migrant work (Jacka, 1997). When migrants talk about jia or huijia (return home), they refer to not only the physical home, but also the village and the home community. When they talk about chuqu (go out), they refer to leaving the

384

C. CINDY FAN

home community to seek work elsewhere. Thus, because of migrant work, the IO model is repackaged as gender, spatial division of labor – the husband is more likely than the wife to pursue migrant work, and the wife is more likely than the husband to stay behind in the village. Children and the elderly also constitute important elements of the rural support system. Children are still the main source of old-age support, and fulfilling obligations to parents is an important means to keep the traditional social support system intact. Siblings and relatives that live nearby, likewise, can help in farming or care-giving when a need arises. By contrast, in the city, peasant migrants are seen and treated as outside labor rather than members of the urban society (Chan, 1996; Fan, 2002; Solinger, 1995; Zhou, 1992). Their social interactions with urban locals are minimal, and they rely mainly on fellow villagers from the same native place (tongxiang or laoxiang) for support. Although there is some evidence of thriving migrant communities, such as the Zhejiang Village in Beijing (Ma & Xiang, 1998; Xiang, 2005), in general peasant migrants’ social support system in the city is much weaker than that in the countryside. In summary, the very consideration of security is an important explanation for peasant migrants’ anticipation to eventually return to the countryside. The terms dagong – literally ‘‘being employed,’’ but more specifically peasants seeking work in industrial and service sectors – and chuqu, for example, highlight migrants’ crossing of the rural–urban border but also connote that the countryside is still their home. For example, some studies have found that migrants’ desire to settle in cities is not strong and the majority wants to eventually return to the countryside (Wang, 2003; Zhu, 2003, 2007). During their dagong tenure – which for some has been over 20 years – migrant workers straddle the urban and rural, seek to benefit from both, and ‘‘earn in the city and spend in the village’’ (Fan & Wang, 2008; Hugo, 2003b). Migrants’ straddling the urban and rural also explains their strategies for work, strategies that center on flexibility. Using household biographies and narratives, the rest of the chapter illustrates how and why rural migrants pursue work flexibility and highlights various forms of household division of labor that support work flexibility.

HOUSEHOLD BIOGRAPHIES AND NARRATIVES Qualitative materials based on small samples tend not to be as representative as macro-level quantitative data, but the former are especially useful in


385

revealing complexity, details, and subtle processes such as family politics. In this study, I examine both household biographies and villagers’ narratives to highlight the strategies of rural–urban labor migrants and their households. Jarvis (1999) has summarized succinctly the advantages of household biographies in social sciences research: ‘‘[Household] biographies can be used to reach beyond revealed or material action to the negotiation of action and thus non-action; that which is taken-for-granted, ruled out, or modified in the process of blending individual narratives within household narratives.’’ Although Jarvis’ research is on urban households, her argument about qualitative life-history is equally applicable to rural households: ‘‘In effect, a biographical representation of household behaviour comprises not only the interweaving of parallel histories (work histories, family mile-stone events, personal relationship histories) but also the negotiation of the interlinkage and temporal ordering of such events y.’’ McHugh (2000) argues eloquently that the conceptualization of migration as a one-way journey is inadequate. Rather, he calls attention to the value of ethnographic research for understanding ‘‘migrations as cultural events rich in meaning for individuals, families, social groups, communities and nations.’’ In addition, personal stories and narratives are powerful means for identifying migrants’ agency, negotiation, and conflicts, and they enable a bottom-up research approach that foregrounds the voices and experiences of marginalized individuals in society (Jacka, 2006, p. 10; Nagar, Lawson, McDowell, & Hanson, 2002). I draw from two related projects, conducted during the Spring Festival in 1995 and 2005. In 1995, the Research Center for Rural Economy of the Ministry of Agriculture conducted in-depth interviews with 300 households from 12 villages – 3 villages each from 2 counties in Sichuan and 2 counties in Anhui. Sichuan and Anhui are major sources of rural–urban labor migrants in China. The counties and villages for the 1995 project were selected based on the following criteria: that in terms of economic development they were representative of the respective provinces; that they had been sending out migrant workers for quite some time; and that migrant workers accounted for at least, respectively, 20% of the county’s labor force and 30% of the village’s labor force (Du, 2000; Du & Bai, 1997, p. 5). In each of the villages, 15 migrant households (where one or more members had had migrant work experience) and 10 non-migrant households were randomly selected. Interviewees’ responses are in the form of narratives and are transcribed verbatim. The 2005 project, which I participated in and which was administered by Renmin University of China, sought to reinterview the same 300 households. Each of the accounts is again transcribed verbatim.

386

C. CINDY FAN

In this chapter, I focus on one village – Village A – selected randomly from the original 12 villages. In 2005, Village A had about 310 households and a total population of approximately 1,200. For the sake of confidentiality, in this chapter, I do not disclose Village A’s location and I use pseudonyms for villagers (see Table 1). Suffice it to say, Village A has many features in common with villages across China that send out migrant workers. First, agricultural activities center mainly on farming and animal husbandry, but labor surplus is persistently large because arable land is of short supply. Second, although some non-agricultural economic activities exist in and near the village, the employment and income they generate are limited. A small brick factory, built in about 1997, for example, hires villagers mostly on a part-time basis. Third, by the mid-1990s, labor outmigration had already become an important source of income for most households in the village, and this trend had further intensified by 2005. Fourth, over time the number and range of labor migrants have increased. In the 1980s and early 1990s, significantly more men than women were engaged in migrant work, largely seen at that time as a new, short-term, economic opportunity. By 2005, migrant work had become an established way of life, and participation in this activity was more extensive, involving more women and almost every household in the village. Finally, despite more than 20 years of migrant work history, very few households had moved out from the village altogether. Officials in Village A estimated that by 2005 only 15 households, or 5% of the village and involving a total of approximately 60 people, had done so. Among them, only four households had moved their hukou elsewhere – thus giving up their contract land in the village. Eleven households – of which eight remained in the same county – continued to keep their village hukou. Of the 25 households in Village A originally interviewed in 1995, 16 were successfully re-interviewed in 2005 (Table 1). Our priority was to interview migrants that returned home for the Spring Festival, but if they were not available then we would interview other family members, most likely migrants’ parents, that were familiar with the household biography and migrants’ work history. In the following, I do not distinguish households originally categorized in 1995 as migrant households from those categorized as non-migrant households, because most of the latter had had some degree of participation in labor migration by 2005. A limitation of the study is that it does not include households where no members stayed in the village or returned during the 2005 Spring Festival. Since only 5% of the village’s households had moved out, that limitation is not expected to affect this study’s findings in significant ways.

IO OO OO OO IO IO OI

II

IO II OO II

IO IO IO OO IO IO IO

IO

IO II II II

OO OO II OI

x3 x4 x5 x6 x7 x8 x9

x10

x11 x12 x13 x14

x15 x16

Pan Yuefang Wei Yufang

Tan Meirong Jiang Mingfang Jiang Xiaobi Li Chenxia

Tan Sumei

Deng Yuling Zhang Shuzheng Wang Mingli Fang Biying Pan Yueqin Hu Wenfen Wang Huilan

Zhao Xiaolan Hu Weiling

Wife

42 38

47 48 33 41

47

37 31 32 30s 38 37 40s

42 40s

Age (Years)

Wang Yuanjian Jiang Fucheng

Jiang Zhengyi Wang Jiankun Wang Yonghua Chen Guowei

Wang Xueqiang

Wang Xinmin Wang Cheng Yu Zhongliang Jiang Guanghua Wang Gen Wei Daming Li Wangping

Wang Guohui Wang Qingping

Husband

43 46

53 49 33 47

53

38 33 34 38 40 39 46

Child x1

I M (non-dagong)

M (non-dagong) M (non-dagong) I O

I I I I I I M (boarding school) I

53 40s O

Age (Years)

19 22

24 22 11 21

25

15 9 11 14 17 12 20

21 18

Age (years)

I

O I I O

12 14

I I I M (boarding school) M (school)

16

22 19 2 19

22

19

13 4

14 14

I

I I

I I I I I I

I

68

73 62

60s 50s 50s 60s 60s 60s

60s

I

I

I I I I I

I

67

50

50s 50s 60s 60s 60s

60s

Age Parent Age Parent Age (years) x1 (years) x2 (years)

I I

I I

Child x2

2004

Inventory of the 16 Sampled Households in Village A. Other Outside

Grandchild, infant

Grandchild, age 1 year

Daughter-in-law, age 20s

Son-in-law, age 20s

Brother Brother’s two and his wife, children, age 12 and 13 years age 40s

Other Inside

Note: Age refers to the age in 2005; I: inside; O: outside; II: wife inside–husband inside; IO: wife inside–husband outside; OI: wife outside–husband inside; OO: wife outside–husband outside; M: miscellaneous.

II OO

OI IO

x1 x2

Household 1994 2004

Table 1.

388

C. CINDY FAN

Table 1 is an inventory of the 16 households that were re-interviewed in 2005. I consider the wife and husband as the core of a household, and include also their unmarried children, married children if living together, parents who live or spend considerable time in the house, and siblings and their families that live in the house. I selected the pseudo names of the wife and husband to be consistent with their real surnames. For example, all the Wangs have the same real surnames. The village is clearly dominated by one lineage – nine of the husbands and two of the wives are Wangs – as is the case in many Chinese villages. In Table 1, I use the IO framework to describe household organization. Inside (I) refers to living in the home village, including working in nearby towns; outside (O) refers to doing migrant work and not living in the village for a significant part of the year. For wife–husband division of labor, the first letter refers to the wife and the second refers to the husband. Thus, inside–inside (II) represents wife inside–husband inside; inside–outside (IO) wife inside–husband outside; outside–inside (OI) wife outside–husband inside; and outside–outside (OO) wife outside–husband outside. For the sake of simplicity, this table indicates for wife–husband divisions of labor for only 1994 and 2004, respectively, the years prior to the two Spring Festival interviews. As the case studies in the next section will show, two points in time are not adequate to show all the changes, but they are used here to describe overall trends. For other family members, Table 1 includes only their inside/outside status for the year 2004. In addition to summarizing demographic and household information, the table is an overall reference for the case studies to be discussed in the next two sections.

FLEXIBLE WORK China’s rural migrants are flexible workers. Their flexibility allows them to tap into urban and rural economic opportunities and to maintain the economic and social bases in the countryside. Below, I identify three features of their work flexibility – frequent changes in work location and activity, changing work location for wage, and switching between agricultural and non-agricultural work – and illustrate them via household biographies and narratives.

Frequent Changes in Work Location and Activity Wang Cheng was 17 years old when he first left home for work (Table 2). In 1989, he decided to pursue migrant work because the family, having

17

21

23 24

25

28

29

33

1989

1993

1995 1996

1997

2000

2001

Plan for 2005

Guangdong (Guangzhou) Same

Town near home

Home Guangdong (Guangzhou) Home; town near home

Guangdong (Guangzhou)

Xinjiang

Husband’s Location of Work

Same

Brick factory; passenger transportation Passenger transportation Construction

Construction; miscellaneous jobs Unknown Construction

Construction; coalmines

Husband’s Work

L to D

D to L

M to N

N to M

D to D L to D

L to D

Husband’s Change of Work Location

M to N N to M

N to M

Husband’s Change of Migrant Status

W

W (twice)

W W

W (twice)

W (twice)

Husband’s Change of Work Type

M to N

N to M

M to N

N to M

Wife’s Change of Migrant Status

OO to IO

II to OO

OO to II

IO to II II to OO

II to IO

Change of Division of Labor

Biography of Household x4 (Zhang Shuzheng and Wang Cheng).

Adoption of daughter

Birth of son

Completed house project Got married

Other

Note: Change of migrant status – N (non-migrant); M (migrant). Change of work location – L (local); D (non-local). Change of work type – W (change). Change of division of labor (see also Table 1) – II (wife inside–husband inside); IO (wife inside–husband outside); OI (wife outside–husband inside); OO (wife outside–husband outside).

Age (Years)

Year

Table 2.

Labor Migration and Rural Families in China 389

390

C. CINDY FAN

completed a house project, was in debt. Being the oldest son, with two younger siblings still in school, Wang Cheng was expected to shoulder the financial responsibility for the family. Wang Cheng’s first destination of migrant work was Xinjiang, where he was hired by a construction team. When the team lost its contract a month later, he stayed in Xinjiang to look for jobs. Between 1989 and 1993, he worked in several coalmines. In 1993, Wang Cheng returned home to get married. After staying home for two months, he left for Guangzhou, Guangdong, while his newly wedded wife, Zhang Shuzheng, stayed home. In Guangzhou, Wang Cheng first found construction work through a laoxiang and later held a variety of miscellaneous jobs. In 1995, Zhang Shuzheng gave birth to a son, and Wang Cheng returned home and stayed for a year. His narratives are not clear on what work he did during that year, but in 1996, both Wang Cheng and Zhang Shuzheng left for construction work in Guangzhou, leaving the infant behind to be cared for by Wang’s parents. A year later, the couple decided to return home. At home, Wang Cheng first worked at the newly built brick factory, but he was soon involved in a legal battle about the factory and was forced to quit the job. He decided to buy his father’s passenger tricycle and worked on passenger transport in a nearby town. In 2000, Wang Cheng and Zhang Shuzheng adopted a baby girl. In 2001, the passenger transport business was increasingly unprofitable, so after four years of staying home, the couple left for Guangzhou to find construction work again, leaving behind both their children – a six-year-old and an infant – and farmland to Wang’s parents. The couple’s goal was to earn and save as much as possible in order to fund a house renovation project. To reduce expenses, for three consecutive years they had forgone returning home for the Spring Festival. At the time of the 2005 interview, the plan was for Zhang Shuzheng to return home later that year in order to relieve the burden on Wang’s parents (see also the section ‘‘Outside–Outside’’). In summary, during a period of 16 years, Wang Cheng had left home for migrant work three times (1989, 1996, and 2001) and returned home twice (1995 and 1997) for extended periods of time, not counting short visits during Spring Festivals and the two months in 1993 when he returned to get married. Each of the two extended returns coincided with the addition of children. Between 1989 and 2005, Wang Cheng was outside for about 11 years and home for about 5 years. He had worked in construction, coalmines, brick factory, and passenger transportation. All told, he had changed work location five times and work type nine times. Other than towns near home, he had


391

worked in two provinces other than Sichuan – Xinjiang and Guangdong – both distant from home. Li Wangping was 24 years old when he first left home for migrant work (Table 3). In 1983, he worked in a brick factory in Hunan. One year later, he returned home to get married and moved into his wife’s (Wang Huilan) village. This practice is much less common than the patrilocal tradition where the wife moves to the husband’s village. Nevertheless, when the wife’s village has more resources, such as arable land, than the husband’s, a matrilocal arrangement may be pursued. Li Wangping explained: ‘‘My home village was poorer than my wife’s village. y Our economic situation was very bad.’’ After getting married, the wife’s village became his home. Another household in the sample (x5) also has a matrilocal arrangement (Table 1). After farming for more than two years and the birth of two daughters, in 1987 Li Wangping decided to go out (chuqu), because ‘‘we had too little farmland. At home I had too much idle time and made too little money. So I wanted to go out again.’’ His wife stayed home to farm and care for their two daughters. This time, Li Wangping worked in a coalmine in Hebei. He made 300 yuan a month, an income much higher than his previous work in Hunan, but he found mining work too demanding and dangerous. After five months, he quit and returned home. From 1988 to 1991, Li Wangping stayed home to farm and build a house. In 1991, through a laoxiang’s connection, he found construction work in Zhangjiagang, Jiangsu. He worked there for two-and-a-half years, and in 1994 he concluded that ‘‘the wages were lower than other places. I wanted to try other places.’’ Again, with the help of a laoxiang, Li Wangping found work at a petrochemical factory in Guangdong. A year later, in 1995, Li Wangping returned home, because he and his wife had decided to lease farmland from fellow villagers who had left for migrant work. Given the prevalence of migrant work, it is not uncommon for rural families to lease out farmland to others.2 After about a year (the narratives are not clear about the exact year) – in 1996 or 1997 – Li Wangping went out again to Zhangjiagang and did construction work there until 2004. The reason for his working in Zhangjiagang for more than seven years, instead of trying other places, was ‘‘my boss was nice, and the job was stable.’’ During the period when Li Wangping was doing migrant work, every year he returned home before the Spring Festival and helped with planting before leaving again. In 2004, he returned home to farm and for the first time his wife Wang Huilan went out. She worked in the garment factory in Zhejiang that her sister had been working in. This ‘‘experiment’’ turned out to be

24 25

26

27

28 29

32

35

36

37

1983 1984

1985

1986

1987 1988

1991

1994

1995

1996 or 1997 2004 Plan for 2005

45 46

Age (Years)

Year

Jiangsu (Zhangjiagang) Home Jiangsu (Zhangjiagang)

Home

Jiangsu (Zhangjiagang) Guangdong

Hebei Home

Same

Same

Hunan Home

Husband’s Location of Work

Table 3. Husband’s Change of Migrant Status

N to M

N to M M to N

Farming Construction

Construction M to N N to M

N to M

Petrochemical factory Farming M to N

Construction

Coalmine Farming

Same

Same

Brick factory N to M Farming M to N

Husband’s Work

D to L L to D

D to D

D to L

D to D

L to D

L to D D to L

L to D D to L


W W

W

W

W

W

W W

W W

Husband’s Change of Work Type

N to M

Wife’s Change of Migrant Status

II to OI OI to OO

IO to II

II to IO

II to IO IO to II

Beginning of II

Change of Division of Labor

Biography of Household x9 (Wang Huilan and Li Wangping).

Leasing out farmland

Leasing farmland from others

House building project

Birth of daughter Birth of daughter

Got married

Other

392 C. CINDY FAN


393

successful, and thus the couple’s plan for 2005 was to both go out but to different places: the husband to Zhangjiagang and the wife to Zhejiang. Unlike the previous case (Wang Cheng of Household x4), Li Wangping’s daughters were older, aged, respectively, 20 and 19 years. Li’s plan was to send them to boarding school. And, since Li Wangping’s and Wang Huilan’s parents had passed away, no one was available to help with farming. They decided, therefore, to lease out their farmland to other villagers and keep their house vacant. Between 1983 and 2005, Li Wangping had left for migrant work 5 times (1983, 1987, 1991, 1996, and 2005), returned home for extended periods 4 times (1984, 1988, 1995, and 2004), changed the location of work 10 times, and changed the type of work 10 times. In addition to working at home, he had worked in four different provinces (Hunan, Hebei, Jiangsu, and Guangdong). In both of the above case studies, the husband’s work location and activity changed frequently, underscoring the following factors that determine the mobility and work of migrants. First, migrants’ returning home for an extended period of time tends to coincide with major life-events such as birth of child, household events such as a house project, and new economic opportunities such as leasing farmland from others (see also Wang & Fan, 2006). This shows that the social and economic activities that physically take place in the countryside may disrupt migrants’ work and explain for long breaks in an otherwise continuous migrant work tenure. Farming is still important. To Li Wangping, for example, farming (the land leased from others) explained one of his extended returns. Second, after each extended return, both husbands decided to go out again. This suggests that despite migrants’ returns and commitment to rural activities, migrant work is still considered a more attractive source of income than those in the countryside, and is something rural Chinese would resort to especially when village-based activities are not successful, as in the case of Wang Cheng. Third, wages, ease of finding jobs, and laoxiang connections appear to be much more important determinants than long-term considerations such as seniority, skill acquisition, and career development, of the location and type of migrant work. Thus, migrants switch from one type of work – be it construction, mining, or manufacturing – to another readily, and they are footloose in terms of work location. In short, rural migrants are highly mobile and flexible. Fourth, the jobs these migrants are channeled to require low skills, accounting in part for the relative ease with which they switch from one type of work to another. Finally, migrants’ moves between jobs and between work locations can be frequent or may not occur for years.

394

C. CINDY FAN

These moves may be seasonal or not seasonal. Returns to help with planting and harvesting and for the Spring Festival are seasonal, but many other moves do not follow a seasonal pattern. Such fluidity is in part due to the labor market segment in which migrants concentrate, where job stability is minimal and job turnover is high.

Changing Work Location for Wage Migrants’ changes in work location highlight the important role of wage. Unlike most interviewees in the survey who gave only scanty information about income, Wang Xinmin provided a detailed longitudinal account of how his wage changed over time (Table 4). Between the ages of 16 and 18 years, Wang Xinmin made about 2 yuan per day from full-time farming. When he was 18 years old, in 1985, his father sent him to learn brickwork. So, Wang Xinmin became an apprentice in a county near home, earning about 2 yuan per day. From that point on, everywhere he went he did brickwork. In 1987, he left for Xinjiang, and his wages increased to 5 yuan per day. In 1988, Wang Xinmin got married and in 1990 he had a son, but other than returning for the Spring Festival and harvesting he continued to work in Xinjiang. In 1991, Wang Xinmin went to Heilongjiang. Although his wages increased to 17 yuan per day, the cold weather there confined work to only the summer months. His overall income was therefore similar to before. Thus, he decided to return to a county near home. In 1992, he left for Zhangjiagang in Jiangsu and continued to work there ever since. His wages were, on a per-day basis: 20 yuan in 1993, 25 yuan in 1994, 30 yuan in 1995, 35 yuan in 1996–1997, 40 yuan in 1998–1999, 45 yuan in 2000, 50 yuan in 2003, and 55 yuan in 2004. During the Spring Festival in 2005, Wang Xinmin returned home and worked on miscellaneous jobs. He explained why he planned to go to Zhangjiagang again after the Spring Festival: The wage at home is only 30 yuan per day, compared to 55 yuan in Zhangjiagang. And, at home, there are very few [non-farm] jobs. y In Zhangjiagang, the weather is nice [compared to Heilongjiang]; we have work throughout the year. More than 10,000 migrants from this county work in Zhangjiagang. With that many laoxiang, it’s easy to find jobs.

Wang Xinmin differed from the previous two cases in that he was trained in a trade and therefore did not switch from one line of work to another. Similar to the first two cases, however, Wang Xinmin tried various locations, as far as Xinjiang in China’s northwest and Heilongjiang in the northeast,

16 18

20 21 23 24

25

37

1987 1988 1990 1991

1992

2004

Wanyuan County (near home); Jiangsu (Zhangjiagang) Same

Xinjiang Same Same Heilongjiang (Daqing)

Home County near home

Age Husband’s (Years) Location of Work

1983 1985

Year

Table 4. Husband’s Change of Migrant Status

Same

Same

Farming Brickwork N to M (apprentice) Brickwork Same Same Same

Husband’s Work

D to L; L to D

D to D

L to D

L to L


W IO

Change of Wife’s Change of Work Type Change of Division of Migrant Labor Status

Other

2 per day 2 per day

Wage (Yuan)

55 per day

5 per day Got married Birth of son Fenjia (see the 17 per day section ‘‘Extended family’’); house renovation Birth of 15 per day daughter

Biography of Household x3 (Deng Yuling and Wang Xinmin).

Labor Migration and Rural Families in China 395

396

C. CINDY FAN

before settling on Zhangjiagang. His experience again supports the observation that wage, ease of finding jobs, and a migrant community are important determinants of the location of migrant work, and that migrants’ flexibility allows them to choose locations that offer attractive wages.

Switching between Agricultural and Non-Agricultural Work Not only are migrants flexible, but villagers who spend considerable time in the countryside are also flexible in terms of location, type, and schedule of work. The following two cases illustrate two different models of combining agricultural and non-agricultural work. Jiang Zhengyi’s (Household x11) experience may not be typical but it shows that urban work is not necessarily the primary income source for a migrant worker. Aged 53 years in 2005, he had done migrant work since 1991, mostly in construction in Fujian and Guizhou. His wife, Tan Meirong, stayed home to farm and care for their two sons. Different from the husbands in the previous case studies, Jiang Zhengyi spent considerable time farming in the home village. Every year, he left home in March or April after planting, and returned home in July for harvest and stayed until the next spring. In other words, he did migrant work for only four to five months a year. The rest of the year, he helped his wife to work on not only their farmland, but also the farmland leased from three families. As a result, farming constituted the leading source of income for Jiang Zhengyi and his wife (excluding the income from their sons), followed by animal husbandry and Jiang’s migrant work. Their diversification of income sources was enabled by the husband’s seasonal migration between home and outside and seasonal switch between agriculture and non-agricultural, migrant work. Among the left-behind, it is not difficult to find examples of combining and switching between agricultural and non-agricultural work. In Household x3, discussed earlier, Deng Yuling stayed in the village all along, while her husband Wang Xinmin did migrant work. Deng’s main responsibilities were to farm and care for their two children. Her agricultural work consisted of farming 2.1 mu (1 mu ¼ 1/15 ha or 1/6 acre) of paddy field and 0.4 mu of ‘‘dry’’ field for wheat and sweet potatoes, as well as raising pigs. At the same time, she worked every other day, from 7 to 11 in the morning and 3 to 5 in the afternoon, in the local brick factory. This schedule allowed her to earn wages from non-agricultural work and meet her agricultural and care-giving duties, simultaneously and without leaving the village. Different

397


from Jiang Zhengyi in Household x11, Deng Yuling’s switch between agricultural and non-agricultural work was not seasonal but daily. The above two cases show that not only are rural Chinese flexible about their location and type of work, but they are also flexible about the tenure and temporal arrangement of the work. They can switch back and forth between different types of work seasonally or even daily. Such flexibility highlights the objectives to diversify income sources, augment income, and balance various household responsibilities. It also reinforces the notion that farming is still an important activity for migrant households, one that may entail migrants’ extended returns.

FLEXIBLE HOUSEHOLD Migrants’ flexibility in the location, type, and tenure of work is possible because of their economic and social bases in the countryside, namely the house and farmland on one hand and the spouse, parents, and other family members on the other. Work flexibility also requires that household organization be flexible. Using the division-of-labor framework, this section focuses on migrants’ household organization. A split-household strategy is the key to understanding how rural households support migrants’ work flexibility. Considering the wife and husband as the core of the rural household, Table 5 summarizes how the wife–husband division of labor had changed between 1994 and 2004. Given the fluidity of migrant work, as described in the last section, examination of two specific points in time may miss important changes in between and before the first point in time. Nevertheless, Table 5 highlights common Table 5.

Changes in Gender Division of Labor.

1994

2004 II

IO

OI

OO

Sum

II IO OI OO

2 1 1 0

0 4 0 0

1 1 0 0

1 3 0 2

4 9 1 2

Sum

4

4

2

6

16

Note: See Table 1.

398

C. CINDY FAN

forms of household organization and the overall trends, while the case studies that follow reveal more detailed changes. In 1994, the most prevalent household arrangement was the traditional IO gender division of labor, which described 9 of the 16 households (Households x2, x3, x4, x5, x7, x8, x9, x10, and x11; see also Table 1). Indeed, during the 1980s and 1990s, the most common form of split households was one where the husband does migrant work and the wife stays behind to farm and raise children (Fan, 2004, 2008, pp. 89–92; Tan, 1996). In 1994, the OO model accounted for only 2 of the 16 households (x6 and x15). By 2004, however, OO had become the most popular arrangement, accounting for 6 households (x2, x4, x5, x6, x13, and x15). In addition, both of the OI households in 2004 planned to switch to OO in the following year (x9 and x16). By contrast, the number of households with an IO arrangement decreased from 9 in 1994 to 4 in 2004. Inside–Outside (IO) The dominance of the IO arrangement in 1994 underscores the prominence of traditional gender norms in the Chinese rural household. Below, narratives selected from three households illustrate the expected gender roles, especially those of women. Wang Gen (Household x7) had done migrant work since 1983 and continued doing so after getting married in 1987. His wife, Pan Yueqin, stayed home to farm and care for their two children. She explained the division of labor they chose: Men should make greater [economic] contribution to the family than women. Just like our family, the husband goes out to make money [and the wife stays home].

Her comment underscores the persistence of gender norms and the expectation for women to play care-giving roles. Likewise, in Household x8, Wei Daming had done migrant work for more than 20 years, while his wife, Hu Wenfen, a teacher at a local nursery school, stayed in the village to take care of their son. Wei Daming explained why his wife did not pursue migrant work: If my wife did migrant work she could probably earn quite a bit, but I do not want her to go out. Children in the countryside must study hard or else their future [livelihood] is bleak.

His comment suggests that not only are wives expected to provide caregiving to children, but a mother’s constant presence is considered very


399

important for children’s education, a reason the IO arrangement persisted in this household. Children’s education is also an important reason why migrants do not transplant their households to the city. Wang Xinmin (Household x3), also discussed earlier, had done migrant work for more than 20 years, while his wife shouldered all village responsibilities, including taking care of two children. Wang Xinmin returned home at least once a year, during the Spring Festival. And, in 1997, he installed a telephone set, the first in the village. By calling home once a week, he was able to maintain close contact with his wife and children. Once a month or every other month, he would send money home. Wang Xinmin explained why he and his wife chose a long-term split-household arrangement, rather than moving the family to the city (Zhangjiagang) altogether: I cannot take the children with me. School is too expensive in Zhangjiagang; the fee is twice what we pay at home. Expenses for the children at home are about 2,000 yuan a year but would be 5,000 yuan over there. y I live at the construction site. If my wife and children go there, we will have to rent a place y

The above three cases all underscore the persistence of traditional gender norms – the wife taking care of the inside and the husband bringing home income from the outside. Through the split-household arrangement, these households can take advantage of higher wages in the city and low cost of living in the village and ensure that the children would be well taken care of (by the wives) and would receive education.

Outside–Inside (OI) Households where the wife does migrant work and the husband stays in the village are not common. When an OI arrangement is selected, it is usually short-lived. For example, in Household x9 (discussed earlier and in Table 3), after one year of staying home while his wife did migrant work (in 2004), Li Wangping planned to leave for migrant work as well in the following year, thus changing the household arrangement from OI to OO. Household x1 illustrates another failed attempt of OO arrangements. In 1994 and 1995, Zhao Xiaolan worked in a sewing machine factory in Jiangsu while her husband Wang Guohui stayed home to farm and take care of their two children. Even though Zhao Xiaolan was proud of her economic achievements, she was pressured by her husband to return home. When interviewed in 1995, they were fighting about this matter. In 1996,

400

C. CINDY FAN

Zhao Xiaolan decided to return to the village for good. Her decision reflected the high priority given to preserving traditional gender norms, despite the lost economic opportunities by not pursuing migrant work. Often, topics such as gender norms are discussed in connection to the children. Zhao Xiaolan explained why she returned: The main reason for my return was I was unable to go out [due to family responsibilities]. Both my daughters were at school age, and my younger daughter in particular needed care [because of her young age]. y In terms of raising children, migrant households – especially those where both the husband and wife are outside – are usually unable to supervise the children’s schoolwork.

By electing to stay in the village, Zhao Xiaolan fulfilled the traditional norm that mothers raise and educate children and managed to safeguard the husband–wife relations. These social accomplishments were made at the expense of economic opportunities for her and standard of living for the family. Nonetheless, the couple was flexible in their work: y a brick factory was built nearby y we began part-time work there. This way, we live at home and there are no additional living expenses [compared to dagong]. Beginning in 2001, we also leased land from relatives to farm.

Like several case studies discussed earlier, this household engaged in both agricultural and non-agricultural work locally, and they took advantage of non-farm economic opportunities in the village. The husband and wife’s flexibility in the type and schedule of work, therefore, helped to mitigate the opportunity cost due to their unwillingness to pursue a split-household arrangement. For both Households x1 and x9, the brevity and failure of the OI arrangement underscores the depth and persistence of gender norms.

Outside–Outside (OO) When both the husband and wife leave home for migrant work, the most common strategy is to recruit their parents’ help to farm and take care of the left-behind children. Among the six OO couples in 2004, all had parents living in the same household. In households that previously had an IO arrangement, the departure of the wife is made possible by the migrants’ parents replacing her as care-givers, thus changing the household arrangement from gender division of labor to intergenerational division of labor. In Household x5, Wang Mingli’s husband, Yu Zhongliang, had for years been doing migrant work in Fujian. In 1996, Wang Mingli left home to join her husband, leaving their two-year-old son to her parents’ care. The


401

migrant couple did not return every Spring Festival. Instead, their communication with the left-behind parents and child was mostly via telephone, especially since they installed a telephone set in the house in 2002. Wang Mingli’s father explained the rationale for choosing intergenerational division of labor: It is impossible for them to take their son with them to the city. Their main goal doing migrant work is to earn money. They don’t have energy to take care of the child. y It’s better that the husband and wife stay in the same place [versus one doing migrant work and another staying home]. y I am a retired teacher, so I know how to teach a child. y I take good care of their son, so they can make more money outside. Unless something happens, I don’t expect and I don’t want them to return home, not even for the Spring Festival. y Train tickets are expensive and hard to get y returning once every three or four years is fine.

Like many grandparents in the countryside, Wang Mingli’s father felt obligated to help raise the grandchild during the migrants’ absence. His explanation shows not only selflessness on his part, but also the pursuit of efficiency – migrants’ social support system in the village can enable them to maximize the time spent outside and also their earnings from migrant work. Intergenerational division of labor, however, is not problem-free. Since 2001, Wang Cheng’s (Household x4) children had been taken care of by his parents, while he and his wife did migrant work. By 2005, the children were, respectively, nine and four years old. Increasingly, however, Wang Cheng’s parents found this arrangement problematic, because of their deteriorating health and inability to supervise the nine-year-old’s schoolwork. They asked the couple to return. After some negotiation, a compromise was reached: Wang Cheng’s wife would return in the summer of 2005 while he would stay in Guangzhou. In other words, intergenerational division of labor would soon be replaced by the traditional IO arrangement. The above cases show that rural Chinese are flexible in not only work location and activity, but also who participate in migrant work. Wives may join the migrant husbands if their parents are available to help. Thus, work flexibility entails flexibility in household division of labor. Nevertheless, if intergenerational division of labor fails, then it is likely replaced by an IO arrangement, once again underscoring the persistence of gender norms.

Extended Family Household division of labor may involve many members of the extended family. The elderly, siblings, and other family members can pool their

402

C. CINDY FAN

resources together to fulfill household and farming responsibilities, such that migrants can maximize their time in and earnings from migrant work. Such a strategy would inevitably affect the traditional practice of fenjia, where adult sons establish their independent households after they get married. In Household x2, two married brothers and their wives and children were living under one roof, together with the brothers’ parents (see Table 1). The brothers and their wives were all migrant workers. Of their four children, the 18-year-old was doing migrant work and the other three – ranging from 12 to 14 years old – stayed in the village with the brothers’ parents. Fenjia was postponed in order to facilitate the intergenerational division-of-labor arrangement. Of the 10 people in the extended family, 5 were migrant workers and 5 were staying in the village. The two brothers’ father commented: We tried fenjia once but changed our minds the next day. The adults are all doing migrant work. My wife and I are taking care of the grandchildren. How can we fenjia?

Because travel was costly, none of the migrant workers in this household returned regularly for the Spring Festival. Instead, a telephone set was installed and the migrants frequently called home. That an increasing number of migrants skip returning for the Spring Festival suggests that over time the economic benefits of migrant work are given priority over social and cultural traditions. Nevertheless, the more widespread use of the telephone – by 2005, about 120 households in the village had telephone sets and 250 households had mobile phones – may have, to a significant degree, substituted for physical travel as the preferred means of maintaining ties with the stay-behind family members. The above case reinforces the observation that migrant households are flexible in their division of labor and suggests that the pursuit of migrant work may increase the appeal of extended family.

Second-Generation Migrant Workers Increasingly, the children of migrants are themselves joining the labor migration streams. In 2005, Jiang Zhengyi (Household x11, also discussed earlier) was 53 years old and among the oldest migrant workers in the village. His oldest son, having graduated from college, had found a technician job in Chongqing and was therefore pursuing a career path different from the vast majority of migrant workers. Jiang Zhengyi’s youngest son, however, chose a path similar to the first-generation migrants.


403

On finishing (senior secondary) school, he found work in factories and later as a painter. Now that both sons were financially independent, Jiang Zhengyi planned to return to the countryside in the near future: I will continue migrant work for two or three more years. When my sons are older I will stop dagong. y We are getting old, and it is our hope that one of our sons will stay in the village and take care of us. The main reason for many years of hard work raising children is for old age security.

His comment underscores the effect of the life cycle and the deep-rooted tradition that grown-up children are part of the social support system and are the main source of old-age security, especially in the countryside. Household x14 is an example of how the second-generation’s migrant work has lifted the family out of poverty. Three years after Chen Guowei began migrant work, in 1990, he was injured and became paralyzed. His wife, Li Chenxia, then became the family’s sole breadwinner. Her farming activity supported their two sons’ education, paid off a debt, and paid for the disabled husband’s medical expenses. Not until their two sons were old enough to do migrant work was there enough food for the family. The oldest son began migrant work at the age of 14 years and the younger son at the age of 17 years. By 2005, Chen Guowei’s younger son, aged 19 years, was married and had an eight-month-old child. Both the son and the daughter-in-law were doing migrant work, leaving the infant to Li Chenxia’s care. Chen Guowei and Li Chenxia were pleased with the remittances sent by their children. Their dream was that both sons would make enough money to build their own houses in the village. In this household, the second-generation’s pursuit of migrant work had motivated intergenerational division of labor. Just like Households x4 and x5 discussed earlier, such household organization is mutually beneficial: the stay-behind grandparents take care of the grandchildren so that migrant workers can focus on making money, building up their economic security and benefiting the entire household. Both the above cases show that migrant work as a means of livelihood is being passed from one generation to the next. Veteran migrants that are physically capable may still be doing migrant work even after their children are old enough to contribute financially to the family, but the second generation’s joining migrant work can provide much needed relief to the parents and pave way for their retirement. The second-generation’s migrant work, in turn, may entail intergenerational division of labor as their parents stay behind in the village and are available to help.

404

C. CINDY FAN

CONCLUSION In this chapter, I have sought to foreground the relationship between labor migration and household organization. In particular, I have highlighted households that are split into two or more places as a result of migrants leaving their family members behind. This split-household arrangement may be due to migrants’ inability to bring along their family, but I have argued that it is also due to migrants’ preference for circulating between the site of work and the home community, a strategy that enables them to obtain the best of both worlds and protect their economic and social security in the countryside. In the case of rural–urban labor migration in China, I have shown that the split-household arrangement is an outcome of migrants’ pursuit of a strategy of flexibility. My analysis of household biographies and narratives from a village in Sichuan highlights migrants’ flexibility in the type, location, and tenure of work as their strength. By changing jobs and the type of work and by being footloose, rural migrants take advantage of regional wage differentials. By switching back and forth between the countryside and urban areas, they play dual roles of farmers and urban workers and enjoy the best of both worlds – urban work to boost income and improve rural living, and the countryside for security, livelihood, and long-term settlement. Their continuing to support agriculture reinforces the notion that access to farmland is still highly valued. Despite agriculture’s low return, migrants still have strong desire to protect and take care of their farmland (Zhu, 2003, 2007). At the time of this chapter’s writing, millions of rural migrants have lost jobs as one after another Chinese manufacturer goes out of business because of the global financial crisis. Many migrants are compelled to make the difficult choice between returning home to farm and facing unemployment in the city. This is a vivid reminder that agriculture remains an important source of subsistence and economic security for rural Chinese. To accommodate migrants’ work flexibility, their households are, accordingly, flexible in organization. Not only are migrants ready to split the household between the city and the countryside, but households frequently change from one form of division of labor to another. The IO model highlights traditional gender ideology that defines women in relation to village and care-giving responsibilities. But, over time, the OO arrangement, where both the husband and wife pursue migrant work, is increasingly popular and may be replacing the IO model. Despite this change, migrants frequently return for extended periods of time, and when the return is for the purpose of caring for children, it is often the wife rather than the husband that returns. Usually, the OO arrangement is facilitated by


405

intergenerational division of labor in the form of help by the elderly and assistance by the extended family. Intergenerational division of labor takes place also when the second generation begins migrant work and replaces first-generation migrants. The readiness to collaborate and negotiate mutually beneficial household organization supports the observation that migration has not undermined the tradition of extended families in the Chinese countryside (Goldstein, Guo, & Goldstein, 1997). On the contrary, the need for and popularity of gender, intergenerational, and even interhousehold divisions of labor suggest that the extended family tradition is strongly appealing.

NOTES 1. Roberts (2007) compares this situation to Mexico–US immigrants who have access to farmland in rural Mexico that cannot be sold and that has become not only an economic asset, but also a base for all household activities. 2. In the Chinese countryside, land ‘‘leasing’’ arrangements are largely informal. A popular arrangement is one where the leaser pays agricultural tax, and the lessee takes all the yield during the period of the lease and pays no rent to the leaser. This arrangement benefits both parties as the lessee is rewarded for his/her time and effort while protecting the farmland’s boundary and keeping it fertile for the leaser’s return.

ACKNOWLEDGMENTS This research was supported by a National Science Foundation award (BCS-0455107) and funding from the Academic Senate and the Asian American Studies Center of the University of California, Los Angeles. I am grateful to Professor Nansheng Bai for his central role in a joint project from which the data for this chapter are drawn. I also thank Dr. Wenfei Winnie Wang for her invaluable assistance and helpful comments.

REFERENCES Alexander, P., & Chan, A. (2004). Does China have an apartheid pass system? Journal of Ethnic and Migration Studies, 30(4), 609–629. Castro, L. J., & Rogers, A. (1983). Patterns of family migration: Two methodological approaches. Environment and Planning A, 15(2), 237–254. Chan, K. W. (1996). Post-Mao China: A two-class urban society in the making. International Journal of Urban and Regional Research, 20(1), 134–150.

406

C. CINDY FAN

Chan, K. W., & Buckingham, W. (2008). Is China abolishing the hukou system? China Quarterly, 195, 582–606. Chant, S. (1991). Gender, migration and urban development in Costa Rica: The case of Guanacastle. Geoforum, 22(3), 237–253. Clark, W. A. V., & Withers, S. D. (2002). Disentangling the interaction of migration, mobility, and labor-force participation. Environment and Planning A, 34(5), 923–945. Croll, E. (1994). From heaven to earth: Images and experiences of development in China. London: Routledge. Deshingkar, P. (2005). The role of circular migration in economic growth. Entwicklung & Landlicher Raum, http://www.rural-development.de/fileadmin/rural-development/volltexte/ 2005-05/ELR_dt_10-12.pdf Deshingkar, P., & Start, D. (2003). Seasonal migration for livelihoods in India: Coping, accumulation and exclusion. Working Paper 220, Overseas Development Institute. Du, Y. (2000). Rural labor migration in contemporary China: An analysis of its features and the macro context. In: L. A. West & Y. Zhao (Eds), Rural labor flows in China (pp. 67–100). Berkeley, CA: Institute of East Asian Studies. Du, Y., & Bai, N. (Eds). (1997). Zouchu Xiangcun [Leaving the village] (in Chinese). Beijing: Jingji kexue chubanshe (Economic Science Press). Entwisle, B., & Henderson, G. E. (2000). Conclusion: Re-drawing boundaries. In: B. Entwisle & G. E. Henderson (Eds), Re-drawing boundaries: Work, households, and gender in China (pp. 295–303). Berkeley, CA: University of California Press. Fan, C. C. (2002). The elite, the natives, and the outsiders: Migration and labor market segmentation in urban China. Annals of the Association of American Geographers, 92(1), 103–124. Fan, C. C. (2004). Out to the city and back to the village: The experiences and contributions of rural women migrating from Sichuan and Anhui. In: A. M. Gaetao & T. Jacka (Eds), On the move: Women and rural-to-urban migration in contemporary China (pp. 177–206). New York: Columbia University Press. Fan, C. C. (2008). China on the move: Migration, the state, and the household. New York, NY: Routledge. Fan, C. C., & Wang, W. (2008). The household as security: Strategies of rural–urban migrants in China. In: R. Smyth & I. Nielsen (Eds), Migration and social protection in China (pp. 205–243). New Jersey, NY: World Scientific. Goldstein, A., Guo, Z., & Goldstein, S. (1997). The relation of migration to changing household headship patterns in China, 1982–1987. Population Studies, 51, 75–84. Hershatter, G. (2000). Local meanings of gender and work in rural Shaanxi in the 1950s. In: B. Entwisle & G. Henderson (Eds), Re-drawing boundaries: Work, household, and gender in China (pp. 79–96). Berkeley, CA: University of California Press. Hugo, G. (2003a). Circular migration: Keeping development rolling?, Migration Policy Institute, http://www.migrationinformation.org/Feature/print.cfm?ID ¼ 129, June 1. Hugo, G. (2003b). Urbanisation in Asia: An overview. Paper prepared for Conference on African Migration in Comparative Perspective, Johannesburg, South Africa, 4–7 June. Hugo, G. (2006). Temporary migration and the labour market in Australia. Australian Geographer, 37(2), 211–231. Jacka, T. (1997). Women’s work in rural China: Change and continuity in an era of reform. Cambridge: Cambridge University Press. Jacka, T. (2006). Rural women in contemporary China: Gender, migration, and social change. Armonk, NY: M.E. Sharpe.


407

Jarvis, H. (1999). The tangled webs we weave: Household strategies to co-ordinate home and work. Work, Employment & Sociology, 13(2), 225–247. Lim, L. (2008). Chinese rural reforms free farmers from land. National Public Radio, http:// www.npr.org/templates/story/story.php?storyId ¼ 96380759, November 6. Lin, G. C. S., & Ho, S. P. S. (2005). The state, land system, and land development processes in contemporary China. Annals of the Association of American Geographers, 95(2), 411–436. Ma, L. J. C., & Xiang, B. (1998). Native place, migration and the emergence of peasant enclaves in Beijing. China Quarterly, 155, 546–581. Mann, S. (2000). Work and household in Chinese culture: Historical perspectives. In: B. Entwisle & G. Henderson (Eds), Re-drawing boundaries: Work, household, and gender in China (pp. 15–32). Berkeley, CA: University of California Press. McHugh, K. E. (2000). Inside, outside, upside down, backward, forward, round and round: A case for ethnographic studies in migration. Progress in Human Geography, 24(1), 71–89. Mincer, J. (1978). Family migration decisions. Journal of Political Economy, 86(5), 749–773. Mitchell, K. (1997). Transnational subjects: Constituting the cultural citizen in the era of Pacific Rim capital. In: A. Ong & D. Nonini (Eds), Ungrounded empires: The cultural politics of modern Chinese transnationalism (pp. 228–256). New York, NY: Routledge. Nagar, R., Lawson, V., McDowell, L., & Hanson, S. (2002). Locating globalization: Feminist (re)readings of the subjects and spaces of globalization. Economic Geography, 78(3), 257–284. National Bureau of Statistics (NBS). (2006). 2005 nian quanguo 1% renkou chouyang diaocha zhuyao shuju gongbao [Announcement of major statistics from the 2005 National 1% Population Sample Survey] (in Chinese), http://www.cpirc.org.cn/tjsj/tjsj_cy_detail.asp?id ¼ 6628 Nelson, J. M. (1976). Sojourners versus urbanites: Causes and consequences of temporary versus permanent cityward migration in developing countries. Economic Development and Cultural Change, 24(4), 721–757. Odland, J., & Ellis, M. (1988). Household organization and the interregional variation of outmigration rates. Demography, 25(4), 567–579. Ong, A. (1999). Flexible citizenship: The cultural logics of transnationality. Durham, NC: Duke University Press. Pan, L. (Ed.) (1998). The encyclopedia of the Chinese overseas. Hong Kong: Joint Publishing (H.K.) Co. Reyes, B. I. (2004). Changes in trip duration for Mexican immigrants to the United States. Population Research and Policy Review, 23(3), 235–257. Roberts, K. D. (2007). The changing profile of Chinese labor migration. In: Z. Zhao & F. Guo (Eds), Transition and challenge: China’s population at the beginning of the 21st century. Oxford: Oxford University Press. Saxenian, A. (2005). From brain drain to brain circulation: Transnational communities and regional upgrading in India and China. Studies in Comparative International Development, 40(2), 35–61. Silvey, R. (2006). Consuming the transnational family: Indonesian migrant domestic workers to Saudi Arabia. Global Networks, 6(1), 23–40. Skeldon, R. (1997). Migrants on a global stage: The Chinese. In: P. J. Rimmer (Ed.), Pacific rim development: Integration and globalization in the Asia-Pacific economy (pp. 222–239). Canberra: Allen & Unwin.

408

C. CINDY FAN

Skinner, G. W. (1976). Mobility strategies in late imperial China: A regional systems analysis. In: C. A. Smith (Ed.), Regional analysis, Vol. I: Economic systems (pp. 327–364). New York: Academic Press. Solinger, D. J. (1995). The floating population in the cities: Chances for assimilation? In: D. Davis, R. Kraus, B. Naughton & E. J. Perry (Eds), Urban spaces in contemporary China (pp. 113–139). Cambridge: Cambridge University Press. Solinger, D. J. (1999). Contesting citizenship in urban China: Peasant migrants, the state, and the logic of the market. Berkeley, CA: University of California Press. Stark, O. (1991). The migration of labor. Cambridge, MA: Basil Blackwell. Stark, O., & Taylor, E. J. (1991). Migration incentives, migration types: The role of relative deprivation. The Economic Journal, 101(408), 1163–1178. Tan, S. (1996). Zhongguo nongcun laodongli liudong di xingbei cha [Gender differences in the migration of rural labor force] (in Chinese). Paper presented at the international conference on rural labor migration, Beijing, China. Wang, F.-L. (2005). Organizing through division and exclusion: China’s hukou system. Stanford, CA: Stanford University Press. Wang, W. (2003). Changes in household registering system against the urbanization background: Shijiazhuang city as a case. Renkou Yanjiu (Population Research) (in Chinese), 27(6), 8–13. Wang, W. W., & Fan, C. C. (2006). Success or failure: Selectivity and reasons of return migration in Sichuan and Anhui, China. Environment and Planning A, 38(5), 939–958. Waters, J. L. (2002). Flexible families? ‘Astronaut’ households and the experiences of lone mothers in Vancouver, British Columbia. Social & Cultural Geography, 3(2), 117–134. Wilkinson, C. (1987). Women, migration and work in Lesotho. In: J. H. Momsen & J. G. Townsend (Eds), Geography and gender in the third world (pp. 225–239). London: State University of New York Press. Willis, K., Yeoh, B. S. A., & Fakhri, S. M. A. K. (2004). Introduction: Transnationalism as a challenge to the nation. In: B. S. A. Yeoh & K. Willis (Eds), State/nation/transnation: Perspectives on transnationalism in the Asia-Pacific (pp. 1–15). London: Routledge. Woon, Y.-F. (1994). Family strategies of prosperous peasants in an emigrant community in South China: A three-year perspective (1988–1991). Canadian Journal of Development Studies, 15(1), 10–33. Xiang, B. (2005). Transcending boundaries: Zhejiangcun, the story of a migrant village in Beijing (translation by Jim Weldon). Leiden, The Netherlands: Brill. Yusuf, S., & Nabeshima, K. (2006). China’s development priorities. Washington, D.C.: World Bank. Zhou, D. (1992). The nonnative labourers of the Pearl River Delta: An analysis of its distribution characteristics and migration tendencies. In: Zhongshan University Research Centre of Pearl River Delta Economic Development and Management (Ed.), Economic development of the Pearl River delta: A retrospect and prospects (pp. 271–280) (in Chinese). Guangzhou: Zhongshan University. Zhou, H. (2004). Zhongguo renkou qianyi de jia ting hua qushi ji yingxiang yinsu fenxi [The analysis on the trend and factors of family migration]. Renkou Yanjiu (Population Research) (in Chinese), 28(6), 60–67. Zhu, Y. (2003). The floating population’s household strategies and the role of migration in China’s regional development and integration. International Journal of Population Geography, 9(6), 485–502. Zhu, Y. (2007). China’s floating population and their settlement intention in the cities: Beyond the hukou reform. Habitat International, 31(1), 65–76.

THE RUBIK’S CUBE STATE: A RECONCEPTUALIZATION OF POLITICAL CHANGE IN CONTEMPORARY CHINA Bai Gao China presents a major puzzle for our time. At first glance, foreign observers are often perplexed by China’s two sharply contrasting images in globalization. On the economic front, China has been the forerunner in expanding trade and global production. Emerging as the world factory in the international division of labor, the might of the Chinese economy can be felt in every corner of the globe. On the political front, however, China is still ruled by an authoritarian regime, remaining untouched by the third wave of democratization, a powerful political current that has swept through many countries in Europe, Latin America, Asia, and elsewhere over the past three decades. According to conventional wisdom, it simply does not make sense for a country to engage its economy in the world capitalist system so deeply but at the same time retain a political regime that is widely perceived as a legacy of past centuries. The ongoing global financial crisis has added a new twist to the debate on China: since many Western countries are debating whether to nationalize their banks and China is suddenly being perceived as the


409

410

BAI GAO

savior of capitalism, the seemingly contradictory nature of China’s political economy has become more intriguing than ever. The China puzzle has stirred a heated debate on the relationship regarding economic development, democratization, and the future prospect of political change in that country. At the empirical level, ‘‘most China experts have subscribed to one of three main theories about the future of China’s government: that it will collapse, democratize, or remain authoritarian’’ (Nathan, 2006, p. 177). Underlying these diverse views are two distinct theoretical perspectives on political change: one emphasizes the structural factors, including socioeconomic and cultural requisites of democracy, and the other regards a political regime as an outcome of strategic choices and interactions by political actors (Colomer, 2000). These two perspectives are sharply divided on the issue of whether economic development will inevitably lead to democratization in China. In this article I first review this debate on China’s prospects for democratization. I point out that structural views have failed to incorporate the strategic responses from China’s party state in their analyses, and thus have major drawbacks. I also point out that while the agency-centric perspective is more powerful in explaining the Chinese case, it needs to further expand its conceptualization to capture the many changes taking place in the party state, transcending the simplistic dichotomy between the authoritarian state and democracy. I then propose an analytical framework of state building informed by the debate on globalization. Viewed through the lens of state building, the changes within China’s party state in the past three decades can be perceived as responses to the challenges of globalization. On the one hand, these responses were neither well planned nor under one unified ideology, and were often reactive to the challenges the regime has faced at each time. On the other hand, the party state determined early on that it would fight for its political survival by promoting economic growth, and democratization would not be an option. In order to achieve this goal it has been willing to incorporate wisdom from various models of the state and build new capacities. When many states in developing, transitional, and advanced industrialized economies have either scaled back or converted themselves to night watchmen in the process of globalization, the party state in China has aggressively rebuilt its capacities, transforming itself into a Rubik’s Cube state that has six distinctive faces: authoritarian, neoliberal, developmental, predatory, refined socialist, and corporatist. It is the combination of these six components that has enabled China to become the forerunner in economic development and prevented it from joining the third wave of democratization.

The Rubik’s Cube State

411

THE STRUCTURAL PERSPECTIVE The structural perspective on China’s prospect of democratization has three variants. The first emphasizes the structural requisites for the survival of the authoritarian state. It argues that the conditions, such as the governing capacity of the state and support from the Chinese people that used to sustain the authoritarian state, have deteriorated significantly and the authoritarian state cannot escape a collapse in the near future (Chang, 2001). The second focuses on the structural requisites for democratization. It holds that the rise of the middle class and the emerging spread of education in China will create favorable conditions for the country to head toward democratization (Gilley, 2004). The third stresses the resilience of China’s authoritarian regime. It argues that the rise of democratic polity in Europe resulted from the special social structures of the continent in the feudal period. Since China’s social structure in its premodern period was quite different, democracy did not become a solution even after the middle class emerged in China. For the same reason, China’s political change will be most likely to move toward rule by law rather than democratization in the future (Pan, 2006). The collapse argument highlights the declining conditions that used to sustain the authoritarian state. According to this view, after China’s party state retreated from the Mao era’s mechanisms of social control and transformed itself from totalitarian to authoritarian, it lost its meaning for the Chinese people. The mounting resentment against authoritarian rule is represented by various social groups, ranging from Falun Gong to Dalai Lama, Uighurs, and Muslims. Widespread corruption has become the greatest threat to the authoritarian state because it ‘‘angers the Chinese people more than anything else’’ (Chang, 2001, p. 269). The inefficient stateowned enterprises (SOEs) will collapse sooner or later. The banks contain huge amounts of hidden nonperforming loans. The impact of China’s entry into WTO will eventually create widespread unemployment (Chang, 2001). ‘‘Despite its awe-inspiring economic growth and progress, a set of selfdestructive dynamics is weakening China most vital political institutions – the state and the ruling party’’ (Pei, 2006, p. 206). In the end these problems ‘‘are likely to result in a breakdown in the communist regime in years rather than decades y all the elements for a really massive collapse are there y. The question is what will the trigger be’’ (MacFarquhar, cited by Shambaugh, 2008, p. 25). The democratization argument shares the structural perspective with the collapse argument but highlights the structural requisites for democratization.

412

BAI GAO

It argues that when GDP per capita reaches US$3,000–10,000 (depending on the study), authoritarian regimes tend to crumble. According to a World Bank estimate, in 2002, GDP at purchasing power parity per capita in China reached around US$4,000. Also, China has witnessed an emergence of a civil society, a growing rule of law, and unprecedented interaction with the outside world through its participation in globalization. At the same time, the Long March generation has passed away and the old guard of the authoritarian regime is gone. Today, even within the Chinese Communist Party (CCP), people debate on the issues related to democracy. In addition, military forces began to be depoliticized; the once politically and economically powerful People’s Liberation Army now has no seat on the Standing Committee of the Politburo. Although Confucianism is indeed the core of public and political values in China and is fairly antidemocratic, the fact that a country is currently nondemocratic does not mean that it cannot become democratic; historically every society that became democratic was once nondemocratic. This is as true of the countries of the West as it is of countries elsewhere in the world (Gilley, 2004). The survival argument contextualizes the relationship between the rise of the middle class and democratization and treats the rise of democracy as an outcome of special historical circumstances rather than a universal trajectory of political change in the process of modernization. It points out that the rise of democracy in modern Europe was associated with its unique social structures in the feudal era. These European social structures lacked economic freedom for certain social status levels to sell lands and labor, and political equality that denied opportunities for the lower classes to participate in the government. Moreover, the competition among the decentralized political entities and fragmented societies nurtured a strong tradition of power politics. At the same time, the unique role played by religion in European societies led to a strong legal tradition that regarded the authority of the court as a contract with God. For the middle class that emerged on the European political horizon in the process of industrialization, democracy was the perfect political solution for these problems (Pan, 2006). In contrast, during the feudal era China had an embedded tradition of economic liberty that allowed people to sell land and labor without obstacle. The imperial examination system sustained a unique tradition of political equality by opening the channels of upward social mobility to the lower classes. Under the influence of the Confucian tradition, governance relied heavily on persuasion through indoctrination of moral principles rather than power politics. Such a style of governance resulted in both a small government and a blurred boundary between the state and society.


413

As a result moral principles, instead of laws of God, were the sources of legitimacy. However, the failures of various political regimes before the CCP came to power were due not to the fact that they were authoritarian, but that they were unable to mobilize Chinese society to resist imperialist innovation and build modern industries. In order to achieve these goals, China eventually chose authoritarianism (Pan, 2006). Many empirical studies of China’s rising middle class seem to support the survival argument. They show that traditional social structures have continued to function in contemporary China. Members in encompassing and embedding solidarity groups in rural China award leaders with moral standing if they comply with these informal institutions and take them away when they fail to do so. This backs up Pan’s claim that traditional governance relied on moral principles, but demonstrates a reverse case: not only do officials use moral indoctrination to persuade people, people also use moral principles to evaluate the officials. Moreover, moral standing is enforced with traditional solidarity groups, such as kinship-based lineage and members in rural temples. These groups continue to exist and function even in the era of economic reform. Through the combination of these two factors, moral principles and solidarity groups, accountability can be achieved without formal institutions of democracy, despite rapid economic development (Tsai, 2007b). Entrepreneurs, one of the major representatives of China’s rising middle class, demonstrate neither clear preference toward political change nor hold progressive or liberal views. Most of these entrepreneurs hold elitist views that ‘‘support greater participation by people like themselves but not more widespread political participation.’’ They unanimously hold the position that China’s party state has played the leadership role in economic reform without societal input. ‘‘Entrepreneurs are much more concerned about the impact of economic competition on stability than are officials, and ironically red capitalists are the most willing to enforce stability at the expense of continued economic growth’’ (Dickson, 2003, p. 135). Contrary to the prediction by modernization theorists, the impact of economic growth on the middle class, measured by individual prosperity, has not significantly affected the attitudes of China’s new middle class with regard to the pace of political reform, or the leadership of the state in initiating political and economic reform (Dickson, 2003). In addition, the party state has already offered business owners informal channels for expressing their concerns (Tsai, 2007a). Thus, despite the fact that China is not definitely excluding the possibilities of democratization, these studies conclude that ‘‘these possibilities are unlikely to grow out of either direct demands from private entrepreneurs or their informal coping strategies’’ (Tsai, 2007a, p. 219).

414

BAI GAO

THE AGENCY PERSPECTIVE The agency perspective highlights the strategic responses by China’s party state to the challenges of globalization. These responses include administrative reform, co-optation of the new elites, building new corporatist mechanisms, and party-building efforts. According to the agency perspective, the party state in China is not a political entity that refuses all change, as the concept of the authoritarian state may suggest. Since the 1980s it has taken many steps in administrative reform. Although Deng Xiaoping’s view of political reform was narrowly defined, he still wanted to reform the party state in an effort to ‘‘maintain and further strengthen Party leadership and discipline, and not to weaken or relax them’’ (cited in Pei, 2006, p. 47). Deng held that China’s political system had four major problems: bureaucratism, overcentralization of power with the top leaders, lifetime tenure of party and government officials, and privileges associated with party and government positions. Since the early 1990s the CCP began to use education, technical expertise, professional competence, and age as the criteria for promoting party and government officials. At the same time, senior party and government officials no longer enjoy lifetime tenure, and mandatory retirement at specified ages at different levels has been strictly enforced (Huang, 2008). Moreover, the procedure of leadership succession has been formally institutionalized. The transitions of the top leadership, both from Deng Xiaoping to Jiang Zemin and from Jiang Zemin to Hu Jintao, were relatively smooth, without major political interruptions (Huang, 2008). To eliminate inefficiency, dysfunction, overstaffing, and degeneration, the government has undergone several rounds of bureaucratic restructuring, usually after the selection of a new premier (Burns, 1989; Zheng, 2004). Beginning in the 1990s the state has introduced a formal mechanism of public hearing, public consultation, and public deliberation in the policy-making process, especially on issues that have major political implications. At both the central and local levels, a certain amount of progress has been made (Yu, 2008). Co-optation has been an adaptive strategy adopted by the party state in the era of economic reform. The CCP has been recruiting new elites who bring new ideas and new goals to the party. After the CCP abandoned its old doctrine of class struggle, its recruitment policy shifted from exclusive to inclusive, bringing in those the party depended on, those who possessed the resources the party needed, and those who might otherwise become a threat to the party. As a result, many former class enemies and counterrevolutionaries were brought into the party because they had the skills for


415

accomplishing the party’s new policy agenda of promoting economic growth (Dickson, 2003). Private entrepreneurs and intellectuals were the two major groups targeted for co-optation. The CCP aggressively recruited college students and promised career and material benefits by promoting young professionals under the age of 45 to college administration, designating young intellectuals as ‘‘reserve cadres’’; and granting professional honor, recognition, and perks to a select group of senior scholars and professionals. By the end of the 1990s nearly 20 percent of the CCP’s members claimed to hold a college or college-equivalent degree (Dickson, 2003; Pei, 2006). Building new corporatist mechanisms within the regime is another major strategy that allows the party state to accommodate a new open market economy and increasingly interest-based politics. As private entrepreneurs and their business organizations gained power in economic sphere, they developed a propensity for asserting their interest in politics and to use their business associations to pursue their individual and collective goods. After the party state transformed itself from totalitarian to authoritarian, it relied less on coercion and propaganda to control society. Instead, it has substituted the manipulation of symbols with the manipulation of organizations, developing links with the chamber of commerce, labor unions, and various professional associations. ‘‘The state has created a dense web of economic and social organizations in order to channel interest articulation, regularize the flow of information between the state and key groups in society, replace direct state controls over the economy and society with at least partial social regulation, and screen out unwanted groups’’ (Dickson, 2003, p. 58). In the Chinese case the party state approached corporatism not as a mechanism for further strengthening the state control over the economy and society, but as a mechanism through which state control could be loosened without creating a major crisis (Unger & Chan, 1995). Party building by the CCP has been another major adaptive strategy. The party-building efforts involve fighting corruption, improving party discipline, rebuilding the local party apparatus, improving extraparty consultation and supervision, increasing inner party democracy, improving cadre competence, providing mid-career training for party and government officials, and expanding the party school system and cadre training academies (Shambaugh, 2008). All of the above steps have helped the party state maintain its control. Although few analysts would go so far as to argue that these steps have offered the CCP an alternative to democratization, they at least help explain why China was able to present a case of ‘‘capitalism without democracy,

416

BAI GAO

political change without regime change, and, indeed, capitalists in a communist party’’ (Tsai, 2007a, p. 221).

THE PROBLEMATIC CONCEPT OF THE AUTHORITARIAN STATE The debate on the future direction of political change in China has centered around the dichotomy between the authoritarian state and democracy. However, the concept of the authoritarian state is not very useful in helping us understand the constantly changing reality of the Chinese political economy. First, the concept of the authoritarian state lacks the sharpness to capture the changing substance of China’s party state over the past three decades. In reality, we face a political entity ‘‘in free-fall transition to some system as yet known’’ (Unger & Chan, 1995, p. 29). In academic discussions, however, we still rely on an old, static concept that ‘‘observers of China find themselves faced with a system y to the point that it often becomes difficult to analytically frame what is occurring at present, let alone attempt analyses of China’s probable future’’ (Unger & Chan, 1995, p. 29). These competing views cite a wide range of evidence that points in very different directions. Nevertheless, the profound changes taking place in the party state over the past three decades have not brought about any change in our conceptualization. Analysts still share the same understanding of the nature of the regime at present, and they differ only in their predictions for the future. For those who subscribe to the collapse argument, the authoritarian state is rigid and cannot make sufficient changes to address the challenges it faces. For those who believe in its survival, the authoritarian state is flexible and adjusts well to the changing environment. In contrast, for those who assert the democratization scenario, whether the authoritarian state changes is inconsequential because its existence will end with the progress of economic development. Second, most analyses on China’s political future adopt the strategy of highlighting one dimension only and make their prediction accordingly. These single dimensions may ‘‘hold strong explanatory value for some of the more important trends’’ (Unger & Chan, 1995, p. 29); however, none of these single dimensions can provide ‘‘an all-encompassing framework for everything occurring in China today’’ (Unger & Chan, 1995, p. 29). This analytical strategy has neglected the fact that many dimensions


417

discussed by different schools in the debate are actually strongly related to each other, and building an argument based on one part of the integral whole often runs into self-contradiction. For example, the efforts of the party state to expand its links to nonstate organizations is often considered corporatist in nature, and these links serve to strengthen the party’s legitimacy. In practice, however, such a corporatist mechanism at the local level often results in a probusiness policy, because a strong alliance between local government and business elites at the local level can turn the state predatory and this often stimulates social tension and political conflict (Pei, 2006). Third, the concept of the authoritarian state is purely political; under this concept the economic dimensions are exogenous to the nature of the party state. Recent literature on democratization considers the economic performance of an authoritarian state as a precondition to democratization (Huntington, 1991). In the Chinese case, however, economic performance is precisely the explanation for the survival of the authoritarian state. Since the ability of the authoritarian state to prevent such economic crises is the precondition for its survival, the state capacities in handling economic affairs is endogenous to the virtue of the authoritarian state itself. Failure to incorporate economic dimensions into our conceptualization of the party state prevents us from effectively explaining the Chinese experience. Moreover, neglecting the economic dimensions in China’s party state neither helps us distinguish China from North Korea or Cuba, or from Russia and other Eastern European countries, nor explains why China has responded to the ongoing global financial crisis more quickly and forcefully. We need a political economy approach that directly conceptualizes state capacities in economic affairs. Both the structural and agency perspectives recognize the changes in China. However, neither of them has conceptualized these changes. Instead, they treat these exogenous changes as independent variables and use them to predict the future of the authoritarian state. In this paper I adopt a different strategy: by drawing on mainstream social science literature on globalization and democratization, I treat the resilience of China’s party state, sustained by its economic performance, as the dependent variable; its efforts in state building, or the rise of the Rubik’s Cube state, as the intervening variable that explains why China’s party state has been able to survive and avoid the path leading toward democratization; and the responding strategy of the CCP to the challenge of globalization – the strategy of one central goal, two basic points – as the independent variable that ultimately explains the rise of the Rubik’s Cube state.

418

BAI GAO

GLOBALIZATION AND STATE BUILDING I propose a state-building approach informed by the globalization debate to reconceptualize the political changes in China’s party state over the past three decades. The fate of the state is the central issue in the globalization debate. The state-building approach rejects the argument that the role of the state in economic governance is being permanently reduced to insignificance in the process of globalization. It argues that the market-building process is also a process of state building. Globalization does not eliminate the role of the state; on the contrary, globalization is one of the driving forces behind the new state-building process. In many cases globalization has reinforced the authority of the state or even expanded state capacities for opportunities (Kahler & Lake, 2003). The state-building approach argues that the role of the state in economic governance can be renewed and replenished, rather than simply eroded in the process of globalization. This approach transcends the debate on whether the state is growing or shrinking because framing the issue in such a way misses a more important change – a shift in the purposes and modalities of contemporary state intervention (Levy, 2006a, 2006b, 2006c). In order to maximize gains in the process of globalization, the development of new state capacities is as important as the reduction of state intervention (Fukuyama, 2004); in fact, state capacities have become an important institutional advantage in international competition (Weiss, 1998). Even when a country decides to join the neoliberal policy wagon and embrace the Washington consensus, the state still has an increasing role, either to break the power of labor unions and producer groups pushing a more procompetition regulation, or to bring the public sector to heel (Gamble, 1994; Richardson, 1994). For developing countries, deregulation will not naturally lead to the increasing inflow of foreign direct investment (FDI). Facing the rigorous competition among nation states for mobile capital, states in developing countries have to make great effort to create the various conditions favorable to attracting international capital. Without the contribution by the state, a developing country cannot even win the neoliberal race. Globalization means the breakdown of national regulatory restraints, the release of market forces, and the integration of various countries into the global production system, sustained by the increasing free flows of capital, technologies, ideas, merchandises, and even labor across national borders. As such, globalization brings not only great opportunities of economic growth but also increasing risks associated with an open market economy. On the one hand, globalization provides developing countries with


419

unprecedented opportunities to promote economic growth. The rise of the global production system has significantly lowered the institutional requisites for developing countries. It has enabled developing countries to engage in the process of economic growth by participating at the low end of the global value chain of production operated by multinational corporations. The inflows of FDIs have brought in not only capital but also technologies, managerial skills, and marketing channels. On the other hand, however, by connecting national economies with international markets, globalization has also brought higher levels of risk to both developed and developing countries. As a result of releasing market forces, the institutions established to prevent the Great Depression have been gradually replaced and social programs have been significantly scaled back. Individuals are exposed to more intense competition in the marketplace and receive less government protection. As the ongoing global financial crisis shows, after national economies became open, economic crisis in one country could easily affect other countries. Through the processes of liberalization, privatization, and deregulation, individuals and individual companies confront increasing risks that they have never encountered in the past. Only the state can ‘‘readily absorb and socialize risks’’ to the extent required by modern technologies and the global production system (Weiss, 1998, p. 6). State building is critical to a country’s performance in globalization in both its effectiveness in grasping opportunities and its management of risks. The outcome of state-building efforts in China was the rise of the Rubik’s Cube state. The Rubik’s Cube state is an analogy. A Rubik’s Cube has six surfaces, each of which independently presents a coherent picture. As soon as play starts, however, the coherent pictures in these six single dimensions are broken, replaced by a combination of various components drawn from different surfaces. I argue that in the game of political change in China, the CCP is the player of the Rubik’s Cube. Responding to the challenge of globalization over the past three decades, China’s party state has developed six distinctive components. It transformed itself from totalitarian to authoritarian. At the same time, it also contains substantial elements identical to the neoliberal, the developmental, the refined socialist, the predatory, and the corporatist states. Each of these six surfaces captures some core properties of the Chinese political economy. How these surfaces combine often depends on the responding strategies adopted by the CCP to the specific challenges under given temporal and spatial conditions. Although only the authoritarian dimension among these six demonstrates the nature of China’s political regime, the other five dimensions also add critical insights to our understanding of the party state in contemporary China.

420

BAI GAO

Despite the fact that the other five dimensions have not changed the authoritarian nature of the party state, their existence has significantly affected and will continue to affect the fate of the party state and the nature of political economy in China. Therefore, without taking these five dimensions into account, any explanation of why China has avoided the path of democratization and any prediction about the country’s future prospects for democratization will miss important points.

THE RESPONDING STRATEGY OF THE PARTY STATE TO THE CHALLENGE OF GLOBALIZATION The state-building approach is not deterministic. It is based on the assumption that globalization creates a new environment to which political actors have to respond. Nevertheless, whether they respond and how they respond are dependent on domestic political structure and institutions because ‘‘nation-states remain central sites of political contestation’’ (Kahler & Lake, 2003, p. 436). To say that the state can respond strategically to the challenge of globalization does not mean that every state will definitely respond or all states respond to the same extent with the same strategy, or that when they respond they can achieve the same outcome of economic performance. The rise of a Rubik’s Cube state in China was driven by the CCP’s strategy of ‘‘one central goal, two basic points’’ as a response to the challenge of globalization. The central goal is economic development, and the two basic points involve reform and opening on the one hand, and maintaining the four basic principles on the other: maintaining Marxism, Leninism, and Mao’s thought; maintaining proletariat dictatorship; maintaining the socialist path; and maintaining the leadership of the CCP. Facing the challenges from the rise of neoliberalism and the third wave of democratization, the essence of the party state’s responding strategy is to embrace the former but reject the latter. In order to rebuild the CCP’s legitimacy on economic performance, the party state is willing to aggressively reform its domestic economic institutions and open itself to international markets. At the same time, however, it firmly rejects the option of moving toward democratization. Such a strategy has three distinct characteristics. First, China’s embrace of globalization has not been driven by a pure economic ideology, but by a political goal of the party state’s survival. For this reason the state in the


421

Chinese economy has never fully given its own power to the market. Second, political forces with diverse ideologies have coexisted within the party state throughout the process of reform and opening. The political strategy of ‘‘one central goal, two basic points’’ is seemingly self-contradictory since it both encourages the reformers to break the restraints of orthodox ideologies, and at the same time provides conservatives with the ideological weapons to fight back. The truth is, however, such diverse political forces have enabled the CCP to cope with the challenge of globalization by using different strategies. Third, the only firm decision China’s party state has made is to avoid the path toward democratization. Other than that, the CCP’s strategy of coping with the challenge of globalization has not been preplanned; rather, it has evolved in piecemeal fashion, often driven by a specific challenge or crisis. The bigger the crisis, the more diverse the adaptations. China is a unique case as far as the relationship between globalization and state building is concerned. In other cases the state tended to be passive, and often became a victim of globalization. Many developing countries in the world were forced to accept the policy paradigm affected by the Washington consensus as a precondition for rescue by the International Monetary Fund or the World Bank during crises. Such a policy paradigm demanded that the state pull back its intervention in the economy and release the market forces. Former socialist countries in Russia and Eastern Europe subscribed to the full package of reform with democratization accompanying marketization and privatization. As a result, the old state intervention was dismantled and the state no longer played an active role in governing the economy. In contrast, the party state in China has played a dual role: it has served as both the agent of globalization and as a counterweighing power to market forces. The six seemingly conflicting facets of the Rubik’s Cube state reflect this complicated role of the party state.

THE RISE OF THE RUBIK’S CUBE STATE Below I discuss these six dimensions of the party state in China. In each dimension, I first present the ideal type; then discuss how the party state, affected by its grand strategy of ‘‘one central goal, two basic points,’’ arrived at this choice, the Chinese variations of the ideal type; and the implications for our central issue, namely the regime survival and regime change.

422

BAI GAO

The Authoritarian State Contrary to the conventional perception as a rigid, changeless entity, China’s authoritarian state was relatively new – an outcome of state building in the era of reform and opening. In order to understand the significance of China’s authoritarian state we need to compare it with its preceding form, the totalitarian state that dominated the country until the end of the Cultural Revolution. According to the orthodox definition, the totalitarian state was driven by ‘‘the totalitarian’s demand for constant agitation, enthusiasm and unanimity and a confrontational foreign policy posture towards an external enemy’’ (Sondrol, 1991, p. 602). The totalitarian state was purely political in nature because ‘‘ideological dedication to utopian futures substitutes for mere materialistic concerns’’ (Sondrol, 1991, p. 603). The totalitarian state extended to every sphere of people’s life, invading such traditionally private spheres as culture, religion, family life, and the relationship between a household and its neighbors. It even affected how free time was spent, and how choices of career and employments were made. An individual account was kept for every resident and employee. The totalitarian state was not subordinate to any stable legal system. The constitution was sufficiently general to grant the government a free hand (Kornai, 1992). In contrast, the authoritarian state lost the desire for a total mobilization and restructuring of society through mass movement. It no longer had the agenda to penetrate and change the thoughts and values of the people. The authoritarian state settled for political control. Rather than forcing people to compete for dedication, the authoritarian state allowed a certain amount of token opposition and the existence of ‘‘limited pluralism’’ (Sondrol, 1991, p. 603). In this sense, the authoritarian state can be considered as more progressive than the totalitarian state. Unless this concept is understood, one cannot fully apprehend why the disaster of the Cultural Revolution has not cost the CCP the loss of political support from the Chinese people since the late 1970s. The turning point of such a transition in China was the 3rd plenum of the 11th national committee of the CCP in late 1978. At this historic meeting, the CCP shifted its policy paradigm from class struggle to economic development. In the past, the legitimacy of the party state was primarily sustained by the personal charisma of Chairman Mao and the Marxist– Leninist ideology of class struggle. From 1978, however, the party state decided to shift the basis of its legitimacy toward economic performance. Between fall 1978 and spring 1979, Deng Xiaoping envisioned the grand


423

strategy of ‘‘one central goal, two basic points’’ for the CCP. Under the guidance of this strategy the party state began to launch its reform and opening programs. To release market forces it loosened its control over the economy and society. In this sense, the adoption of the ‘‘one central goal, two basic points’’ strategy directly triggered the transformation of China’s party state from totalitarian to authoritarian. One of the most distinctive characteristics of China’s authoritarian state in comparison with others is its economic focus. According to the existing literature, part of the reason for the third wave of democratization was that the old authoritarian states in Southern Europe and Latin America could no longer rely on economic performance to maintain their legitimacy after the collapse of the Bretton Woods system and the first Oil Shock in the early 1970s (Huntington, 1991). In contrast, the transition of the party state in China from totalitarian to authoritarian in the late 1970s was driven by the belief that, affected by the Marxist–Leninist ideology of class struggle, the party state had never taken the issue of economic performance seriously and that its strong state capacities should enable it to outperform capitalist competitors when it concentrated its efforts in economic development. Furthermore, in order to achieve the goal of improving economic performance, furthermore, the authoritarian state was determined to embrace the market forces that used to be considered as capitalist principles. The transformation from totalitarian to authoritarian and the new focus on economic performance helped the party state survive the new environment brought about by globalization. Since the late 1970s China has confronted the challenges from the third wave of democratization twice: one at the beginning of the reform, represented by the Xidan Democratic Wall movement in 1978–1979; and the other on the eve of the collapse of the Berlin Wall, represented by the Tiananmen Democratic Movement in 1989. The party state took a hard stand against both challenges, cracking down on these democratic movements. Without the loosened control over the economy and society and the focus on economic growth, these crackdowns would have provoked the Chinese who had been highly mobilized by political missions in the Mao era, and China might not have been able to avoid the alternative of democratization. The authoritarian state in China is not just a static entity that sticks to old traditions. On the contrary, it has demonstrated great adaptive capacity. Facing the emerging new middle class, the party state has tried to consolidate its power base by co-opting the new elites into the CCP.

424

BAI GAO

The Neoliberal State The neoliberal state regards the mobility of capital between sectors, regions, and countries as the most essential factor to generate economic growth, and asserts that ‘‘all barriers to that free movement (such as tariffs, punitive taxation arrangements, planning and environmental controls, or other locational impediments) have to be removed’’ (Harvey, 2005, p. 66). In the neoliberal state, the assignment of private property rights is considered the best way to protect against the tragedy of the commons. The neoliberal state emphasizes that ‘‘each individual is held responsible and accountable for his or her own actions and well-being. This principle extends into the realms of welfare, education, health care, and even pensions y. Individual success or failure are interpreted in terms of entrepreneurial virtues or personal failings rather than being attributed to any systematic property’’ (Harvey, 2005, p. 65). In the policy paradigm of a neoliberal state, liberalization, privatization, and deregulation are the common components. Neoliberalism, which has swept the world in the past three decades, is often considered to have been invented in the West, especially in the AngloSaxon world. The truth is its contemporary origin was as much Chinese as Western. Today’s Chinese are still very sensitive about the concept of neoliberalism in domestic political discourse because it has a strong connotation of bourgeoisie spiritual pollution. Nevertheless, among the major milestones in the contemporary history of neoliberalism, the earliest was the CCP’s decision to launch its reform and opening at the 3rd Plenum of the 11th National Congress in late 1978 under the de facto leadership of Deng Xiaoping, well ahead of Margaret Thatcher’s election as British Prime Minister in May 1979, Paul Volcker’s selection as the chairman of the Federal Reserve in July 1979, and Ronald Reagan’s election as the U.S. president in 1980 and their subsequent policy innovations (Harvey, 2005). The rise of neoliberalism in China was driven by the party state’s grand strategy of ‘‘one central goal, two basic points.’’ After the Cultural Revolution the CCP came to believe that the poor performance of the planned economy resulted from both the state regulations that aimed at restraining market forces and the inefficiencies associated with various institutional arrangements of social protection. To improve its economic performance in order to strengthen its legitimacy, the party state fully embraced market-oriented ideology. In the historical context of the postCultural Revolution era, neoliberalism became a political ally and useful policy tool for the authoritarian state to promote economic growth in an effort to boost its legitimacy for political dominance. The neoliberal

425


assertions of the market principle spoke to the heart of the party state that had to bear the burden of providing these services in the prereform era. A unique characteristic of the neoliberal state in China is its relation with the authoritarian state. When states in other developing countries began to embrace neoliberalism represented by the Washington consensus, most had engaged themselves in the third wave of democratization. As a result, domestic resistance in democratic politics often became a political barrier to the implementation of neoliberal policies. In the Chinese case, however, the authoritarian state often used its political power to implement neoliberal policies. At the same time, the authoritarian state has never allowed neoliberalism to completely run its own course because after all the latter was just a political tool of the former. The impact of the neoliberal state on the survival of China’s party state is mixed. On the one hand, policies affected by neoliberal ideology have effectively released market forces in China. Domestically, the prices for more than 90 percent of goods and merchandise are determined by the market rather than government planning. SOEs, which used to dominate the national economy, today constitute only 20 percent of the total number of companies. The openness of China’s domestic markets to foreign products has gone far beyond many of its East Asian neighbors. FDI has played a significant role in China’s economy, contributing to around 60 percent of China’s total trade and between 35 and 40 percent of China’s total GDP. In terms of efficiency and productivity, China has reached a level with which few developing economies and former socialist economies can compete. These achievements have significantly improved the economic performance of the party state and helped strengthen its legitimacy. On the other hand, the release of market forces has also brought economic inequality to a dangerous level. In many cases, issues related to economic inequality have become the seeds for social tension and political conflict in contemporary China. Affected by the neoliberal policies, China has been left far behind by its former socialist counterparts that had joined the third wave of democratization and marketization in providing public goods and services. These policies have created new legitimacy issues for the party state in China. The CCP began to address the issues of economic inequality under the policy paradigm of harmonious society only a few years ago.

The Developmental State The classic developmental state represented by Japan and South Korea forms a dichotomy with the regulatory state represented by the United States. The

426

BAI GAO

regulatory state concerns itself with rules of economic competition, but not with substance matters. Which industries prosper and which industries decline are determined by the market. In contrast, the developmental state, which tends to emerge in latecomer countries in industrialization, directly takes on the developmental functions. It gives greatest precedence to industrial policy with the goal of promoting a structure of domestic industry that would optimize the country’s international competitiveness (Johnson, 1982). Different from the neoliberal ideology that regards promoting fair competition among individual firms in the market as the top priority, developmental ideology approaches the economic phenomena with a unit of analysis of nation state. In contrast to neoliberal ideology that emphasizes the interests of consumers, developmental ideology demonstrates a strong orientation toward production, giving producers a higher priority. Developmental ideology takes a strategic view of the economy; it favors organized competition and rejects the profit principle in corporate governance (Gao, 1997). The CCP’s grand strategy of ‘‘one central goal, two basic points’’ coordinated well with two distinctive characteristics of its East Asian counterparts: the practice of industrial policy to promote a desired industrial structure, and the aggressive promotion of exports. Since the central goal for the party state was to promote economic growth, the states at both central and local levels have used various capacities developed in the planned economy in this new mission. In order to attract FDI, the developmental states at both levels have offered various incentives to multinational corporations, such as tax reduction and low-priced or even free lands. They have also worked hard to build infrastructure, improve services, and ensure the supply of a cheap yet better-educated labor force. The state has also offered various incentives for domestic companies to promote exports through its taxation, monetary, trade, and exchange rate policies. Many local governments have done even more, helping local companies sell their products to international markets. Without this assistance and support from the development state, China would not have become the world factory. Because of the neoliberal facet, the Chinese developmental state clearly differentiates itself from its East Asian counterparts. Instead of protecting domestic markets as other development states in East Asia did in their early stages of industrialization, China’s developmental state aggressively opened its domestic markets to the inflows of foreign capital and products. Since 1993 China has been the biggest FDI recipient among developing countries, and has attracted more than $700 billion FDI. Today China is not only the second-largest exporting country in the world, but also the third-largest importing country. Rather than promoting the exports of value-added

427


products by domestic firms that have intellectual property rights and independent brands as Japan and South Korea have done, China concentrated its comparative advantage in cheap labor, willing to work at the low end of the global value chain of production operated by multinational corporations, often through original equipment manufacturing and processing trade. The multinational corporations have contributed 60 percent of China’s total exports and imports. In these dimensions the Chinese model of the developmental state demonstrates significant variation from the classic model of the developmental state in East Asia represented by Japan and South Korea. In recent years, the developmental state in China has moved closer to the classic East Asian model. Much of its earlier efforts tended to concentrate in labor-intensive industries and processing trade. From 2004, however, the developmental state in China began to shift its policy paradigm toward promoting independent innovation under the strong international political pressures to appreciate the value of China’s currency and anticipating the declining competitiveness of China’s labor-intensive industries. In the past several years the state has significantly increased the inputs to research and development, and it now sponsors 5,000 Ph.D. students to study at the world’s leading universities every year in an effort to improve its human capital. A major change in the state’s policy toward the major SOEs has been to force them to create their own independent brands, indicating a departure from its early policy that focused on creating joint ventures to produce foreign brands. The policies adopted by the developmental state in China have helped the country compete in the globalized economy by linking China to the global production network and by borrowing external dynamics for the growth of the domestic economy. The significantly increased volume in exports has helped promote domestic economic growth, and this in turn has strengthened the CCP’s legitimacy.

The Predatory State The predatory state refers to the state that extracts large amounts of otherwise investable surplus and provides little public goods; ‘‘those who control the state apparatus seem to plunder without any more regard for the welfare of the citizenry than a predator has for the welfare for its prey’’ (Evans, 1989, p. 562). In the typical predatory state, state services are exchanges between those who occupy the office and their political supporters. Government officials can distribute resources directly to supporters in the

428

BAI GAO

form of subsidies, cheap loans, jobs, contracts, dams, water, and so on. Alternatively, they can also use their lawmaking authority to create rents by restricting the ability of market forces to operate by rationing imports, licensing a limited number of producers, prohibiting the introduction of new products, and so on. These officials also extract a share of the rent for themselves (Evans, 1989). As a result, corruption represents a type of private behavior that exploits public authority for essentially private material motivation and advantage. The symbiotic relationship involved in rent creation is self-reinforcing. Supporters whose original economic power derived from productive activities are likely to become increasingly dependent on rents and therefore increasingly committed to the expansion of ‘‘rent haven.’’ In this sense, the rise of the predatory state results from rent creations and rent seeking by government officials. All government officials require political supporters in order to survive, and they in turn must provide incentives sufficient to retain political support (Evans, 1989; Sondrol, 1991). Thus, the special phenomenon of the strong allying with the strong (among political, economic, and academic spheres) is not just a form of co-optation, as analysts who examine the situation from the angle of the authoritarian state would argue. If we examine the situation from the angle of the predatory state, it is also an exchange between those who occupy government offices but need political support and financial gain, and those who control the business world but need favorable treatment by the government and can offer financial benefits to those who help them. In contrast to other angles of the party state discussed in this chapter, the rise of the predatory state never occurred by design but as the unintended consequence of state building. Nevertheless, it was still closely related to other angles of the party state. Under the strong pressure of achieving the central goal of promoting economic growth, the promotion of government and party officials was often linked directly to the growth rate of local GDP. The corporatist mechanisms have served to link party and government officials with business elites, creating many opportunities for rent seeking. The lack of checks and balances and freedom of speech under authoritarian rules make the corruptions uncontrollable. The fact that the state still exercises too much control in the economy has also provided the political power for making economic gains. The strong pressures for economic growth have often legitimized favorable policies to private companies. The rise of the predatory state at the local level is largely an outcome of individual adaptation to the era of economic growth and China’s opening to international markets. Under authoritarian rule, political power can be

429


utilized for essential private ends. ‘‘Lacking the constraining ideological goals and roles which normatively bind totalitarians, the authoritarians treat the national patrimony essentially as a huge private domain’’ (Sondrol, 1991, p. 603). In the Chinese context, the predatory state is a local-level phenomenon. Decentralization has much to do with the rise of the predatory state. Moreover, after the reform of the banking industry in 1994, local governments could no longer obtain capital at their own discretion. They found a new cash cow, selling land for commercial development. As a result, land is usually the sphere in which most predatory behavior took place in contemporary China (Pei, 2006). Apparently the predatory behavior of various local governments has become the most destabilizing factor to the party state because this behavior often stimulated collective protests from the locals. In many cases these protests ended in violence.

The Refined Socialist State The classic socialist state directly holds ownership of firms and exercises property rights of control over SOEs. Moreover, the SOEs occupy the ‘‘commanding heights’’ of the socialist economy, the positions that allow the state to dominate the important sectors in the economy such as mining, energy, transportation, banking, insurance, manufacturing, and foreign trade. In the socialist ideology, SOEs belong to ‘‘the whole of the people’’ or ‘‘the whole of society.’’ As the owner of SOEs, the socialist state holds the rights of disposal of residual income from their utilizations in its own budget (Kornai, 1992). Such national ownership is closely associated with state-led coordination in the production process. According to orthodox thinking, both classic socialism and capitalism represent a coherent system in which the nature of ownership is compatible with the pattern of coordination. An affinity exists both between public ownership and bureaucratic coordination, and between private ownership and market coordination. In both systems the two components of the pair reinforce each other (Kornai, 1992). In contrast, the refined socialist state in China no longer controls all firms; in fact, it sold off most of the medium and small SOEs. At present the central government controls only a group of large corporations. The state has also abandoned government planning. It no longer intervenes in the coordination of the production process in the SOEs, and has substantially reduced the scope of economic intervention. Nevertheless, the state is still determined to maintain the dominance of the SOEs in industries that are

430

BAI GAO

crucially important to the economy. It invests heavily in these industries, constantly improving the competitiveness of these SOEs. It helps these SOEs to maintain monopoly or oligopoly positions in their respective industries, thus generating high levels of profits. At the same time, the state still controls the prices of the products and services offered by the SOEs to serve its policy objectives. The SOEs remain the most important source of tax revenue for the state; these tax revenues are indispensable to the party state for maintaining its political dominance. The rise of the refined socialist state experienced a twist resulting from the internal conflicts contained in the grand strategy of ‘‘one central goal, two basic points.’’ Affected by the reform-and-opening side in the two basic points, the governance of the Chinese economy reform in the 1980s was accompanied by a process of decentralization, which aimed at mobilizing the political support of local elites for the reform–and-opening program (Shirk, 2007). As a result, the power of the central government was weakened significantly. The central government revenues as a percentage of GNP fell from 31.2 in 1978 to 14.7 in 1992; according to the pace of decline at the time it went as low as 11.3 by 2000 (Fewsmith, 2001). After the crackdown on the Tiananmen democratic movement and the fall of the Berlin Wall, the party state began to realize that maintaining the state’s extract power would be crucial to the regime’s survival, and the marketoriented economic reform could undermine the CCP’s political dominance if the state became too weak. It was argued that ‘‘it seems possible that once the ‘political strong man’ at the center (for example, Deng Xiaoping) passes away, a situation like that in Yugoslavia after the death of Tito could take place in China’’ (cited by Fewsmith, 2001, p. 135). In the early 1990s the party state recognized that globalization did not mean the end of the state. In the Chinese context, ‘‘reform does not necessarily mean weakening state capacity, and a market economy does not necessarily connote the elimination of state intervention; modernization certainly does not need a weak central government’’ (Hu and Wang, cited in Fewsmith, 2001, p. 136). In other words, the party state was determined to abandon the planned economy. Nevertheless, it has never attempted to abandon all SOEs because SOEs were an important instrument for the party state for controlling critical resources that it would need in its struggle for political survival. A refined socialist state began to emerge in China in the late 1990s through an aggressive reform of the SOEs called ‘‘grasp the bigs and abandon the smalls.’’ The party state had two goals in this reform: releasing its heavy financial burden by selling most medium and small SOEs; and strengthening the positions of some large SOEs in 16 industries that were

431


critical to the Chinese economy. Through such a reform, the party state was able to shift the resources from the social protection mechanisms associated with the old SOEs toward the production-oriented infrastructure building. The profound implications of this reform for the party state became increasingly clear after China joined WTO in 2001. The heavy investment in the 16 industries greatly helped China to attract a new wave of inflow of FDI in the rigorous international competition for mobile capital. More importantly, as China emerged as the world factory, sustained by a huge amount of inflow of FDI, the SOEs in the 16 industries that provided infrastructure services became the backbone of the world factory. As these FDI firms helped China expand its trade more than 30 percent and grow its economy more than 10 percent each year after it joined WTO, these SOEs also made a huge amount of profits, more than 1 trillion yuan in 2007. The outstanding performance of the SOEs in the past several years brought about a record inflow of tax revenue to the state. In 2005–2007, the tax revenue of the central government increased at a pace of 1 trillion extra yuan per year. This has substantially strengthened the financial position of the party state in handling various kinds of crises. In 2008 alone we saw large amounts of money spent on the storm in January, the earthquake in May, the Olympics in August, and the rescue effort during the financial crisis in October. Without solid control of the tax base through these SOEs in strategic industries, the party state would not have had the financial power to respond to these unprecedented crises. In this sense, the refined socialist state has strengthened the party state’s capacity to respond to crises. On the other hand, these SOEs hold monopoly or oligopoly positions in the infrastructure-related industries such as energy, utility, communication, and transportation. Practicing a favorable competition policy toward these SOEs often occurs at the cost of victimizing the interests of ordinary consumers who have often suffered from the high prices of these services. In addition, the income gap between the employees who work in these SOEs and others has also contributed to the rise of economic inequality and generated dissatisfaction toward the state.

The Corporatist State An ideal type of corporatist state recognizes only one organization as the sole representative of the interests of individuals, companies, or institutions that comprise that organization’s assigned constituency in one specific sector. These sector-based associations often get involved in the

432

BAI GAO

policy-making processes and help implement the state policy on the government’s behalf. Corporatist mechanisms emphasize organized consensus and cooperation. Harmony is often the catchword of a corporate system, regardless of whether this harmony is truly consensual or imposed from above. Generally speaking, corporatist solutions are apt to be sought during wartime or by regimes that stress rapid economic development, guided and spurred by a state simultaneously dedicated to enforcing political and social stability (Unger & Chan, 1995). The national mobilization for the total war in the 1930s and 1940, for example, triggered the rise of corporatist mechanisms in major industrialized countries. In order to mobilize citizens of underprivileged social groups to take on their political responsibility in fighting for national survival, they had to be treated equally. As a result, the war functioned to improve political and economic equality in these countries (Yamanouchi, 1995). This is why the warfare state is often associated with the welfare state in the existing literature. Corporatist mechanisms, however, ‘‘do not define a political system; a polity can contain corporatist elements and at the same time be a dictatorial Communist Party regime, or an authoritarian Third World government, or a liberal parliamentarian state’’ (Unger & Chan, 1995, p. 31). In the comparative political economy literature, corporatism in Western Europe is often depicted as counterpoised to pluralist interest-group models of organization and free market forces in the United States that rely on divisive competitions and conflict in politics. In recent decades, various corporatist institutional arrangements have occurred in both democratic and nondemocratic regimes. The increasing impact of corporatist mechanisms on Chinese politics was partly a result of the party state’s strategy of ‘‘one central goal, two basic points.’’ The major goal of this strategy was to carry out economic reform and opening, and at the same time maintaining authoritarian rules. Without the introduction of the corporatist element, however, this goal was deemed difficult to achieve. Before the economic reform, Chinese society was highly organized and politically constructed in accordance to Mao’s perceptions of what a society should be, and realized by forceful organizational weapons. Relying on such a social structure, the party state was able to mobilize social groups onto the political arena and created a new power base to implement its political missions in land reform, collectivization, and nationalization of business and commerce. Although the party state began to transform itself from totalitarian to authoritarian after the Cultural Revolution, exemplified by the shift of the


433

party’s focus from ‘‘the continuing revolution under the proletarian dictatorship’’ to four modernizations at the 3rd plenum of the 11th Central Committee, it was not accompanied by a correspondent transformation in China’s social structure. The Tiananmen democratic movement and the fall of the former Soviet Union and Eastern European countries at the end of the 1980s, however, indicate that without a private exit the members of society had to fight for what they wanted in a highly politicized public arena; this could bring major challenges to the stability of the regime. To reduce the potential risks, the party state began to build an interest-based social order (Zheng, 2004). Put differently, an interest-based social order was an inevitable outcome of the rise of a market economy during the reform and opening. Corporatism became an indispensable component in this interest-based social order. On the one hand, the newly emerged interest groups in China demanded political representation of their interests in domestic politics. On the other hand, the party state would not choose the option of democratization. Under such circumstances corporatist mechanisms became a good compromise. The notion of state corporatism is less threatening to the party state in China than democratization: instead of having autonomy from the state, corporate groups in China are embedded in the state, ‘‘where they can be manipulated, their leaders replaced, and their finances controlled’’ (Dickson, 2003, p. 67). In this sense, state corporatism in China serves not as a mechanism for further strengthening the state’s grip over the economy and society; instead, it has served as the reverse mechanism through which the state’s grip can be gradually loosened. The rise of state corporatism represents the changing nature of the party state in China from totalitarian to authoritarian (Unger & Chan, 1995). The impact of state corporatism on the party state has been mixed. Generally speaking, the corporatist mechanisms have helped the party state consolidate its power base among the elites. Over the past two decades, the policy-making process at the national level and local level began to witness increasing opportunities for diversified interests to voice their concerns. This helped the party state reduce political risk. At the same time, however, these corporatist mechanisms have also led to the rise of the predatory state at the local level, creating tension and conflict that often weaken the legitimacy of the party state. Since the corporatist state in China often sustained the strong links among the elites, business elites with strong political connections could borrow the power of the authoritarian state in achieving their economic gains.

434

BAI GAO

CONCLUSION This chapter shows that the concept of the Rubik’s Cube state provides an important clue to solving the China puzzle because its multidimensional capacities developed in the state-building process, especially those associated with economic dimensions, have helped China’s party state survive the challenges of globalization. In addition to the authoritarian, predatory, and corporatist angles that have drawn much attention in the existing literature, the Rubik’s Cube state incorporates three economic angles into our conceptualization of the party state in China. Each of these three angles deepens our understanding of the Chinese political economy: the neoliberal angle tells us why the Chinese economy has been more dynamic than other former socialist economies in the past 30 years; the developmental angle reveals the source of China’s competitive advantage in competing for FDI and upgrading its economic structure in comparison with other developing countries; and the refined-socialist angle adds to our understanding of why the party state has more resources for supporting its policy objectives than other states. More importantly, these economic angles demonstrate that the precondition for the third wave of democratization asserted by Huntington, namely that an authoritarian regime can no longer rely on its economic performance to sustain its legitimacy, has never existed in China. The two seemingly conflicting images discussed at the beginning of this chapter are in fact complementary because the party state’s success in promoting economic growth has directly strengthened its legitimacy and power base. The strong connection between the state capacities in promoting economic growth and the resilience of the authoritarian state has not only raised serious questions about the structural perspective that relies on the middle class and level of education to predict democratization, but also added more analytical components to the agency perspective that highlights only the political responses from the party state. The performance of the Rubik’s Cube state highlights the importance of state building in the process of globalization. Without embracing neoliberalism, the party state would not be able to release market forces, open China’s domestic markets to foreign investments, and lead the country to participate in the international division of labor that transformed China into the world factory. Without adding the developmental angle, the party state would not be able to win the international competition for mobile capital, could suffer badly from being locked in at the lower end of the global production chain, and would not be able to achieve aggressive industrial upgrading. Without building a refined-socialist component, the


435

party state could remain poor while the country becomes rich since it might run short of the resources for crisis prevention. The Chinese case confirms that state intervention can develop in new directions, through new instruments, and on behalf of new objectives (Levy, 2006b). The Rubik’s Cube state represents a distinct pattern of state response to the challenge of globalization. Such a state has never seen itself as a passive recipient of the changes in the environment. Instead, it has always taken preemptive actions to build new state capacities in order to meet the various challenges that emerge. These newly developed state capacities have helped the party state to not only refuse the call for democratization, but also lead China to become the third-largest economy and trading nation in the world. This analysis also shows that the dominant political party’s responding strategy to the challenge of globalization plays a crucial role for statebuilding efforts in China. The seemingly self-contradictory principle of ‘‘one central goal, two basic points’’ has allowed the party state to fully benefit from the process of globalization, and at the same time counter any factor coming with globalization that could weaken its own political dominance. By focusing the regime’s attention on economic performance, the party state has been able to create a political environment in which internal groups with different ideologies and ideas can coexist within the CCP. Tension and conflict still provide dynamics for Chinese politics. Nevertheless, they are no longer allowed to interrupt the process of economic growth as they often did in the regime’s totalitarian stage. Contrary to the conventional wisdom that the authoritarian state in China is a rigid political corpse that refuses to change, this elastic political structure created under the CCP’s grand strategy has enabled the party state to develop new capacities without ideological constraints, borrowing wisdom from various models of the state. What are the implications of the ongoing global economic crisis to the Rubik’s Cube state? China’s responses to the ongoing global financial meltdown show that some capacities of the Rubik’s Cube state can be very useful at a time of crisis. The quick adoption of a huge stimulus plan and the rapidly improving bank loans demonstrate the effectiveness of the refined socialist state; the aggressive investments in overseas energy and raw materials by China’s SOEs and the adoption of the revitalization plan for 10 major industries portray the strengths of the developmental state; the major reform of rural areas’ health care system and wide coverage of a rescue package for the poor in urban areas highlight the corporatist strength of the party state. Of course, the tightened policies toward the media illustrate the increasing control of the authoritarian state. For the

436

BAI GAO

time being, the neoliberal angle of the state retreated; nevertheless, it by no means signals the end of the market orientation of public policies. Does the Rubik’s Cube state represent a new model of political development replacing democratization? The jury is still out on this question. We do know, however, that China still lacks two basic conditions for a peaceful transition toward democracy – equality and capital mobility – that have been stressed in the recent literature (Boix, 2003). On the one hand, China suffers from a very high level of economic inequality and its Gini index between urban areas and rural areas is considered as the highest in the world. On the other hand, capital mobility remains under control and the ongoing global financial crisis will further delay China’s opening of capital account. Without these two conditions the elites in China are less likely to accept a proposal for democratization. Moreover, China is not yet in a situation that could lead to democratization through an alternative channel. Democratization may also take place when the costs and risks of cracking down on the demand for democratization become too high, and the elites in an authoritarian state agree to compromise in an effort to avoid a social revolution (Acemoglu and Robinson, 2006). Economic crises are often the crucial moments at which the elites are forced to make such choices. There are increasing signals of economic hard times in China, as 20 million migrant workers, 10 million urban residents, and 1 million new college graduates are now unemployed, and the economic contractions of China’s major trading partners have led to the decrease of the country’s imports and exports to 43 percent and 17 percent, respectively, in January 2009. The increasingly violent protests in China also indicate that the cost of cracking down on domestic demands for democracy has become very high, and the risks of a failure of such cracking down in generating a social revolution have increased substantially. However, the party state in China is still in a much better position than many other states to fight this global economic downturn. Since the Rubik’s Cube state is a product of a specific historical environment, it has its own limitations. So far, the adaptive capacities of the Rubik’s Cube state have been sustained by the upward trend of globalization. The CCP’s grand strategy of one central goal, two basic points has been based on the assumption that the opportunities provided by the process of globalization would enable China to achieve a fast pace of economic growth, and as long as it works the party state would be able to strengthen its legitimacy. Nevertheless, when globalization hits a major downturn or this ongoing global economic crisis becomes a long-lasting one, the question remains whether the party state is prepared both politically and

437


ideologically to meet future challenges. The political structure that has allowed the coexistence of diverse interest groups within the party state is beneficial when economic growth can be maintained at a high speed because this releases the dynamics of the market forces and sustains entrepreneurship in the Chinese economy. When the country needs to turn toward a totally new direction, however, such a political structure can be a major obstacle because it undermines the leadership that is badly needed for a major transformation. In this sense, global depression may still create a situation in which democratization becomes an option.

REFERENCES Acemoglu, D., & Robinson, J. A. (2006). Economic origins of dictatorship and democracy. New York: Cambridge University Press. Boix, C. (2003). Democracy and redistribution. New York: Cambridge University Press. Chang, G. G. (2001). The coming collapse of China. New York: Random House. Colomer, J. M. (2000). Strategic transitions: Game theory and democratization. Baltimore, MA: Johns Hopkins University Press. Dickson, B. J. (2003). Red capitalists in China: The party, private entrepreneurs, and prospects for political change. New York: Cambridge University Press. Evans, P. (1989). Predatory, developmental and other apparatuses: A comparative political economy perspective on the third world state. Sociological Forum, 4(4), 561–587. Fewsmith, J. (2001). China since Tiananmen: The politics of transition. New York: Cambridge University Press. Fukuyama, F. (2004). State building: Governance and world order in the 21st century. Ithaca, NY: Cornell University Press. Gamble, A. (1994). The free economy and the strong state. Houndmills, U.K.: Macmillan. Gao, B. (1997). Economic ideology and Japanese industrial policy: Developmentalism from 1931 to 1965. New York: Cambridge University Press. Gilley, B. (2004). China’s democratic future: How it will happen and where it will lead. New York: Columbia University Press. Harvey, D. (2005). A brief history of neoliberalism. New York: Oxford University Press. Huang, J. (2008). Institutionalization of political succession in China: Progress and implications. In: C. Li (Ed.), China’s changing political landscape: Prospects for democracy (pp. 80–97). Washington, DC: Brookings Institution Press. Huntington, S. P. (1991). The third wave: Democratization in the late twentieth century. Norman: University of Oklahoma Press. Johnson, C. (1982). MITI and the Japanese miracle. Stanford: Stanford University Press. Kahler, M., & Lake, D. A. (2003). Governance in a global economy: Political authority in transition. Princeton, NJ: Princeton University Press. Kornai, J. (1992). The socialist system: The political economy of communism. Cambridge: Harvard University Press.

438

BAI GAO

Levy, J. D. (2006a). The state also rises: The roots of contemporary state activism. In: J. D. Levy (Ed.), The state after statism: New state activities in the age of liberalization (pp. 1–30). Cambridge: Harvard University Press. Levy, J. D. (2006b). The state after statism: From market direction to market support. In: J. D. Levy (Ed.), The state after statism: New state activities in the age of liberalization (pp. 367–394). Cambridge: Harvard University Press. Levy, J. D. (Ed.) (2006c). The state after statism: New state activities in the age of liberalization. Cambridge: Harvard University Press. Nathan, A. (2006). Present at the stagnation: Is China’s development stalled. Foreign Affairs (July/August), 177–182. Pan, W. (2006). Toward a consultative rule of law regime in China. In: S. Zhao (Ed.), Debating political reform in China: Rule of law vs. democratization (pp. 3–40). Armonk: M.E. Sharpe. Pei, M. (2006). China’s trapped transition: The limits of developmental autocracy. Cambridge: Harvard University Press. Richardson, J. (1994). Doing less by doing more: British government, 1979–1993. West European Politics, 17(3), 178–197. Shambaugh, D. (2008). China’s Communist Party: Atrophy and adaptation. Washington, DC: Woodrow Wilson Center Press. Shirk, S. L. (2007). China: Fragile superpower. New York: Oxford University Press. Sondrol, P. C. (October 1991). Totalitarian and authoritarian dictators: A comparison of Fidel Castro and Alfredo Stroessner. Journal of Latin American Studies, 23(3), 449–620. Tsai, K. S. (2007a). Capitalism without democracy: The private sector in contemporary china. Ithaca, NY: Cornell University Press. Tsai, L. L. (2007b). Accountability without democracy: Solidary groups and public goods provision in rural China. New York: Cambridge University Press. Unger, J., & Chan, A. (1995). China, corporatism, and the East Asian model. Australian Journal of Chinese Affairs, 33, 29–53. Weiss, L. (1998). The myth of powerless state. Ithaca, NY: Cornell University Press. Yamanouchi, Y. (1995). Hohoteki Joron: Soryokusen to shisutemu togo (On the Methodology: The Total War and the Integration of the System). In: Y. Yasushi (Ed.), Soryokusen to Geidaika (pp. 9–56). Tokyo: Kashiwa Shobo. Yu, K. (2008). Ideological change and incremental democracy in reform-era China. In: C. Li (Ed.), China’s changing political landscape: Prospects for democracy (pp. 44–60). Washington, DC: Brookings Institution Press. Zheng, Y. (2004). Globalization and state transformation in China. New York: Cambridge University Press.

Work and Organizations in China After Thirty Years of Transition (Research in the Sociology of Work)

Diversity in the Work Force, Volume 14 (Research in the Sociology of Work) (Research in the Sociology of Work)

Globalism Localism at Work, Volume 13 (Research in the Sociology of Work) (Research in the Sociology of Work)

Gender, Time, and Reduced Work (SUNY Series in the Sociology of Work and Organizations)

Years Work in English

NATO after thirty years

Personality and Work: Reconsidering the Role of Personality in Organizations

Personality and Work Reconsidering the Role of Personality in Organizations

Cambridge Handbook of Culture, Organizations, and Work

HRM, Work and Employment in China

Institutional Work: Actors and Agency in Institutional Studies of Organizations

Worker Participation: Current Research and Future Trends, Volume 16 (Research in the Sociology of Work)

Work, Employment & Transition

Spirituality, leadership, work and organizations

Under Construction: Work and Alienation in the Building Trades (Suny Series in Sociology at Work)

The Sociology of Gambling in China

Educational Work of Women's Organizations, 1890-1960

World of Work in Transition: Growing Social Inequalities

A Sociology of Work in Japan (Contemporary Japanese Society)

Transition and Development in China

Political Work: The Lifeline of All Work

Inside Organizations: Anthropologists at Work

The Democratisation of China (Routledge Studies on China in Transition)

Creatures - Thirty Years of Monsters

The Future of Work

Thirty Years in a Red House: A Memoir of Childhood and Youth in Communist China

The Work of Mourning

Creatures - Thirty Years of Monsters

Work and Organizations in China After Thirty Years of Transition (Research in the Sociology of Work)

Diversity in the Work Force, Volume 14 (Research in the Sociology of Work) (Research in the Sociology of Work)

Globalism Localism at Work, Volume 13 (Research in the Sociology of Work) (Research in the Sociology of Work)

Gender, Time, and Reduced Work (SUNY Series in the Sociology of Work and Organizations)

Years Work in English

NATO after thirty years

Personality and Work: Reconsidering the Role of Personality in Organizations

Personality and Work Reconsidering the Role of Personality in Organizations

Cambridge Handbook of Culture, Organizations, and Work

HRM, Work and Employment in China

Institutional Work: Actors and Agency in Institutional Studies of Organizations

Worker Participation: Current Research and Future Trends, Volume 16 (Research in the Sociology of Work)

Work, Employment & Transition

Spirituality, leadership, work and organizations

Under Construction: Work and Alienation in the Building Trades (Suny Series in Sociology at Work)

The Sociology of Gambling in China

Educational Work of Women's Organizations, 1890-1960

World of Work in Transition: Growing Social Inequalities

A Sociology of Work in Japan (Contemporary Japanese Society)

Transition and Development in China

Political Work: The Lifeline of All Work

Inside Organizations: Anthropologists at Work

The Democratisation of China (Routledge Studies on China in Transition)

Creatures - Thirty Years of Monsters

The Future of Work

Thirty Years in a Red House: A Memoir of Childhood and Youth in Communist China

The Work of Mourning

Creatures - Thirty Years of Monsters

Recommend Documents