Trade Policy, Inequality and Performance in Indian Manufacturing (Routledge Advances in South Asian Studies)

Trade Policy, Inequality and Performance in Indian Manufacturing The relationship between trade policy and economic pe...

Author: Kunal Sen

21 downloads 650 Views 2MB 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

Trade Policy, Inequality and Performance in Indian Manufacturing

The relationship between trade policy and economic performance is one of the oldest controversies in economic development. In recent years, there has been a revival of interest in the debate on the implications of trade reforms for productivity growth and domestic pricing behaviour due in part to the current phase of widespread trade liberalisation in developing countries and in part to developments in economic theory, notably endogenous growth theories and theories of imperfect competition. Also, there has been considerable interest in whether trade reforms can lead to higher wage inequality and regional inequality in developing countries. Both in academic and policy arenas the interest in international trade as a powerful positive force for reducing poverty in low-income countries has increased. In this book, the author examines the implications of trade reforms with specific reference to the Indian manufacturing sector. In particular, the book explores the evolution of regional and wage inequality, employment, productivity and prices from the import substitution phase of the 1970s to the period of radical reforms of the 1990s. The strength of the book is the careful and systematic examination of the various aspects of the trade–development nexus using rigorous empirical methods and a detailed data-set of Indian industries from 1975 to 2000. Economists in general and South Asian scholars in particular will find this thorough study interesting and useful. Kunal Sen is Professor of Development Economics and Policy at the Institute for Development Policy and Management, University of Manchester, UK. He specialises in macroeconomics and international trade, with over a decade of experience in research on economic reforms in India. Among his many publications are International Competitiveness, Investment and Finance: A Case-study of India (with A.G. Kumar and R. Vaidya, 2003), Saving, Investment and Growth in India (with P. Athukorala, 2002), and Economic Restructuring in East Asia and India: Perspectives on Policy Reform (with P. Agrawal, S. Gokarn, V. Mishra, and K.S. Parikh, 1995).

Impartial analytical accounts of the impact of India’s liberalisation programme on development including employment and poverty are rare. Kunal Sen’s book is an exception. It provides a highly readable and succinct account of the controversies on the impact of India’s trade policy on development carefully avoiding commitment to the cause of either the liberals or the left wing critics of India’s economic reforms. The conclusions of the book that the impact of trade reforms on employment in the organised manufacturing sector leaves much to be desired and that they may have increased wage inequalities are grounded in a carefully designed econometric model. The most useful and instructive part of the book, that should be of interest to both students of the Indian economy and policy makers, is its detailed discussion of the so called jobless growth phenomenon in India. No book on the economy of India can please one and all; Sen’s book comes close to doing so, mostly because of its scholarly analysis of the issues grounded in the extensive literature on growth and development. V.N. Balasubramanyam, Professor of Development Economics (Emeritus) Department of Economics, Lancaster University Management School, UK Kunal Sen’s book is distinguished from the vast majority of numerous recent books on India by its depth of research and analysis. He provides the most detailed and systematic quantitative analysis of how India’s trade liberalization in the 1980s and 1990s impacted efficiency, growth and poverty outcomes. The book—a must-read for all those interested in the impact of India’s economic reforms—will place Sen among the leading trade economists studying India. Arvind Panagariya, Professor of Economic & Jagdish Bhagwati Professor of Indian Political Economy Columbia University, New York, USA

Routledge Advances in South Asian Studies Edited by Subrata K. Mitra South Asia Institute, University of Heidelberg, Germany

South Asia, with its burgeoning, ethnically diverse population, soaring economies, and nuclear weapons, is an increasingly important region in the global context. The series, which builds on this complex, dynamic and volatile area, features innovative and original research on the region as a whole or on the countries. Its scope extends to scholarly works drawing on history, politics, development studies, sociology and economics of individual countries from the region as well as those that take an interdisciplinary and comparative approach to the area as a whole or to a comparison of two or more countries from this region. In terms of theory and method, rather than basing itself on any one orthodoxy, the series draws broadly on the insights germane to area studies, as well as the tool kit of the social sciences in general, emphasising comparison, the analysis of the structure and processes, and the application of qualitative and quantitative methods. The series welcomes submissions from established authors in the field as well as from young authors who have recently completed their doctoral dissertations. Perception, Politics and Security in South Asia The compound crisis of 1990 P.R. Chari, Pervaiz Iqbal Cheema and Stephen Philip Cohen Coalition Politics and Hindu Nationalism Edited by Katharine Adeney and Lawrence Saez The Puzzle of India’s Governance Culture, context and comparative theory Subrata K. Mitra India’s Nuclear Bomb and National Security Karsten Frey Starvation and India’s Democracy Dan Banik Parliamentary Control and Government Accountability in South Asia A comparative analysis of Bangladesh, India and Sri Lanka Taiabur Rahman

Political Mobilisation and Democracy in India States of emergency Vernon Hewitt Military Control in Pakistan The parallel state Mazhar Aziz Sikh Nationalism and Identity in a Global Age Giorgio Shani The Tibetan Government-in-Exile Politics at large Stephanie Roemer Trade Policy, Inequality and Performance in Indian Manufacturing Kunal Sen

Trade Policy, Inequality and Performance in Indian Manufacturing

Kunal Sen

First published 2009 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Simultaneously published in the USA and Canada by Routledge 270 Madison Ave, New York, NY 10016 Routledge is an imprint of the Taylor & Francis Group, an informa business This edition published in the Taylor & Francis e-Library, 2008. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” © 2009 Kunal Sen All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data Sen, Kunal 1961– Trade policy, inequality and performance in Indian manufacturing / Kunal Sen. p. cm. – (Routledge advances in South Asian studies) Includes bibliographical references and index. 1. India – Commercial policy. 2. Manufacturing industries – India. 3. Income distribution – India. I. Title. HF1589.S3893 2008 382′.30954–dc22 2008001047 ISBN 0-203-89437-5 Master e-book ISBN

ISBN10: 0–415–41335–4 (hbk) ISBN10: 0–203–89437–5 (ebk) ISBN13: 978–0–415–41335–0 (hbk) ISBN13: 978–0–203–89437–8 (ebk)

In memory of my father

Contents

List of figures List of tables Preface 1

Trade policy, inequality and performance – the key analytical issues

xi xiii xv

1

2

Trade policy in India

14

3

Growth and structural change in Indian manufacturing, 1975–99

30

4

Trade policy and economic growth

47

5

Trade policy, productivity and prices

64

6

International trade and employment

79

7

Trade policy and wage inequality

99

8

Trade policy and regional inequality

120

9

Conclusion

134

Appendix Notes References Index

141 150 156 165

Figures

2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.10 2.11 2.12 2.13 2.14 2.15 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8

Openness, exports plus imports of goods and services, and goods only as ratios of GDP, India Manufacturing exports and imports as shares of total merchandise exports and imports, India Manufacturing trade balance, India The nominal exchange rate, Indian rupee to US dollar, India Effective Rates of Protection (ERP) by sector, India Effective Rates of Protection (ERP) in the capital goods sector, by industry, India Effective Rates of Protection (ERP) in the intermediate goods sector, by industry, India Effective Rates of Protection (ERP) in the consumer goods sector, by industry, India Import Coverage Ratios (ICR) by sector, India Import Coverage Ratios (ICR) in the capital goods sector, by industry, India Import Coverage Ratios (ICR) in the intermediate goods sector, by industry, India Import Coverage Ratios (ICR) in the consumer goods sector, by industry, India Changes in the price ratio, capital goods sector, India Changes in the price ratio, intermediate goods sector, India Changes in the price ratio, consumer goods sector, India Overall manufacturing employment in India Aggregate real wages in India Aggregate labour productivity in India Aggregate unit labour costs in India Number of factories in India Average factory size in India Manufacturing exports, India Manufacturing imports, India

17 18 19 19 22 22 23 24 25 25 26 26 27 27 28 30 31 32 32 39 39 41 43

xii

Figures

3.9

Index of intra-industry trade for capital and intermediate goods, India Growth of Indian GDP, 1955–2003 The investment rate and its components, 1955–2003, India Total fixed investment and its components, 1955–2003, India Total public fixed investment and its components, 1955–2003, India Total private fixed investment and its components, 1955–2003, India Total private corporate fixed investment and its components, 1955–2003, India Total household fixed investment and its components, 1955–2003, India The relative price of equipment and structures, 1955–2003, India The relative price of equipment in 1980 for India and other developing countries Aggregate Total Factor Productivity growth in Indian manufacturing Aggregate price-cost margin in Indian manufacturing Factor content of India’s manufacturing exports Factor content of India’s manufacturing imports Employment coefficients, exports and import-competing production, India Contribution of wearing apparel to change in employment coefficient, India Contribution of other manufacturing goods (including jewellery, sporting goods and toys) to the change in employment coefficient, India Wage inequality in Indian manufacturing – aggregate Relative employment of skilled workers in Indian manufacturing – aggregate Reduction in Effective Rates of Protection (ERP) between 1980 and 1999 and the share of unskilled workers in 1980, India Relationship between unskilled employment growth and skill intensity, India Relationship between skilled employment growth and skill intensity, India

4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 5.1 5.2 6.1 6.2 6.3 6.4 6.5

7.1 7.2 7.3

7.4 7.5

43 49 50 51 52 53 54 55 56 62 70 71 88 89 92 93

94 109 110

110 115 115

Tables

3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 3.11 3.12 4.1 4.2 4.3 4.4 4.5 4.6 4.7 5.1 5.2 5.3

5.4 5.5

Overall trends – Indian manufacturing Structure of employment, India Employment growth, India Real wage growth, India Structure of female employment, India Structure of production, India Growth of real value added, India Share of factories, India Structure of Indian exports Structure of Indian imports Import penetration ratios, India Export orientation ratios, India Growth rates of output, by sector and aggregate Share of major sectors in GDP, India Gross investment and its components as a per cent of GDP, India Components of fixed investment by sector as a per cent of GDP, India Components of private fixed investment by sector as a per cent of GDP, India Summary statistics of variables used in the empirical analysis, India Regression results – economic growth – India Price-cost margins and total factor productivity growth, India Summary statistics, Indian manufacturing Total Factor Productivity – regression estimates of equation (5.8a) employing the fixed-effects procedure, India Total Factor Productivity – regression estimates of equation (5.8b) employing the fixed-effects procedure, India Domestic prices – regression estimates of (5.13) using the fixed-effects procedure, India

31 33 34 35 36 37 38 40 42 44 45 46 49 50 51 53 54 59 60 71 72

75 76 77

xiv

Tables

6.1

Structure of total manufacturing exports and imports, India 6.2 Structure of total manufacturing exports, Bangladesh, Kenya, South Africa and Vietnam 6.3 Employment coefficients of manufacturing exports and import-competing domestic manufacturing production, Bangladesh, Kenya, South Africa and Vietnam 6.4 Decomposition of manufacturing employment changes, India 6.5 Decomposition of manufacturing employment changes, Bangladesh, Kenya, South Africa and Vietnam 6.6 Labour demand – regression results – India 7.1 Wage inequality by industry (NIC 2-digit), India 7.2 Employment shares (percentage), India 7.3 Relative skill intensity and measures of trade policy – regression results, pooled ordinary least squares and between effects, India 7.4 Relative employment of skilled workers by industry, India 7.5 Wage inequality and measures of trade policy – regression results, fixed effects, India 8.1 Categorisation of Indian states 8.2 List of NIC 2-digit industries, India 8.3 Employment and production structure, Eastern Indian states 8.4 Employment and production structure, Western Indian states 8.5 Employment and production structure, Northern Indian states 8.6 Employment and production structure, Southern Indian states 8.7 Employment and production structure, coastal vs inland states and states with and without metropolises, India 8.8 Employment and production structure, all fourteen major states of India 8.9 Structure of exports, NIC 2-digit, India 8.10 Structure of imports, NIC 2-digit, India A1 Complete list of industries used for Chapters 5 and 7 A2 Complete list of industries used for Chapters 3 and 6 A3 Matching ISIC 4-digit codes to NIC 1987 codes A4 US producer price index matched to NIC (1987) classification

90 91

94 95 96 97 111 114

116 117 118 123 123 124 126 127 128 130 131 132 133 142 146 147 149

Preface

My interest in the relationship between trade policy and economic development has been long-standing – and so has my interest in Indian industry. In this book, I have attempted to merge these two separate but nonetheless interconnected research interests. I started working on issues relating to the impact of trade policy on the Indian economy when I was at the Australian National University in 1995–97, in a series of papers with Satish Chand. Subsequently, my interest in international trade carried on with further collaborative research with Rhys Jenkins in the School of Development Studies, University of East Anglia, Norwich, where I was based from 1999 to 2006. During this period, I realised that there was not a single body of research that has attempted to address the multifaceted and complex ways that international trade has impacted on Indian economic development within a unified framework and method of analysis. This book is therefore an attempt to redress this limitation in the literature. I would like to thank Darlene Mutalemwa for providing me with efficient research assistance at an early part of the project, and to Frances Bowcock for editorial assistance. During my professional life, several scholars have influenced my thinking on trade policy and Indian economic development, and I would like to acknowledge in particular the intellectual debt I owe to Prema-chandra Athukorala, V.N. Balasubramanyam, Pranab Bardhan, Jagdish Bhagwati, Krishna Bharadwaj, Ira Gang, Vijay Joshi, Anne Krueger, Sanjaya Lall, Ian Little, Arvind Panagariya and Kirit Parikh. I have also benefited from my collaborations with Satish Chand, A. Ganesh Kumar, Rhys Jenkins, Sumit Majumdar and Rajendra Vaidya, and from the comments received from participants in conferences in Brussels, Helsinki, New York and Singapore, where parts of the book were presented. The research that has resulted in this book has been partly funded by a small research grant from the British Academy, for which I am grateful. The School of Environment and Development in the University of Manchester, which has been my academic home since the summer of 2006, provided me with a congenial intellectual environment for the duration of this study. Finally, I would like to thank my family – my wife, Shilpi, and my children, Ayushi and Kushan – who put up with the long hours that I have spent in writing this book, for their patience, understanding and good humour. Kunal Sen Manchester December 2007

1

Trade policy, inequality and performance – the key analytical issues

1 Introduction The relationship between trade policy and economic performance is one of the oldest controversies in economic development. In the past few decades, most developing countries have abandoned import substitution policies that were followed extensively in the 1950s and 1960s for more outward-oriented trade regimes. It is not clear, however, whether the trade reforms that these countries have undergone in recent years has led to clear benefits in terms of higher economic growth and better economic performance. An influential view has concluded that countries with a more open trade orientation have tended to grow faster through time (Krueger 1997, Michely et al. 1991). This view has been contested by Rodriguez and Rodrik (1999), who have argued that there is little systematic evidence linking inward orientation and growth, and that the evidence linking outward orientation and growth has overstated the relationship between the two. There has also been considerable debate on whether international trade can be a powerful positive force in the reduction of poverty and inequality in developing countries by creating jobs, especially for unskilled workers, and by reducing the inequality between workers of different skills and educational levels, and between different regions in the same country (Venables 2003, Jenkins and Sen 2006, Goldberg and Pavcnik 2007). There have been concerns that trade reforms can, on the contrary, lead to job destruction and higher wage inequality in developing countries, similar to what has been observed in several developed countries, and that greater openness can exacerbate the inequality between regions in a country. However, neither the theoretical nor the empirical literature has yielded an unambiguous conclusion on whether trade reforms have a positive impact on economic performance, and whether trade liberalisation increases employment and decreases wage inequality between skilled and unskilled workers, and reduces inequality in the distribution of manufacturing employment and production between regions. Common to other developing countries, the Indian economy has undergone significant trade liberalisation since the late 1970s, and especially since

2

The key analytical issues

1991. The aim of this book is to examine the implications of trade reforms on efficiency and equity in India’s manufacturing sector. We will ask whether trade reforms have been able to deliver the efficiency outcomes that were the original aim of the policy-makers who initiated the reform process by examining the effects of the reforms on total factor productivity growth and whether the reforms have had pro-competitive effects on domestic pricing in the Indian manufacturing sector. We will also assess whether the trade reforms have had any role to play in India’s recent economic growth. We will then be concerned with the equity outcomes of the trade reforms. We will examine whether trade reforms have increased or decreased employment in Indian manufacturing. Following this, we will ask what the effects of the trade reforms have been on wage inequality in the manufacturing sector. We will also examine the implications of the trade reforms for regional inequality in India. India is particularly suitable as a case-study in understanding the relationship between trade policy and economic development. Beginning in the 1970s, the trade regime pertaining to the Indian manufacturing sector was one of the most restrictive in the world. A process of gradual reform was initiated in the late 1970s, quickened in pace in the 1980s, culminating in 1991 in a radical set of reforms that dismantled much of the import licensing system. The reforms first targeted the capital and intermediate goods sectors, with restrictions on the consumer goods sector only being relaxed in the mid1990s. The gradualist nature of the reform process, the move from a highly restrictive policy regime to a relatively open one, and the unevenness of the reforms across sectors and over time make the Indian economy a highly relevant context for understanding the welfare implications of trade reforms. Furthermore, the availability of data at a disaggregated industry level for a large number of industries and years encompassing the pre-reform and post-reform periods allows for a rigorous examination of the issues at hand. In the remainder of this chapter, we provide the analytical background and empirical context of the book. In the next two sections, we set out the key analytical issues relating to the links between trade policy and economic performance, and between trade policy and inequality in turn. We then discuss how we propose to measure the restrictiveness of trade policy. We next describe the empirical context of the research including the focus on the manufacturing sector and the choice of the period of analysis. We end with a description of the organisation of the book.

2 Trade reforms and performance – the key analytical issues The relationship between trade policy and economic growth has been extensively studied, but most of the research is relatively recent. This is due in part to the current phase of widespread trade liberalisation in developing countries and in part to developments in economic theory – most notably, the endogenous growth theories pioneered by Romer (1986) and Lucas (1988).


3

The endogenous growth theories identify several mechanisms by which trade reforms may have a sustained impact on economic growth. First, trade liberalisation increases the variety of goods and raises productivity by providing higher-quality intermediate and capital goods. Second, trade liberalisation leads to the exploitation of scale economies as firms in the reforming economy expand into world markets. Finally, trade reforms may lead to greater technological progress in the reforming economy as firms in this economy are able increasingly to capture new ideas being generated in the rest of the world. While the endogenous growth theories provide new and important insights on the dynamic effects of trade on growth, the theoretical literature does not yield an unambiguous conclusion on whether trade reforms have a positive impact on economic growth (Rodrik 1988, Tybout 1992). Furthermore, systematic attempts at quantification have failed to single out trade policy as a major factor in economic growth (Rodrik 1992). Both the ambiguity in the theoretical literature and the weak empirical evidence have led trade liberalisation sceptics to argue that ‘the effect on trade liberalisation on growth is, at best, very tenuous, and at worst, doubtful’ (Edwards 1993: 383). A key point of departure of our empirical analysis from previous studies, is that we pay particular attention to the explicit mechanisms by which trade policy can affect economic growth in the Indian context. This is important as an a-theoretical approach to testing for the effects of trade reforms on growth does not make clear the precise ways in which trade policy has impacted on economic growth, and is, therefore, not amenable to policy analysis. Trade reforms can primarily affect economic growth through growth in total factor productivity. However, the available empirical evidence on this issue has been far from conclusive – studies for developing countries that use firm- or industry-level data do not find an unequivocal positive relationship between trade reforms and productivity growth (see surveys by Havrylyshyn 1990, Nishimizu and Page 1990, Rodrik 1995). Moreover, most of these studies have been plagued by conceptual shortcomings. These studies rarely pay attention to the explicit theoretical mechanisms through which trade policy may impact on productivity growth. As Rodrik notes, ‘since the conceptual issues are rarely sorted out as a prelude to empirical analysis, the hypothesised cause-and-effect are difficult to interpret’ (1995: 2935). In this study, we will pay particular attention to the theoretical links between trade reforms and productivity growth when undertaking the empirical analysis. Another important way that trade reforms can enhance efficiency in the liberalising economy is by exerting a pro-competitive effect on domestic market power. Since protection often allows firms to enjoy super-normal profits, and market structures in developing countries significantly depart from perfect competition, it is an open question whether the reduction in protection will make firms in the domestic industry reduce their market power and, by doing so, the prices they charge their customers. Here, the relevance of imperfect competition for analysing the relationship between international

4


trade and domestic market structure has received increased attention in recent years. Theoretical studies of trade policy under imperfect competition predict that exposure to international trade can exert competitive pressures on hitherto-protected firms in the domestic market (Helpman and Krugman 1989). In line with the theoretical prediction, several empirical studies have found clear evidence of the disciplining effect of imports on domestic market power of firms, particularly in concentrated industries.1 In this study, we will examine the effect of trade liberalisation on domestic market power in Indian manufacturing, looking at the ways that trade reforms can bring about both price and quantity competition. This issue is particularly important in the developing country context, where widespread government controls can lead to highly non-competitive conditions among domestic producers, such that a sustained episode of trade liberalisation may exert significant competitive pressures on domestic firms, bringing about a decline in market power in several industries.

3 Trade reforms and inequality – the key analytical issues Whether international trade can be a strong positive force for reducing world poverty is one of the most controversial development issues of the day. The labour market is the key channel by which globalisation can impact on poverty. Increased integration with the world economy can potentially reduce poverty through the creation of new jobs in export industries. However, greater openness also brings increased competition from imports for previously protected industries. This can lead to job losses in certain sectors, with workers falling into poverty as a result of retrenchment. Whether globalisation creates or destroys jobs, and who are the winners and losers in employment is ultimately an empirical issue. In spite of its importance in understanding the links between trade and poverty, there has been scant empirical research on the impact of globalisation on employment in developing countries.2 Much of the initial work in this area was undertaken in a multi-country study sponsored by the National Bureau of Economic Research in the early 1980s (Krueger et al. 1981). Recently, Jenkins and Sen (2006) examined the impact of trade flows on employment in four developing countries – Bangladesh, Kenya, South Africa and Vietnam. This research showed that integration with the world economy has led to a significant increase in the number of unskilled jobs, particularly for women, in Bangladesh and Vietnam. However, job creation as a result of greater openness has been minimal in Kenya and South Africa and is biased towards more skilled workers (Jenkins and Sen 2006). This research suggests a potential continental divide in the pro-poor outcomes of globalisation, with Asian economies benefiting more from increased integration with the world economy than African economies. Yet it is not clear whether such a finding is robust, and what may explain such a continental divide, if it exists. An influential line of argument in this regard is


5

that of Wood (2003), who predicts that Africa’s long-term development path would be more like that of the land-abundant Americas than land-scarce Asia. Wood argues that Africa’s high land/labour ratio relative to that of Asia implies that Africa has less of a comparative advantage in labour-intensive manufacturing activities than Asia. Thus, differences in factor endowments between Africa and Asia explain why Africa’s exports structure seems to be biased towards natural resource based commodities rather than unskilled labour-intensive manufacturing exports. A corollary to this argument is that it is unlikely that globalisation can exert the same positive direct impact on employment creation in African countries as in Asia, given that labourintensive manufacturing seems to be the key sector in developing countries that has benefited most from the increase in world trade that has occurred in the past few decades. Wood’s argument provides a possible explanation of the results that have been obtained by Jenkins and Sen. However, it leaves several questions unanswered. First, how robust is the finding of an apparent ‘continental divide’? If factor endowment is the crucial driving force behind the differences in employment outcomes between Bangladesh and Vietnam on the one hand, and Kenya and South Africa on the other, does this hold for other Asian countries which also have a plentiful supply of unskilled labour relative to land? Second, how important is the policy regime in mediating the relationship between factor endowments and favourable employment outcomes in the manufacturing sector? Could it be argued that favourable policies towards export-oriented foreign direct investment in Bangladesh and Vietnam may have been more important in explaining the significant increase in labour-intensive exports in these two countries, a phenomenon that is not observed amongst some other neighbouring countries in the region? Third, does the manner in which increased international integration is undertaken matter in determining the magnitude of impact of globalisation on the labour market? There is preliminary evidence that Vietnam’s gradualist trade reforms may have limited job losses due to import penetration as compared to South Africa, where radical trade liberalisation was undertaken more rapidly. Finally, does international trade necessarily have a positive impact on employment, even in sectors or industries which are unskilled labourintensive? As industries seek to compete against imports or in international markets, job losses due to trade-induced technological change may occur. Evidence from South Africa and Vietnam suggests that sectors subject to import penetration have been particularly prone to labour-saving technological advances. The above discussion implies that a proper assessment of the impact of international trade on employment outcomes in developing countries can only be arrived at by undertaking a more complete comparative study that includes countries that are different from the four countries previously studied either in terms of factor endowments or the policy regime. In this book, we build on the research undertaken by Jenkins and Sen (2006) to

6


augment our understanding of the complex and contradictory ways by which international trade globalisation impacts on manufacturing employment in developing countries such as India. The key question here is whether trade integration has created or destroyed jobs in the Indian manufacturing sector. A second issue that has attracted a significant amount of interest is the impact of international trade on the inequality in earnings between skilled and unskilled workers. The increased flow of goods and services from developing to developed countries has coincided with adverse labour market outcomes for less-skilled workers in developed countries. In the United States, real wages for less-skilled workers fell both in absolute and relative terms during this period, while in Continental Europe there was a sharp increase in unemployment, particularly among the less skilled (Slaughter 1998). There is a strong consensus that both the higher wage inequality in the United States and the rise in unemployment rates in Continental Europe can be ascribed to a fall in the relative demand for less-skilled labour in these economies since the 1980s. The difference in labour market outcomes between the United States and Continental Europe with regard to the same phenomenon – a decline in the relative demand for less-skilled workers – can be attributed mainly to the differences in labour market institutions in these two regions. There has been considerable debate among economists as to what caused the shift in the relative demand for less-skilled labour in the advanced market economies. The two most commonly studied causes are globalisation in the form of increased international trade in goods and technological change biased towards skilled labour. In a survey of the empirical literature, Slaughter concludes that ‘the empirical evidence does not rule out either of these explanations, but on balance seems . . . to give more support to the globalisation hypothesis’ (1998: 1478). This conclusion is disputed by Krugman who argues that ‘there are serious difficulties with an explanation of growing inequality that places the main weight on trade’ (2000: 15). The main theoretical framework that guides research on the link between international trade and wage inequality is the one-cone Heckscher–Ohlin (H–O) model. This model predicts that increased trade between developed and developing countries will lead to increasing wage inequality between lessskilled and more-skilled workers in developed countries but decreasing wage inequality between these two groups of workers in developing countries. In recent years, several researchers have called into question the extreme assumptions of the H–O model, pointing out that more realistic versions of this theory do not allow an unambiguous prediction of the effect of international trade on wage inequality. At the same time, there has been the development of an alternate theoretical perspective which argues that international trade may have adverse effects on the wages and employment of unskilled workers relative to skilled workers by bringing about skill-biased technological change that leads to a displacement of unskilled workers (Wood 1994, 1997). This is often called the ‘defensive innovation’ hypothesis in the literature.


7

In contrast to the earlier emphasis in the literature on examining the impact of international trade on wage inequality in developed countries, in recent years there have been several studies that have addressed this issue from a developing country perspective. Such a shift in emphasis is welcome as the rapid progress of globalisation in developing countries has raised significant concerns about the effect of globalisation on poverty and inequality in these countries. Furthermore, given that the H–O model – the dominant theoretical framework used in the literature – predicts that international trade will reduce wage inequality in developing countries, it is indeed pertinent to ask whether the increasing integration of many developing countries in the global economy has had such a favourable outcome. On the other hand, if the ‘defensive innovation’ hypothesis is empirically valid, then wage inequality will increase among workers in developing countries with trade liberalisation. In this book, we will examine whether trade reforms in India have had an impact on wage inequality in the Indian manufacturing sector, and if so, whether there has been a reduction in inequality, as predicted by H–O theory. A final issue we examine is how regions within countries adjust to trade liberalisation. The new economic geography theorists argue that international trade can exert a strong impact on the location of economic activity by opening up new markets outside the country and thereby providing an incentive for firms to relocate closer to those markets (Hanson 1998). International trade can also lead to a concentration of economic activity if the industries expanding due to greater openness are characterised by increasing returns to scale, or if exporting activities lead to strong external economies linked to localisation and urbanisation (Lall et al. 2003). However, Venables (2003) argues that a move from an autarkic to a more open economy can also lead to a deconcentration of economic activity. We will examine the regional effects of trade reforms by looking at the evolution of spatial distribution of manufacturing employment and production across the major states in India, and assessing whether trade reforms have had a role to play in the changing spatial structure of the manufacturing sector in India. This is particularly relevant in the Indian context, where some Indian states have experienced strong economic growth while others have languished (Sachs et al. 2002). The question that the book will attempt to address is whether disparities in regional manufacturing growth in India can be linked to the trade liberalisation process.

4 Measuring trade policy An important methodological question that comes up when assessing the effects of trade policy is how to measure the restrictiveness of trade policy (Pritchett 1996, Anderson and Wincoop 2004). There has been a lack of reliable measures of trade policy changes (Edwards 1993). In earlier studies that have investigated the link between trade liberalisation and total factor productivity growth, for example (such as Krueger and Tuncer 1982),

8


causality was attributed merely by association – if there was evidence that total factor productivity increased in the post-reform period, then it must be due to the reforms. More sophisticated analyses in the same vein that have used dummy variables to demarcate the post-reform period from the prereform one (such as Ahluwalia 1991and Harrison 1994) are open to the same criticism. More importantly, the use of dummy variables to measure trade reforms implicitly assumes (a) that the trade reform was a once and for all event, and (b) that it was complete. Neither of these two conditions may be satisfied in most trade liberalisation episodes that one observes in developing countries (see the survey of recent trade liberalisation episodes by Michaely et al. 1991). The usefulness of dummy variables to capture trade policy shifts is greatly diminished if the trade liberalisation has been gradual over time, or if the reforms have been proceeding at an uneven pace across sectors. This is certainly true in the Indian case, as trade reforms have been uneven both across sectors and over time. Additionally, dummy variables may be picking up other changes in the economy not directly attributable to trade policy changes. In this book, we avoid the use of dummy variables to measure changes in trade policy at the industry level.3 Instead, we measure trade policy in a variety of ways, using both policy and outcome measures of trade policy changes. Thus, for policy-based measures, we use effective rates of protection and import coverage ratios (a fuller discussion of these measures must wait till the next chapter). For outcome-based measures, we use import penetration and export orientation ratios. There are limitations in the use of both sets of measures. Import penetration and export orientation ratios measure the outcome of changes in trade policies rather than the policies themselves. Furthermore, significant trade reforms may affect productivity growth without being reflected in import volumes. On the other hand, while effective rates of protection and import coverage ratios are more direct measures of trade policy, their usefulness in empirical work is hindered by the unavailability of enough observations over time to undertake any meaningful econometric analysis. We will use both sets of measures in our empirical analysis, but supplement them, wherever appropriate, with a measure of protection that is less susceptible to the criticisms made of other measures and is also sufficiently informative in that it can capture changes in trade policy both across industries and over time. This measure is based on a comparison of the domestic and international price of a particular industry – the price wedge.

5 The empirical context In this book, we study the effect of trade reforms on the Indian organised manufacturing sector from the mid-1970s to the late 1990s, using industries as units of analysis and using quantitative economic methods. Our focus on the manufacturing sector and not on other sectors such as agriculture and services, the restriction of the empirical analysis to the organised sector, and our


9

choice of the time-period and the unit of analysis need further elaboration and justification. Why manufacturing? The focus on the manufacturing sector in this book follows a long and illustrious tradition in the economic analysis of international trade pertaining to India, such as Bhagwati and Desai (1970) and Bhagwati and Srinivasan (1975), which principally studied the relationship between trade and industrialisation in India, with less attention paid to the relationship between trade and other important sectors such as agriculture and services. In part, the interest in trade and industrialisation in India in the work of economists of the stature of Jagdish Bhagwati, Padma Desai and T.N. Srinivasan could be linked to the need to understand the complex economic effects of the highly distortionary trade regime relating to the manufacturing sector that has characterised the Indian economy for much of the post-independence period. These economists and others were also interested in examining whether the avowed aims of Indian planners to develop a strong and self-reliant manufacturing under a very protective trade regime and stringent regulations on industrial activity had met with any significant success. Our interest in the growth prospects of India’s manufacturing sector is also influenced by our belief that the key to India’s future economic growth and poverty reduction remains the growth of a dynamic outward-oriented manufacturing sector which is significantly labour-absorbing. While the agricultural sector remains the sector in India with the highest numbers of workers employed, it is well known that primary commodities have been facing a long-run decline in prices in the world market (Sarkar and Singer 1991, Bleaney and Greenaway 1993, Sapsford and Chen 1998). Furthermore, since 1980, a novel feature of this phase of globalisation has been the rapid expansion of two-way North–South trade in manufactures, with many developing countries emerging as major exporters of manufactured goods and opening up their protected manufacturing industries to import competition and foreign investment (Jenkins and Sen 2006). Clearly, for the Indian economy to grow at rates that can have a sizeable impact on poverty reduction in India, the prospects for outward-oriented economic growth where agriculture is the leading sector is bleak. This is particularly the case in the current context, where there is limited possibility of agricultural growth being the driver of poverty declines in many regions of the country (Palmer-Jones and Sen 2003). While the service sector has made an important contribution to India’s strong economic growth witnessed in recent years, the role of trade policy in impacting on the growth of the service sector is limited, given that large segments of the service sector are non-tradable. Furthermore, by its very nature, the export-oriented segment of the service sector – the information technology sector – is skilled labour intensive and cannot be expected to have

10


a significant impact in reducing the large pool of surplus labour in India, employed in low-productivity work in backward agriculture or in the urban informal sector. In contrast, the manufacturing sector offers much greater prospects for job creation, especially for the unskilled and poorly educated segment of the labour force, which comprises most of India’s working poor. As has been recognised both among academics and policy-makers, the capacity of India to increase the standard of living of its citizens in the long term depends fundamentally on whether the Indian manufacturing sector can grow at rates characteristic of the manufacturing sector of India’s main competitor – China – in the world economy (Howes and Singh 2000, Fanelli and Medhora 2002, Ganesh-Kumar et al. 2003). Why the organised segment of the manufacturing sector? We confine our analysis to the organised manufacturing sector in India, as has been the case with most empirical studies on Indian industry (e.g. Ahluwalia 1991, Srivastava 1996).4 Given that the large proportion of the manufacturing workforce is in the unorganised (informal) sector, this is an important limitation.5 However, the limitation is to a large extent data-driven – data on unorganised manufacturing is available only for a few years, and not on a continuous basis. Unlike the data on the organised manufacturing sector, which is obtained from a full census of manufacturing units that are in the organised sector, the available data on unorganised manufacturing is from surveys conducted by the National Sample Survey Organisation (NSSO) that use stratified random sampling methods similar to the expenditure surveys conducted on households by the same organisation. Thus, the data on the unorganised manufacturing sector, by virtue of the method of data collection used, does not include all informal units. Thus, there is also non-comparability in the data on organised and unorganised manufacturing sectors. An important methodological strength of our empirical analysis is the rich data-set we use, which contains annual observations for a large number of industries for twenty-five years. In most cases, the number of observations we use exceeds 2,000, and the largest number of observations we work with is 3,475. The large number of observations used in the analysis, along with the comparability of the data over the period of the study, makes our empirical findings robust, which would not have been possible if we had included the unorganised manufacturing sector in the empirical analysis. The choice of the time period We use industry-level data for the period 1973/74 to 1997/98 for Chapters 5, 7 and 8, and for the period 1975 to 1999 for Chapter 6. For the macroeconomic analysis of Chapter 4, we use data that goes back to the mid-1950s and ends in 2003/04. It is important for the analysis to start in the early to mid-1970s, as trade reforms started occurring, albeit in an incremental manner, from the


11

late 1970s. The fact that the period of analysis ends in the late 1990s for the industry analysis implies that we are able to assess the impact of the 1991 reforms on efficiency and equity outcomes in the Indian manufacturing sector. Thus, our period of analysis encompasses both pre-reform and postreform phases. The reason we end in the late 1990s (and do not use more recent data) for the industry-level analysis is due to certain peculiarities of the data, and the need to use disaggregated industry data for the analysis of the effects of trade policy on productivity, prices and relative wages. For assessing the impact of trade policy on these three variables, the use of disaggregated industry data is essential for meaningful analysis. The data we use for the empirical analysis of the effects of trade policy on productivity, prices and relative wages is at the National Industrial Classification (NIC) 3-digit level, which is the highest level of disaggregation that the data will permit. The period of analysis for this data is 1973/74 to 1997/98. The data-set is compiled by the Central Statistical Organisation (CSO) of the Government of India, which used a particular classification (NIC 1987 classification scheme) for the pre-1998 data.6 From the year 1998/99, the CSO radically changed the industry classification, and it is virtually impossible to match the industries in the new NIC 3-digit classification of the post-1998 data with the industries at the NIC 3-digit level in the pre-1998 data. While we could have conducted the empirical analysis at a higher level of aggregation at the NIC 2-digit level, and by doing so used the post-1998 data, this would have meant that the empirical analysis would have been considerably weaker, as trade policy has differed a great deal across NIC 3-digit level industries within the same 2-digit classification. To evaluate the impacts of trade policy, it will be necessary to construct robust measures of trade policy changes for the period of the study that can capture both the across-industry and over-time variation in trade policy in India. The ability of the trade policy measures that we use to reveal the ‘true level of protection’ in a particular industry depends critically on the level of disaggregation at which the measures are computed. As we will see in Chapter 2, in a highly complex trade regime as was the case in India, changes in trade policy have differed widely across industries. Aggregating over industries with disparate levels of protection can lead to a significant loss in the information content of our measure.7 Thus, it would have been difficult to precisely estimate the effects of trade policy on economic outcomes at the NIC 2-digit level.8 For this reason, we have preferred to end our empirical analysis in 1997/98 wherever we have used the NIC 3-digit data (and to make the industry-level analysis comparable, we have ended the analysis in the 1990s where we use more aggregated industry data as in Chapters 3 and 6). Industries as units of analysis In our empirical analysis, we use industries and not firms as units of analysis, and we use panel data methods wherever appropriate. The focus on industries

12


as units of analysis separates our analysis from some of the recent empirical literature both for India and for other countries. This needs explanation. There are two reasons why we choose to use industry-level and not firm-level data. First, there is a data constraint. Firm-level data in India essentially comes from three sources – the PROWESS database of the Centre for Monitoring Indian Economy, the Reserve Bank of India’s (RBI) Survey of Public Limited Companies, and the original unit-level data obtained by the CSO that they can compile in the different Annual Survey of Industries. The PROWESS and RBI data track the same firms over time; however, data on key variables, such as employment and value added, that is needed in the empirical analysis, is not collected. Moreover, the firm-level data does not go as far back as we would like it to, to capture the trade reforms that were incorporated in the late 1970s and early 1980s. The unit-level census data of the CSO is clearly an extensive and rich data-set; however, its weakness is that the data is in the nature of repeated cross-sections, and therefore, not amenable to panel data analysis, at least at the firm level. The superiority of panel data methods over simple regression analysis (ordinary least squares) is well recognised in the empirical literature that assesses the impact of trade policy using quantitative methods, and is crucial in isolating the effects of trade policy from time-invariant omitted variables that are unobservable to the econometrician (such as intrinsic industry-specific time-invariant technological progress) in the quantitative analysis. The industry-level data that we use has the advantage of going as far back as the early 1970s and also encompasses a large cross-section of industries. The second reason for using industries as units of analysis is that, in our view, many of the questions that we are asking about the relationship between trade policy and economic development make more sense when analysed at the industry level in the Indian case. Trade (and industrial) policy in India differed greatly across industries at a point in time, and for the same industries, over time. While not minimising the importance of firm-level heterogeneity in understanding the welfare effects of trade reforms, policy-induced industry-level heterogeneity is clearly of significant empirical importance in the Indian context. The variation in trade policy across industries and over an extended period of time in India provides a fertile empirical ground where we can analyse important issues, such as the effect of trade policy on productivity, market power, employment and relative wages in a systematic and rigorous manner – perhaps more thoroughly than has been possible for any other developing country.

6 Organisation of the book The book is organised as follows. In Chapter 2, we provide a brief overview of the evolution of India’s trade and related policies, and of various measures of the trade policy regime pertaining to Indian manufacturing. Our objective is to highlight the important phases in the evolution of the policy regime and


13

its implications for the empirical analysis that we will undertake later in the book. We also provide a short discussion of the political economy of India’s trade policies. In Chapter 3, we set out the ‘stylised facts’ of the Indian manufacturing sector, with reference to the 1975–99 period, as a prelude to the empirical analysis in later chapters. In this chapter, we identify the key patterns and trends in different aspects of manufacturing, such as employment, wages and production. We begin the empirical analysis of the effects of trade policy on the manufacturing sector by attempting to answer a broader question: what has been the role of trade policy in India’s strong economic growth of the 1980s and 1990s? We do this in Chapter 4. We then examine in Chapter 5 the efficiency impacts of trade policy by looking at the effects of trade reforms on productivity and domestic prices in Indian manufacturing. In the next three chapters, we analyse some of the equity issues relating to international trade, by first examining the labour market effects of international trade, and then how different states in India have adjusted to trade liberalisation. In Chapter 6 we examine whether international trade has had any discernible impact on manufacturing employment. In Chapter 7 we assess what we know about the impact of trade reforms on wage inequality. In Chapter 8 we describe the evolution of the pattern of employment and production in major Indian states, and ask whether trade reforms have had an effect in the changing spatial distribution of industry. The concluding chapter summarises the key findings of the analysis in the book and examines their policy implications.9

2


1 Introduction By the mid-1970s, India’s trade regime had evolved into one of the most complex in the world. Since the late 1970s, there have been incremental attempts to reform the trade regime, with more significant attempts to do so, first in 1985 and then more comprehensively in 1991. In this chapter, we provide an overview of India’s trade policies since independence, which we do in the next section. We also summarise in the Section 3 developments in exchange rate and industrial policies, both of which influence the way in which trade policy impacts on the economy. We then describe in Section 4 patterns and trends in different measures of trade restrictiveness. We end with a discussion of the political economy of India’s trade policy in Section 5.

2 An overview of trade polices The import and exchange rate regime that Indian policy-makers followed since independence was aimed at the comprehensive, direct control over foreign exchange utilisation, with an overwhelming reliance on quotas rather than tariffs (Bhagwati and Srinivasan 1975). Import licences allocated reflected two major criteria: (1) the principle of ‘essentiality’; and (2) the principle of ‘indigenous non-availability’. Thus imports, in terms of both magnitude and composition, were to be permitted only if the firm in question certified to the government that they were ‘essential’ (as inputs or equipment for production). At the same time, the government had to clear the imports from the viewpoint of indigenous availability: if it could be shown that there was domestic production of the imports demanded, then the imports were not permitted (regardless of cost and quality considerations). Nearly all imports were subject to discretionary import licensing, or were ‘canalised’ by government monopoly trading organisations. The only exceptions were commodities listed in the Open General Licence (OGL) category. Capital goods were divided into a restricted category and the OGL category. While import licences were required for restricted capital goods, those in the OGL could be imported without a licence subject to several conditions. Intermediate goods


15

were also classified as falling into the banned, restricted and limited permissible categories plus an OGL category. As these names suggest, the first three lists were in order of import licensing stringency. OGL imports of intermediate goods were also governed by the ‘actual user’ condition. The import of consumer goods was banned, however (except those that were considered ‘essential’ and could only be imported by the designated government canalising agencies). Beginning with the export–import policy of 1977–8, there was a slow but sustained relaxation of import controls. Several capital goods that previously were not allowed to be imported without an import licence were steadily shifted to the OGL category. The number of capital goods on the OGL list increased from 79 in 1976 to 1,170 in April 1988. These changes were made with the intention of allowing domestic industries to modernise. Moreover, during the 1980s the import licensing of capital goods in the restricted list was administered with less stringency (Pursell 1992). As a consequence, the import penetration ratio in the capital goods sector increased from 11 per cent in 1976–7 to 18 per cent in 1985–6 (Goldar and Renganathan 1990). In the case of intermediate goods, too, there was a steady shift of items from the restricted and limited permissible categories to the OGL category. However, in practice a capital or an intermediate good was placed on the OGL list only if it was not being domestically produced. Thus, import liberalisation during this period may have led to some degree of competition to established producers of intermediate and capital goods in India (though in several instances, the goods that were allowed to be imported were imperfect substitutes of domestically produced goods). Furthermore, there was an increase in tariff rates across all commodities, in particular, on capital goods. By 1987/88, the unweighted average of tariffs on manufactured goods was 147 per cent, with most tariff lines for manufacturing clustered around a range of 140–60 per cent.1 The pace of the trade reforms – in particular, the shift from quantitative import controls to a protective system based on tariffs – initiated in the mid-1970s was considerably quickened by the new government (led by Rajiv Gandhi) that came into power in November 1985. Restrictions on the import of capital goods were further eased to encourage technological modernisation. Also, beginning in the mid-1980s, there was a renewed emphasis by the new administration on export promotion. The number and value of incentives offered to exporters were increased and their administration streamlined. The allotment of REP (replenishment) licences – tradable import entitlements awarded to exporters on a product-specific basis – became increasingly generous (Agarwal et al. 1995). Finally, the duty exemption scheme for imported inputs was extended to cover all imported inputs for both direct and indirect exporters.2 In 1991, as a part of the comprehensive economic reform programme initiated that year, there was a significant liberalisation of the trade regime with respect to capital and intermediate goods. Import licensing was virtually

16


abolished with respect to the imports of most machinery and equipment and manufactured intermediate goods (Ahluwalia 1999). There was also a significant cut in tariff rates, with the peak tariff rate reduced from 300 per cent to 150 per cent, and the peak duty on capital goods cut to 80 per cent.3 Import-weighted custom duty rates fell from an average of 97 per cent in 1990–91 to 29 per cent in 1995–6. There was, however, little change in trade policy with respect to consumer goods, which remained on the ‘negative’ (banned) list (Balasubramanyam 2003).4 What has been the effect of the trade policy changes on openness in the Indian economy? Exports plus imports as a ratio of GDP is often used as an outcome-based measure of openness (Figure 2.1). We compute this measure both for goods and services and for goods only. As is well-known, India has been a major exporter of information technology services in recent years. This is reflected in the larger increase in the ratio of exports plus imports of both goods and services to GDP as compared to ratio of exports plus imports of only goods to GDP. However, both indicators have shown a steady increase since the late 1980s, and the ratio of exports plus imports of goods to GDP stood at around 25 per cent in 2003. Clearly, since the trade reforms of the 1980s and early 1990s, the Indian economy has significantly increased its integration with the world economy. Manufacturing goods have comprised an increasing share of total exports and imports for the Indian economy (Figure 2.2). This is particularly evident in the case of manufacturing exports, which now comprise around 75 per cent of all of India’s merchandise exports. The manufacturing trade balance also shows an increase in the 1990s, after persistent deficits in the 1980s (Figure 2.3). Along with the increase in merchandise exports as a ratio of GDP, this indicates the increasing importance of manufacturing both in India’s economic activities and in the country’s relationship with the rest of the world.

3 An overview of exchange rate and industrial policies 5 Exchange rate policies In the period 1947–71, the rupee was pegged to the pound sterling. After a brief period when the rupee was pegged to the US dollar, there was a switch back to a sterling peg from December 1971. The link to a weak currency was in part deliberate. In spite of the oil shock of 1973 and the severe droughts of 1972 and 1974, with the sterling falling in value in relation to other currencies in the early 1970s, there was a real effective devaluation of the rupee during this period. In September 1975, the peg was altered to a basket of currencies with undisclosed weights, though sterling continued to be currency of designation and intervention. For the first two years from September 1975, the nominal exchange rate (NER) was kept fixed, but allowed to fall in 1978 and 1979 as the dollar was weak at that time (Joshi and Little 1994). From 1980 to 1982, the NER was again kept fixed leading to an appreciation of the RER,


17

Figure 2.1 Openness, exports plus imports of goods and services, and goods only as ratios of GDP, India. Source: World Development Indicators, World Bank.

with inflation in India high during this period. During 1983–5, the NER was made more flexible with a view to keeping the RER at its 1982 level. With the emphasis on export promotion from the mid-1980s, there was a steady devaluation of the Indian rupee during this period. Effectively, India operated an ‘active’ crawling peg from 1986 onwards to produce a sharp real depreciation of the rupee in the period 1986–90. In July 1991, as a part of the stabilisation package, the rupee was devalued against the dollar by 18 per cent. From 1991 to 1993, India moved gradually to full current account convertibility of the exchange rate, first in March 1992, replacing the tradable import entitlements, i.e. the Eximscrips, by a dual exchange rate system, under which exporters could sell 60 per cent of their export earnings to authorised foreign exchange dealers at a ‘marketdetermined’ rate (the remaining 40 per cent at the official exchange rate); and then, in March 1993, moving to a unified ‘market-determined’ exchange rate system (i.e. a managed float). However, strict controls over the capital account, especially capital outflows remain. This substantially mutes the influence of movements in the capital account on the exchange rate. Further, even on the current account, the ‘market-determined’ exchange rate is not completely free

18


Figure 2.2 Manufacturing exports and imports as shares of total merchandise exports and imports, India. Source: World Development Indicators, World Bank.

of intervention by the Reserve Bank of India (RBI). In line with the reforms, the nature of RBI intervention has undergone a qualitative change. Unlike in the past, when the RBI used to ‘announce’ the official exchange rate, it now influences to a great extent the ‘market-determined’ exchange rate through direct market operations as a seller or a buyer of foreign exchange, and also through other policy instruments such as the interest rate. Indeed, it may be argued that in the recent past the RBI has been protecting the rupee through a combination of market operations and interest rate changes. Figure 2.4 plots the nominal rupee–US dollar exchange rate. It is clear that the Indian rupee has been depreciating significantly against the US dollar, especially since 1991. Industrial policy For first four decades after independence, the government intervened in almost all aspects of the activities of industrial firms. Industry in India was subject to rather formidable legal barriers to entry. Investments, both in terms of expansion of capacity of existing firms and creation of new firms, was controlled by the government through its licensing policies that were in turn determined according to plan priorities. Following an initial attempt in 1975–6, in 1985–6, the liberalisation of


Figure 2.3 Manufacturing trade balance, India. Source: International Economic Data-bank, Australian National University.

Figure 2.4 The nominal exchange rate, Indian rupee to US dollar, India. Source: World Development Indicators, World Bank.

19

20


industrial controls gathered momentum when some industries were taken out of the purview of industrial licensing. Modernisation of equipment was also allowed in a limited manner. Expansion of capacity up to a mandated efficient scale was also permitted in industries where economies of scale were considered to be significant. Finally, the definition of the industrial licensing act was also relaxed to allow not only small-sized firms but also medium-sized firms to fall outside its purview. On 24 July 1991, a path-breaking Industrial Policy Statement was tabled in Parliament. Industrial licensing was abolished irrespective of the level of investment except for sixteen industries (Mookherjee 1995). The number of industries reserved for the public sector was significantly reduced. The industrial licensing act was amended such that it now only seeks to curb anticompetitive behaviour on the part of firms. Also, under the new policy guidelines on foreign investment, automatic permission is granted for foreign equity participation up to 51 per cent in a specified list of high technology and high investment priority industries.

4 Measuring the restrictiveness of trade policy in India In this section, we present three measures of protection – two of which are standard in the literature – and a third which is a usual supplement to these standard measures in a complex trade regime such as India. The first two are the Effective Rate of Protection and the Import Coverage Ratio, and the third is the Price Wedge. The Effective Rate of Protection (ERP) is defined as the percentage excess of domestic value added vis-a-vis world value added, introduced because of tariff and other trade barriers.6 This measures the distortions introduced due to tariffs on input prices as well as tariffs on final output prices, and therefore measures protection to domestic factors of production. As Bhagwati and Desai (1970) point out, effective rates of protection are more relevant than nominal tariffs as guides to the effects of protection on the allocation of domestic resources and the relative outputs of different commodities. Along with ERPs, we also present estimates of the Import Coverage Ratio (ICR), which measures the proportion of commodities in a particular industrial sector, the importation of which is restricted by a Non-Tariff Barrier.7 In the Indian case, given the wide prevalence of nontariff barriers till 1991, such a measure is particularly relevant in measuring the restrictiveness of trade policy, in conjunction with a tariff-based measure such as the ERP. The final measure of protection that we use is the ‘price wedge’ – the deviation of the domestic price of the output produced by a particular industry from the world free trade price for that industry. The price wedge measures the degree of slack that protection provides to domestic producers relative to their international competitors, and any such slack would be reflected as a positive deviation of the domestic price from the international price.8


21

We approximate the world free trade price of a particular industry by the price prevailing in the United States for that industry. Thus, we define the price wedge, Rit, for industry i at time t as: Rit = (PINDIA / PUSA ) * et it it where PINDIA and PUSA are the prices of industry i at time t for India and the it it United States respectively and et is the exchange rate (Indian rupees per US dollar). According to our measure, protection increases in a particular industry in India if its domestic price increases relative to that of the US or if the nominal exchange rate depreciates (that is, if e increases). We present estimates of ERP obtained from Das (2003) for the period 1980–99 for the three major sectors in India – capital goods, intermediate goods and consumer goods in Figure 2.5. The most protected sector with respect to tariff protection in the beginning of the 1980s was the intermediate goods sector, where the average ERP was 147 per cent. The ERPs for the capital goods and consumer goods sectors for the same period were 63 and 102 per cent respectively. There was little change in ERPs in the 1980s, as the gradual relaxation of import licensing over this decade coincided in most instances with an increase in tariffs. However, since the 1991 trade reforms, there has been a clear decrease in ERPs across all three sectors. In fact, by the end of the 1990s, the average ERP for the intermediate goods sector at 40 per cent was actually less than the average ERP for the consumer goods sector at 48 per cent. However, the least protected sector in the late 1990s was the capital goods sector, with an average ERP at 33 per cent. While the intermediate goods sector may have been the most protected and the capital sector the least protected in the aggregate with respect to tariff protection, there are clear differences in the degree of protection within each sector. This is evident from Figures 2.6, 2.7 and 2.8 where we provide ERPs for individual industries in the capital goods, intermediate goods and consumer goods sectors respectively. Within the capital goods sector, an industry such as agricultural machinery (Industry Code: 350)9 has an ERP of 30 per cent in 1980–85 as compared to an ERP of 200 per cent for batteries (Industry Code: 362) in the same period. Similarly, in the intermediate goods sector, fabricated metal products (Industry Code: 340) has an ERP of 428 per cent in 1980–85 compared to fertilisers and pesticides (Industry Code: 301) which has an ERP of 51 per cent in 1980–85. Finally, in the consumer goods sector, waterproof textile fabrics (Industry Code: 268) has an ERP of 161 per cent in 1980–85 compared to drugs and medicines (Industry Code: 304) which has an ERP of 80 per cent in 1980–85. However, by the late 1990s, there is significantly lower variation of ERPs within each sector, with the rationalisation of tariffs and the move to a more uniform tariff rate across industries. Effective rates of protection may not adequately measure protection in Indian industry, given the extensive use of non-tariff barriers (NTBs) in

22


Figure 2.5 Effective Rates of Protection (ERP) by sector, India. Source: Das (2003).

Figure 2.6 Effective Rates of Protection (ERP) in the capital goods sector, by industry, India. Source: Das (2003). Note: See Appendix, Table A1 for industry codes.


23

Figure 2.7 Effective Rates of Protection (ERP) in the intermediate goods sector, by industry, India. Source: Das (2003). Note: See Appendix, Table A1 for industry codes.

Indian trade policy. A more accurate measure of the restrictiveness of trade policy in this case is the Import Coverage Ratio (ICR), which is the share of imports subject to non-tariff barriers. We find that the use of ICRs as a measure of the restrictiveness of trade policy suggests that all the three sectors were equally protected by the use of NTBs in the early 1980s, and that NTBs covered almost all manufacturing goods. The average ICR by sector was at 95 per cent or more in 1980–85 (Figure 2.9). However, by the late 1990s, the most spectacular decline in NTBs was in the capital goods sector, where the ICR fell to 8 per cent. The ICRs in the case of intermediate goods and consumer goods sectors were 28 and 33 per cent in 1996–9 respectively. When we examine the presence of NTBs at the individual industry level, we find that for the whole of the 1980s, in most industries, the ICRs stood at 100 per cent, indicating the widespread use of quotas in Indian trade during this period (Figures 2.10, 2.11 and 2.12). However, there were a few industries where no NTB was applied – these are machine tools (Industry Code: 357), leather tanning (Industry Code: 290) and cotton textiles (Industry Codes: 232, 233 and 234). In the 1990s, there is a significant decline in ICRs across most industries. In the case of intermediate goods and capital goods sectors, in several industries, the ICR falls to zero, as quotas are completely removed

24


Figure 2.8 Effective Rates of Protection (ERP) in the consumer goods sector, by industry, India. Source: Das (2003). Note: See Appendix, Table A1 for industry codes.

in these industries. This is not the case, however, in the consumer goods sector, where NTBs still remain, though with much reduced coverage. Overall, the patterns in ERP and ICR suggest that it is better to use both measures of protection in the empirical analysis when we attempt to relate economic outcomes to trade policy changes since the two measures seem to capture somewhat different aspects of the trade regime in India. Finally, we present price wedge computations for some select industries in India. As we have seen earlier in this section, the price wedge is composed of two elements – the ratio of the price of the commodity in India to its price in the US, and the nominal exchange rate. We have observed in Figure 2.4 that the nominal exchange rate has steadily depreciated since the mid-1980s, thereby providing greater protection to domestic industry in India, all other variables held constant. We now look at the price ratio – the ratio of the price of the commodity in India to its price in the US, and plot changes in the price ratio for twelve industries – four in each of the use-based categories. The price ratios for the capital, intermediate and consumer goods sectors are presented in Figures 2.13, 2.14 and 21.5 respectively. We find that in several industries, there is clear evidence of a decrease in the price ratio, particularly since 1991. This is particularly true for the capital and consumer goods industries.


25

Figure 2.9 Import Coverage Ratios (ICR) by sector, India. Source: Das (2003).

Figure 2.10 Import Coverage Ratios (ICR) in the capital goods sector, by industry, India. Source: Das (2003). Note: See Appendix, Table A1 for industry codes.

26


Figure 2.11 Import Coverage Ratios (ICR) in the intermediate goods sector, by industry, India. Source: Das (2003). Note: See Appendix, Table A1 for industry codes.

Figure 2.12 Import Coverage Ratios (ICR) in the consumer goods sector, by industry, India. Source: Das (2003). Note: See Appendix, Table A1 for industry codes.


Figure 2.13 Changes in the price ratio, capital goods sector, India. Source: Author’s calculations. Note: See Appendix, Table A1 for industry codes.

Figure 2.14 Changes in the price ratio, intermediate goods sector, India. Source: Author’s calculations. Note: See Appendix, Table A1 for industry codes.

27

28


Figure 2.15 Changes in the price ratio, consumer goods sector, India. Source: Author’s calculations. Note: See Appendix, Table A1 for industry codes.

5 The political economy of protection in India To what extent has the structure of protection in India been due to political motives, and less to do with economic reasons? There is a large literature that examines the political economy of trade policy in general – Magee (1994) provides an excellent overview. Among the economic factors, the most important are infant industry considerations – policy-makers would like to protect nascent industries that may be relatively inefficient at the beginning so that they can achieve world standards in efficiency levels. This has certainly been true in the Indian case, and infant industry protection was a cornerstone of trade policy from the beginning of the planning process in the 1950s (Chakravarty 1987). The greater protection of the capital and intermediate goods industries as compared to the consumer goods industries was also linked to the fact that Indian planners were keen to replicate the Soviet model of industrialisation based on the growth of heavy industries, which were seen to have greater positive externalities. The political economy of trade protection literature also identifies lobbying by workers and capitalists in certain industries as being important in explaining protection accorded to those industries. We have seen quite significant variation in both the level and rate of change of protection within each industrial classification that suggests that political factors may have also


29

played a part in explaining inter-industry protection levels, and why protection in some industries was removed faster than for others. Gang and Pandey (1996) undertake a detailed empirical analysis of the factors determining protection in India in the 1980s and find that political factors such as lobbying by workers and capitalists seem to explain protection policy in India better than the infant industry argument. However, in our view, political factors cannot provide a convincing argument why some industries were more protected than others, or why the pace of reform differed across industries so significantly. This is for two reasons. Trade unions in India are established at the national level, so they are not present only at the industry level. Furthermore, most trade unions in India are linked to national political parties, for example the Centre for Indian Trade Unions is affiliated to the Communist Party of India–Marxist and the Indian National Trade Union Congress is affiliated to the Congress Party. It is difficult to see why a national-level trade union (and the political parties that they are affiliated to) would lobby for protection for some of its workers in one industry and not in other industries. Second, a large proportion of firms in India are members of business groups, where the owner of the group is a wealthy family (Majumdar and Sen 2007). These family-owned business groups are highly diversified entities, and are actively present in several quite disparate industries. Again, it is difficult to see why business groups would lobby for protection in one industry versus another, and their more general concern would be the protection of domestic Indian industry in its entirety. In our view, both the levels of, and changes in protection that have occurred in Indian manufacturing since the late 1970s has been policy driven and exogenous to the political and economic factors occurring at the individual industry level. At the beginning of the planning period, Indian policy-makers accorded the capital goods and intermediate goods industries the most protection in the early phases of industrialisation, linked to their belief that heavy industries will provide the backbone of India’s industrial development. These were also the industries that underwent the most significant reforms, with the clear recognition by the Indian policy-makers since the late 1970s, and especially in the 1990s, that access to capital and intermediate goods from abroad at international prices were crucial in Indian industry’s modernisation drive after a period of industrial stagnation in the late 1960s and early 1970s. In the empirical analysis that we conduct in this book when we examine the effects of trade reforms on different aspects of Indian industrial development, we will take the trade reforms themselves as being exogenous to political-economic factors.10

3

Growth and structural change in Indian manufacturing, 1975–99

1 Introduction This chapter describes the evolution of the Indian manufacturing sector from the mid-1970s to the late 1990s, the period of analysis of this study. We will examine the patterns and trends in real wages, employment, production, exports and imports for the period 1975–99, in order to set the stage for the empirical analysis in the ensuing chapters. In the next section, we examine the patterns in employment and real wages. In Section 3, we examine the structure and growth of production. In Section 4, we describe the structure of exports and imports. We make some concluding remarks in Section 5.

Figure 3.1 Overall manufacturing employment in India. Source: Author’s calculations, data from United Nations Industrial Development Organization (UNIDO).

Growth and structural change, 1975–99

31

2 Patterns and trends in real wages and employment We plot overall manufacturing employment in Figure 3.1. After a period of increase in the second half of the 1970s, there was stagnation in employment levels in the first half of the 1980s. From the second half of the 1980s, there has been steady increase in employment, with the level of employment in 1999 standing at 9.1 million (as compared to 5.8 million in 1975). Employment growth has been particularly strong in 1991–5, at an annual average rate of growth of 4.2 per cent (Table 3.1). Aggregate real wages have increased steadily since the early 1980s, with a 66 per cent increase in the period 1975–99 (Figure 3.2). However, the Table 3.1 Overall trends – Indian manufacturinga Average growth rate

1976–80

1981–5

1986–90

1991–5

1996–9

Real value added Employment Labour productivity Factoriesb Real wages

1.6 3.7 −2.1 N/A 3.5

7.4 −1.0 6.4 −0.8 3.7

9.3 2.2 7.1 2.0 2.0

11.1 4.2 6.9 3.6 1.7

5.3 0.9 4.4 0.4 −0.3

Source: Author’s calculations, UNIDO data. Notes a Average annual growth rates in per cent. b Only for 1996–7.

Figure 3.2 Aggregate real wages in India. Source: Author’s calculations, UNIDO data.

32


Figure 3.3 Aggregate labour productivity in India. Source: Author’s calculations, UNIDO data.

Figure 3.4 Aggregate unit labour costs in India. Source: Author’s calculations, UNIDO data.


33

strongest growth in real wages has occurred in the second half of the 1970s and the first half of the 1980s, when the annual rates of growth of real wages were 3.5 and 3.7 per cent respectively (Table 3.1). After a period of stagnation in the 1970s, labour productivity has increased at a rapid rate since the early 1980s (Figure 3.3). Labour productivity growth was strongest in the period 1986–95, increasing at an annual rate of 7 per cent in this period (Table 3.1). The rapid increase in labour productivity, which has out-paced real wage growth, has meant that unit labour costs have fallen steadily in Indian manufacturing since the early 1980s (Figure 3.4). This has meant that the Indian manufacturing sector has significantly increased

Table 3.2 Structure of employment, India (percentage share) Industries

1976–80

1981–5

1986–90

1991–5

1996–9

311 313 314 321 322 323 324 331 332 341 342 351 352 353 354 355 356 361 362 369 371 372 381 382 383 384 385 390

18.5 0.5 5.2 25.1 0.8 0.5 0.4 1.1 0.1 1.8 2.3 2.6 4.2 0.3 0.4 1.3 0.5 0.4 0.9 3.8 7.6 0.8 3.0 6.0 4.5 6.4 0.6 0.5

16.2 0.6 5.3 22.7 0.8 0.5 0.4 1.0 0.1 2.0 2.3 2.8 4.6 0.3 0.4 1.3 0.7 0.5 0.9 4.7 8.3 0.8 2.8 6.4 5.0 7.4 0.6 0.4

14.9 0.7 5.9 20.4 1.3 0.6 0.6 0.9 0.1 1.9 2.1 3.1 5.0 0.3 0.5 1.5 1.0 0.4 0.8 5.0 7.5 1.5 3.0 6.5 5.4 7.9 0.7 0.5

15.2 0.8 6.1 18.0 2.5 0.6 0.7 0.8 0.1 2.0 2.0 3.2 5.1 0.3 0.5 1.6 1.2 0.3 0.8 4.7 5.9 2.3 3.1 6.3 5.2 9.3 0.7 0.7

15.0 0.7 6.0 17.3 3.2 0.6 0.7 0.8 0.1 2.1 1.8 3.9 5.3 0.4 0.4 1.7 1.5 0.3 0.5 4.4 5.2 2.3 3.3 5.7 4.9 10.4 0.7 1.0

TCF NON TCF

26.3 73.7

24.0 76.0

22.2 77.8

21.1 78.9

21.1 78.9

Source: Author’s calculations, UNIDO data. Note: TCF = Textile, Clothing and Footwear; Industries 321, 322 and 324.

34


its international competitiveness in the 1980s and 1990s, with both India’s manufacturing exports as a share of world manufacturing exports and India’s manufacturing trade ratio (the ratio of manufacturing exports to manufacturing imports) showing steady increases since the mid 1980s (Ganesh-Kumar et al. 2003). Examining the distribution of employment across industries in the Indian manufacturing sector and the changes in this distribution over time, we find that the relative ranking of industries in the share of employment has stayed stable over time (Table 3.2). The two leading industries in terms of number of workers employed are ISIC 311 – Food Manufacturing – and ISIC 321 – Textiles – over the period 1975–99. These two industries comprised 32.3 per cent of total employment in 1996–9. Transport equipment (ISIC 384) moved to third place in 1996–9, employing 10.4 per cent of the manufacturing workforce in this period. Perhaps the most noteworthy finding is

Table 3.3 Employment growth, India (average, per cent per annum) Industries

1976–80

1981–5

1986–90

1991–5

1996–9

311 313 314 321 322 323 324 331 332 341 342 351 352 353 354 355 356 361 362 369 371 372 381 382 383 384 385 390

4.4 8.5 13.4 0.8 13.0 7.8 4.3 1.6 −4.2 5.1 1.7 5.3 7.4 5.6 7.3 2.4 13.4 6.5 0.9 4.1 3.4 2.6 2.0 3.4 4.0 6.7 5.5 −0.3

−5.6 3.7 −2.7 −2.4 4.1 4.3 4.4 −1.3 −1.9 0.2 0.6 1.9 2.1 9.6 5.8 −0.8 4.7 0.1 −0.4 5.7 0.8 0.8 −0.3 1.9 2.1 −0.3 1.2 0.6

2.8 3.8 7.9 −0.3 12.8 3.8 10.4 −1.9 −4.6 2.6 −1.1 1.7 −0.2 7.0 1.0 5.9 8.3 −1.5 −1.0 0.6 −2.2 34.4 4.1 1.5 2.1 6.4 3.6 7.1

3.3 4.6 2.2 2.8 19.6 3.7 5.1 3.0 4.0 4.3 5.0 6.6 7.1 7.2 5.8 4.1 7.9 −0.2 3.5 2.4 2.4 8.4 4.5 3.4 2.9 5.6 6.4 13.7

1.4 −1.1 3.5 −0.8 2.1 1.6 −1.6 −0.8 8.8 3.0 −3.2 6.0 2.6 4.3 −1.9 2.5 4.3 −0.4 −5.7 1.2 −2.2 −1.3 2.1 −1.5 −0.3 4.2 −2.2 3.8

Source: Author’s calculations, UNIDO data.


35

that the share of industries which are conventionally defined as labour intensive – textiles, clothing and footwear (TCF) – has fallen from 26.3 per cent in 1976–80 to 21.1 per cent in 1996–9. Therefore, the pattern of employment in Indian manufacturing has been such that more workers are employed in the capital-intensive industries over time. Though employment growth has been positive since the mid-1980s, the disaggregated growth rates suggest that several industries have seen job losses, especially in 1996–9 (Table 3.3). The fastest growing industry in terms of employment has been ISIC 322 – Wearing Apparel – which has witnessed an annual growth rate of 19.6 per cent in 1991–5 alone. Other industries which have seen strong employment growth, at least for part of the period under study, are ISIC 353 – Petroleum Refineries, ISIC 356 – Plastic Products, and ISIC 372 – Non Ferrous Metals. Interestingly, textiles (ISIC 3210) – the largest industry in terms of employment – has seen job

Table 3.4 Real wage growth, India (average, per cent per annum) Industries

1976–80

1981–5

1986–90

1991–5

1996–9

311 313 314 321 322 323 324 331 332 341 342 351 352 353 354 355 356 361 362 369 371 372 381 382 383 384 385 390

2.3 −3.5 1.8 4.3 4.4 5.0 4.0 4.8 10.2 3.4 4.5 4.9 2.8 −4.0 4.2 4.9 7.8 8.7 3.0 3.2 2.7 3.9 5.5 4.8 3.3 3.9 5.8 8.5

11.1 5.1 7.5 1.1 −0.3 −1.5 0.4 2.5 3.2 0.9 3.2 2.7 2.2 17.8 0.8 −0.6 3.3 0.1 5.9 1.7 2.0 2.2 2.1 3.5 2.5 3.3 1.8 −0.3

4.8 2.6 −2.2 1.9 1.6 3.3 −1.8 2.3 2.5 2.9 4.3 3.3 2.4 −1.3 2.0 1.8 1.2 6.0 3.4 1.6 3.6 −3.3 1.4 2.6 1.9 1.4 1.8 0.3

1.4 1.1 0.0 −0.5 1.5 −0.1 0.1 0.4 0.7 1.2 3.6 1.4 −0.2 11.4 −0.7 0.9 1.2 −1.0 3.1 2.9 5.3 4.7 2.6 1.8 2.6 2.1 1.6 2.8

−0.8 −0.5 0.2 −1.5 −0.2 0.9 −3.1 −1.1 2.1 −1.1 4.4 −0.7 0.4 −6.0 2.6 −1.5 4.7 4.2 8.1 4.1 0.0 −4.5 −0.1 2.1 0.6 −2.2 2.7 2.9


36


Table 3.5 Structure of female employment, India Industries

Share in total female employment (per cent)

Share in total employment in industry (per cent)

311 313 314 321 322 323 324 331 332 341 342 351 352 353 354 355 356 361 362 369 371 372 381 382 383 384 385 390

23.4 0.5 21.1 12.4 14.3 0.8 2.7 0.6 0.0 0.7 0.6 0.3 9.0 0.0 0.2 0.5 0.6 0.3 0.2 3.0 0.5 0.2 0.5 0.8 3.6 1.0 0.9 1.2

13.1 5.4 25.2 6.0 30.5 10.7 26.4 6.5 1.4 2.9 3.0 0.7 14.0 0.4 4.7 2.7 4.2 7.4 2.4 6.1 0.9 0.7 1.5 1.2 6.3 0.9 11.1 11.5


contraction over the period 1975–99. A feature of employment growth across industries has been the higher volatility of employment growth as evident in the increase in the coefficient of variation from 0.8 in 1976–80 to 3.1 in 1996–9. Similar to employment growth, real wage growth rates differ significantly across industries (Table 3.4). Again, in 1996–9, several industries have witnessed decreases in real wages, reversing the trend in increases in real wages in the earlier sub-periods. Cross-industry variation in real wages has increased over the period under consideration – the coefficient of variation increased from 0.7 in 1976–80 to 5.0 in 1996–9. The share of female employment in total employment differs significantly across industries. The industries which employ the largest number of female workers are food (ISIC 311) – 23.4 per cent, tobacco (ISIC 314) – 21.1 per


37

Table 3.6 Structure of production, India (percentage share) Industries

1976–80

1981–5

1986–90

1991–5

1996–9

311 313 314 321 322 323 324 331 332 341 342 351 352 353 354 355 356 361 362 369 371 372 381 382 383 384 385 390

9.9 0.7 2.4 18.6 0.2 0.2 0.1 0.9 0.2 2.3 2.3 7.7 6.7 0.8 0.3 2.1 0.8 0.3 0.6 3.3 9.7 1.3 3.9 8.3 7.4 7.8 0.4 0.8

12.2 0.8 2.7 14.5 0.1 0.3 0.1 0.7 0.1 1.8 2.3 8.8 5.9 0.5 0.3 2.4 1.0 0.3 0.6 4.3 8.1 1.0 3.6 8.9 10.3 7.7 0.4 0.4

10.7 0.7 2.4 11.2 0.2 0.2 0.0 0.5 0.1 1.9 2.0 11.2 6.8 1.2 0.2 1.9 1.3 0.3 0.5 3.6 6.7 1.3 3.5 7.3 16.9 6.5 0.7 0.2

9.2 0.6 2.1 9.4 0.1 0.2 0.0 0.4 0.1 2.0 2.1 13.9 7.5 1.4 0.2 1.5 1.5 0.2 0.5 4.0 6.8 1.6 2.2 7.0 18.7 6.0 0.5 0.2

8.6 0.6 1.7 8.1 0.0 0.2 0.0 0.3 0.1 1.8 2.0 11.9 8.4 1.1 0.2 1.6 1.6 0.3 0.7 4.8 10.3 1.6 1.7 6.8 18.8 6.3 0.4 0.2

TCF NON TCF

20.5 79.5

15.7 84.3

14.0 86.0

13.2 86.8

11.0 89.0


cent, textiles (ISIC 321) –12.4 per cent and wearing apparel (ISIC 322) – 14.3 per cent (Table 3.5). The most female labour-intensive industries are wearing apparel, where close to one out of every three workers is female, tobacco and footwear (ISIC 324) – where a quarter of the workforce is female. Female workers are a small proportion of the workforce in the non-TCF and nonagro-based industries.

3 Patterns and trends in production Real value added has increased at a rapid rate since the early 1980s, with annual growth rates of over 9 per cent for the period 1985–95 (Table 3.1). Similar to what we have observed for employment, the share of TCF

38


Table 3.7 Growth of real value added, India (average, per cent per annum) Industries

1976–80

1981–5

1986–90

1991–5

1996–9

311 313 314 321 322 323 324 331 332 341 342 351 352 353 354 355 356 361 362 369 371 372 381 382 383 384 385 390

−5.5 6.6 −8.6 −7.6 14.7 −5.5 −2.7 −1.9 −4.4 1.7 −2.2 −2.1 −1.7 −2.9 4.4 −6.6 16.3 10.0 −4.8 0.5 −2.7 −13.6 1.4 1.6 −0.5 1.5 0.0 2.5

16.0 19.6 4.1 3.9 14.6 3.0 7.3 0.7 2.5 7.1 5.9 13.4 5.6 24.8 15.4 10.8 19.9 −8.5 14.8 20.8 3.3 14.3 3.2 10.5 5.9 4.8 11.4 23.7

6.8 9.8 11.8 1.5 29.6 5.1 14.0 −3.1 6.7 27.0 0.8 9.9 5.2 30.3 −1.9 11.5 7.3 6.7 4.0 9.6 8.8 63.0 −6.3 2.3 8.8 12.3 0.2 −9.7

5.3 7.7 −1.1 2.1 25.6 −10.9 6.3 −2.1 10.1 9.3 12.8 22.7 5.9 10.7 8.5 −1.5 6.4 1.5 17.2 7.7 11.8 19.9 6.7 6.1 6.7 10.5 9.6 35.4

6.2 8.9 11.9 0.9 −0.6 6.2 8.0 6.1 36.5 −9.5 −1.6 2.5 9.7 56.5 10.4 16.6 13.6 7.6 13.7 11.1 16.1 −3.5 3.3 0.5 7.4 4.3 −7.0 24.3

Source: Author’s calculations, UNIDO data. Note: Gross Value Added for each industry is deflated by the wholesale price index for all manufacturing.

industries in total production has decreased from 20.5 per cent in 1976–80 to 11.0 per cent in 1996–9 (Table 3.6). However, the decline in share of output for the TCF industries has been more rapid than the decline in the share of employment, suggesting that labour productivity levels have also been declining in TCF industries relative to non-TCF industries. The industry with the highest share in output in 1976–80 – Textiles at 18.6 per cent – saw a decline in its relative importance over the period. Its share in output in 1996–9 was a mere 8.1 per cent. Industrial Chemicals (ISIC 351) and Electrical Machinery (ISIC 383) witnessed a significant increase in their shares in total output in the period 1975–99. The share of Industrial Chemicals increased from 7.7 per cent in 1976–80 to 11.9 per cent in 1996–9, while


Figure 3.5 Number of factories in India. Source: Author’s calculations, UNIDO data.

Figure 3.6 Average factory size in India. Source: Author’s calculations, UNIDO data.

39

40


Table 3.8 Share of factories, India (percentage share) Industries

1976–80

1981–5

1986–90

1991–5

1996–7*

311 313 314 321 322 323 324 331 332 341 342 351 352 353 354 355 356 361 362 369 371 372 381 382 383 384 385 390

NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA

18.7 0.4 8.4 12.9 1.2 0.7 0.2 3.6 0.3 1.6 3.4 1.9 4.5 0.0 0.4 1.7 1.9 0.5 0.6 7.0 5.3 1.0 6.5 7.9 4.1 3.2 0.9 1.0

18.5 0.5 7.5 12.1 1.5 0.8 0.3 3.2 0.3 1.9 3.1 2.0 4.7 0.1 0.5 1.8 2.3 0.5 0.6 7.7 4.4 1.7 6.4 7.7 4.4 3.8 0.8 0.9

18.1 0.5 6.2 12.6 2.4 0.9 0.4 2.8 0.3 2.1 2.7 2.1 4.9 0.1 0.5 1.9 2.8 0.5 0.6 7.9 2.8 2.6 6.2 7.3 4.4 4.9 0.7 1.0

17.5 0.5 6.3 12.6 2.7 0.9 0.3 2.6 0.2 2.3 2.2 2.2 5.0 0.2 0.4 2.1 3.1 0.5 0.5 7.9 3.0 2.6 6.4 7.0 4.5 4.8 0.7 1.1

Note: * Data not available after 1997.

the share of Electrical Machinery increased from 7.4 per cent in 1976–80 to 18.8 per cent in 1996–9. Some industries have seen significant growth in real value added over the period 1975–99 (Table 3.7). The industries which have witnessed the most significant expansion are Wearing Apparel, Petroleum Refineries and Plastic Products. Other Industries (ISIC 390) (which includes jewellery) has seen strong growth in the 1990s. In keeping with the expansion of production and employment in Indian manufacturing since the early 1980s, the total number of factories has shown a steady increase in the same period (Figure 3.5). However, the average factory size with respect to employment in Indian manufacturing has remained stable – a typical factory in India employs around 70 workers (Figure 3.6). With respect to the distribution of factories across industries, we do not see


41

Figure 3.7 Manufacturing exports, India. Source: Author’s calculations, data from International Economic Data-bank, Australian National University.

significant differences over time, with about one-third of all factories located in food and textile industries (Table 3.8).

4 Patterns and trends in exports and imports There has been a significant increase in manufacturing exports over the period 1975–1999. In 1975, manufacturing exports was US $3.0 billion. In 1999, manufacturing exports had reached US $23.5 billion. Much of the increase occurred from the mid-1980s onwards (Figure 3.7). However, there seems to be a stagnation in manufacturing exports in the period 1996–9. The structure of exports shows a slight increase in TCF exports from 33 per cent in 1976–80 to 37.8 per cent in 1996–9 (Table 3.9). Among other exports, ISIC 352 – Other Chemical Products – has increased its share from 1.8 per cent in 1976–80 to 8.2 per cent in 1996–9. Manufacturing imports also show a steady increase – however, the increase in the case of manufacturing imports occurs from the mid-1970s as compared to manufacturing exports, which show an increase only in the 1980s (Figure 3.8). Within manufacturing imports, the largest increases are in transport equipment (ISIC 384) and professional and scientific equipment (ISIC 385) – these two industries comprise close to one-third of all manufacturing imports (Table 3.10). One important feature of manufacturing imports since the mid1980s has been the increase in differentiated capital and intermediate goods,

42


Table 3.9 Structure of Indian exports (percentage share) Industries

1976–80

1981–5

1986–90

1991–5

1996–9

311 313 314 321 322 323 324 331 332 341 342 351 352 353 354 355 356 361 362 369 371 372 381 382 383 384 385 390

17.0 1.5 0.0 0.2 21.7 10.1 8.4 1.3 0.4 0.1 0.1 0.2 1.8 2.9 0.5 0.0 0.6 0.3 0.1 0.4 1.0 6.5 2.5 0.1 4.7 3.1 2.4 4.6

15.1 1.3 0.0 0.5 19.5 13.7 6.6 2.9 0.3 0.1 0.1 0.3 2.0 5.0 4.2 0.0 0.9 0.2 0.1 0.3 0.5 1.2 0.4 0.0 3.7 3.7 2.4 4.2

8.7 0.9 0.0 0.3 16.1 18.0 5.8 3.7 0.1 0.0 0.1 0.2 4.7 4.6 6.9 0.0 0.9 0.2 0.1 0.2 0.4 1.7 0.5 0.0 2.3 3.9 2.5 3.1

9.6 0.7 0.1 0.2 16.5 19.5 3.5 2.9 0.2 0.0 0.2 0.2 6.3 4.3 2.5 0.0 1.3 0.8 0.1 0.3 1.0 3.8 0.9 0.0 2.6 3.2 2.1 4.1

7.9 1.0 0.0 0.1 19.0 18.0 2.3 0.8 0.4 0.4 0.1 0.3 8.2 2.1 1.2 0.0 1.1 1.9 0.1 0.4 1.2 4.1 0.9 0.0 3.7 3.0 2.9 3.3

TCF NON TCF

33.0 67.0

36.1 63.9

37.7 62.3

38.9 61.1

37.8 62.2

Source: Author’s calculations, data from Internationl Economic Data-bank, Australian National University. Note: TCF = Textile, Clothing and Footwear; Industries 321, 322 and 324.

as the trade regime increasingly allowed the importation of these goods via the OGL policy. This is evident in the increase in the Grubel–Lloyd measure of intra-industry trade in capital and intermediate goods, which is a good measure of the extent of trade in differentiated products (Chand and Sen 2002).1 This can be seen in Figure 3.9, where the Grubel–Lloyd measure shows a sharp increase since 1986. Examining trends in import penetration ratios, we observe that there has generally been an increase in import penetration ratios since the mid-1980s, though there are significant differences in the degree of import penetration across industries (Table 3.11).2 Export orientation ratios also vary a great deal across industries – industries such as Wearing Apparel (ISIC 322), Furniture


43

Figure 3.8 Manufacturing imports, India. Source: Author’s calculations, data from International Economic Data-bank, Australian National University.

Figure 3.9 Index of intra-industry trade for capital and intermediate goods, India. Source: Author’s calculations, data from International Economic Data-bank, Australian National University.

44


Table 3.10 Structure of Indian imports (percentage share) Industries

1976–80

1981–5

1986–90

1991–5

1996–9

311 313 314 321 322 323 324 331 332 341 342 351 352 353 354 355 356 361 362 369 371 372 381 382 383 384 385 390

12.6 0.3 0.1 0.0 1.2 0.0 0.0 0.0 0.1 0.1 2.7 0.5 16.7 2.3 12.6 0.0 0.4 1.0 0.0 0.3 1.1 9.9 5.1 0.9 3.0 12.1 3.9 7.3

9.5 0.3 0.0 0.0 1.3 0.0 0.0 0.0 0.1 0.1 2.6 0.5 15.8 2.1 14.0 0.1 0.6 1.0 0.0 0.3 0.8 11.1 3.7 1.6 3.7 12.3 5.0 6.6

5.5 0.3 0.0 0.0 1.7 0.1 0.2 0.0 0.2 0.2 2.8 0.6 15.9 2.6 12.0 0.2 0.5 1.5 0.0 0.5 0.3 9.3 4.2 3.2 3.0 12.3 7.1 6.1

3.1 0.5 0.0 0.0 2.2 0.1 0.4 0.0 0.2 0.2 2.4 0.5 19.2 3.2 17.1 0.4 0.6 1.2 0.0 0.4 0.3 6.3 3.8 2.9 2.2 10.9 6.1 6.2

6.1 0.4 0.1 0.1 2.4 0.3 0.4 0.0 0.2 0.4 2.2 0.7 12.2 3.5 4.4 0.6 0.8 1.6 0.1 0.5 0.4 5.1 2.8 1.3 3.3 18.5 10.2 9.6

Source: Author’s calculations, data from International Economic Data-bank, Australian National University.

(ISIC 332) and Other Manufacturing Industries (ISIC 390) are significantly more export oriented than others (Table 3.12).3

5 Concluding remarks In this chapter, we set out the ‘stylised facts’ of Indian manufacturing for the period 1975–99 with respect to employment, real wages, production and trade. It is clear that after a period of stagnation in labour productivity in the 1970s, there has been strong growth in labour productivity and real value added since the early 1980s. After a period of rationalisation when employment fell in the first half of the 1980s, there has been growth in employment, especially in the first half of the 1990s. Real wages have also increased in this period, though lagging behind labour productivity growth. Manufacturing exports and imports have also increased in the


45

Table 3.11 Import penetration ratios, India (per cent) Industries

1976–80

1981–5

1986–90

1991–5

1996–9

311 313 314 321 322 323 324 331 332 341 342 351 352 353 354 355 356 361 362 369 371 372 381 382 383 384 385 390

0.8 0.1 0.0 0.1 0.0 0.0 0.0 0.2 2.7 1.5 0.4 2.6 0.3 2.4 0.0 0.2 1.6 0.2 0.7 0.5 1.0 3.1 1.2 1.9 0.6 1.4 4.7 10.2

0.8 0.1 0.0 0.1 0.1 0.1 0.0 0.3 3.3 1.7 0.5 2.6 0.4 2.2 0.1 0.3 1.6 0.2 0.9 0.3 1.2 3.0 2.0 2.4 1.1 1.3 6.6 15.5

0.5 0.1 0.0 0.2 0.1 0.4 0.1 0.6 7.2 1.7 0.7 2.5 0.5 2.0 0.2 0.3 1.5 0.2 1.6 0.1 1.0 2.5 1.7 2.5 1.4 1.1 6.4 22.6

0.3 0.1 0.0 0.3 0.1 1.0 0.1 0.7 9.7 1.6 0.6 3.0 0.7 4.2 0.5 0.4 1.1 0.3 1.3 0.1 0.8 1.8 1.2 2.3 1.3 1.2 5.6 13.0

0.5 0.2 0.2 0.3 0.2 0.9 0.1 0.8 16.3 1.4 0.8 1.5 0.6 1.0 0.9 0.5 1.1 1.1 1.6 0.1 0.6 1.3 1.5 3.6 2.0 1.3 8.2 9.5


same period. However, a worrying feature of manufacturing performance has been the slow-down in the late 1990s in employment growth and a fall in real wages. With respect to the structure of production and employment, there has been a gradual contraction of the textile, clothing and footwear industries, which are the more labour-intensive industries in the manufacturing sector. However, these industries have increased their shares in manufacturing exports. The shift away from the key labour-intensive industries in the period of the study is a disquieting feature of structural change in the Indian manufacturing from an employment point of view. At the same time, the strong growth in labour productivity is one of the positive features of manufacturing performance in the same period. The important question that we will attempt to answer is whether trade policy changes have had any role to play

46


Table 3.12 Export orientation ratios, India (per cent) Industries

1976–80

1981–5

1986–90

1991–5

1996–9

311 313 314 321 322 323 324 331 332 341 342 351 352 353 354 355 356 361 362 369 371 372 381 382 383 384 385 390

0.9 0.0 0.1 1.1 15.7 8.4 5.1 0.7 1.8 0.1 0.1 0.2 0.3 0.1 0.0 0.3 0.5 0.6 1.0 0.3 0.6 1.3 1.5 0.4 0.3 0.7 0.8 14.9

0.7 0.0 0.2 1.0 15.2 7.4 7.8 0.4 1.2 0.0 0.2 0.2 0.5 0.3 0.0 0.3 0.2 0.2 0.5 0.1 0.1 0.2 1.1 0.4 0.3 0.5 1.0 19.1

0.5 0.0 0.2 1.2 17.5 7.8 10.2 0.3 0.6 0.0 0.2 0.5 0.6 0.8 0.0 0.4 0.1 0.5 0.4 0.1 0.1 0.2 0.9 0.6 0.3 0.4 1.0 30.8

0.9 0.1 0.2 2.0 17.5 8.6 9.5 0.7 2.7 0.2 0.2 1.0 0.9 0.7 0.0 0.9 0.7 1.5 0.9 0.4 0.5 0.4 1.5 0.7 0.5 0.8 1.1 23.4

0.7 0.0 0.1 2.2 14.6 5.8 2.7 1.9 16.2 0.1 0.3 1.1 0.4 0.3 0.0 0.7 1.4 0.7 1.1 0.5 0.5 0.4 1.8 0.6 0.6 0.5 1.2 18.8


in explaining either of these two important ‘stylised facts’ of Indian manufacturing performance since the mid-1970s.

4


1 Introduction In the 1980s and 1990s, the Indian economy grew at a rate of 5.8 per cent per annum, considerably higher than the rate of economic growth that was witnessed in the previous decades.1 At the same time, as we have described in the previous chapter, there was significant trade liberalisation in the 1980s and 1990s in India. Can we explain the higher economic growth observed in the Indian economy since the beginning of the 1980s by pointing to the changes in trade policy that occurred during the same period? In this chapter, we examine the relationship between trade policy and economic growth, paying particular attention to how the trade policy changes may have impacted on the manufacturing sector. Trade policy can positively impact on economic growth either by increasing the rate of investment or by increasing the efficiency of investment.2 In this chapter, we focus our attention on the former mechanism. In particular, we examine the relationships between equipment investment and growth, and between the relative price of capital and equipment investment in this chapter. This is for the following reason. As we have seen in Chapter 2, the major reforms that occurred in the trade regime early on in trade liberalisation process were to do with the liberalisation of capital and intermediate goods imports and less to do with the liberalisation of consumer goods imports. As we will argue in this chapter, liberalisation measures that specifically target capital and intermediate goods imports may bring about a fall in the relative price of capital goods. This may lead to an increase in the rate of investment in equipment, and hence in the rate of economic growth, and provide a clear causal mechanism to relate the trade policy changes that occurred in India in the 1980s and 1990s to economic growth. We will, however, examine the effect of trade reforms on the efficiency of investment when we study the impact of these reforms on total factor productivity growth and prices in Chapter 5. The rest of the chapter is divided into five sections. In the next two sections, we set the stage for the empirical analysis by describing the Indian growth experience and then attempting to identify patterns in economic growth and investment behaviour in India during the period of our study. Section 4

48


presents the conceptual framework that relates the relative price of capital, equipment investment and economic growth. Section 5 discusses the empirical results. Section 6 concludes.

2 Economic growth in India – a brief overview We begin with a graph of GDP growth (Figure 4.1). As is expected of an economy where climactic factors play an important role in determining total output, output growth shows a great deal of variation from year to year. There is, however, a clear increase in the trend rate of growth of output in the 1980s and 1990s as compared to the earlier period. This is also clear from Table 4.1, where one can see that the high growth rates of the 1980s and 1990s had not been witnessed in the earlier decades, except briefly in the early 1960s. It is also clear that much of the higher economic growth of the 1980s was primarily due to the manufacturing sector, which was the fastest growing sector in the 1980s. The high growth of the manufacturing sector was predominantly due to the registered manufacturing sector. In contrast, the unregistered manufacturing sector’s share in total output remained more or less constant during the 1980s and 1990s (Table 4.2). Beginning in the mid-1980s, trade, hotels and restaurants, along with financial services (TRFIN), also became an important source of growth, and in the late 1990s the infrastructural sector also grew rapidly. Over the period 1955–2000, the agricultural sector steadily declined in importance, as its share in total GDP fell from 55 per cent in 1960 to 27 per cent in 2000. The manufacturing sector’s share of GDP increased from 11 per cent to 17 per cent for the same period. Trade, hotels and restaurants along with financial services also increased from 16 per cent to 27 per cent in the same period.

3 Investment behaviour in India: trends and patterns In this section, we present a brief overview of investment behaviour in India over the period 1955–2003, highlighting the key features of the data. There has been an increase in gross fixed capital formation (GFCF) as a ratio of GDP over the period 1955–2003, with the rate of gross domestic capital formation standing at 23.7 per cent in 2000–3 (Figure 4.2 and Table 4.3). Much of this increase has been due to a steady rise in total fixed investment. The increase in the fixed investment to GDP ratio in this period can be clearly attributed to a rapid increase in the ratio of equipment investment to GDP, particularly since the 1980s. In contrast, the ratio of structures investment to GDP has been stagnant since the mid-1970s (Figure 4.3 and Table 4.3). The rate of equipment investment increased from 8 per cent in 1975–9 to 16 per cent in 2000–3. At the same time, the rate of structures investment fell from 10 per cent to 7 per cent for the same period. The increase in fixed investment as a ratio of GDP since the 1980s cannot


49

Figure 4.1 Growth of Indian GDP, 1955–2003. Source: National Accounts Statistics, Central Statistical Organisation, various issues.

Table 4.1 Growth rates of output, by sector and aggregate (per cent, per annum) Average over years

AMQ MF REGMF UNREGMF CON TRFIN INF

CSPS GDP

1955–9 1960–4 1965–9 1970–4 1975–9 1980–4 1985–9 1990–4 1995–9 2000–3

1.9 3.4 1.8 1.2 1.5 5.8 3.5 3.6 2.7 2.4

4.0 6.0 4.5 4.1 4.1 5.1 7.1 3.6 9.7 5.0

6.2 8.1 3.7 3.4 5.2 6.3 7.8 5.4 6.6 6.1

8.2 10.1 4.8 2.7 5.8 7.8 7.9 6.4 6.0 6.7

4.4 5.8 2.3 4.3 4.6 4.1 7.5 3.7 7.6 5.1

7.4 8.6 5.7 −1.4 5.3 4.1 5.6 4.7 6.6 6.2

4.5 5.6 3.3 3.2 5.4 6.2 8.3 7.2 8.9 6.8

7.7 7.4 6.0 5.5 6.6 5.9 7.2 6.5 8.6 10.7

3.8 5.2 3.0 2.1 3.7 5.6 6.3 4.9 6.5 5.4

Source: National Accounts Statistics, Central Statistical Organisation, various issues; our calculations. Note: AMQ = Agriculture, Mining and Quarrying; MF = Manufacturing; REGMF = Registered Manufacturing; UNREGMF = Unregistered Manufacturing; CON = Construction; TRFIN = Trade, Hotels, and Restaurants + Financing, Insurance, Real Estate and Business Services; INF = Electricity, Gas and Water Supply + Transport, Storage and Communication; CPS = Community, Social and Personal Services.

50


Table 4.2 Share of major sectors in GDP, India (percentage share) Years

AMQ

MF

REGMF UNREGMF CON

TRFIN INF

CSPS

1960 1970 1980 1990 2000

54.7 47.4 39.9 35.3 27.4

10.8 12.9 14.9 16.6 16.7

5.6 7.8 8.8 10.6 10.8

16.0 16.8 19.4 22.1 27.3

9.3 10.6 12.1 12.3 13.4

5.1 5.2 6.0 6.0 5.9

4.7 6.4 5.8 5.2 5.1

4.5 5.9 8.0 8.5 10.1

Source: National Accounts Statistics, Central Statistical Organisation, various issues; our calculations. Note: AMQ = Agriculture, Mining and Quarrying; MF = Manufacturing; REGMF = Registered Manufacturing; UNREGMF = Unregistered Manufacturing; CON = Construction; TRFIN = Trade, Hotels, and Restaurants + Financing, Insurance, Real Estate and Business Services; INF = Electricity, Gas and Water Supply + Transport, Storage and Communication; CPS = Community, Social and Personal Services.

be said to have resulted from an increase in public fixed investment, which has been falling as a ratio of GDP (both in the aggregate and in each of the components) since the mid-1980s (Figure 4.4). Instead, the rise in aggregate fixed investment could be attributed to the sharp increase in private fixed investment since the mid-1980s (Figure 4.5). The ratio of private fixed investment

Figure 4.2 The investment rate and its components, 1955–2003, India. Source: National Accounts Statistics, Central Statistical Organisation, various issues.


51

Table 4.3 Gross investment and its components as a per cent of GDP, India Average over years

GDCF

Gross fixed capital formation

Total investment in structures

Total investment in equipment

1955–9 1960–4 1965–9 1970–4 1975–9 1980–4 1985–9 1990–4 1995–9 2000–3

17.4 18.8 19.9 20.5 21.8 21.4 22.2 22.5 23.3 23.7

15.6 16.7 18.3 17.2 18.2 19.0 19.8 21.0 23.4 23.3

8.8 9.3 11.3 10.1 10.2 8.7 7.7 7.6 7.0 7.4

6.8 7.3 7.0 7.1 8.0 10.3 12.1 13.4 16.4 15.9

Source: National Accounts Statistics, Central Statistical Organisation, various issues; our calculations. Note: GDP and Investment data are in constant prices.

Figure 4.3 Total fixed investment and its components, 1955–2003, India. Source: National Accounts Statistics, Central Statistical Organisation, various issues.

to GDP increased from 9.5 per cent in 1980–84 to 18 per cent in 2000–3, while the public fixed investment rate fell from 9.5 per cent to 5.4 per cent in the same period (Table 4.4). The steady increase in the private fixed investment rate was primarily due to a sharp rise in the rate of private equipment investment, with the private structures investment rate showing no signs of buoyancy. The private equipment investment rate rose from 5.8 per cent in 1975–9

52


Figure 4.4 Total public fixed investment and its components, 1955–2003, India. Source: National Accounts Statistics, Central Statistical Organisation, various issues.

to 13.8 per cent in 2000–3. In contrast, the private structures investment rate remained at around 3–5 per cent for the period 1980–2003. The increase in the private equipment investment rate was primarily due to an increase in the private corporate equipment investment rate, which increased at a remarkable pace from 1.3 per cent in 1975–9 to 8.9 per cent in 1995–9, after declining to 5.7 per cent in 2000–3 (Figure 4.6 and Table 4.5). On the other hand, the household equipment investment rate showed no sign of buoyancy in the period 1980–2000, increasing from 3.1 per cent in 1975–9 to 4.1 per cent in 1995–9 (Figure 4.7 and Table 4.5). It is only after 2000 that it increased rapidly to 7.7 per cent of GDP for the period 2000–3. Structures investment by the private corporate and household sectors did not show any significant increase after the 1980s. One possible explanation of the sharp rise in private equipment investment could be the fall in the relative price of equipment investment. The latter showed a negative trend from the late 1970s, in contrast to the relative price of structures investment which showed a significant increase in the 1980s (Figure 4.8). To sum up, the behaviour of gross fixed investment in India suggests a clear upward trend in the series as a ratio of GDP since the 1980s, driven primarily by a spectacular increase in the private equipment investment rate, in spite of


53

Figure 4.5 Total private fixed investment and its components, 1955–2003, India. Source: National Accounts Statistics, Central Statistical Organisation, various issues.

Table 4.4 Components of fixed investment by sector as a per cent of GDP, India Average Public Public over years total fixed investment investment in structures

Public Private investment total fixed in investment equipment

Private investment in structures

Private investment in equipment

1955–9 1960–4 1965–9 1970–4 1975–9 1980–4 1985–9 1990–4 1995–9 2000–3

2.0 2.7 2.6 2.5 3.4 4.4 5.0 4.7 3.5 2.5

4.6 4.0 6.2 5.4 5.5 3.7 3.3 4.0 4.1 4.6

4.8 4.6 4.4 4.7 4.6 5.8 7.1 8.7 12.9 13.4

6.2 8.1 7.7 7.1 8.2 9.5 9.5 8.3 6.4 5.4

4.2 5.4 5.1 4.6 4.8 5.0 4.5 3.6 2.9 2.8

9.4 8.6 10.6 10.1 10.0 9.5 10.3 12.7 17.0 18.0


stagnant or declining rates of investment in structures and public investment in equipment. At the same time, there has been an increase in the average annual growth rate of output and a fall in the relative price of equipment in the same period. In the econometric analysis, we will explore more rigorously whether the behaviour of these three series – the relative price of equipment,

54


Figure 4.6 Total private corporate fixed investment and its components, 1955–2003, India. Source: National Accounts Statistics, Central Statistical Organisation, various issues.

Table 4.5 Components of private fixed investment by sector as a per cent of GDP, India Average over years

Private corporate total fixed investment

Private corporate investment in structures

Private Household Household Household corporate total fixed investment investment investment investment in in in equipment structures equipment

1955–9 1960–4 1965–9 1970–4 1975–9 1980–4 1985–9 1990–4 1995–9 2000–3

2.2 3.0 1.7 1.8 1.6 3.5 3.6 6.1 9.7 6.4

0.4 0.7 0.4 0.3 0.2 0.4 0.4 0.6 0.9 0.6

1.8 2.3 1.3 1.5 1.3 3.1 3.3 5.5 8.9 5.7

7.3 5.6 8.9 8.3 8.5 6.0 6.7 6.6 7.3 11.6

4.2 3.3 5.8 5.1 5.2 3.3 2.9 3.5 3.2 3.9

3.0 2.3 3.1 3.2 3.3 2.7 3.8 3.1 4.1 7.7



55

Figure 4.7 Total household fixed investment and its components, 1955–2003, India. Source: National Accounts Statistics, Central Statistical Organisation, various issues.

the rate of equipment investment and the growth rate of output – can be causally linked.

4 Trade policy and economic growth: conceptual issues and empirical framework 3 The argument that trade policy can positively affect economic growth via an increase in equipment investment is based on three behavioural relationships. These are: (i) the link between equipment investment and economic growth; (ii) the link between the relative price of equipment and the rate of equipment investment; and (iii) the link between trade policy and the relative price of equipment. We discuss each of these three relationships in turn. The link between equipment investment and growth A vast number of cross-country studies have examined the determinants of economic growth using a wide range of explanatory variables. While the set of factors that has been found to be important in explaining economic growth has differed from study to study, Levine and Renelt (1992) have found that the significance of the investment rate in explaining economic growth

56


Figure 4.8 The relative price of equipment and structures, 1955–2003, India. Source: National Accounts Statistics, Central Statistical Organisation, various issues.

remains robust to different specifications. There seems to be little doubt arising from the empirical literature that the investment rate is a crucial determinant of economic growth, if not the key determinant. Furthermore, the new growth literature has argued that, among the three components of total investment – investment in equipment, investment in structures and investment in inventories – the most important for growth is equipment investment (De Long and Summers 1991, 1992, 1993, Temple 1998). According to this view, the social return to equipment investment is far greater than that for other types of investment. The reasons are not very clear, though as De Long and Summers argue, ‘historical accounts of economic growth invariably assign a central role to mechanization’ (1991: 447). It could be that the role of external economies may be greater for equipment investment than that for structures investment, possibly due to the greater amount of research and development expenditures in the machinery sector. Among the countries in the De Long and Summers sample, India has had one of lowest rates of equipment investment for the period 1960–85. India also had one of the lowest levels of income per capita in 1980 in the same sample of countries.4 De Long and Summers show that an increase an extra 1 percentage point devoted to equipment investment is associated with an 0.302 percentage point increase in the annual GDP per worker growth rate.5


57

To assess the impact of equipment investment on the growth rate of output, we use a simple empirical formulation similar to that used by De Long and Summers. As they do, we take the growth rate of output to be a linear function of the rates of equipment and structures investment, along with the third component of gross investment – inventory investment (changes in stocks). However, in our context, it would be useful to take the rate of private equipment investment separately from the rate of public equipment investment in the output growth equation. This is for two reasons. First, the rate of return on private equipment investment may be different from that on public equipment investment. Second, our purpose in the next stage of the analysis is to explain the behaviour of equipment investment and it can be argued that public investment is in great part exogenously determined, and is influenced more by political and institutional factors than by economic variables. Thus, the determinants of public investment may be quite different from those of private investment. The final specification is as follows:6 GY = a1PVEQI + a2PBEQI + a3STI + a4CST

(4.1)

where: GY = growth rate of GDP; PVEQI = ratio of private sector investment in equipment to GDP; PBEQI = ratio of public sector investment in equipment to GDP; STI = ratio of investment (both private and public) in structures to GDP; CST = ratio of changes in stocks (inventories) to GDP. The link between the relative price of equipment and the rate of equipment investment Since we are interested in examining the relationship between the relative price of equipment and the rate of equipment investment, we confine our empirical analysis for this section to the determinants of private equipment investment (as we have argued earlier, public investment may be taken to be exogenously determined). We model the rate of equipment investment by the private sector as follows: PVEQI = b1RPE + b2FINT + b3RI + b4PBI + b5 D91+ b6PVEQI (− 1) (4.2) where: RPE = Logarithm of relative price of equipment (price deflator for equipment investment as a ratio of the GDP deflator); RI = the real interest rate (the bank lending rate minus the inflation rate); FINT = financial deepening, measured by real domestic credit to the private sector;

58

Trade policy and economic growth PBI = total public fixed investment as a ratio of GDP; D91 = dummy variable to capture possible effects of the 1991 reforms.

We would expect from theory that the sign of the coefficient for the relative price of equipment, b1, will be negative – an increase in the latter will decrease the rate of equipment investment in the economy. Financial deepening is expected to have a positive impact on equipment investment. Financial deepening can increase both the volume and efficiency of investment. The ‘debt accumulation’ hypothesis of Gurley and Shaw (1955), formalised more recently by Bencivenga and Smith (1991), argues that the spread of organised finance can help overcome indivisibilities in investment through the mobilisation of otherwise unproductive resources. Moreover, financial intermediaries and markets play an important role in selecting the most promising firms and households for lending purposes and thus contributing to the more efficient use of capital (Levine 1997). Financial intermediaries may also enhance the quality of investment by identifying those entrepreneurs with the best chances of successfully initiating new goods and production processes (King and Levine 1993). In the Indian case, Bell and Rousseau (2001) find conclusive evidence of the positive impact of financial deepening on gross domestic fixed investment in the post-1950 period. Following Bell and Rousseau, we use the ratio of real domestic credit to the private sector as a ratio of GDP as our preferred measure of financial deepening (FINT).7 An increase in the real interest rate (RI) is expected to have a negative effect on the rate of equipment investment via an increase in the cost of capital. Public investment (PBI) may affect private investment via both supply and demand sides. On the supply side, the private sector relies on public investment for most of the infrastructure, because this is either a natural or a legal monopoly of the government. Public infrastructural investment can affect private equipment investment by influencing its rate of return – poor roads, an erratic supply of electricity or inadequate communication facilities can negatively affect the amount of output that is possible to obtain from a given amount of investment. Thus public investment in infrastructure and private investment should be complementary (Blejer and Khan 1984). On the demand side, the relationship is ambiguous. If there is some slack in the economy one would expect a change in public investment to push private investment in the same direction. Otherwise, some private investment will probably have to be ‘crowded out’ (Athukorala and Sen 2002). Finally, to capture the costs to adjusting the capital stock and gestation lags in the adjustment process, which would imply that firms gradually respond to changes in the desired capital stock (as is standard in the neoclassical model of investment behaviour; see Jorgenson 1967), we include the oneperiod lagged private equipment investment rate. We also include a dummy variable taking the value one from 1991 onwards (D91) to capture the effects of the 1991 reforms.

59

Trade policy and economic growth The link between trade policy and the relative price of equipment

Machinery is a key tradable commodity and its relative price will be greatly influenced by trade policy. We choose a very simple specification for the relative price of equipment, modelling it purely as a function of the trade policy reform dummies of 1985 and 1991. We also add a time-trend to the specification to capture the upward drift in the relative price series from the mid-1950s to the late 1970s, as evident from Figure 4.8. The lagged relative price of equipment investment term captures the high persistence of the series. RPE = c0 + c1TIME + c1D85 + c2D91 + c3 RPE(− 1)

(4.3)

where: TIME = time trend; D85 = dummy variable for the 1985 trade reforms (value of one from 1985 onwards, zero otherwise); D91 = dummy variable for the 1991 trade reforms (value of one from 1991 onwards, zero otherwise);

5 Results and findings Our period of analysis is 1955/56 to 2003/04. A summary of the data used in the regressions is provided in Table 4.6 and the final results in Table 4.7. We first estimate equation (4.1) using Ordinary Least Squares (OLS). There is a possibility that equation (4.1) may be subject to simultaneity bias as we have equipment and structures investment rates as explanatory variables in the regression. This is because positive and significant coefficients on the rates of private equipment investment and structures investment could imply that higher investment rates are the result of economic growth, not the other way around. Therefore, we re-estimate equation (4.1) using Two Stage Table 4.6 Summary statistics of variables used in the empirical analysis, India Statistics

GY

GYM PBI

Mean 4.6 5.9 Median 5.1 6.6 Maximum 9.9 14.9 Minimum −5.3 −3.6 Std Dev. 3.0 3.8

7.7 7.7 10.4 5.1 1.4

PBEQI PVEQI STI

CST RPE

3.4 3.2 5.3 1.4 1.1

2.1 −0.19 20.1 1.9 −0.17 12.6 4.9 0.12 77.9 −1.9 −0.58 1.8 1.4 0.15 20.1

6.9 5.7 13.8 2.9 3.3

8.8 8.7 11.6 6.7 1.4

FINT RI 0.05 0.06 0.16 −0.16 0.06

Source: National Accounts Statistics, Central Statistical Organisation, various issues; International Financial Statistics, International Monetary Fund, for FINT and RI. Note: GY = Growth Rate of Real GDP; GYM = Growth Rate of GDP in Manufacturing; PBI = Total Public Investment (ratio of GDP); PBEQI = Public Investment in Equipment (ratio of GDP); PVEQI = Private Investment in Equipment (ratio of GDP); STI = Total Investment in Structures (ratio of GDP); CST = Inventory Investment (ratio of GDP); RPE = Relative Price of Equipment (in logs); FINT = Real Bank Credit to the Private Sector; RI = Real Interest Rate.

Table 4.7 Regression results – economic growth – India Independent variables

Dependent variable GY Col. (1)

GY Col. (2)

GYM Col. (3)

PVEQI Col. (4)

RPE Col. (5)

Intercept

–

–

–

–

PBEQI

0.40 (1.33) 0.30*** (3.03) 0.15 (1.09) −0.06 (0.17) –

0.66 (1.50) 0.20 (1.62) 0.22 (1.22) 0.12 (0.24) –

– –

–

–

–

–

–

PBI

0.48* (1.61) 0.27*** (3.15) 0.14 (0.99) −0.05 (0.15) –

−0.12** (2.17) –

–

RPE

–

–

–

PVEQI(−1)

–

–

–

RPE(−1)

–

–

–

0.21*** (3.79) −2.30** (2.26) 0.52*** (4.40) –

FINT

–

–

–

RI TIME

– –

– –

– –

0.05*** (2.94) – –

D85

–

–

–

–

D91

–

–

–

1.46** (2.33) 0.93 1.79

0.004* (1.87) −0.09** (2.22) −0.06* (1.64) 0.71 1.83

PVEQI STI CST

R-square Durbin Watson statistic (Durbin’s h statistic for Col. (4) Estimation method Instruments used

0.06 2.42

– – 0.67*** (4.51) – –

0.06 2.42

0.02 1.46

OLS

TSLS

TSLS

OLS

OLS

–

RI, FINT, RPE

RI, FINT, RPE

–

–

Note: GY = Growth Rate of Real GDP; GYM = Growth Rate of GDP in Manufacturing; PBI = Total Public Investment (ratio of GDP); PBEQI = Public Investment in Equipment (ratio of GDP); PVEQI = Private Investment in Equipment (ratio of GDP); STI = Total Investment in Structures (ratio of GDP); CST = Inventory Investment (ratio of GDP); RPE = Relative Price of Equipment (in logs); FINT = Real Bank Credit to the Private Sector; RI = Real Interest Rate. OLS = Ordinary Least Squares; TSLS = Two Stage Least Squares. T-ratios in Brackets; Heteroskedasticity-consistent standard errors and covariance; ***, ** and * denote significance at the 1, 5 and 10 per cent levels.


61

Least Squares (TSLS) and instrumenting private equipment and structures investment with the real interest, financial deepening and public fixed investment. In both the OLS and TSLS estimates, we find that that the coefficient on private equipment investment rate is positive and statistically significant at the 1 per cent level. In contrast, the coefficient of public equipment investment is not statistically significant in the TSLS estimates though its magnitude is higher at 0.40 compared to 0.30 for private equipment investment. Structures and inventory investment do not seem to have a significant impact on economic growth. Our finding that an increase in the rate of private equipment investment has a far stronger positive impact on the growth of output than an increase in the rate of structures investment is consistent with what has been observed in the cross-country studies. We also re-estimate equation (4.1) using TSLS and substituting the growth rate of manufacturing instead of the growth rate of the economy as the dependent variable. This is to see if the private equipment investment rate has a similar effect on the manufacturing sector as it has on the entire economy. We find a weaker effect in this case, as compared to the estimates presented in Table 4.7 Col. (2). We now move to the estimate of equation (4.2). In the case of private equipment investment, the relative price of equipment plays a decisive role in its determination – its coefficient is negative and highly significant at the 5 per cent level. The coefficient on the relative price of capital is −2.30. Financial deepening has also been a key determinant of the private equipment investment rate in India. Public investment seems to play a complementary role to private investment, as indicated by the positive and significant coefficient on the public investment rate. In addition, the dummy for the post1991 period indicates a clear increase in equipment investment associated with that period. This may suggest that the deregulation of industrial policy in 1991 has played an important role in the observed increase in the equipment investment in the post-1991 period. We found that the coefficient on the real interest rate was not significant in all experimental runs of equation (4.2) and it was omitted from the final specification. Also, we included dummies for 1980 and 1985 to pick up increases in private equipment investment due to the reforms of these periods. Political scientists like Kohli (2006), and economists like De Long (2003) and Rodrik and Subramanian (2004), argued that the acceleration in India’s growth occurred primarily due to a change in the attitudes of the national government towards the private sector under the prime ministership of Indira Gandhi when she returned to power in 1980, from being anti-business to being pro-business, and less to do with economic policies and with the pro-reform stance of the Rajiv Gandhi government of 1985. However, we did not find that these dummies were significant at conventional levels of significance. Thus, in contrast to what has been argued by previous commentators on India’s growth acceleration, we do not seem to find that the reforms of the 1980s have had a positive impact on the private equipment investment rate independent of the effects they may have had via

62


the relative price of equipment.8 Finally, we see that there are high adjustment costs to changing the private capital stock, as indicated by the coefficient of 0.52 on lagged private equipment investment, which is significant at the 1 per cent level. Estimating equation (4.3), we see that while the trade liberalisation episode of 1985 has led to a larger fall in the relative price of equipment than the trade liberalisation episode of 1991, the coefficient on the latter is −0.09 as compared to a coefficient of −0.07 for the former. The fact that the more muted 1985 liberalisation episode has resulted in a larger fall in the relative price of equipment than the more comprehensive 1991 reforms is indeed surprising, and we will return to this issue later in the book. Using the coefficients estimated from equations (4.1), (4.2) and (4.3), it is possible to obtain an estimate of the contribution of the trade reforms of 1985 and 1991 to the average annual growth of GDP following the reforms. Our estimates indicate that the combined effect of the trade reforms of 1985 and 1991 was to increase the average annual growth rate of GDP during this period by 1.2 percentage points. The important role that the relative price of equipment can play in economic growth has also been pointed out by Jones (1994), who shows that there is a strong negative relationship between the relative price of equipment and economic growth, using a cross-section of 65 countries over the period 1960–85. At the beginning of the 1980s, India had one of the highest levels of the relative price of equipment investment among developing countries (Figure 4.9). Clearly, the subsequent improvement in India’s economic

Figure 4.9 The relative price of equipment in 1980 for India and other developing countries. Source: Jones (1994), data available at: http://elsa.berkeley.edu/~chad/datasets.html


63

growth rates can be linked to the trade policy reforms which brought about a significant decline in the relative price of equipment, boosting equipment investment and hence, growth.

6 Concluding remarks In this chapter, we have examined whether the growth acceleration that occurred in the Indian economy in the 1980s and 1990s can be causally linked to the trade policy changes that occurred in the same period. We have shown that the high growth rates of the 1980s and 1990s can mostly be attributed to the sharp increase in private equipment investment, and that the latter has a significantly more growth-enhancing effect than public equipment and structures investment. The increase in private equipment investment can itself be causally linked to the fall in the relative price of equipment that occurred due to the trade policy changes in 1985 and 1991. Our results suggest that, for the purpose of accelerating growth in the Indian economy, it is the more conservative trade policy reforms of 1985 that seem to have a stronger impact. This finding needs further discussion and we will return to this important issue in the concluding chapter, where we assess what we have learnt from the preceding chapters with respect to the welfare effects of the trade reforms that have occurred in the Indian economy in the 1980s and 1990s, and why these effects may have differed across the two sets of reforms.

5


1 Introduction Whether an economy gains from international trade is intimately linked to the question of whether trade reforms bring about positive productivity gains and declines in domestic prices in the economy in question. This chapter studies the effect of trade liberalisation on the Total Factor Productivity growth of Indian manufacturing. It also examines whether trade reforms have led to a decline in domestic market power in Indian industry, thereby causing a fall in real domestic prices. The rest of the chapter is structured as follows. Sections 2 and 3 present the analytical frameworks for studying the relationship between trade policy and productivity, and trade policy and prices respectively. In Section 4, we review recent literature that studies the effect of trade reforms on productivity and prices, especially in the Indian context. Section 5 summarises trends in productivity growth and in domestic market power. Section 6 undertakes the empirical analysis. Section 7 concludes.

2 Trade policy and productivity growth: the analytical framework 1 The mechanics linking trade and productivity is as yet an open question in the theoretical literature, as noted previously. To motivate the empirical analysis, we suggest two channels via which trade liberalisation impinges on Total Factor Productivity (TFP) growth. The first relies on the X-inefficiency literature where trade reform leading to increased international competition brings about a reduction in ‘slack’ in labour input (see Vousden 1993, Horn et al. 1996). The second channel via which trade reform raises productivity uses the ‘love of variety’ formulation of Dixit and Stiglitz (1977), where access to a greater variety of specialised inputs raises TFP growth when these inputs are imperfect substitutes for one another (Romer 1987). The above two mechanisms are incorporated within the standard neoclassical production function to deduce a reduced form that gives trade liberalisation a role in growth.


65

Let the sectoral production function be of the form Y = A.F(K,L*)

(5.1)

where Y is output, A is an index of Hicks neutral technological progress and is a constant, K is stock of physical capital and L is labour input with * denoting the effective quantity of the factor used in production. We observe Y, K and L where L = E(R) L.

(5.2)

E is an index for the quantity of effort put in by labour. R could be a vector comprising a host of variables that determine the level of effort put in by workers and the utilisation rate of capital, but here we will confine ourselves to considering the role of trade only. We measure the degree of protection by R. We would expect that E will be negatively related to R – that is, the lower the level of protection, the higher will be the effort put in by workers. We assume a simple linear function for E such that: E = E0 − e1R

(5.3)

Log linearising (5.1) and substituting in (5.3) gives Y = A + sk (K) + sL (E0 − e1R + L)

(5.4)

If the production function in (5.1) is linearly homogeneous, then sk and sL denote factor shares, the sum of which should equal one. Access to specialised inputs is another channel via which trade liberalisation can impinge on TFP growth. The role of intermediate inputs in productivity growth can be significant via two channels. First, if the production function takes the form Q = F(K,L,M)

(5.5)

where Q is output gross of material inputs, M, which in turn is a composite of imperfect substitutes, m of type j; ∞

M=

1 ρ

冤冱α m( j)冥 , 冱α = 1 j

j=0

j

(5.6)

j

where the elasticity of substitution between the individual inputs types is given by 1 −1 ρ. Now, an increase in access to differentiated inputs is also going to bring about a rise in TFP. Alternatively, we could think of K in equation (1) as an aggregate,

66

Trade policy, productivity and prices ∞

K=

1

冤冱γ k( j)冥 , 冱γ = 1, ρ

j

j=0

j

(5.7)

j

where j could for example be an index of the specificity and/or vintage of a particular type of capital. Output in equation (5.1) rises through gains in access to specialised inputs even when the overall quantity of inputs is held constant (see Grossman and Helpman 1991). Hence, liberalisation of the intermediate and/or capital goods sector(s) in the above framework leads to gains in TFP. Now incorporating either (5.6) or (5.7) into (5.5) gives an estimable equation of the form Y = α0 + α1K + α3L − β1 R + β2 J, α, β > 0

(5.8a)

where J denotes the number of intermediate or capital inputs available at time t, the αs denote technology coefficients, the βs measure the responsiveness of output growth to changes in policy-related variables. A constant returns to scale technology would imply that α1 + α3 = 1. The literature on trade policy and TFP growth suggests that there are a number of channels via which trade policy can impinge on growth, the robust finding from all these specifications is that that trade reforms lead to higher productivity growth, as implied by (5.8a) above. Controlling for K and L gives an alternate specification of (5.8) as TFP = α’0 − β’1 R + β’2 J,

(5.8b)

where TFP is Total Factor Productivity and is measured explicitly. We will estimate both variants of equation (5.8) in the empirical analysis that we will undertake in Section 6. We will capture R by alternate measures of protection – effective rates of protection (ERP), Import Coverage Ratios (ICR) and the price wedge as discussed in Chapter 2. We will capture J by the Grubel–Lloyd measure of intra-industry trade applied to intermediate and capital goods, as discussed in Chapter 3.

3 Trade policy and domestic prices: the analytical framework 2 To motivate the empirical analysis, we consider a simple mark-up formulation where domestic prices are determined by a function with two arguments, unit costs (c) and the level of mark-up (h). That is: PD = p(c, h)

(5.9)

where PD is real domestic price of the good in question. A simpler formulation, where domestic prices are a constant mark-up over unit costs is:

Trade policy, productivity and prices PD = ch

67

(5.10)

is admissible in the formulation given in equation (5.9) above. Log-linearising equation (5.9) gives: pD = a0 + a1c + a2h

(5.11)

Let the foreign good (F) be an imperfect substitute for the domestic good (D). We draw on models of industrial organisation to explain adjustments in mark-up in terms of the degree of market concentration, the extent of substitutability between domestic and foreign goods, and the magnitude of quantity and price competition from abroad.3 Hence, we let mark-up (h) be a function of the levels of the nominal exchange rate (e), the price of the foreign good (pF), the foreign firms’ share of the domestic market (m), and the domestic market power of firms (s). That is: h = ho + h1e + h2pF − h3m + h4s.

(5.12)

Substituting (5.12) in (5.11), the estimable equation can then be written as: PD = δ0 + δ1c + δ2e + δ3PF + δ4m + δ5s

(5.13)

The mark-up rule given in (5.10) would imply δ1 = 1, while a perfectly competitive market structure would, in addition to the above, also have δj = 0 for j ≠ 1. The interpretation for the rest of the coefficients in equation (5.13) are as follows: δ2 is the exchange-rate pass-through coefficient which is expected to be positive and bounded by zero and one; δ3 is the elasticity of domestic prices to foreign prices, and is expected to be a positive fraction; δ4 is the degree of responsiveness of domestic prices to quantity competition from abroad and is expected to be negative; δ5 is the elasticity of domestic prices to some measure of domestic market power, this coefficient is expected to be positive. The extent of exchange-rate pass-through is determined by the degree of homogeneity of the import to domestic output. The magnitude of δ2 can be judged by considering two limiting cases; one of a small open economy facing perfect competition such that exchange-rate pass-through is complete (within the time-frame considered) giving a unit value to δ2; the polar opposite case is when there is imperfect competition and where foreign suppliers absorb all changes in exchange rate in their profits giving a zero value to δ2. Furthermore, with strategic behaviour between suppliers, as is the case when there are few foreign and domestic firms selling differentiated products in the domestic market, the constant mark-up rule given in equation (5.10) is not justified (see Dornbusch 1986). In this case, realistic for the Indian manufacturing sector, δ2 will be strictly less than unity. The magnitudes of δ1 and δ2, therefore, are crucial for the interpretation of the results.

68


In our empirical analysis, we will measure foreign prices by the US producer price index at the industry level (see Chapter 2 and the Appendix for more details). We will use the India–US rupee/dollar nominal exchange rate for e. We define the exchange rate as the amount of rupees per US dollar, so an increase is a depreciation of the exchange rate. We will use two alternate measure of quantity competition from abroad – the Import Coverage Ratios discussed in Chapter 2, and import penetration ratios (as defined in Chapter 3) where we match the ISIC 4-digit classification of the trade data to the NIC 3-digit classification of the production data.

4 Previous empirical evidence on productivity and prices 4 In a survey of the literature in 1990 that examines the direct link between trade policy on one hand, and productivity and domestic market power on the other, Havrylyshyn (1990) observed that there were relatively few such studies. Since then, there has been an explosion of research on the relationship between trade policy and productivity, and on the effects of trade reforms on domestic market power, especially relating to developing countries. In this section, we will summarise the main findings of the empirical literature in relation to India. There have been several studies that have examined the effects of trade reforms on the efficiency of firms and industries in India. These are Khambhampati (2003), Driffield and Khambhampati (2003), Topalova (2004), Chand and Sen (2002) and Milner et al. (2007). The first three use firm-level data and the second two industry-level data. Khambhampati (2003) and Driffield and Khambhampati (2003) use stochastic frontier production function methods to estimate firm-level efficiency, and relate this variable to trade reforms, along with other variables. The first paper looks at the cotton textile industry and the second at five more industries. The authors find that average efficiency has increased in the post-reform period. There is some evidence that trade reforms may have contributed to the increase in efficiency. This is also found by Topalova (2004) in her detailed study of plant-level efficiency in Indian manufacturing. Unlike the first two papers, which measure trade reforms by outcome based indicators such as export and import intensities, Topalova uses nominal tariffs, which are a more accurate measure of trade policy. Among the industry-level studies of the link between trade reforms and productivity, Chand and Sen (2002) find clear evidence that trade reforms, in the form of a reduction in price distortions and an increase in the availability of specialised intermediate and capital goods, had a positive effect on Total Factor Productivity growth in the 1980s. This chapter is an extension of the Chand–Sen methodology to more recent data, especially the trade reforms of the 1990s. Milner et al. (2007) compute Total Factor Productivity growth in Indian industry at the NIC 3-digit level for the more recent period, and find evidence that it has increased in the post-reform period. They, however, use a


69

simple before/after trade reform methodology, which leaves the analysis open to alternate interpretations of the productivity increase. With respect to the effect of trade reforms on profitability, Khambhampati and Parikh (2003, 2005) study this phenomenon using firm-level data. They allow for the possibility that trade reforms can reduce profits via higher competition, but also increase profits via higher firm efficiency. They find that trade liberalisation has significantly influenced firm profitability via a procompetitive effect of exports on profit margins. Two other papers that study the effect of trade reforms on market power and efficiency simultaneously are Krishna and Mitra (1998) and Balakrishnan et al. (2002). The first paper finds evidence that, following the 1991 reforms, there is a reduction in domestic market power and an increase in the growth rate of productivity. The second paper finds the opposite with respect to market power and argues that it has increased since the reforms due to less competition as the number of firms contracted following the reforms. This occurred due to ‘the rationalisation of industry structure’ in the post-reform period, according to the authors.

5 Trend and patterns in productivity and prices in Indian manufacturing We plot Total Factor Productivity growth and price-cost margins for the entire manufacturing sector in Figures 5.1 and 5.2. Total Factor Productivity is calculated as the residual obtained from subtracting a Tornquist index of the primary inputs – capital and labour from the growth rate of real value added.5 We obtain real value added using the double-deflation method. Thus, real value added for industry i is defined as: Yi =

Oi Pi

−

Mi PMi

where Oi is Gross Value of Output and Pi is the product price of the industry, and Mi is the cost of total inputs and PMi is the price of materials for the industry. The use of the double-deflation method avoids the possible bias to the measurement of real value added emanating from changes in the relative price of inputs (Bruno 1978, 1984). As Balakrishnan and Pushpangadan (1994) have argued, there is significant variation in the latter in the Indian case. This makes the use of double-deflation preferable to single-deflation methods (where gross value added is deflated by the industry price) in the Indian context. Price-cost margins are defined as the difference between Gross Value Added and Wages of all production and non-production workers as a ratio of Gross Value Added. It is a measure of domestic market power – an increase in the price-cost margin will suggest an increase in the latter.

70


We find that there is no clear trend in Total Factor Productivity growth, which shows significant variation in the period under consideration. However, the aggregate price-cost margin shows a clear increase in the 1990s. The averages for price-cost margins and Total Factor Productivity growth may mask variations across industries, especially by the major use-based industrial groups. This is an important limitation – as we have seen in Chapter 2, there have been quite clear differences in the evolution of trade policy between consumer goods, intermediate goods and capital goods industries. We provide averages by these industrial groups and in the aggregate by sub-period in Table 5.1. We find that there is evidence of an increase in domestic market power (as captured by price-cost margins) across all three industrial groups over the period 1973–97. We also find that Total Factor Productivity growth for the capital and intermediate goods sectors was highest in 1985–90 and for the consumer goods sector, in 1973–8.

6 Empirical analysis In Table 5.2, we present the summary statistics of the key variables used in the regression analysis. The first column of the table reports the mean and the next two columns decompose the standard deviation into its betweenindustry and within-industry components, followed by the overall standard deviation. We adopt the terminology conventionally employed by panel data

Figure 5.1 Aggregate Total Factor Productivity growth in Indian manufacturing. Source: Author’s calculations. Note: Weighted average, weights are share of each industry in Gross Value Added.


71

Figure 5.2 Aggregate price-cost margin in Indian manufacturing. Source: Author’s calculations. Note: Weighted average, weights are share of each industry in Gross Value Added.

Table 5.1 Price-cost margins and total factor productivity growth, India Price-cost margins

1973–7 1978–84 1985–90 1991–7

Consumer goods

Intermediate goods

Capital goods

Average

0.70 0.70 0.73 0.79

0.74 0.75 0.78 0.84

0.71 0.72 0.73 0.80

0.71 0.72 0.75 0.82

Consumer goods

Intermediate goods

Capital goods

Average

6.8 3.9 5.6 5.1

5.0 4.9 7.0 5.7

6.7 6.0 7.5 5.7

6.1 4.7 6.5 5.2

Total Factor Productivity growth

1973–7 1978–84 1985–90 1991–7

Source: Author’s calculations. Note: Weighted average, weights are share of each industry in Gross Value Added; Price-Cost Margins = (Gross Value Added − Wages)/Gross Value Added.

72


studies: ‘between-industry’ refers to the differences in industry-specific averages across industries, where the averages are computed over time, and ‘within-industry’ refers to deviations of variables from these industry-specific means. We find that both within-industry and between-industry variation accounts for a significant proportion of the total variance of most of the variables to be used in the regression analysis (except of course the exchange rate and the index of intra-industry trade in intermediate and capital goods, which only vary over time). This indicates that studies that use measures of protection that are cross-sectional in nature to decipher a negative relationship between protection and Total Factor Productivity may not meet with much success in the Indian context, given the importance of time variation in the variables of interest. We first estimate equation (5.8a). We employ the fixed-effects estimator to allow for intrinsic differences across industries with respect to the rate of technological progress. We use the standard procedure of sweeping out the fixed effects by transforming variables to deviations from their industryspecific means. We assume that the impact of changes in protection on output growth in manufacturing is equal across the component NIC 3-digit industries. This assumption would be particularly valid in the event of complete factor mobility between these industries. We present our results in Table 5.3. Col. (1) is an estimate of equation (5.8a) without the protection variables. The rest are augmented with the variables as suggested in Section 2 above. Col. (2) includes the Import Coverage

Table 5.2 Summary statistics, Indian manufacturing Variables

Real value added TFP Price-cost margin Prices Labour Capital Materials J ICR ERP Exchange rate US price Factories

Mean

9.7 490.2 −0.34 4.9 9.5 5.4 9.5 46.6 64.3 101.7 2.7 4.7 6.6

Standard deviation Between industry

Within industry

Overall

1.76 916.5 0.09 0.09 1.52 1.85 1.84 0.00 28.01 43.14 0.00 0.09 2.01

0.72 623.4 0.24 0.57 0.42 0.95 1.19 10.57 33.37 41.17 0.55 0.24 1.28

1.89 1107.6 0.27 0.57 1.58 2.07 2.18 10.57 43.92 63.13 0.55 0.26 2.37

Note: Real value added, price-cost margins, prices, labour, capital, materials, the exchange rate, US prices and number of factories are in logs. Total Factor Productivity (TFP), J – the index of intra-industry trade in intermediate and capital goods, the Import Coverage Ratio (ICR), and effective rate of protection (ERP) are not in logs.


73

Ratio. Col. (3) includes the effective rate of protection. Col. (4) is an estimate of the complete specification of equation (5.8a), where we use the price wedge (as defined in Chapter 2) as a measure of R and the Grubel–Lloyd (1975) index of intra-industry trade has been used to proxy for extent of trade in intermediate inputs as a measure of J. In Col. (5), we include year dummies. In Col. (6), we estimate a more general form of the production function underpinning the empirics, regressing real output on labour, capital and material inputs and the protection measure, R (as captured by the price wedge) and J. The estimates of the technology coefficients in the augmented models are more plausible than that in Col. (1). The assumption of constant returns to scale is not rejected in any of the augmented models, though the point estimates suggest increasing returns to scale. Most of the β coefficients, the coefficient of interest to this study, have signs and significance that are in accord with the theoretical priors. The estimate in Col. (2) suggests that an increase in quantitative restrictions as captured by an increase in the ICR has a strong negative effect of Total Factor Productivity (TFP). The coefficient on ERP is positive, when we would expect a negative relationship between ERP and TFP. However, the coefficient is not significant at conventional levels of significance. In Col. (4), we see that a rise in the price wedge has a statistically significant negative impact on growth while a rise in extent of intra-industry trade in intermediate and capital goods has a positive impact. We also find that in all three estimates, the fixed effects are significant (probability that u_i = 0 is rejected). We examined the robustness of this finding by carrying out three further estimates. First, we introduced year dummies in Col. (5) of Table 5.3 to incorporate temporary economy-wide shocks to productivity. Second, in Col. (6) of Table 5.3, we relaxed the restricted nature of the production function in equation (5.8a) and estimated a gross output production function instead of a value added production function. Finally, in Table 5.4, imposing the statistically acceptable restriction of constant returns to scale on the parameters of K and L, we estimated equation (5.8b) where a direct measure of TFP growth using the Tornquist index formula is used as the dependent ˆ and L ˆ arising out of variable. We do this to control for the endogeneity of K their possible correlation with demand shocks included in the error term. In Cols (1) to (3), we included the protection variables in the same order as we did for the previous set of estimates in Table 5.3. We find that in all our estimates, there is no difference to our qualitative results with the coefficients on the variables of interest retaining their correct signs and statistical significance. The results indicate that trade reforms that brought about a reduction in price distortions and an increase in the access to specialised intermediate and capital goods from abroad has led to a significant increase in Total Factor Productivity in India. The R square adjusted coefficient of variation is low for all the estimates in Table 5.4. The low coefficient of variation is not surprising given that

74


estimates are for growth rates, panel data is used and a fixed-effects procedure has been employed which employs dummy variables for each of the industries. We note that use of the OLS procedure gives an adjusted coefficient of variation of approximately 20 per cent in the models estimated in Table 5.4. If we use the model in Col. (1) of Table 5.3 as the benchmark, then the augmented models are considerably better at explaining the variation in the data. Furthermore, the objective here is to investigate the role of trade liberalisation in TFP growth, hence the ability of the model to explain the variation is not of primary concern. We now turn to the estimation of the price equation (5.13). While estimating this equation, we add an aggregate price level index, given by the GDP deflator, since sectoral domestic prices, pD, may vary across time purely due to movements in the aggregate price level. We measure import penetration as the ratio of imports to domestic sales plus imports. The exchange rate is the rupee–US dollar exchange rate (rupees per US dollar) while the world price of the commodity (pF) is proxied by the US domestic price of that commodity. The variable s measures changes in market structure; however, most measures of market structure such as the four-firm Concentration Ratio or the Herfindahl index are rarely available in time-series form. One possible measure of market structure in a time-series context that is available for the data-set we are using is the number of factories in the particular industry – the larger the number of factories, the more competitive the industry will be. We use this rather imperfect measure of market concentration. Equation (5.13) is estimated using the fixed-effects estimator to allow for intrinsic differences in the growth of mark-ups across industries (say, due to technological progress or changes in the elasticity of demand). The results of the estimation are presented in Table 5.5. Col. (1) is an estimate of equation (5.13), without the price and quantity competition variables. We then estimate the full version of equation (5.13), with the estimates presented in Col. (2). In Col. (3), we add time dummies to incorporate temporary economy-wide shocks to demand and/or supply. In the estimates presented in Cols (2) and (3), we find that the quantity competition variable (Import Coverage Ratio) and one of the price competition variables – the exchange rate – have the right signs and are statistically significant. An increase in the Import Coverage Ratio leads to greater protection for domestic firms, and hence in domestic market power and ceteris paribus, domestic prices. An increase (depreciation) in the nominal exchange rate leads to greater protection for domestic industry and, therefore, an increase in domestic market power and prices. However, the sectoral foreign price has the wrong sign in Col. (1); once we introduce time dummies, it has the right sign – positive – and is statistically significant at the 5 per cent level. An estimate of the price formulation of equation (5.10), which restricted the coefficient of unit cost to unity, was rejected by the data. There is, thus, no evidence of a one-to-one pass-through of costs to prices in Indian manufacturing. We also find that in all three estimates the fixed effects are significant (probability that u_i = 0 is rejected).


75

Table 5.3 Total Factor Productivity – regression estimates of equation (5.8a) employing the fixed-effects procedure, India Variable

Col. (1) Col. (2)

Col. (3) Col. (4)

Col. (5)

Intercept

Price wedge

0.51*** (4.3) 0.48*** (74.3) 0.70*** (48.7) –

1.17*** (5.8) 0.59*** (35.3) 0.57*** (21.7) –

1.07*** (5.3) 0.60*** (38.2) 0.56*** (22.2) –

J

–

–

–

ICR

–

ERP

–

Ln M

–

−0.001*** – (3.59) – 0.0003 – (1.64) – – –

0.92*** −3.41*** (3.7) (26.7) 0.55*** −0.08*** (14.1) (4.1) 0.64*** 0.26*** (20.0) (15.7) −0.34*** −0.35*** (6.10) (24.9) 0.020*** 0.004*** (4.5) (7.1) – –

Ln K Ln L

Year dummies? No No 2 R 0.91 0.92 Probability 61.0*** 43.5*** that all u_i = 0 Number of 3,745 1,094 observations Number of 139 60 industries

0.67*** (3.2) 0.58*** (19.8) 0.63*** (21.9) −0.079*** (3.16) 0.004*** (3.63) –

Col. (6)

–

–

–

No No 0.92 0.91 40.8*** 40.5***

Yes 0.92 35.5***

0.73*** (52.2) No 0.11 0.0579

1,094

1,380

1,380

420

60

94

94

94

Notes a Dependent variable is Ln Real value added – obtained from double-deflation method in Cols (1) to (5), and Ln Real Output in Col. (6). b L = Labour; K = capital; M = Materials; J = Grubel–Lloyd Index of Intra-industry trade in intermediate and capital goods; ICR = Import Coverage Ratio; and ERP = Effective Rate of Protection. c ***, ** and * denote significance at the 1, 5 and 10 per cent levels.

Among other variables, coefficients on unit costs and the GDP deflator are positive and significant as expected. The coefficient on the number of factories is positive and significant in Col. (1) but negative and significant in Col. (3), when we include time dummies. There is clearly a concern whether this variable is truly capturing the extent of domestic competition and the results for this variable should be interpreted with a degree of caution. We then estimate fixed-effects regressions where we directly regress pricecost margins against the price and quantity competition variables. In Col. (4) we use Import Coverage Ratios as the measure of quantity competition, and in Col. (5) we experiment with a different measure of quantity competition – the

76


Table 5.4 Total Factor Productivity – regression estimates of equation (5.8b) employing the fixed-effects procedure, India Variable

Col. (1)

Col. (2)

Col. (3)

Intercept Price wedge

1603.3.*** (36.1) –

1505.1*** (27.3) –

J

–

–

ICR

–

ERP

−10.6*** (17.3) –

−786.7*** (9.6) −4.82*** (2.42) 2.54* (1.70) –

R2 Probability that all u_i = 0 Number of observations Number of industries

0.01 71.9*** 1,092 60

−6.32*** (12.6) 0.01 67.6*** 1,075 60

– 0.02 79.4*** 1,380 94

Notes a Dependent variable is Total Factor Productivity – obtained from double-deflation method. b J = Grubel–Lloyd Index of Intra-industry trade in intermediate and capital goods; ICR = Import Coverage Ratio; and ERP = Effective Rate of Protection. c ***, ** and * denote significance at the 1, 5 and 10 per cent levels.

sectoral import penetration ratio. While the exchange rate remains positive and significant, the foreign price is not significant in either of the two estimates. Interestingly, the coefficient on the import coverage rate is not significant, but the coefficient on the import penetration ratio is negative and statistically significant as expected by theory. Overall, the findings suggest that an increase in quantity competition from abroad has a statistically significant and negative impact on domestic market power in all of the estimates. However, there is weak evidence of a dampening effect of price competition from abroad on domestic pricing behaviour.6 A limitation of the estimates presented in Table 5.5 is that there may be a reverse causality from the domestic price to import penetration as a lower price of the domestic good may imply less demand for the imported good (assuming the two are imperfect substitutes). The estimate of m may, therefore, be biased. We did not, however, pursue instrumental variables estimation to correct for the simultaneity problem for two reasons. First, we could not obtain instruments that were adequate explanatory variables for m. Second, the reverse causality from p to m pointed out above should bias the estimated coefficient of m towards zero. Given our finding of a negative and significant coefficient on m, estimating equation (5.13) using instrumental variables would not change the qualitative results obtained in Table 5.5.


77

Table 5.5 Domestic prices – regression estimates of (5.13) using the fixed-effects procedure, India Variable

Col. (1)

Intercept

2.25*** 3.03*** 3.35*** −0.75*** (26.6) (15.9) (18.9) (2.3) 0.77*** 0.55*** – – (51.9) (11.8) 0.05*** 0.03*** – 0.04*** (3.5) (2.7) (1.3) 0.02** −0.01 −0.02* 0.03 (2.9) (1.3) (1.8) (1.3) – 0.24*** 0.79*** 0.09*** (6.3) (28.4) (3.7) – −0.11*** 0.08** −0.001 (2.7) (2.2) (0.01) – 0.0003*** 0.0004*** 0.0001 (2.4) (3.2) (0.6) – – – –

Ln GDP deflator Ln unit costs Ln factories Ln exchange rate Ln US prices ICR IMP Year dummies? R2 Probability that all u_i=0 Number of observations Number of industries

Col. (2)

Col. (3)

Col. (4)

Col. (5) −0.79*** (4.4) – – 0.04** (2.2) 0.09*** (6.2) −0.007 (0.02) –

No 0.92 11.8***

No 0.93 12.9***

Yes 0.94 25.6***

No 0.09 16.8***

−0.002** (2.0) No 0.06 22.8***

3,473

645

645

640

540

139

45

45

45

36

Notes a Dependent variable is Domestic Prices in Cols (1) to (3), and price-cost margins in Cols (4) and (5). b ICR = Import Coverage Ratio; IMP = Import Penetration Ratio. c ***, ** and * denote significance at the 1, 5 and 10 per cent levels.

7 Conclusions The question investigated in this chapter has been whether trade reform in Indian manufacturing has had a positive impact on TFP growth and a negative impact on domestic prices. We find strong evidence that this has been the case. A reduction in quantitative restrictions has had a positive effect on Total Factor Productivity. More interestingly, a reduction in price distortions, along with an increase in intra-industry trade in intermediate and capital goods, has had a strong positive impact on Total Factor Productivity. We also find that an increase in quantity competition from abroad has a statistically significant and negative impact on domestic market power, and in domestic prices. With respect to price competition from abroad, we find that

78


the exchange rate depreciation of the mid-1980s to 1990s may have had a countervailing effect on domestic prices by providing more protection for import-competing sectors. Thus, the net effect of price competition from abroad was such that it brought about an increase in domestic prices and in domestic market power, in spite of a reduction in tariffs since 1991. On the whole, the empirical findings here provide strong support for the efficiencyenhancing effects of trade liberalisation in the Indian context.

6


1 Introduction For a country such as India with a large pool of surplus labour both in the agricultural and informal manufacturing sectors, the question of whether the trade reforms of the 1980s and 1990s have led to an appreciable increase in employment in the formal manufacturing sector is obviously of significant policy concern. Clearly, one important criterion of success for the trade reforms would be the creation of jobs in the formal manufacturing sector, apart from the productivity and price effects of the reforms which we discussed in the previous chapter. Job creation in the formal manufacturing sector is important not only because it contributes to overall employment growth, but also because jobs in the formal manufacturing sector are considered to be of better quality than jobs created in the agricultural or informal manufacturing and services sectors, where workers do not have same level of protection against unfair dismissal or adverse working conditions as workers in India’s formal manufacturing sector. This chapter examines the trade–employment relationship for the Indian formal manufacturing sector by using three different methods for assessing the impact of trade on employment – factor content, growth decomposition and labour demand modelling. It also compares the experiences of the manufacturing sectors of four developing countries – Bangladesh, Kenya, South Africa and Vietnam – with trade policy changes, thus providing a comparative dimension to the evidence we present in this chapter. The rest of the chapter is in seven sections. The next section sets out the theoretical framework. Section 3 introduces the three empirical methods. Section 4 discusses previous empirical findings on the relationship between trade and employment. Sections 5, 6 and 7 apply, respectively, the factor content, growth decomposition and labour demand approaches to Indian industry and trade data. Section 8 makes some concluding remarks.

80


2 Trade and manufacturing employment: theoretical linkages 1 The overall level of manufacturing employment in an economy is by definition equal to the level of manufacturing output times the weighted average employment coefficient for the manufacturing sector. L = Q.Σ wi (L/Q)i

(6.1)

where L is total manufacturing employment Q is total manufacturing output wi = Qi/Q i refers to branches of manufacturing. The impact of trade on manufacturing employment can therefore be decomposed into three elements represented in Equation (6.1). First, it may have an impact on the total output of the manufacturing sector (Q). Increased exports have a positive effect on the level of output, tending to increase employment, while greater import penetration depresses output and displaces labour. Second, trade influences the shares of different industries in overall manufacturing output (wi), increasing the output of exportables and reducing output of import competing industries. Finally, trade can have an impact on employment by changing labour coefficients within industries (L/Q)i. These three impacts are referred to in this chapter as the scale effect, the composition effect and the process effect of trade. Theory suggests that trade might influence manufacturing employment through each of these effects. One determinant of the size of the manufacturing sector is a country’s comparative advantage, which may in turn reflect factor endowments. In the model first proposed by Krueger (1977) and extended by Leamer (1987), the crucial variable determining trade and production structure is the land/labour ratio. Thus, land-abundant developing countries such as those in Africa and Latin America, would be more likely to specialise in primary commodities while developing countries in Asia would be more likely to specialise in (labour-intensive) manufactures. Wood (2003) finds persuasive evidence for the Krueger–Leamer variant of the Heckscher–Ohlin model – differences in factor endowments between Africa and Asia seem to explain why Africa’s export structure is biased towards natural resource based commodities rather than labour-intensive manufacturing exports. Increased trade would therefore tend to lead to slower growth (or even contraction) of the manufacturing sector in African countries compared to Asian countries. An alternative view would explain a country’s comparative (dis)advantage in manufacturing in Ricardian terms, where differences in technology across sectors explain the effects of trade. In this case, the size of the manufacturing sector in a country is determined by its overall competitiveness, which in turn


81

is partly a result of technological capabilities in manufacturing. In this case it is the acquisition of technological capabilities that determines the impact of trade on manufacturing employment rather than factor endowments. For countries which have a comparative disadvantage in manufacturing, this view is less optimistic that a contraction of the manufacturing sector as a result of increased trade is matched by expansion of other non-manufacturing sectors. This is particularly true when greater trade openness occurs in the presence of specific factors or labour market rigidities. We refer to the impact of trade on employment via the overall size of the manufacturing sector as the scale effect, irrespective of the ultimate causes of changes in manufacturing output. The composition effect of trade depends on the impact of trade on the share of different branches in total manufacturing output. An increase in the share of labour-intensive industries in aggregate output would tend obviously to raise the overall level of manufacturing employment. A key prediction of the standard two-factor Heckscher–Ohlin model is that with international trade, developing countries with plentiful supplies of labour will export labour-intensive commodities and import commodities with relatively higher capital requirements. Thus, as a developing country gradually integrates with the world economy, it will observe a change in the composition of its output towards more labour-intensive activities. This will shift the national demand for labour curve to the right, and, under an assumption of a fairly elastic supply of labour, will lead to an increase in overall employment.2 Not all trade theories suggest that increased trade will necessarily lead to a more labour-intensive composition of output in developing countries. This is mainly applicable to cases of inter-industry trade and far less relevant where there is intra-industry trade.3 Indeed, as Feenstra and Hanson have shown, it is possible in the latter case that, contrary to the orthodox Heckscher–Ohlin prediction, trade increases the demand for the scarce factor in developing countries.4 In any case, intra-industry trade may be more a reflection of economies of scale and product differentiation than of factor endowments. Once again we refer to the effects of changes in the weights of different branches of manufacturing on employment as the composition effect, irrespective of the factors which have contributed to such changes. The final way in which international trade can impact on manufacturing employment is that it can lead to change within a sector which affects the quantity and kind of labour required to produce a given output. Within the standard trade theory such changes are due to a shift in relative factor prices brought about by changes in relative factor demand as the economy opens up (the Stolper–Samuelson effect). These in turn lead to factor substitution in production. Industry-level impacts on employment may also occur via induced productivity effects, as firms shed labour in response to external competitive pressures, due to either greater export orientation or increased import penetration (Macdonald 1994, Greenaway et al. 1999). Such a trade-induced productivity

82


effect could be due to a decrease in X-inefficiency as trade reform leading to increased international competition brings about a reduction in ‘slack’ in labour input (Horn et al. 1996). It could also be due to trade-induced technological transfers (for example, via an increase in the importation of capital goods). We refer to all these effects of trade on employment within industrial branches as the process effect of international trade. In this chapter, we attempt to assess the importance of the scale effect, the composition and the process effect of international trade on employment for the four countries under consideration. In order to do this, we need to implement a set of methodologies that allow us to capture all three effects.

3 The methodological approaches The chapter employs three commonly used methodological approaches to study the impact of international trade on employment. These are factor content, growth accounting and labour demand approaches. Factor content approach Factor content studies have been widely used both in order to test theories of international trade and to estimate the employment effects of trade, particularly between developed and developing countries.5 This approach allows us to examine whether a change in the structure of production as a result of greater outward orientation leads to an increase in the labour-intensity of production, and hence, overall employment. This it does by computing direct and indirect labour requirements per unit of exports and import substitutes, with indirect labour requirements calculated using input-output tables. In this chapter, we will only examine the direct labour requirements per unit of exports and import substitutes, as we lack the requisite input-output tables for the countries in question for the more recent periods. Growth accounting approach Factor content studies only consider the impact of trade on employment, but growth accounting can be used to go beyond this to analyse the impact of different forces on changes in employment. This approach decomposes changes in employment into the effects of changes in domestic demand, exports, imports and productivity. Starting from the basic accounting identity that Qit = Dit + Xit − Mit where Dit is domestic absorption of industry i at time t Qit is domestic production of industry i at time t

(6.2)


83

Xit is exports of industry i at time t Mit is imports of industry i at time t Employment can be calculated as Lit = lit(Dit + Xit − Mit)

(6.3)

where Lit is employment in industry i at time t lit = Lit/Qit Changes in employment between t = 0 and t = 1 can then be decomposed using the equation: ∆Li = li1(1 − mi0)∆Di + li1∆Xi + li1 (mi0 − mi1)Di1 + (∆li) Qi0

(6.4)

where mit = Mit/Dit The first term on the right-hand side measures the impact of changes in domestic demand on employment, the second the effect of changes in exports, the third the impact of changes in import penetration and the final term indicates the effect of productivity changes. This corresponds to a Chenery type decomposition. This approach assumes that increases in exports create additional employment while increased import penetration reduces employment. The growth accounting approach has been subject to methodological criticisms, including the arbitrariness of the decompositions involved and the fact that since they derive from basic accounting identities, they cannot be interpreted in a causal way (Martin and Evans 1981). Moreover, as Wood (1994) has argued, part of the technological change which occurs may be defensive, where firms respond to increased competitive pressure from imports. Therefore it is invalid to assume that reduced employment as a result of increased productivity is independent of trade. Nevertheless, despite these limitations, growth accounting has been extensively used in the literature, both on developed and developing countries, and as such provides a useful first approximation to considering the impacts of trade flows on employment. Furthermore, as was shown above, growth accounting provides a useful way of separating out the scale effects from the composition effect of trade on employment. Labour demand modelling The first two approaches estimate the effect of international trade on employment via changes in the labour-intensity of production across industries (as in

84


the factor content approach) or via the expansion or contraction of output due to export expansion or import penetration (as in the growth accounting approach). However, as we have argued earlier, international trade can also lead to changes in the efficiency of labour use within the same industry. This can be captured by the estimation of labour demand equations at the industry level, where employment regressed at the industry level against a number of explanatory variables, derived from a standard labour demand framework. This approach has been used by Hine and Wright (1998) and Greenaway et al. (1999) to analyse the impact of trade on employment in UK manufacturing and in a developing country context by Milner and Wright (1998) for Mauritius. Consider a standard derived demand for labour equation at the industry level, augmented by a variable that captures the extent of integration of the industry with the world market. Lit = α + β1 Wit + β2 Qit + φ Zit

(6.5)

where Lit is employment in industry i at time t, Wit is real wage in industry i at time, and Qit is real output in industry i at time t, and Zit measures the degree of openness of industry i in time t. We will estimate the equations using the natural logarithms of L, W and Q, so that the coefficients on W and Q in equation (6.5) can be interpreted as the wage and output elasticities of labour demand. As is standard in the literature, we capture the degree of openness (Zit) by the import penetration ratio (IM) and the export-output ratio (EO) defined at the industry level (Hine and Wright 1998, Greenaway et al. 1999). The use of these two variables also allows us to separate the effects of import competition from export orientation on the efficiency of labour use. Thus, we can rewrite (6.5) as: Lit = α + β1 Wit + β2 Qit + φ1 IMit + φ1EOit

(6.6)

This approach can take account of the indirect impact of trade on employment via trade-induced productivity changes. In fact, since the output variable incorporates the direct effects of changes in exports and imports, the import and export penetration variables capture the indirect effects. We would expect that β1 < 0, and β2 > 0. Also, following our discussion in the previous section, we would expect that φ1 < 0 and φ2 < 0. We use dynamic panel data methods to allow for short-run rigidities to labour adjustment to its optimal level, following shocks to domestic demand, productivity and India’s export markets.


85

4 Previous empirical analysis Africa Currie and Harrison (1997) use micro-level data on individual enterprises for Morocco to examine the wage and employment outcomes of trade reforms implemented in that country between 1984 and 1990. The paper finds evidence of a significant employment response to the trade reforms, with the sectors most affected by the trade reforms witnessing a decline in employment. However, there is a great deal of heterogeneity in the firms’ responses to trade reform. In fact, a significant proportion of manufacturing firms did not adjust either wages or employment in response to the trade reforms. Currie and Harrison argue that this could possibly be due to the high degree of market power that incumbent firms had at the beginning of the reform process, and also due to endogenous technological change. Their results suggest that many firms responded to reforms by cutting profit margins and raising productivity, and not adjusting wages or employment. These results provide support for Wood’s and Rodrik’s respective arguments that tradeinduced technological progress and changes in the elasticity of labour demand are crucial in explaining the effect of international trade on labour outcomes in the South (Wood 1994, Rodrik 1997). However, the Currie and Harrison study seems to suggest that the manner in which the two mechanisms may actually work could be in contradiction to those originally proposed by their proponents. Milner and Wright (1998) examine changes in inter-industry employment and wages for the Mauritian economy pre- and post-liberalisation. They find that employment and wages have increased in both the exportable and importable sectors following the trade liberalisation initiated in the 1980s. That employment and wages should increase in the exportable sector for a labour-abundant country like Mauritius is very much in line with the H–O model. However, the increase in employment and wages in the importable sector is in contradiction with the H–O model. Milner and Wright appeal to non-H–O elements to support their findings, arguing that the latter may have occurred due to trade-induced productivity gains in the importable sector and a rapid increase in the supply of labour as women entered the labour market in Mauritius in large numbers. Latin America Revenga (1997) investigates the effect of trade reform implemented in the 1980s on wages and employment in Mexican manufacturing, again using firm-level data. Revenga finds a negative response of employment and wages to trade liberalisation. Trade reforms led to the reduction in wages both by bringing about a fall in the firm’s output (and consequently, labour demand) and by a weakening in the workers’ bargaining powers with a decline in the rents accruing to sectors that were protected prior to the reforms. Revenga

86


argues that a significant fraction of the rents generated by trade protection in the manufacturing sector were absorbed by workers in the form of a wage premium. Consequently, the fall in wages was more substantial than the fall in employment: an average tariff reduction of 20 percentage points implied a negative wage response in the order of 5–6 per cent. Two studies that examine the effect of trade reforms on employment are Levinsohn (1999) and Moreira and Najberg (2000). Levinsohn examines changes in employment patterns in the Chilean manufacturing sector at the firm level following trade reforms and finds evidence of substantial heterogeneity in employment adjustment among firms in the same industries. His study indicates the use of industry-level data to ascertain the effect of trade reform on employment may be misleading. In the case of Chile, the paper finds that there was substantial churning of jobs among firms within the same industries, but which left net employment in these industries unaffected. Levinsohn also finds that the effect of trade liberalisation on the firm’s employment did not depend on whether the firm was in the import-competing or export sectors but was dependent on the firm’s size and whether the firm was in the tradable or non-tradable sector. With respect to size, job creation was disproportionately more in the larger firms, while job destruction was disproportionately more in the small firms. With respect to the firm’s location in tradable or non-tradable sectors, firms in the non-tradable sector had higher job growths during periods of real exchange rate appreciation and lower rates during periods of depreciation. The latter also indicates that a change in employment patterns in a newly reforming economy is also linked to macroeconomic factors. Moreira and Najberg (2000) examine the changes in the structure and level of employment in the Brazilian economy in the 1990s, following trade reforms. They find that, even though there was an initial negative impact on employment, there was a trade-induced shift to more labour-intensive activities such as agriculture and services and away from manufacturing, leading to a more labour-intensive output mix in the long run. This finding is corroborated by Amadeo and Pero (2000), who argue, however, that such a change may not be in the interests of workers in Brazil, as many of the new jobs were the result of a rise in the number of workers in the informal sector and self-employed workers. Thus, most of the new jobs created were of lower quality compared to the jobs lost in the manufacturing sector. Ghose (2000) examines whether international trade has led to a reallocation of labour from export-oriented industries to import-competing industries in the newly industrialising countries in the 1980s and 1990s, as predicted by the H–O model.6 Ghose finds instead that trade liberalisation accelerated employment growth in both export-oriented and importcompeting sectors for most of the countries in the sample. Ghose also finds evidence of a strong growth in real wages in these countries, which seems to be correlated with the growth in trade. Thus, unlike the evidence that comes from the developed countries, international trade seems to have had


87

a substantial positive effect on wages and employment in the manufacturing sectors of developing countries. India There are relatively few studies that examine the impact of international trade (or trade policy) on manufacturing employment in India. Khambhampati et al. (1997) examine the effect of the 1991 trade reforms on employment in five import-competing industries using firm-level data over the period 1987–93 and find that trade reforms have a small and insignificant effect on employment, overall and in each of the five industries. Examining the relationship between labour demand and mark-ups, they find support for the pro-competitive effects of trade reforms, as firms cut mark-ups and increased output and labour demand following the reforms. This may have at least partially offset the reduction in labour demand caused by other factors.

5 A factor content approach We begin this section by examining the factor-intensity of manufacturing exports, as a prelude to the factor content calculations.7 In order to do so, we apply Krause’s (1982) classification of ISIC manufacturing industries according to their dominant factor input.8 This distinguishes between natural resource intensive, labour-intensive, technology-intensive and human capital intensive industries. The natural resource intensive industries are further subdivided into agricultural and mineral-based industries. Unskilled labourintensive industries are those with the lowest value added per worker. The remaining industries are divided into technology intensive and human capital intensive, with the industries with a high ratio of R&D to value added being classified as technology intensive.9 We find that unskilled labour-intensive commodities are the most important in India’s manufacturing exports, comprising 45 per cent of total manufacturing exports in 1996–9 (Table 6.1 and Figure 6.1). However, while the share of unskilled labour-intensive commodities in total manufacturing exports has increased over the period 1975–99, the increase has not been substantial, from 37 per cent in 1975–80. In contrast, the increase of the share of unskilled labour-intensive exports in total manufacturing exports of Bangladesh and Vietnam has been far more striking (Table 6.2). However, in comparison to Kenya and South Africa, India’s share of unskilled labourintensive exports is substantially higher. India’s human capital intensive and technology-intensive exports have also increased as a share of total manufacturing exports over the period 1975–99, from 10 and 17 per cent in 1975–80, to 15 and 24 per cent in 1996–9 respectively. On the other hand, India’s agricultural-intensive exports as a share of total manufacturing exports have fallen quite dramatically from 31 per cent in 1975–80 to 12 per cent in 1996–9.

88


With respect to imports, technology-intensive exports remain the dominant set of commodities in India’s manufacturing import basket, followed by human capital intensive goods (Figure 6.2). There has been a slight increase in the share of human capital intensive goods in India’s manufacturing imports, with no significant change in the share of technology-intensive imports in total manufacturing imports, in spite of the trade liberalisations of the 1980s and 1990s which were mostly biased towards import liberalisation of capital and intermediate goods. Employment coefficients of exports and import-competing domestic production The discussion so far has focused solely on the composition of exports and imports, without explicit computations of their labour-intensities. However, in order to examine the impact of trade on employment, we need to look at the labour-intensity of both exports and imports. To do this, we derive

Figure 6.1 Factor content of India’s manufacturing exports. Source: Author’s calculations; International Economic Data-bank, Australian National University.


89

Figure 6.2 Factor content of India’s manufacturing imports. Source: Author’s calculations; International Economic Data-bank, Australian National University.

employment coefficients at the industry level which are then weighted by the share of each industry in exports and imports. The employment coefficients are presented in Figure 6.3.10 We see that the employment coefficients of exports and imports have consistently fallen over the period. Moreover, the difference between the employment coefficient of exports and that of imports has narrowed over time. The findings suggest that a unit increase in manufacturing exports matched by an identical increase in manufacturing imports will lead to a smaller positive effect on employment in 1996–9 as compared to 1975–80. To understand what explains this puzzling phenomenon during a period when India attempted to re-align its trade regime in line with its own comparative advantage in unskilled labour-intensive commodities, we look at the changes in the contribution of two sets of commodities which comprise the bulk of India’s manufacturing exports. These are wearing apparel (ISIC Code 322) comprising 18 per cent of total manufacturing exports in 1996–9

90


Table 6.1 Structure of total manufacturing exports and imports, India Percentage share (except total exports) Exports Agricultural resource intensive Mineral resource intensive Unskilled labour intensive Technology intensive Human capital intensive Total Manufacturing Exports (in US$ million) Imports Agricultural resource intensive Mineral resource intensive Unskilled labour intensive Technology intensive Human capital intensive Total Manufacturing Imports (in US$ million)

1975–80

1981–5

31

24

5 37 10 17 3,861

1986–90

1991–5

1996–9

16

15

12

5 39 11 21 5,442

8 40 13 23 10,270

5 43 13 24 17,890

4 45 15 24 22,916

16

14

10

7

10

24 4 41 15 4,575

24 5 40 17 8,834

22 6 39 24 12,433

27 6 39 21 16,387

11 8 44 27 21,019

Source: Author’s calculations; International Economic Data-bank, Australian National University.

and other manufacturing goods (ISIC Code 390), which are principally jewellery, sporting goods and toys, comprising 19 per cent of total manufacturing exports in 1996–9. These two sets of commodities can contribute to changes in the overall employment coefficient either by a change in their own individual employment coefficients or by a change in their share of total manufacturing exports. We scale the employment coefficients and export shares for wearing apparel and other manufacturing goods to 100 for the period 1975–80. The results are presented in Figures 6.4 and 6.5 respectively. Interestingly, the employment coefficients for both sets of commodities have actually decreased over the period 1975–99. Export shares for these two sets of commodities have, on the other hand, increased from the late 1970s to the mid-1980s and then remained stable in the case of wearing apparel. Export shares of other manufacturing goods have also increased over the period, in particular for the sub-period 1996–9. These findings suggest that the principal reason for the decrease in the overall employment coefficient for India’s exports over the period 1975–99 is the fall in employment intensity of production, and not in the lack of specialisation in labour-intensive products. The fact that the employment-intensity of production has fallen during a period where Indian policy-makers reformed the trade regime to remove some of the biases in incentives towards the production of capital and intermediate goods is an issue that needs to be investigated in further detail. What is instructive to note


91

Table 6.2 Structure of total manufacturing exports, Bangladesh, Kenya, South Africa and Vietnam Percentage share (except total exports) Bangladesh Agricultural resource intensive Mineral resource intensive Unskilled labour intensive Technology intensive Human capital intensive Total Manufacturing Exports (in US$ million) Kenya Agricultural resource intensive Mineral resource intensive Unskilled labour intensive Technology intensive Human capital intensive Total Manufacturing Exports (in US$ million) South Africa Agricultural resource intensive Mineral resource intensive Unskilled labour intensive Technology intensive Human capital intensive Total Manufacturing Exports (in US$ million) Vietnam Agricultural resource intensive Mineral resource intensive Unskilled labour intensive Technology intensive Human capital intensive Total Manufacturing Exports (in US$ million)

1976–80

1981–5

1986–90

1991–5

1996–8

25.8

27.1

21.7

9.9

7.0

6.8 63.8 3.4 0.2 287.8

7.6 62.3 2.3 0.6 423.0

2.8 72.8 2.3 0.4 981.1

1.5 84.7 3.4 0.6 2,340.3

0.7 89.9 1.7 0.7 4,008.2

65.8

64.8

73.9

63.4

64.7

16.8 4.1 8.5 4.9 285.9

15.6 5.0 10.0 4.6 283.9

2.5 7.6 11.7 4.4 319.8

3.2 15.9 13.1 4.4 384.9

4.2 15.8 8.6 6.8 403.2

26.9

20.3

19.5

17.9

16.1

28.4 4.8 9.1 30.7 4,432.6

32.5 6.5 15.0 25.7 4,704.8

38.0 6.6 11.1 24.9 6,640.0

32.5 9.4 11.4 28.8 8,654.8

29.4 9.9 13.1 31.5 12,643.7

63.5

83.9

80.6

38.6

21.0

5.9 21.4 6.8 2.3 34.6

2.4 10.2 2.4 1.1 56.2

1.2 14.2 1.5 2.5 210.8

4.3 49.7 1.9 5.6 1,637.5

2.1 58.7 5.6 12.7 4,941.4

Source: Jenkins and Sen (2006).

is that the employment coefficients for manufacturing exports for Bangladesh and Vietnam are significantly higher than the employment coefficients for import-competing production for the same countries (Table 6.3). In fact, the ratio of India’s employment coefficient for exports to that for importcompeting production is closer to those for Kenya and South Africa, a

92


surprising finding given that India’s relative factor endowments are closer to Bangladesh and Vietnam than to South Africa and Kenya.

6 Decomposition of employment changes As was seen above, there have been substantial changes for India in terms of openness in recent years, with both exports and imports growing rapidly. A first stab at estimating the effects of increased openness on manufacturing employment can be made using a growth accounting methodology which divides employment changes over a period of time into that attributable to changes in domestic demand, exports, import penetration and productivity.11 Employment has increased in the period 1985–99, following a decline in 1980–85 (Table 6.4). Much of the employment increase has been driven by increases in domestic demand. Increases in labour productivity all through the 1980s and 1990s have led to labour shedding for the entire 1980s and 1990s. The contribution of exports to employment growth has been greater in the period 1985–95 than in the preceding and ensuing periods. Import penetration has led to few jobs for much of the period under consideration. The analysis of the sources of employment growth in Table 6.4 and Table 6.5 reinforces the contrast between India and the two comparator Asian countries, and its similarity with the two African economies, described in the

Figure 6.3 Employment coefficients, exports and import-competing production, India. Source: Author’s calculations; International Economic Data-bank, Australian National University.


93

Figure 6.4 Contribution of wearing apparel to change in employment coefficient, India. Source: Author’s calculations; International Economic Data-bank, Australian National University.

previous section. In Bangladesh, the marked increase in the growth of manufacturing employment from the late 1980s was led by the growth in exports of labour-intensive manufactures. Manufacturing employment also grew in Vietnam during the 1990s and the data for the latter half of the decade indicates that, as in Bangladesh, this was driven primarily by exports, followed by the growth of domestic demand. The net effect of trade on employment was reduced by increased import penetration during this period, but was still highly positive. Kenya, in contrast, experienced very limited manufacturing employment growth throughout the period, with a tendency to slow down in the 1990s. Trade liberalisation led to increased import penetration in the manufacturing sector without any compensating increase in employment being generated by exports. Finally, for South Africa, the rate of growth of manufacturing employment in South Africa has declined, decade on decade, since the 1970s, turning negative in the 1990s. During the first half of the 1990s, as in Kenya, trade liberalisation led to increased import penetration with negative effects on employment, though exports become a more significant factor in employment generation than it had been in earlier decades.

94


Figure 6.5 Contribution of other manufacturing goods (including jewellery, sporting goods and toys) to the change in employment coefficient, India. Source: Author’s calculations; International Economic Data-bank, Australian National University.

Table 6.3 Employment coefficients of manufacturing exports and import-competing domestic manufacturing production, Bangladesh, Kenya, South Africa and Vietnam Exports

Import-competing

Female Male Total

131 128 259

7 84 91

Female Male Total

7 30 37

5 30 35

Total

16

21

156 56 214

49 47 96

Bangladesh (1997)

Kenya (1996)

South Africa (1996) Vietnam (1998) Female Male Total Source: Jenkins and Sen (2006).


95

Table 6.4 Decomposition of manufacturing employment changes, India Total Domestic Productivity Export employment demand growth growth effect

Import Net penetration employment growth from trade

Absolute numbers (in thousands) 1975–80 1980–5 1985–90 1990–5 1995–9

1,122 −333 639 848 870

1,444 1,898 1,981 2,034 1,935

−263 −2,227 −1,883 −1,687 −1,154

5 76 388 655 15

−63 −80 152 −154 75

−59 −4 541 501 90

128.69 −569.45 310.08 239.77 222.35

−23.47 668.22 −294.73 −198.87 −132.65

0.43 −22.83 60.80 77.23 1.69

−5.64 24.06 23.85 −18.14 8.60

−5.22 1.23 84.65 59.10 10.30

Percentage contribution 1975–80 1980–5 1985–90 1990–5 1995–9

Source: Author’s calculations; International Economic Data-bank, Australian National University. Note: We exclude other manufacturing industries from our sample of industries (ISIC Code 390).

7 Labour demand estimation The previous section examined the direct effects of international trade on manufacturing employment via trade-induced adjustments in output. In this section, we study the indirect impact of international trade on employment via changes in the efficiency of labour use. To capture the indirect effects of trade, we estimate constant-output labour demand equations at the industry level, augmented by variables that capture trade orientation. We have twenty-seven industries at ISIC 3-digit level for the period 1975–99. The exports and imports data matched to the ISIC 3-digit level is obtained from the International Economic Databank at the Australian National University. In labour demand modelling, it is usually assumed that due to large adjustment costs related to hiring and firing of workers, employment adjusts to output and wage changes slowly over time. This implies there are lagged employment terms in equation (6.6) and the possible correlation between these terms and country-specific time-invariant effects (fixed effects). In this case, the preferred estimator is the Generalised Method of Moments (GMM) estimator proposed by Arellano and Bond (1991) which differences the data to get rid of country specific effects or any time-invariant country-specific variable (such as labour-saving technological progress that may differ across industries). This also eliminates any endogeneity that may be due to the

96


Table 6.5 Decomposition of manufacturing employment changes, Bangladesh, Kenya, South Africa and Vietnam Total Domestic Productivity Export Import Net employment demand growth growth penetration employment effect growth from trade Bangladesh 55 56 559 864

3 75 277 435

18 −49 27 −316

60 51 247 802

−26 −21 8 −57

34 30 255 745

39 19 25 10 10

53 45 46 7 −26

−23 −43 −37 8 49

4 5 3 5 −8

5 12 13 −10 −5

9 17 16 −5 −13

354 103 −125 −169

386 94 123 14

−160 −69 −230 −255

16 64 108 78

112 14 −126 −6

128 78 −18 72

340

435

−570

699

−224

475

1975–80 1980–5 1985–90 1990–7 Kenya 1975–80 1980–5 1985–90 1990–4 1994–8 South Africa 1970–80 1980–90 1990–5 1996– 2001 Vietnam 1995–9

Source: Jenkins and Sen (2006). Note: Figures in thousands.

correlation of the country-specific effects and the independent variables. The estimator also allows for possible endogeneity of the independent variables, by using lags of the right-hand side variables as instruments for the possible endogenous variables. Thus, the GMM estimator allows for the possibility that some of the independent variables in equation (6.6) – output, wages, import penetration and export orientation – may be correlated with the error term. For example, positive productivity shocks may lead industries to withstand import competition better. To test whether the Arellano–Bond GMM estimator is correctly specified, two diagnostic statistics are normally reported – tests for first- and second-order serial correlation. The GMM estimator is appropriately specified if the test for first-order serial correlation cannot reject the null on no correlation, but the test for second-order serial correlation does reject the null of no correlation by any standard levels of significance.12


97

The regression results are presented in Table 6.6. All the diagnostic statistics reported in the table are satisfactory in all cases. The absence of firstorder serial correlation is rejected and the absence of second-order serial correlation is not rejected. The significance of the two lagged employment terms suggest high adjustment costs to changing employment levels – a finding that is expected, given India’s restrictive labour laws in the manufacturing sector (Besley and Burgess 2004).The coefficients on real output and real wage have the expected signs and are statistically significant in all cases at the 1 per cent level. A 1 per cent increase in output leads to a 0.39 per cent increase in employment, and a 1 per cent increase in the real wage rate leads to a fall in employment by around 0.37 per cent. We find that the coefficients on the import penetration and export orientation variables are statistically not significant at conventional levels of significance. Thus, the evidence seems to be that neither import penetration nor export orientation has had a discernible effect on employment via the indirect route of changes in labour productivity. Table 6.6 Labour demand – regression results – India Variables

Col. (1)

Constant

−0.009* (1.91) 0.47*** (9.30) 0.13*** (4.75) 0.39*** (6.11) −0.37*** (5.64) 0.04 (1.12) −0.005 (0.89) Arellano–Bond Dynamic Panel Estimator −2.83 (0.005) 0.19 (0.84) 27 594

Log L (−1) Log L (−2) Log Q Log W IM EO Estimation method First-order serial correlation (p value in brackets) Second-order serial correlation (p value in brackets) Number of industries Number of observations

Notes a Log L is the dependent variable. b *, ** and *** denote statistical significance at the 10, 5 and 1 per cent level respectively. c T-ratios in brackets, except where mentioned otherwise. d Robust standard errors; one-step ahead residuals.

98


8 Concluding remarks This chapter has examined the effect of international trade on manufacturing employment in India, comparing it with four developing countries, two in Africa and two in Asia, utilising a common set of approaches. Using the factor content approach, we find that the share of unskilled labour-intensive goods in India’s export basket has increased over time, but not at the same rate of growth as has been observed in Bangladesh and Vietnam. We also find that the employment coefficients of exports and imports in India have consistently fallen over the period, along with the fact that the difference between the employment coefficient of exports and that of imports has narrowed over time. This suggests that the employment impacts of trade for a given change in output may have actually been less in the post-reform period than in the pre-reform period. The growth accounting approach suggests that some of the employment growth in the 1990s can be linked to the growth of exports. However, most of the employment increase that has occurred over the period 1975–99 can be attributed to increases in domestic demand and less to international trade. Finally, estimating a labour demand equation that allows for trade to affect employment via changes in labour productivity, we find that there is no clear impact of trade on employment via the latter route. Our surprising finding is that the impact of international trade on manufacturing employment in India mirrors the two African countries for which we can undertake comparisons rather than the two Asian countries. This is an issue which we explore in more detail in the concluding chapter, as the latter finding has a broader significance that goes beyond the domain of employment to a wider assessment of the development impact of trade policy changes in India.

7


1 Introduction While a country’s exposure to international trade may increase the size of its resources due to both static and dynamic gains from trade, it may also affect the distribution of its resources (Goldberg and Pavcnik 2007). In this chapter, we first critically survey recent research on the impact of international trade on labour markets of developing countries. We ask the question: what happens to wages, particularly among the less skilled, in a developing country in the face of increasing international integration of its goods, services and capital markets? We attempt to answer this question by assessing what we know about the relationship between international trade and wage inequality in developing countries both from the perspective of economic theory and empirical evidence. We then examine the relationship between international trade and wage inequality in the Indian manufacturing sector for the period 1973–99. Given India’s relative abundance of unskilled labour relative to skilled labour, it is pertinent to ask whether India’s rapid integration with the world economy, following trade reforms undertaken in the 1980s and 1990s, has led to a narrowing of wage rate differences between unskilled and skilled labour in the Indian manufacturing. The rest of the chapter is in six sections. In the next section, we critically assess alternate theoretical perspectives on the effects of globalisation on labour markets in developing countries via international trade. Section 3 surveys the empirical studies that have examined the effect of international trade, investment and labour flows on labour markets of developing countries and summarises the main findings. Section 4 sets out the trends in wage inequality and related variables in Indian manufacturing. In Section 5, we evaluate whether the Heckscher–Ohlin theoretical perspective can explain the trends in the data that we observe in the previous section. Section 6 evaluates the validity of the ‘defensive innovation’ hypothesis of Adrian Wood. Section 7 concludes.

100


2 International trade and wage inequality: a synthesis of the theoretical literature This section reviews the three main theoretical perspectives on the link between international trade and labour market outcomes – the Heckscher– Ohlin model and the trade-induced technological progress and elasticity of labour demand arguments proposed by Wood and Rodrik respectively. The Heckscher–Ohlin model As we have noted, the one-cone H–O model has been the main theoretical framework guiding research in the area of international trade and labour markets. In this framework, all countries make the same sufficiently diversified mix of products (i.e. all countries’ endowments lie within the same ‘cone of diversification’ – the one-cone assumption) under perfect competition and all factors are perfectly mobile across industries such that international trade fully determines national factor prices as predicted by the Stolper–Samuelson (S–S) theorem. The S–S theorem provides the theoretical underpinning to the link between international trade and wage inequality in the H–O framework. The theorem’s core is the link between product prices and factor prices imposed by the ‘zero-profit conditions’ equating price with average cost that must hold in all perfectly competitive industries with actual production. These zero-profit conditions imply a systematic relationship between the entire set of product prices facing domestic producers and the entire set of factor prices paid by these producers. Any trade-induced change in a country’s product prices alters the relative profit opportunities facing its price-taking firms, who respond by shifting their resources towards the industries whose relative profitability has risen. This leads to an increase in country-wide demand for the factors used relatively intensively in the now relatively profitable sectors. Given fixed factor supplies, these demand shifts change factor prices until zero profits are restored in all sectors. The key implication of the S–S theorem is that trade impacts on the terms of trade, and by doing so, affects relative factor prices. How may the S–S theorem be used to explain the observed increase in wage inequality in the developed countries? Consider two countries, one in the North (developed) and the other in the South (developing), producing two commodities, each with two factors of production – skilled and unskilled labour. The North has comparative advantage in the production of the skilled labour-intensive commodity (that is to say, the relative price of the skilled labour-intensive commodity is lower in the North than in the South before trade). Suppose that there is now trade between the North and the South. With trade, domestic prices in these two commodities will be equalised, thereby stopping any further movement of goods. This would imply an increase in the relative price of the skill-intensive commodity in the North


101

and a fall in the relative price of the same commodity in the South. Via the S–S theorem, wages of skilled workers would rise in the North relative to those of unskilled workers, leading to increasing wage inequality in developed countries. By the same logic, international trade will cause decreasing wage inequality in developing countries. There are now several studies that attempt to analyse the issue of rising wage inequality in the North from an S–S perspective. One set of studies has focused on changes in the relative prices of products in developed countries to see whether there has been a decline in the prices of unskilled labourintensive products as predicted by the S–S theorem. The evidence from the United States is not very supportive of the S–S argument, as the prices of these products show no clear decline relative to those of skilled labourintensive products since the 1980s, a period in which inequality across skills started rising in that country. Another set of studies examines trade volumes in the light of the H–O model’s insight that trade embodies international flows of factor services. These studies attempt to deduce from changes in the factor content embodied in the net exports of the country in question (in most cases, the United States) the relative shifts in demand for and supply of skilled and unskilled labour, and, consequently, the implied changes in the relative wages of skilled versus unskilled labour (Wood 1994, Krugman 1995, Borjas et al. 1997). However, these studies have led to a substantial methodological debate about the conditions under which trade volumes can correctly identify the effect of trade on relative factor prices and whether factor-content studies have much empirical value.1 Despite unresolved methodological differences, the current consensus is that international trade accounts for a positive yet relatively small share of the increase in wage inequality and/or unemployment witnessed in the developed countries in recent decades (Slaughter 1998). The H–O model has come under increasing criticism from economists who have questioned some of the extreme assumptions of the model. Researchers have focused on the following assumptions of the model – (a) perfect competition prevails in all markets; (b) there is perfect mobility of factors across sectors of the economy; (c) countries do not specialise in production after trade; (d) trading countries produce homogeneous products in the same industry; (e) constant returns to scale; (f) the presence of non-traded goods; and (g) significant technological differences can exist between countries. Relaxing one or more of these assumptions in the H–O model may lead to a weakening or complete reversal of the stark prediction of the S–S theorem that international trade causes wage inequality to change asymmetrically in developed and developing countries (Winters 2000). To see how a relaxation of the H–O assumptions brings about a weakening of the S–S results, consider the case of domestic and foreign suppliers producing differentiated goods in the same industry, where the domestic varieties are imperfect substitutes of foreign varieties. In this case, the change in the prices of foreign varieties of this product may not have much of an effect on the prices of

102


domestic varieties, and, consequently, lead to little change in relative domestic factor prices. A second case is the presence of non-traded goods, which form a substantial proportion of total output in the country in question. In the case of non-traded goods, prices of these goods are determined in domestic markets, and changes in traded goods prices may not have much of an impact on non-traded goods prices. If non-traded goods use a particular factor more intensively (in the case of developing countries, it is usually unskilled labour), then changes in the price of that factor may not be greatly influenced by international trade but by changes in domestic demand and supply for non-traded goods.2 From the perspective of the South, there are two main implications for research that can be drawn from the discussion on the theoretical and empirical relevance of the H–O model in explaining wage inequality in the North. First, there are significant conceptual problems that come up in the empirical implementation of the H–O model to analyse the relationship between relative factor prices and actual trade flows between developed and developing countries. Second, in contrast to the simple versions of the H–O theory that have guided much of the research in the subject, more realistic versions of the H–O model do not allow an unambiguous prediction on the effect of international trade on wage inequality. Arguably, this implies the use of a great deal of caution in following a theory-guided approach to understanding changes in wage inequality in developing countries, if the theoretical framework is largely derived from the H–O model. Trade-induced technological progress An alternate and extremely influential explanation of the rising wage inequality observed in the North is that there has been significant technological change biased towards skilled labour in the developed countries in the recent past (principally due to increased computerisation of the workplace in these countries).3 Such technological change increases the marginal productivity of skilled labour relative to that of unskilled labour, increasing the relative demand for skilled labour, and, consequently, the wage premium on skilled work. While much of the literature has treated changes in technology as exogenous to international trade, Robbins (1995) and Wood (1994, 1995, 1997) argue that the latter may well be a driving force behind the former. For a developed country, trade could cause technological progress by inducing firms in the import-competing sector to look for new methods of production that economise on unskilled labour. This ‘defensive innovation’ may enable firms in the sector to fight off imports but would reduce the demand for unskilled labour (Wood 1995). For a developing country, international trade may be skill-enhancing by leading to an augmentation in the production technology available.4 This would occur if domestic firms obtain greater access to imported better-quality capital goods, or through opportunities for exporters to learn from foreign buyers and be exposed to foreign markets


103

(Robbins 1995). In both cases, trade-induced technological progress will lead to higher wage inequality. In the case of a developing country, the effect that international trade will have on wage inequality could be contrary to the predictions of simple versions of the H–O model.5 Furthermore, if wages are not flexible in the labour market, by contracting the demand for unskilled labour via improvements in technology, trade may lead to an increase in unemployment among unskilled workers. One problem with the arguments above is that trade-induced technological change is not necessarily skill biased. Consider the case of a Cobb–Douglas production function, Y = AU αS β

(7.1)

where A is an index of Hicks-neutral technological progress, U is unskilled labour and S is skilled labour, and the production function is characterized by decreasing returns to scale, that is, α + β < 1.6 If w u and w s are wages of unskilled and skilled labour respectively, the demand for unskilled labour will be: α U* = U w

1−β γ

β wS

β γ

冢冣冢冣冢Ap冣

1 γ

(7.2)

where p is the product price, and γ = 1 − α − β. It is clear from the above that an increase in A will unambiguously increase U*. The index of technological progress, A, may itself be positively related to trade reform – greater openness could lead to improvements in productivity for firms and industries in the economy as we have seen in Chapter 5. Thus, Wood’s argument is premised on a certain type of technological progress – skill-biased – occurring following trade reforms, and it not obvious why this should be so. Trade-induced technological progress that is factor neutral will, on the other hand, lead to an increase in the employment of unskilled workers in the economy. Trade and the own-price elasticity of labour demand The argument that international trade may have adverse outcomes for the wages and employment of unskilled workers in developing countries by increasing the own-price elasticity of labour demand finds its clearest statement in Rodrik (1997). This argument can be best understood by appealing to some simple notions in the theory of labour demand. Consider the standard formulation of firm’s own-price labour-demand elasticity: ηLLj = −[1 − s] σLL − sηj

(7.3)

104


where ηLL is firm j’s labour’s own-price demand elasticity (i.e. the percentage change in labour demanded by firm j in response to a 1 per cent change in labour’s price), and is defined to be negative; s is labour’s share of the firm’s total revenue; σLL is the constant-output elasticity of substitution between labour and all other factors of production, and ηj is the product-demand elasticity facing firm j in its output market. As written in (1), ηLLj consists of two parts. The first, − [1 − s] σLL is the ‘substitution effect’ and tells, for a given level of output, how much the firm substitutes away from labour towards other factors after a wage increase. The second part of (7.3), − sηj, is the ‘output effect’ or ‘scale output’, and tells how much labour demand changes after a wage change due to a change in the firm’s level of output. Higher wages imply higher costs and thus, for a given product-market demand schedule, lower firm output. Both the substitution and scale effects reduce labour demand in the face of an increase in wages, as the firm substitutes away from labour towards other factors because labour is now relatively more expensive, and with higher costs the firm produces less output such that it demands less of all factors. For these reasons, ηLL is negative. International trade may affect the firm’s own-price elasticity of labour demand, ηLLj , in two ways. First, if international trade leads to increased product-market competitiveness, product demand will become more elastic (that is to say, ηj will increase), and so too will the firm’s own-price elasticity of labour demand. As Rodrik (1997: 16) argues, ‘since the demand for labour is a derived demand, which varies proportionately with the elasticity of demand for goods, the integration of goods markets alone makes the demand for domestic labour more elastic’. There is strong theoretical and empirical support for the proposition that international trade decreases the market power of incumbent firms and, by doing so, increases the product elasticity of demand (Levinsohn 1993, Harrison 1994, Katics and Petersen 1994, Chand and Sen 1999). Second, international trade can also increase the firm’s own-price elasticity of labour demand by increasing the elasticity of substitution between labour and other factors, σLL. The reason is ‘that employers and the final consumers can substitute foreign workers for domestic workers more easily – either by investing abroad or by importing the products made by foreign workers’ (Rodrik 1997: 16). Interestingly, from (7.3), it is clear that the larger the share of labour in the firm’s total revenue/costs, s, the larger the impact on ηLLj of an increase in ηj, but an increase in σLL will have a smaller effect on ηLLj.7 Rodrik draws three implications from the argument that international trade leads to an increase in the own-price elasticity of labour demand. First, increased trade and investment opportunities will make it more costly for workers to achieve a higher level of labour standard and benefits. Second, higher elasticities will trigger more volatile responses of wages and/or employment to any exogenous shock to labour demand. Finally, higher elasticities may shift from labour towards capital the bargaining power over rent


105

distribution in firms which enjoy extranormal profits, resulting ‘in a lower share of the enterprise surplus ending up with workers’ (Rodrik 1997: 24). It will also mean that unions will become weaker. Recently, Rodrik’s argument that international trade will unambiguously increase the elasticity of demand has come under attack from Panagariya (1999). Using both Heckscher–Ohlin and the specific factors trade models, Panagariya shows that there is no obvious link between openness and the economy-wide elasticity of labour demand.8 This criticism is, in our view, based on a misunderstanding of one of the main elements of Rodrik’s theory – that it applies to the labour demand of an individual firm, and not of the entire country. This misunderstanding is compounded by a looseness in Rodrik’s own exposition of his theory of what he means by ‘labour demand’ – thus, he states, ‘in an economy that is more open to foreign trade and investment, the demand for labour will generally be more responsive to changes in the price of labour, or more elastic’ (1997: 16). Panagariya’s arguments would be correct if ‘the demand for labour’ referred to here is the national demand for labour, when clearly Rodrik’s theory applies only to the firm’s demand for labour.9 There is a more significant problem in the inference that Rodrik draws from his argument that international trade has increased the own-price elasticity of the firm’s labour demand. To Rodrik, the outcome of such an increase in elasticity of labour demand is necessarily negative for wages and employment of unskilled workers. However, it is not obvious why this should be so. An increase in the product-elasticity of demand, ηj, following an increase in product market-competitiveness brought about by trade liberalisation, may allow firms to expand production and increase labour demand (in the next section, we will review studies for developing countries that find empirical evidence for such a mechanism). Contrary to Rodrik’s assertion, there is no clear theoretical link between higher elasticities of labour demand following the opening up of the economy and adverse wage and employment outcomes for unskilled workers. It should also be noted that, as in the case of the changing relative price of unskilled labour, it is still very much an open-ended issue whether international trade has led to an increase in the elasticity of labour demand in the North. Using industry-level data for the US for 1960–91, Slaughter (1997) finds that while demand for production labour has become elastic in several industries, the hypothesis that trade has contributed to increased elasticities has mixed support, at best.

3 Previous empirical evidence In contrast to the vast literature on international trade and wage inequality for the developed countries, there are relatively few such studies for developing countries. We briefly survey some of the more significant studies in this area.

106


Among the most significant empirical pieces that examine the effect of trade reforms and foreign direct investment on wage and employment outcomes for developing countries are studies by Robbins (1995, 1996), Currie and Harrison (1997), Feenstra and Hanson (1997), Khambhampati et al. (1997), Revenga (1997), Milner and Wright (1998), Beyer et al. (1999), Harrison and Hanson (1999) and Ghose (2000). With the exception of Robbins and Ghose, all these studies are confined to individual countries, and most of these studies are concentrated in Latin America. We begin our review with a discussion of the single-country studies, organising the discussion by the continents within which these countries are located. A study on the Mexican manufacturing sector following the trade reforms in the 1980s is by Hanson and Harrison (1999). This study finds strong evidence of an increase in the wage gap between skilled and unskilled workers. Hanson and Harrison argue that the trade reforms could be considered to be the driving force behind the increasing wage inequality. According to Hanson and Harrison, the mechanism by which it operated was very much along S-S lines. Prior to the reforms, Mexico had extended trade protection preferentially to industries that make relatively intensive use of unskilled labour, even though it had ‘an intermediate abundance of skilled labour’. Thus the trade liberalisation had a disproportionately large impact on the non-skilled intensive sectors, as the exposure to China and other countries that have abundant unskilled labour contributed to a decrease in the relative wages of unskilled workers. Similar to the experience of Mexico, Beyer et al. (1999) find evidence of increasing wage inequality in Chile following the trade reforms in the 1970s and 1980s. They argue that this could be linked to a change in the productive structure of the country, although they are unclear what these changes may be and how the latter could be causally linked to the changes in the wage premium on skilled labour. Another factor that could have contributed to increasing wage inequality in Mexico was the dramatic increase in foreign direct investment (FDI) since the capital market deregulations of the 1980s. This increase in FDI mostly occurred in assembly plants known as maquiladoras. Using a rich data-set that combines regional and industry level data on maquiladoras, Feenstra and Hanson (1997) find that the increase in the relative demand for skilled labour is positively correlated with the change in the number of maquiladoras, suggesting that greater wage inequality could also be linked to the increasing flow of capital, and not just to the flow of goods. In one of the few papers that has studied whether trade reforms lead to higher own-wage elasticities of labour demand, Fajnzylber and Maloney (2005) use dynamic panel data techniques to estimate labour demand relations for manufacturing establishments in Chile, Colombia and Mexico. Their results do not strongly support the hypothesis that trade reforms have a direct impact on own-wage elasticities.


107

Comparative studies The papers by Robbins (1995, 1996) provide a much-needed comparative perspective to the relationship between trade liberalisation and labour market outcomes in developing countries. From the two studies by Robbins, there seems to be some evidence of a continental divide in the way globalisation affects wage inequality in developing countries. Controlling explicitly for changes in domestic factor supply, Robbins finds that greater openness has led to decreasing wage inequality between skilled and unskilled workers in the majority of the East Asian countries in his sample.10 However, for the five Latin American countries studied (Argentina, Chile, Colombia, Costa Rica and Uruguay), there is evidence of a widening inequality in wages associated with greater openness. The finding that greater openness may have led to higher wage inequality in Latin America seems to be fairly robust, given that the studies have been for several countries and used a variety of approaches (studies have been both at firm and industry level, and also economy-wide). It is not clear why this has been the case, as compared to the more beneficial effects of greater openness on the wage gap in the East Asian countries. Wood (1997) argues that the difference in outcomes may be explained by the different periods used in Robbins’ comparative studies – those on East Asia are for the 1960s and 1970s, while those for Latin America are for the 1980s and 1990s. The difference between these two periods is that in the latter period, the international trading system witnessed the entry of large low-income exporters, such as China, which had abundant supplies of unskilled labour. This may help explain why trade reforms may have hurt the interests of unskilled workers in countries in Latin America, whose ratio of unskilled labour to skilled labour is below the average for developing countries. A different argument is offered by Leamer at al. (1999) who argue that Latin American countries are abundant not in labour (whether skilled or unskilled) but in natural resources. The abundance of natural resources implies a more capital-intensive development strategy (due to a shift towards capital-intensive extraction and permanent agriculture) as these countries specialise following reforms. This led to a greater relative demand for skilled labour, and consequently, higher wage inequality in these countries. Previous studies on India Two previous studies on India that have examined the relationship between trade liberalisation and wage inequality in India are Mishra and Kumar (2005) and Chamarbagwala (2006). A third paper has assessed the ‘elasticity of labour demand’ hypotheses that we have discussed in the previous section, The first paper evaluates the impact of the 1991 trade liberalisation on the industry wage structure. It finds a strong negative relationship between changes in trade policy and changes in industry wage premiums over time.

108


The second paper finds a widening skill gap and narrowing gender wage differential between 1983 and 2000 for the economy as a whole, and investigates whether these can be attributed to economic liberalisation. It finds large increases in within-sector demand for skilled men and women, which were not solely generated by trade liberalisation, but other domestic sector reforms as well. Both papers use individual level micro-data collected by the Indian National Sample Survey Organisation (NSSO). Hasan et al. (2003) find that the trade reforms have a positive effect on labour-demand elasticities, which would imply that volatility in productivity and output will get translated into larger wage and employment volatility. Our paper is similar to Chamarbagwala (2006) in its focus on the skill premium, rather than on inter-industry wage differences as in Mishra and Kumar (2005). However, unlike Chamarbagwala (2006), we use industries rather than workers as units of analysis, as in Hanson and Harrison (1999). Since the data on wage rates by skill for industries is continuous and available annually (and spans over twenty years in our case) this allows us to explore the time dimension of the changing skill premium in a more rigorous manner. In contrast, the wage data for individuals is available only for a few years, coinciding with the years that the NSSO conducted employment surveys in the country. This is particularly relevant in the Indian case, where trade reforms have been incremental and have taken place very gradually over time and unevenly across sectors.

4 Relative wages and employment in India As is standard in the literature, we define skilled workers to be white-collar workers (that is, non-production workers such managers and supervisors, along with clerical and administrative staff) and unskilled workers to be bluecollar workers (that is, production workers). The classification of production workers as unskilled workers and non-production workers as skilled workers is also supported by the data on relative wages in India – as we will see later in this section, in all industries, wage rates of non-production workers are between one and a half to three times the wage rates of production workers. We begin with a graph of aggregate wage inequality in Indian manufacturing (Figure 7.1). We see that after a period of falling wage inequality in the early 1970s, and then a period of sharply rising wage inequality in the late 1970s, there has been a period of slow but a steady increase in wage inequality since the mid-1980s. However, the relative wage of skilled workers in the end of the 1990s still remains below what it was in the early 1970s. The increase in wage inequality seems to have gone hand in hand with an increase in the relative employment of skilled workers across all industries over time – the ratio of skilled to unskilled workers has increased from 0.30 in 1981 to 0.34 in 1997 (Figure 7.2). The upward movement in both series since the early to mid-1980s seems to suggest that relative demand shifts in favour of skilled workers have been dominant in both periods. Whether this is due to


109

Figure 7.1 Wage inequality in Indian manufacturing – aggregate. Source: Annual Survey of Industries, various issues; our calculations. Note: Wage inequality is the wage rate of skilled workers relative to the wage rate of unskilled workers.

international trade or simply skill-biased technological change (SBTC) that was exogenous to external factors is a question that we need to address later in the chapter. However, H–O theory may also explain the increase in wage inequality if the decline in protection has been mostly in the unskilled labourintensive industries. Is there a relationship between the decline in protection and degree of skill intensity of the industry? A priori, we may have expected that there would be a negative relationship between the decline in ERPs (or ICRs) over the period 1980–99 and the share of unskilled workers in total employment in 1980, since the largest reductions in ERPs occurred in the intermediate goods sector, which would tend to be more skilled labour-intensive than the consumer goods sector. However, we find from Figure 7.3 that there is in fact a weak but positive relationship between the share of unskilled workers in an industry and the decline in protection in that industry. This has also been observed by Mishra and Kumar (2005) who find that tariff reductions have been the largest in industries with a greater share of unskilled labour in India. This phenomenon has also been observed in several Latin American countries such as Argentina, Colombia and Mexico (Hanson and Harrison 1999, Galiani and Sanguinetti 2003, Attanasio et al 2004,), where the largest tariff reductions occurred in the unskilled labour-intensive industries. We now undertake a more detailed examination of wage inequality by NIC 2-digit industry, averaged over the periods 1973–7, 1978–84, 1985–90 and

110


Figure 7.2 Relative employment of skilled workers in Indian manufacturing – aggregate. Source: Annual Survey of Industries, various issues; our calculations. Note: Relative employment of skilled workers equals the employment of skilled workers relative to the employment of unskilled workers.

Figure 7.3 Reduction in Effective Rates of Protection (ERP) between 1980 and 1999 and the share of unskilled workers in 1980, India. Source: Annual Survey of Industries, various issues; our calculations.


111

1991–7 (Table 7.1). For the last three periods, the starting years were years when important changes in the trade regime occurred – in 1978, there was an initial relaxation of import controls, a further relaxation in 1985, and then the major reforms that occurred in 1991. We note that wage inequality in the pre-trade reforms period, 1973–7, was higher in all industry groups than in the period of the radical trade reforms, 1991–7. However, wage inequality in 1991–7 was higher in several industries than in 1985–90. It is also striking that increases in wage inequality after the 1991 reforms do not seem to be correlated with the sectors where the major reforms occurred – namely, the capital goods and intermediate goods sectors. At first glance, there does not seem to be any obvious relationship between trade policy changes and changes in wage inequality both across industries and over time.

5 Testing for Stolper-Samuelson effects As noted in Section 2, the link between trade liberalisation and changes in relative wage inequality is provided by the workhorse model of international Table 7.1 Wage inequality by industry (NIC 2-digit), India Industries

1973–7

1978–84

1985–90

1991–7

Food products (21) Beverages and tobacco (22) Cotton textiles (23) Wool, silk and man-made fibre textiles (24) Jute and other vegetable fibre textiles (25) Textile products including wearing apparel (26) Wood products (27) Paper products (28) Leather products (29) Chemical products (30) Rubber, plastic and petroleum products (31) Non-metallic mineral products (32) Basic metal and alloy industries (33) Metal products and parts (34) Machinery and equipment (35 and 36) Transport equipment and parts (37) Other manufacturing industries (38)

2.75 3.10 2.51 2.08

2.17 2.54 2.21 1.87

2.06 2.11 1.75 1.86

1.95 2.21 1.87 1.92

2.09

1.70

1.71

1.56

2.20

1.75

1.80

2.03

2.02 2.19 2.55 3.01 3.37

2.13 1.66 2.39 2.18 2.02

1.86 1.62 2.33 2.00 1.90

1.83 1.75 2.14 2.00 1.92

2.61 2.26 2.23 2.37

2.03 1.79 1.92 1.83

2.09 1.73 1.84 1.76

1.91 1.79 1.83 1.82

1.87 1.99

1.69 1.84

1.63 1.86

1.63 1.84

Source: Annual Survey of Industries, various issues; our calculations. Notes a Skill Premium equals the wage rate of skilled workers relative to the wage rate of unskilled workers. b NIC 1987 2-digit industry codes in brackets.

112


trade, the Heckscher–Ohlin model, and its companion theorem, StolperSamuelson (S-S). According to the Heckscher–Ohlin model, a developing country will export relatively unskilled labour-intensive commodities and import relatively skilled labour-intensive commodities. The S-S theorem predicts that, with trade liberalisation, the price of the import sector will fall and the price of the export sector will rise. This will hurt the factor that is used intensively in the import sector (skilled labour) and benefit the factor that is used intensively in the export sector (unskilled labour). Thus, with trade liberalisation, wage inequality between skilled and unskilled workers will decrease. An important point to keep in mind in the above discussion of the S-S theorem is that trade liberalisation will affect economy-wide and not sectorspecific wage inequality. This is because factors of production are assumed to be mobile across uses within the country, so that their returns are equalised across sectors. Thus, labour mobility is an essential ingredient of the H–O argument (Attanasio et al. 2004). We begin our evaluation of the applicability of H–O theory to explain movements in wage inequality in India by revisiting our findings in Chapter 6 on the changing factor content of India’s exports and imports. According to H–O theory, India should be exporting more unskilled labour-intensive commodities and importing more skilled labour- and capital-intensive commodities, especially after the trade reforms of the 1980s and 1990s, which should have aligned India’s comparative advantage with her factor endowments. We found in Chapter 6 that unskilled labour-intensive commodities are the most important in India’s manufacturing exports, comprising 45 per cent of total manufacturing exports in 1996–9 (see Table 6.1). However, while the share of unskilled labour-intensive commodities in total manufacturing exports has increased over the period 1975–99, the increase has not been substantial, from 37 per cent in 1975–80. India’s human capital intensive and technology intensive exports have also increased as a share of total manufacturing exports over the period 1975–99, from 10 and 17 per cent in 1975–1980 to 15 and 24 per cent in 1996–9 respectively. On the other hand, India’s land-intensive exports as a share of total manufacturing exports has fallen quite dramatically from 31 per cent in 1975–80 to 12 per cent in 1996–9. With respect to imports, capital-intensive exports remain the dominant set of commodities in India’s manufacturing import basket, followed by skilled labour-intensive goods. There has been a slight increase in the share of skilled labour-intensive goods in India’s manufacturing imports, with no significant change in the share of capital-intensive imports in total manufacturing imports, in spite of the trade liberalisations of the 1980s and 1990s which were mostly biased towards import liberalisation of capital and intermediate goods. The above findings do not seem to suggest that there has been a major re-alignment of India’s structure of trade such that there has been a significant increase in the share of unskilled labour-intensive goods in total


113

manufacturing exports and a significant increase in the share of capitalintensive and skilled labour-intensive goods in total manufacturing imports. Though it would be wrong to conflate the quantity effects of international trade with price effects, there is reason to believe that there may not be strong S-S effects on relative wage inequality in Indian manufacturing since the trade reforms of the 1980s and 1990s. Do we observe significant changes in the shares of employment in each industry such that there has been a re-allocation of labour from the skilled labour-intensive industries to unskilled labour-intensive industries? We look at industry shares in employment over the period 1973–97 (Table 7.2). We find that employment shares in some of the industries which are intensive in unskilled labour have increased since the mid-1980s, such as wearing apparel (NIC 2-digit code: 26) and leather products (NIC industry code: 29). At the same time, the employment shares of some skilled labour-intensive industries such as chemical products (NIC 2-digit code: 30) and rubber and petroleum products (NIC 2-digit code: 31) have also increased. There is no obvious pattern that emerges from the changes in employment shares, especially since the trade reforms of the mid-1980s. There is a remarkable stability in employment shares in industries across time – the correlation coefficient between employment shares at the NIC 3-digit level in 1973 and 1999 is 0.77. Such stability of employment shares may not be surprising in the Indian context, where there are significant restrictions on the movement of labour from one sector to another, and given the labour laws in the organised segment of the manufacturing sector which do not easily permit retrenchment of workers and their redeployment from one location to another without formal approval of the government (Ganesh-Kumar et al. 2003). This also clear from Figures 7.4 and 7.5 where we plot unskilled and skilled employment growth over the period 1973–99 in the vertical axis respectively against the ratio of skilled to unskilled workers in 1973 in the horizontal axis. It is clear that there is no clear pattern in the growth of skilled and unskilled employment – it is not the case that unskilled labour-intensive industries have seen the most growth in unskilled labour and the same is true for skilled labour-intensive industries. We next proceed to test for S–S effects more rigorously by estimating regressions as in Hanson and Harrison (1999), where we regress the relative share of skilled workers in a particular industry against real price changes in that industry. If S–S effects are present in the data and if trade reforms occurred mostly in the skilled labour-intensive industries, we should expect a negative relationship between real output price and relative skill intensity. As output prices increase in the unskilled labour-intensive industries relative to output prices in the skilled labour-intensive industries, there should be a decrease in overall relative skill intensity of employment. We regress the log of the ratio of skilled to unskilled workers against real domestic prices and present the estimates in Col. (1) of Table 7.3. We find a strong negative relationship between the two variables, as expected from the predictions of

114


Table 7.2 Employment shares (percentage), India Industries

1973–7

1978–84

1985–90

1991–7


17.8 5.0 19.4 0.9

17.7 6.3 16.4 0.8

15.7 6.9 13.7 0.9

16.6 7.8 11.8 0.9

5.1

4.6

4.0

3.6

1.5

1.6

2.1

4.0

1.5 3.8 0.8 6.0 1.8

1.3 3.7 1.0 6.7 1.8

1.2 3.7 1.4 7.9 2.1

1.1 3.8 1.6 9.0 2.4

5.3 9.3 3.2 10.7

5.8 9.9 2.9 10.8

7.0 10.0 3.1 11.7

7.2 8.4 3.1 11.0

6.9 0.8

7.5 0.8

7.2 1.1

5.9 1.6

Source: Annual Survey of Industries, various issues; our calculations. Notes a Employment shares are the employment of skilled and unskilled workers in each industry as a ratio of total employment of skilled and unskilled workers in the manufacturing sector. b NIC 1987 2-digit industry codes in brackets.

the S–S theorem. The coefficient on real price is significant at the 1 per cent level. However, output price changes may not only be caused by trade policy changes and could also be due to the dismantling of the industrial licensing system in 1991. We directly regress the log of the ratio of skilled to unskilled workers against different measures of trade policy. In Col. (2), we regress relative skill intensity on ERP and ICR, in Col. (3), on import penetration and export orientation. We find that there is a strong negative relationship between policy-based measures of trade policy such as ERP and ICR on one hand, and relative skill intensity on the other. With respect to outcome-based measures of trade policy, the negative relationship holds true for export orientation but not for import penetration. Finally, we estimate a betweeneffects regression on industry means, to take account of the possibility that the negative relationship between trade policy measures and relative skill intensity is due to purely trade-induced SBTC, which may be leading to the substitution of unskilled labour for skilled labour within industries. The estimates are presented in Col. (4) and we see that the negative relationship


115

Figure 7.4 Relationship between unskilled employment growth and skill intensity, India. Source: Annual Survey of Industries, various issues; our calculations. Note: Change in Unskilled Workers, 1973–1997 equals log of Unskilled Workers in 1997 − log Unskilled Workers in 1973.

Figure 7.5 Relationship between skilled employment growth and skill intensity, India. Source: Annual Survey of Industries, various issues; our calculations. Note: Change in Skilled Workers, 1973–1997 equals log of Skilled Workers in 1997 − log Skilled Workers in 1973.

116


between relative skill intensity and real prices remains valid even at the crosssectional level. Thus, there seems to be evidence of reasonably strong S–S effects in the data in the period under consideration.

6 Testing for trade-induced skill-biased technological change Our finding in the previous section that S–S effects are present in the Indian manufacturing sector during a period when this sector gradually opened due to trade reforms does not necessarily mean that trade-induced technological change has not occurred at the same time. When we examine relative employment of skilled workers by NIC 2-digit industry (Table 7.4), we do not seem to see a clear pattern emerging for the whole period 1973–99 across all industries – the relative employment of skilled workers is higher in some industries in 1991–7 compared to 1973–7, and lower in some other industries for the same periods of comparison. However, we observe that the relative Table 7.3 Relative skill intensity and measures of trade policy – regression results, pooled ordinary least squares and between effects, India Variables

(1)

(2)

(3)

(4)

Intercept

–

–

–

ERP

–

–

ICR

–

–

–

IMP

–

−0.006*** (17.8) −0.005*** (9.1) –

−0.30 (0.64) –

−0.0005 (0.76)

–

EO

–

–

–

P

−0.27*** (132.2) 0.83 Pooled, OLS

–

−0.06*** (7.6) –

0.64 Pooled, OLS

0.005 Pooled, OLS

3,468

1,000

1,219

R-square Method

Number of observations

−0.21 (2.08)** 0.002 Between effects on industry means, OLS 3,468 (number of industries: 139)

Notes a Dependent Variable: Ratio of Skilled to Unskilled Workers (in logarithmic form). b ERP = Effective Rate of Protection; ICR = Import Coverage Ratio; IMP = Import Penetration Ratio; EO = Export Orientation Ratio; P = Real Output Price. c ***,**, * denote that the independent variable is significant at 1, 5 and 10 per cent respectively. d OLS is Ordinary Least Squares. e T-ratios in parentheses.


117

employment of skilled workers is higher in all industries in 1991–7 as compared to 1985–90. This is a striking finding, and seems to suggest strong SBTC across the board in the Indian manufacturing sector since the economic reforms of 1991. Can this be linked to international trade? To test for the presence of tradeinduced STBC, we estimate fixed effect regressions of wage inequality against measure of trade policy – the use of fixed effects allows us to capture only within industry variation in wage inequality, and to see if it has been impacted on by trade reforms. By doing so, we can isolate the effect of trade-induced STBC from the Stolper–Samuelson effects, which should apply both over time and across industries. We present estimates of fixed effect regressions in Table 7.5. In Col. (1) we include both ERP and ICR, in Col. (2) only ERP, in Col. (3) only ICR, and in Col. (4) import penetration and export orientation. Though the coefficients of ERP and ICR are not statistically significant in Col. (1), there is possibility of collinearity between the two measures of trade policy. When we include the two measures in turn, we find a clear negative and significant relationship between the degree of protection and wage Table 7.4 Relative employment of skilled workers by industry, India Industries

1973–7

1978–84

1985–90

1991–7


33.7 28.3 17.1 23.2

32.0 25.2 15.5 22.8

30.8 26.9 18.4 22.1

31.1 26.9 18.7 25.2

27.1

26.1

22.9

26.8

21.4

25.3

25.3

28.5

26.1 33.7 23.2 39.2 29.6

24.8 34.7 21.3 43.0 31.9

24.3 35.7 18.1 42.7 30.7

25.9 41.2 20.4 45.3 33.5

22.7 35.4 30.5 39.8

27.9 33.0 28.8 44.3

24.7 33.5 29.6 50.5

28.6 35.7 32.6 52.3

33.6 35.5

41.9 39.8

31.6 34.6

31.9 40.4

Source: Annual Survey of Industries, various issues; our calculations. Notes a Relative employment of skilled workers equals the employment of skilled workers relative to the employment of unskilled workers. b NIC 1987 2-digit industry codes in brackets.

118


Table 7.5 Wage inequality and measures of trade policy – regression results, fixed effects, India Variables

(1)

(2)

(3)

(4)

Intercept

0.66*** (22.5) −0.005 (2.15)** –

0.67*** (27.9) –

0.63*** (97.9) –

–

−0.005* (1.78) –

–

IMP

0.67*** (21.8) −0.004 (0.97) −0.004 (1.35) –

EO

–

–

–

R2 Method Number of observations

0.006 Fixed effects 1,000



ERP ICR

−0.0001 (0.59) −0.0003 (0.15) 0.0003 Fixed effects 1,219

Notes a Dependent Variable: Ratio of Wage Rates of Skilled to Unskilled Workers (in logarithmic form). b ERP = Effective Rate of Protection; ICR = Import Coverage Ration; IMP = Import Penetration Ratio; EO = Export Orientation Ratio. c ***,**, * denote that the independent variable is significant at 1, 5 and 10 per cent respectively. d OLS is Ordinary Least Squares. e T-ratios in parentheses.

inequality. The outcome-based measures of trade policy are not significant, however. On the whole, we find evidence of a decline in protection leading to an increase in wage inequality within industries, suggesting the presence of trade-induced skill biased technological progress.

7 Conclusions This chapter has tried to assess what we know about the relationship between international trade and wage inequality in developing countries from the perspective of economic theory and empirical evidence. Among the many ways international trade affects labour markets in developing countries, perhaps the strongest is the shifts in labour demand that follow from tradeinduced changes in product prices or output. The theoretical literature generates ambiguous results on whether international trade has had adverse outcomes on labour markets in developing countries by increasing wage inequality or negatively affecting wages and employment. Three theoretical perspectives have been proposed in the literature – the Stolper-Samuelson theorem, trade-induced technological progress and the elasticity of labour demand arguments. We have argued that there is a lack of robustness in the


119

results that come out of all these theories, so that no theory can unequivocally predict what will happen to wage inequality or, more generally, to wages and employment in a developing country in the face of increased international trade. We then examined the Indian case. After a period when the relative wages of skilled (non-production) workers fell relative to those of unskilled (nonproduction) workers in the early 1970s, there has been a period of slow but a steady increase in wage inequality in the Indian manufacturing sector in the 1990s. The increase in wage inequality seems to have gone hand in hand with an increase in the relative employment of skilled workers across all industries since the mid-1980s. We investigated whether the co-movement of relative employment and wages of skilled workers can be attributed to the changes in trade policy that have occurred in the Indian economy since the mid-1980s using disaggregated industry-level data for Indian manufacturing for the period 1973–97. We focused on the two dominant theoretical perspectives on why international trade may impact on skill intensity and wage inequality – the Heckscher–Ohlin theory and trade-induced skilled-biased technological progress. We found evidence of the validity of both theoretical perspectives to explain the co-movement in wage inequality and relative skill intensity in Indian manufacturing, with both variables increasing in the 1990s. Tradeinduced technological progress has led to an increase in relative skill intensity and wage inequality within industries. At the same time, the decline in protection that seems to have occurred more in unskilled labour-intensive industries has led to a relative fall in the economy-wide return to unskilled labour relative to skilled labour. Therefore, trade reforms have led to a widening of wage gap between skilled and unskilled workers, and an increase in relative skill intensity in Indian manufacturing.

8


1 Introduction How regions adjust to trade liberalisation is a topic that has only recent attracted attention amongst researchers and policy-makers. An important reason why there is an increasing interest in the regional impacts of trade policy changes is due to recent theoretical developments in the ‘new economic geography’ that stress the importance of agglomeration factors that may explain why some regions develop at a different rate than others (Fujita 1988, Krugman 1991). This new literature builds on the insights of an earlier literature that has highlighted the importance of climactic factors and other innate geographical factors in explaining spatial disparities in economic development (Sachs 2003). A second reason why there is an interest in spatial aspects of trade reforms is that the experience of large countries like China in integrating with the world economy has shown that some regions within the country have benefited from trade integration more than others (Demurger et al. 2002, Kanbur and Zhang 2005). In this chapter, we examine the evolution of the spatial structure of production and employment in the manufacturing sector in India and assess to what extent the patterns and trends we find can be explained by changes in the trade policy regime. In the next section, we briefly review the theories of the new economic geography, highlighting the key predictions of these theories. In Section 3, we examine the patterns and trends in employment and production in the fourteen major states in India. In Section 4, we assess to what extent these patterns and trends can be explained by trade policy and its changes. Section 5 concludes.

2 The relationship between trade policy and regional economic development As we have seen in the previous chapters, the H-O model makes strong predictions on the impact of trade policy on the sectoral allocation of resources – between labour-intensive and capital-intensive industries, for example – and on the relative demand for different factors of production – skilled and


121

unskilled workers, for example. It is only recently that trade economists have been interested in the regional impacts of trade policy. H-O theory suggests that trade will cause an economy to be more specialised. Regions that are relatively specialised in import-competing industries will contract while regions specialised in exporting industries will expand. Alternate models of regional economic activity which build on the theoretical insights of Henderson (1974) and Fujita (1988) emphasise the importance of increasing returns to scale, transport costs and congestion costs in explaining the relationship between international trade and industry location (Hanson 1998). The interaction of scale economies and transport costs creates a centripetal force such that firms tend to agglomerate in industry centres and large cities. The presence of internal economies implies that firms economise on both transport costs and production costs by locating near a large market. The presence of external economies implies that firms benefit from spillovers by locating near other firms in their industry. On the other hand, with concentration of economic activity, land rents increase, and workers located in or close to industry centres demand higher wages, which operates as a centrifugal force. Higher land rents and wages are congestion costs and work against agglomeration. The degree of concentration of economic activity will, therefore, depend on whether centripetal forces dominate over centrifugal forces. How does international trade affect the relative importance of centripetal forces versus centrifugal forces? One set of models predicts that when an economy moves from being closed to being open, it is more likely that a balanced city structure may develop (Puga 1998, Krugman and Livas 1992). In a more open economy, firms are less reliant on local markets and local suppliers, so within-city agglomeration forces are weaker. Therefore, it is possible that, with international trade, there could be a deconcentration of economic activity. However, there would be a tendency for firms to relocate near ports, and therefore in coastal regions or border areas (Venables 2003). This will be due to the lower transport costs involved in exporting if firms are located near ports or close to their nearest markets across the border. The greater the returns to agglomeration and the larger the presence of external and internal economies relative to congestion costs, the more likely will it be that coastal regions will prosper relative to inland regions. However, if the economy in question is large enough, there may be other industrial centres that are not near ports or border areas where the firms in these centres are producing essentially for the home market. The clear theoretical prediction that emerges from the above discussion is that coastal regions will tend to do better with respect to manufacturing production and employment than inland regions. However, it is not clear how international trade will impact on regional economic convergence – will regions with already developed industrial bases do better than regions with underdeveloped industrial bases with the opening up of the economy? And will regions with large industrial centres find that with deconcentration of economic activity, their importance in the economy recedes over time? The

122


prevalent theoretical perspective drawn from the new economic geography does not provide clear guidance on what we may expect, as it will depend on the balance of forces for and against agglomeration. Thus, what happens to the spatial structure of economic activity after opening up seems to be an empirical matter.

3 Trends and patterns in regional distribution of manufacturing in India The patterns of production and employment in the manufacturing sector in recent years in India have been relatively under-studied in the literature. Most studies that look at regional economic development examine disparities in total incomes in Indian states (e.g. Marjit and Mitra 1996, Sachs et al. 2002, Marjit and Kar 2007).1 We describe trends and patterns in production and employment for fourteen major Indian states in this section. To relate the descriptive statistics that we present in this section to the theoretical insights we have discussed in the previous section, we classify the fourteen states in three different ways – by regional grouping (East, West, North and South), coastal versus inland, and states with a large metropolis (metrop.) versus states without a large metropolis. We provide the categorisation of states using these three classification schemes in Table 8.1. The table is self-explanatory, except the inclusion of Madhya Pradesh as a Northern state, when it is a state in the Central region of the country. This we did to eliminate a possible fifth regional grouping – states in the Central region, in order to keep the regional groupings simple. However, our results do not change if we were to include Madhya Pradesh in a different regional group. With respect to the metropolises, we confine ourselves to the largest of Indian cities: Ahmedabad, Bangalore, Chennai, Delhi, Kolkata, Hyderabad and Mumbai. The data that we use is from the Annual Survey of Industries (see Appendix for more details). Unfortunately, state-level industry data is only available at the NIC 2-digit level, so the analysis in this chapter is undertaken at that level. We provide a list of the NIC 2-digit industries in Table 8.2. Examining the structure of employment and production in the Eastern states first (Table 8.3), we find that these three states have seen their aggregate share in employment and output contract over the period 1979–97. The share of Eastern states in manufacturing employment was 14.9 per cent in 1979. This fell to 12.5 per cent in 1997. Similarly, the share of Eastern states in manufacturing output was 14.7 per cent in 1979. The share fell to 8.0 per cent in 1997. The difference between average employment growth in the Eastern states and average employment growth across all states was −1.8 per cent in the period 1979–97. At the same time, the difference between average output growth in the Eastern states relative to average output growth across all states was −4.1 per cent in 1979–97. The decline in employment and output seem to occur across all industries (except beverages for employment and textiles for output).


123

Table 8.1 Categorisation of Indian states Region

Name of state

Coastal or inland?

With or without metrop.?

East

Assam Orissa West Bengal

Inland Coastal Coastal

No No Yes, Kolkata

West

Maharashtra Gujarat Rajasthan

Coastal Coastal Inland

Yes, Mumbai Yes, Ahmedabad No

North

Punjab Haryana

Inland Inland

No Yes, Faridabad and Gurgaon (National Capital Region of Delhi)

Madhya Pradesh Uttar Pradesh

Inland Inland

No Yes, NOIDA (National Capital Region of Delhi)

Karnataka Kerala Tamil Nadu Andhra Pradesh

Coastal Coastal Coastal Coastal

Yes, Bangalore No Yes, Chennai Yes, Hyderabad

South

Table 8.2 List of NIC 2-digit industries, India Industry code

Name of industry

20 22 23 26 27 28 30 31 32 33 37

Food Products Beverages, Tobacco and Related Products Cotton Textiles Textile Products including Wearing Apparel Wood and Wood Products Paper and Paper Products Basic Chemicals and Chemical Products Rubber, Plastic, Petroleum and Coal Products Non-Metallic Mineral Products Basic Metal and Alloy Industries Transport Equipment and Parts

Moving now to the structure of employment and production in the Western states (Table 8.4), we also find that these states have also seen their aggregate share in employment and output decline over the period 1979–97. The share of Western states in manufacturing employment was 33.5 per cent in 1979. This fell to 25.4 per cent in 1997. The share of Western states in manufacturing output was 41.3 per cent in 1979. The share fell to 33.8 per cent in 1997. The difference between average employment growth in the Western states and average employment growth across all states was −0.9 per cent in 1979–97. At the same time, the difference between average output growth in the Western

14.7 1.9 6.3 9.3 26.1 22.2 10.8 18.7 13.3 35.2 27.1

14.9

20 22 23 26 27 28 30 31 32 33 37

Total

13.0

13.0 1.4 6.0 4.4 23.5 17.8 10.2 10.7 12.5 35.1 18.8 12.5

17.4 3.5 4.8 1.6 18.6 16.2 8.1 9.6 13.0 29.7 15.7

1997

14.7

17.5 4.6 4.9 7.3 21.1 23.4 7.2 23.0 15.2 29.2 19.4

1979

14.2

20.5 2.0 5.3 5.6 14.2 16.5 6.7 13.0 10.9 31.3 14.0

1985

12.4

12.5 2.2 2.9 2.3 16.2 14.3 7.4 13.1 7.7 39.7 10.3

1991

8.0

9.7 2.5 5.4 1.7 9.8 11.8 4.1 11.8 4.5 16.4 6.1

1997

Output – Gross Value Added (per cent)b

−3.3 −3.5 0.5 −8.0 −4.2 −3.8 −3.1 −3.7 −6.8 −3.2 −6.4 −4.1

1.0 3.4 −1.5 −9.8 −1.9 −1.8 −1.6 −3.7 −0.1 −0.9 −3.0 −1.8

Relative Employment Relative Output Growth Growth – Annual per cent – Annual per cent changec changed

Notes a Per cent of total employment in the industry across all states. b Per cent of total production in the industry across all states. c Relative Employment Growthij = Employment Growth of Industry i in Region j − Average Employment Growth of Industry i in all regions – from 1979 to 1997. d Relative Output Growthij = Output Growth of Industry i in Region j − Average Output Growth of Industry i in all regions – from 1979 to 1997.

Source: Author’s calculations, Annual Survey of Industries, Central Statistical Organisation.

14.5

15.3 1.9 7.5 7.0 20.6 19.4 12.2 10.5 12.1 34.6 20.0

1991

1979

Year

1985

Employment (per cent)a

Industries

Table 8.3 Employment and production structure, Eastern Indian states


125

states and the average output growth across all states was −1.0 per cent in 1979–97. The decline in employment and output was uneven across industries, with the largest contraction in textile products (including wearing apparel). Wood products, on the other hand, expanded rapidly, especially in employment. With respect to the Northern states, there has been an increase in both employment and output over the period 1979–97 (Table 8.5). The share of Northern states in manufacturing employment was 19.4 per cent in 1979. The share in employment increased to 21.1 per cent in 1997. The share of Northern states in manufacturing output was 18.6 per cent in 1979. The share increased to 26.4 per cent in 1997. The difference between average employment growth in the Northern states and average employment growth across all states was 0.7 per cent in 1979–97. The difference between average output growth in the Northern states and the average output growth across all states was 1.5 per cent in 1979–97. The increase in employment and output occurred in all industries, except food products, beverages and non-metallic mineral products for employment and food products and non-metallic mineral products for output. Finally, examining the structure of employment and production in Southern states, these states have expanded significantly in both employment and output over the period 1979–97 (Table 8.6). The share of Southern states in manufacturing employment was 32.3 per cent in 1979. The share in employment increased to 41.1 per cent in 1997. The share of Southern states in manufacturing output was 25.4 per cent in 1979. The share increased to 31.9 per cent in 1997. The difference between average employment growth in the Southern states and the average employment growth across all states was 1.1 per cent in 1979–97. The difference in average output growth in the Southern states and average output growth across all states was 0.8 per cent in 1979–97. The increase in employment and output was the most noticeable in textiles (NIC 23) and textile products (NIC 26). In textiles, employment and output increased by 3.4 and 4.2 per cent respectively over the period 1979–97. In textile products, employment and output increased by 3.7 and 3.5 per cent respectively over the period 1979–97. We find that there is no clear correlation between geographical and agglomeration factors and the pattern of industrialisation in India. For example, the share of coastal states in employment was 74.1 per cent in 1979 and 74.6 per cent in 1997 (Table 8.7). Their share in output was 77.1 per cent in 1979 and 73.8 per cent in 1997. Thus, coastal states do not seem to have benefited from the opening up of the economy, as would be expected if production decisions were being determined by location costs, which is the prediction of the ‘new economic geography’ theoretical perspective. Similarly, states with large industrial centres (metrop.) have not expanded their shares in total employment and production. The share of states with metropolises in total employment was 80.7 per cent in 1979. Their share in total employment was 79.6 per cent in 1997.

19.9 14.3 53.8 40.5 11.9 31.6 48.9 34.1 37.5 23.5 29.9

33.5

20 22 23 26 27 28 30 31 32 33 37

Total

26.3

20.7 14.8 38.8 20.3 23.9 21.4 34.2 39.0 26.3 23.1 23.9 25.4

19.4 15.9 33.6 11.9 25.8 23.2 36.5 37.6 32.7 21.5 25.2

1997

41.3

27.1 22.8 55.7 41.5 26.3 26.5 60.3 38.6 29.9 26.2 35.4

1979

38.7

24.3 22.5 48.0 36.6 23.7 32.7 49.6 55.9 30.2 20.8 45.7

1985

31.4

27.0 12.4 39.1 37.9 27.3 22.5 40.2 35.7 33.8 13.8 37.8

1991

33.8

27.9 18.0 23.3 19.9 32.2 29.2 57.1 45.3 22.1 19.3 37.5

1997


0.2 −1.3 −4.8 −4.1 1.1 0.5 −0.3 0.9 −1.7 −1.7 0.3 −1.0

−0.9

Relative Output Growth – Annual per cent changed

−0.2 0.6 −2.6 −6.8 4.3 −1.7 −1.6 0.5 −0.8 −0.5 −1.0

Relative Employment Growth – Annual per cent changec



29.2

19.7 19.3 42.3 29.3 21.6 28.0 36.9 43.1 28.4 21.8 27.5

1991

1979

Year

1985


Industries

Table 8.4 Employment and production structure, Western Indian states

26.8 11.7 15.9 14.5 8.6 19.6 12.2 13.0 22.8 26.1 19.9

19.4

20 22 23 26 27 28 30 31 32 33 37

Total

22.5

29.3 10.1 17.5 17.4 6.3 26.4 18.8 19.2 27.4 24.2 31.7 21.1

24.3 6.8 17.8 16.6 8.9 26.1 16.8 19.7 20.8 31.0 34.3

1997

18.6

25.2 23.4 15.4 22.0 5.1 19.6 12.5 12.0 22.1 29.8 16.9

1979

20.6

25.4 25.6 14.4 23.0 7.5 19.8 17.6 10.8 27.1 37.3 13.0

1985

25.0

26.8 37.5 13.3 21.4 4.8 24.6 28.8 12.1 30.6 29.6 28.0

1991

26.4

24.8 33.6 20.0 23.0 10.4 25.8 22.6 14.3 10.3 51.9 37.8

1997


0.7

−0.5 −3.0 0.6 0.8 0.2 1.6 1.8 2.3 −0.5 0.9 3.0


1.5

−0.1 2.0 1.5 0.2 4.0 1.5 3.3 1.0 −4.3 3.1 4.5




22.4

29.5 11.0 18.6 16.0 7.3 23.5 16.9 14.5 31.6 28.1 24.4

1991

1979

Year

1985


Industries

Table 8.5 Employment and production structure, Northern Indian states

38.6 72.1 24.0 35.8 53.4 26.5 28.1 34.2 26.4 15.2 23.1

32.3

20 22 23 26 27 28 30 31 32 33 37

Total

38.2

37.1 73.7 37.7 57.9 46.3 34.4 36.8 31.1 33.8 17.7 25.6 41.1

38.9 73.9 43.9 69.9 46.7 34.5 38.6 33.1 33.5 17.8 24.8

1997

25.4

30.2 49.2 24.0 29.2 47.5 30.6 20.0 26.4 32.8 14.9 28.3

1979

26.5

29.9 49.8 32.4 34.7 54.5 31.0 26.1 20.3 31.8 10.6 27.4

1985

31.2

33.7 47.9 44.7 38.4 51.7 38.6 23.6 39.1 27.9 16.9 24.0

1991

31.9

37.6 45.9 51.3 55.3 47.6 33.3 16.2 28.6 63.1 12.4 18.6

1997


1.2 −0.4 4.2 3.5 0.0 0.5 −1.2 0.4 3.6 −1.0 −2.3 0.8

1.1


0.0 0.1 3.4 3.7 −0.7 1.5 1.8 −0.2 1.3 0.9 0.4




33.9

35.5 67.9 31.6 47.7 50.6 29.1 34.0 31.8 27.9 15.5 28.1

1991

1979

Year

1985


Industries

Table 8.6 Employment and production structure, Southern Indian states


129

The share of states with metropolises in total output was 81.2 per cent in 1979. This share remained virtually constant at 82.6 per cent in 1997. To what extent does the aggregation of individual states in regional groupings or by geographical characteristics mask the variation in employment and output trends within each of the groupings that we have used? We examine the structure of employment and production for the fourteen major Indian states in Table 8.8. We find considerable heterogeneity in the patterns of employment and production, with no clear pattern emerging from the regional data. For example, among Southern states, it is clearly Andhra Pradesh and Tamil Nadu that have witnessed the highest growth in manufacturing employment and output, with Karnataka and Kerala benefiting much less. Among the Northern states, Haryana and Punjab have seen the highest growth in employment and production, with Rajasthan not showing similar dynamism. Manufacturing employment in Uttar Pradesh, on the other hand, seems to have contracted. Among the Eastern and Western states which have seen contractions in employment and production overall, the states with the largest declines in absolute terms are Maharashtra and West Bengal. In particular, Maharashtra has seen a decline in employment share from 18.1 per cent in 1979 to 14.9 per cent in 1997, and a fall in output share from 27.3 per cent in 1979 to 23.8 per cent in 1997. Further, there is no correlation between changes in employment and production and whether the state is coastal/inland or with/without a metropolis. Coastal states like Andhra Pradesh have expanded and inland states like Haryana have also done so. Coastal states like Maharashtra and West Bengal have contracted, and inland states like Assam have also done. States with large metropolises like Karnataka have remained stagnant in industrialisation while states without similar metropolises like Punjab have seen expanding industrial sectors. Neither is it clear that states with existing industrial capacities have built on these capacities to expand over time – two of the most industrialised states in 1979 were Maharashtra and West Bengal, and we have observed their relative decline over time. Thus, the overall finding seems to be that changes in manufacturing employment and production over time in the major Indian states may have less to do with factors that can be ascribed to the new economic geography, whether directly or indirectly influenced by international trade.

4 The role of trade policy in explaining regional patterns of industrialisation While theoretical perspectives drawn from the new economic geography may not able to explain the evolving regional patterns of industrialisation in India, it is possible that trade reforms may have played a role by influencing the patterns of comparative advantage, such that Indian states with exportoriented industries prospered and states with import-competing industries remained stagnant or declined. To see if this is true, we present the structure

74.1 25.9 80.7 19.3

71.5 28.5 79.0 21.0

74.6 25.4 79.6 20.4

77.1 22.9 81.2 18.8

74.8 25.2 80.8 19.2

1985 70.0 30.0 77.3 22.7

1991 73.8 26.2 82.6 17.4

1997 0.0 −0.1 −0.08 0.31

Relative Employment Growth – Annual per cent changea

−0.2 0.7 0.10 −0.46

Relative Output Growth – Annual per cent changeb

Notes a Relative Employment Growthj = Employment Growth of Region j − Average Employment Growth – all regions – from 1979 to 1997. b Relative Output Growthj = Output Growth of Region j − Average Output Growth – all regions – from 1979 to 1997. Region is Coastal/inland or States with metropolises/states without metropolises.

71.5 28.5 79.6 20.4

1997

1979

1991

1979

1985

Output – Gross Value Added (per cent)

Employment (per cent)

Source: Author’s calculations.

Coastal Inland States with metrops States without metrops

Classification of states

Table 8.7 Employment and production structure, coastal vs inland states and states with and without metropolises, India

3.4 1.6 10.3 5.2 5.6 11.6 11.1 18.1 10.6 2.3 3.0 1.9 3.4 11.9

1979

3.1 2.1 7.9 5.2 5.2 13.7 13.3 15.2 8.8 2.3 4.1 3.0 4.2 11.9

1991 2.8 2.0 8.7 5.5 5.8 14.9 13.5 14.9 9.3 2.2 4.2 2.9 3.5 9.8

1997 3.0 3.0 9.2 5.4 3.6 11.6 4.9 27.3 12.1 2.6 3.6 2.5 3.1 8.1

4.5 2.0 8.3 5.1 3.4 11.9 5.7 26.7 11.7 2.4 3.9 3.1 3.1 8.3

3.1 3.2 5.9 5.6 4.2 13.5 6.9 20.8 10.0 3.0 4.5 3.4 4.7 11.3

2.0 2.0 4.1 5.6 3.3 11.7 10.8 23.8 12.6 2.8 4.1 3.7 3.2 10.4

−1.1 1.0 −1.0 0.3 0.2 1.4 1.1 −1.1 −0.7 −0.3 1.9 2.3 0.2 −1.1

Output – Gross Value Added (per cent) Relative Employment Growth – Annual per cent changea 1979 1985 1991 1997

−2.2 −2.3 −4.5 0.2 −0.5 0.0 4.3 −0.7 0.2 0.4 0.8 2.2 0.2 1.4

Relative Output Growth – Annual per cent changeb

Notes a Per cent of total ea. Relative Employment Growthj = Employment Growth of State j − Average Employment Growth – all states – from 1979 to 1997. b Relative Output Growthj = Output Growth of State j − Average Output Growth – all states – from 1979 to 1997.

3.4 2.0 9.5 5.1 4.8 13.3 10.6 16.4 9.9 2.5 4.0 3.0 3.8 11.8

1985

Employment (per cent)

Source: Author’s calculations.

Assam Orissa West Bengal Karnataka Kerala Tamil Nadu Andhra Pradesh Maharashtra Gujarat Rajasthan Punjab Haryana MP UP

States

Table 8.8 Employment and production structure, all fourteen major states of India

132


of exports and imports in the NIC 2-digit classification (by re-classifying the ISIC 3-digit trade data to the NIC 2-digit classification) in Tables 8.9 and 8.10 respectively. The largest increase in the share of exports has been in textile products/wearing apparel (NIC 26) and in basic chemicals (NIC 30). The largest decline in the share of imports has been in food products (NIC 2) and rubber and petroleum products (NIC 31). An export-oriented industry such as wearing apparel production was mostly concentrated in the Southern states in 1979 (36 per cent), and these states have seen a further increase in concentration of this industry, with the share going up to 70 per cent in 1997. On the other hand, the Southern states had a large share of the importcompeting food products industry in 1979 (39 per cent) and this share has remained more or less the same over the period 1979–97. In contrast, a large share of basic chemicals – another export-oriented industry – was in the Western states in 1979 (49 per cent), and the importance of the chemicals industry in these states has actually declined over the period 1979–97, with the share in 1997 falling to 37 per cent. Thus, there is not enough strong evidence that suggests that states with a larger share of the export-oriented industries have benefited over time, and states with a larger share of the import-competing industries have seen declines in their industrial status over time. We seem to see that trade policy has not played a large role in influencing regional patterns of industrialisation, either through conventional H-O mechanisms or through agglomeration/deconcentration effects. It is possible that the institutional and business environment and the role of the state government in attracting private investment may have played a more important part in determining the evolution of the spatial structure of production and employment in Indian manufacturing.2 Table 8.9 Structure of exports, NIC 2-digit, India NIC 1987 2-digit code

1976–80

1981–5

1986–90

1991–5

1996–9

20 22 23 26 27 28 30 31 32 33 37 Others

18.4 0.2 21.7 10.1 0.5 0.3 4.7 1.5 1.5 13.7 4.6 22.7

16.5 0.5 19.5 13.7 0.3 0.4 7.0 5.3 0.8 5.4 4.2 26.4

9.6 0.3 16.1 18.0 0.2 0.3 9.2 8.0 0.6 4.6 3.1 30.1

10.3 0.2 16.5 19.5 0.2 0.4 10.6 4.6 1.4 7.4 4.1 24.6

8.9 0.1 19.0 18.0 0.8 0.4 10.3 4.3 1.7 8.7 3.3 24.6

Source: International Economic Data-bank, Australian National University; author’s calculations.


133

Table 8.10 Structure of imports, NIC 2-digit, India NIC 1987 2-digit code

1976–80

1981–5

1986–90

1991–5

1996–9

20 22 23 26 27 28 30 31 32 33 37 Others

12.9 0.1 1.2 0.0 0.2 3.1 19.0 14.0 1.4 18.9 7.3 21.9

9.7 0.1 1.3 0.0 0.2 3.1 17.9 15.7 1.1 20.1 6.6 24.1

5.8 0.1 1.7 0.1 0.4 3.4 18.5 14.2 0.8 19.8 6.1 29.3

3.6 0.0 2.2 0.1 0.3 2.9 22.4 19.3 0.8 15.2 6.2 26.9

6.5 0.3 2.4 0.3 0.6 2.9 15.7 7.4 1.0 12.5 9.6 40.8

Source: International Economic Data-bank, Australian National University; author’s calculations.

5 Concluding remarks This chapter has examined the role of trade policy in influencing regional patterns of industrialisation in India. We have found that states in the Southern and Northern regions of the country have seen an expansion in manufacturing employment and production while states in the Eastern and Western regions of the country have seen a decline in manufacturing employment and production over the period 1979–97. There does not appear to be a clear correlation between geographical and agglomeration factors and spatial economic development – coastal states and states with large cities have not observed either a significant increase or decrease in their shares of industrial employment and production over the period under consideration. When examining the patterns of industrialisation by individual states, we see a marked heterogeneity in industrial performance. States like Andhra Pradesh and Tamil Nadu have seen significant increases in manufacturing employment and production. In contrast, states like Maharashtra and West Bengal have seen a decline in industrial performance over time. We also do not seem to see evidence of the manufacturing sector expanding in states with exportoriented industries and declining in states with import-competing industries. We conclude that the institutional and business environment rather than trade policy per se may have played a crucial role in determining which states have done better in manufacturing performance relative to other states.

9

Conclusion

The impact of trade policy on a country’s economic development has remained a topic of considerable interest both in the academic and policy domains. In this book, we examined the efficiency and equity outcomes with respect to the trade policy changes that have occurred in the Indian economy since the late 1970s, with particular reference to the Indian manufacturing sector. With respect to the efficiency dimension of trade policy, we studied the impacts of trade reforms on economic growth, and on productivity and prices. With respect to the equity dimension of trade policy, we studied the impacts of trade on employment, wage inequality and regional inequality. This chapter summarises the key findings and draws some overall lessons to inform policy in the future.

1 Key findings Trade policy in India By the mid-1970s, India’s trade regime had evolved into one of the most complex in the world. The import and exchange rate regime that Indian policy-makers followed since independence was aimed at the comprehensive, direct control over foreign exchange utilisation, with an overwhelming reliance on quotas rather than tariffs. From the late 1970s onwards, there were incremental attempts to reform the trade regime, with more significant attempts to do so, first in 1985 and then more comprehensively in 1991. Most of the reforms occurred in the intermediate and capital goods sectors, and, in 1991, almost all quotas on capital and intermediate goods were removed, and there was a reduction in peak tariff rates as well. There was less progress with trade reforms in the consumer goods sector. An important aspect of trade policy in India was that the level of protection till the 1980s (and the pace of reforms subsequently) differed significantly within sectors – whether these were the capital, intermediate or consumer goods sectors.

Conclusion

135

Growth and structural change in Indian manufacturing After a period of stagnation in the 1970s, Indian manufacturing showed consistent growth in all indicators from the early 1980s to the mid-1990s. There was strong growth in labour productivity and real value added since the early 1980s. Employment too increased substantially, especially from the mid-1980s. Real wages also increased in this period. However, unit labour costs fell as labour productivity growth overtook real wage growth. Perhaps as a consequence, manufacturing exports also increased in the same period. The period of growth in Indian manufacturing was not sustained till the end of the 1990s, with a slow-down in employment growth in the late 1990s and a fall in real wages. Surprisingly, the labour-intensive textile, clothing and footwear industries gradually contracted from the mid-1970s to the late 1990s in production and employment, though not in manufacturing exports. This may suggest that some of the production (and perhaps a large proportion of employment) in these industries may have shifted to the unorganised/informal sector. The shift away from the key labour-intensive industries suggests a weakness in the growth patterns of the organised manufacturing sector in the period of the study. At the same time, the strong growth in labour productivity is one of the positive features of manufacturing performance in the same period. Trade policy and economic growth The Indian economy has been growing at a brisk pace since the late 1970s. An important contributing factor to the growth acceleration that occurred in the Indian economy in the 1980s and 1990s is trade reforms. The high growth rates of the 1980s and 1990s can be mostly attributed to the sharp increase in private equipment investment, which has a significantly more growth-enhancing effect than public equipment and structures investment. The increase in private equipment investment was due to the fall in the relative price of equipment that occurred as a result of the trade policy changes in 1985 and 1991. Interestingly, for the purpose of accelerating growth in the Indian economy, it was the more conservative trade policy reforms of 1985 that seemed to have a stronger impact than the more radical reforms of 1991. This may be due to the nature of trade policy changes, and the manner these changes impacted on total factor productivity. We discuss this next. Trade policy, productivity and prices Trade reforms in India have had a strong positive impact on total factor productivity. The trade reforms may have led to a decrease in X inefficiency in the manufacturing sector, decreasing slack in the workplace as workers and managers both increased their efforts following the pro-competitive effects of the reforms. More importantly, the access to specialised capital and

136

Conclusion

intermediate goods from the rest of the world as the quotas on the importation of these goods were gradually removed led to a productivity boost to Indian manufacturing firms. Such mechanisms by which trade policy impacts on productivity have been extensively discussed in the endogenous growth literature, and our evidence suggests that these mechanisms were responsible for the way trade policy impacted on performance in Indian manufacturing. The large shift from quotas to tariffs in the case of intermediate and capital goods came with the 1985/86 trade policy changes implemented by the Rajiv Gandhi government. The 1991 reforms could be seen to have been the culmination of this policy shift. Our evidence suggests that the 1985 reforms seemed to have a stronger effect on economic growth, both directly, by boosting productivity growth of Indian manufacturing firms, and indirectly, by increasing investment in machinery and equipment by the private sector, as real equipment prices fell (both due to the productivity increases and greater external competition faced by Indian capital goods producers). We also found that an increase in the quantity of competition from abroad had a strong negative impact on domestic market power, and therefore, on domestic prices. However, the exchange rate depreciation of the mid-1980s to 1990s may have had a countervailing effect on domestic prices by providing more protection for import-competing sectors. Thus, the net effect of price competition from abroad was such that it brought about an increase in domestic prices and in domestic market power, in spite of a reduction in tariffs since 1991. International trade and employment We examined the effect of international trade on manufacturing employment in India, comparing the outcomes with four developing countries, Kenya and South Africa in Africa and Bangladesh and Vietnam in Asia. We found that the share of unskilled labour-intensive goods in India’s export basket has increased over time, but not at the same rate of growth as has been observed in Bangladesh and Vietnam. We also found that the employment coefficients of exports and imports in India have consistently fallen over the period, and that the difference between the employment coefficient of exports and that of imports has narrowed over time. Thus, the employment impacts of trade for a given change in output may have been actually less in the post-reform period than in the pre-reform period. In fact, most of the employment increase that has occurred over the period 1975–99 can be attributed to increases in domestic demand and less to international trade. Neither was there evidence of an indirect effect of trade on employment, as the increase in labour productivity that we have observed occurred in the Indian manufacturing sector from the 1980s onwards and cannot be directly attributed to international trade. The finding that the impact of international trade on manufacturing employment in India has been limited and mirrors the two African countries

Conclusion

137

for which we can undertake comparisons rather than the two Asian countries seems to be in contrast to the prediction of Heckscher–Ohlin theory. India’s combination of resources would have made the country particularly competitive in labour-intensive manufacturing – in comparison with other regions, it has a low level of education and limited natural resources relative to its supply of labour (Wood and Calandrino 2000, Mayer and Wood 2001). Our findings suggest that a Ricardian perspective, which stresses technological and policy factors, may provide more explanatory power, as these have proved important in India’s inability to compete successfully in labourintensive manufacturing goods, where the country seems to have an innate natural advantage. We will return to this point later in the chapter. Trade policy and wage inequality After a period when the relative wages of skilled (non-production) workers had fallen relative to those of unskilled (non-production) workers in the early 1970s, there was a period of slow but a steady increase in wage inequality in the Indian manufacturing sector in the 1990s. The increase in wage inequality seems to have gone hand in hand with an increase in the relative employment of skilled workers across all industries since the mid-1980s. Trade policy changes can explain both the increase in wage inequality and the increase in the relative employment of skilled workers since the 1980s in the Indian manufacturing sector. The largest decreases in protection were in the unskilled labour-intensive sectors, which led to a decrease in the relative demand for unskilled workers and depressed their relative wages. At the same time, trade reforms led to skilled-biased technological change, as industries across the board attempted to fight the heightened external competition with greater recourse to skilled labour. As a consequence, the relative demand for skilled labour also increased, along with an increase in the skill premium. Therefore, trade reforms seem to have led to a widening of the wage gap between skilled and unskilled workers, and an increase in relative skill intensity in Indian manufacturing. Trade policy and regional inequality Regional disparity in manufacturing performance has increased since the late 1970s. States in the Southern and Northern regions of the country have seen an expansion in manufacturing employment and production, while states in the Eastern and Western regions of the country have seen a decline in manufacturing employment and production. However, trade policy does not appear to have played a significant role in influencing regional patterns in industrialisation. There does not appear to be a clear correlation between geographical factors that may be influenced by international trade and spatial economic development – coastal states and states with large cities have not seen either a significant increase or decrease in their shares of industrial

138

Conclusion

employment and production over the period under consideration. Neither is there evidence of the manufacturing sector expanding in states with export-oriented industries and declining in states with import-competing industries. While further research is necessary to understand the regional patterns of economic development, it seems to be the case that the institutional and business environment specific to the state rather than trade policy per se may have played a crucial role in determining state-level manufacturing performance.

2 Policy implications Our study has three main policy implications. First, our research shows that international trade may have played an important contributing role in explaining the phenomenon of ‘jobless growth’ in Indian manufacturing observed in the 1980s and 1990s (Bhalotra 1998, Nayyar 2006). Trade reforms have unambiguously increased economic growth since the early 1980s, but have not had similar positive effects on employment growth. There are two possible reasons why this has been the case. One, in contrast to the other labour-surplus economies in South Asia, India has built up ‘dynamic comparative advantage’ in technology-intensive manufacturing commodities such as drugs and pharmaceuticals – this is reflected in the factor content of India’s manufacturing exports discussed in Chapter 6. The capabilities that India has acquired in the technology-intensive sectors have been due in part to the large public investment in science and technology institutions of higher learning, and in part to the ‘infant industry’ protection that was provided to these industries at the early stages of development (Lall 1998, 2001). Thus, the pattern of trade in India since the 1980s has been more skill- and technology-intensive than may be expected from countries with similar levels of economic development and factor endowments. Second, in spite of favourable unskilled labour to land ratios, India has not been able to export labour-intensive commodities to the same extent as China because of strong policy impediments still in place, which has constrained the growth of the labour-intensive manufacturing sector (Panagariya 2007a). Chief among these has been the presence of restrictive labour laws that create a strong disincentive for firms in the organised sector to expand in the face of a more volatile macroeconomic environment since the economic reforms of 1991, where flexibility and speedy adjustment to shocks in the product and factor markets are the keys to firms’ survival strategies (Panagariya 2007b).1 Reform of labour laws is seen to be controversial and difficult to implement in the era of coalition governments that have characterised the Indian political system in the 1990s and beyond. Small-scale sector reservation policies, which did not allow entry of large firms and foreign firms into several industries, many of which were in the labour-intensive sector, were also important in explaining the lack of dynamism of the labour-intensive manufacturing sector in India in the 1990s (Mohan 2003). These policies have now been considerably

Conclusion

139

weakened, and foreign firms allowed to invest in the industries previously reserved for the small-scale sector. Both labour law reform and small-scale sector reservation policies belong to what the noted political scientist Ashutosh Varshney has termed ‘mass politics reforms’ – which are reforms that may be considered anti-populist and are therefore difficult to implement in India’s current political context. The slow pace of change in labour laws in particular can explain to a large extent India’s weak performance in labourintensive manufacturing exports, and, consequently, the low rate of tradeinduced job creation in the Indian manufacturing sector. This also implies that the role of international trade in reducing poverty in India may be less than its comparators (such as other economies in South and East Asia), as long as current policy impediments in the labour-intensive manufacturing sector remain. On the positive side, if these impediments are removed, then trade can be a powerful engine of poverty reduction, by bringing about labour-absorbing economic growth. Such a process of economic growth would transfer surplus labour from unproductive and poor-paying agricultural activities to more remunerative jobs in the industrial sector, and provide a more viable route out of poverty in rural India than radical land reform, as proposed by political scientists such as Kohli (1989). The second policy implication is to do with the widening disparities between regions and between workers of different skills that we have observed to have occurred in the manufacturing sector since the 1980s. With respect to the increase in regional disparities in manufacturing production and employment, it is clear that simplistic accounts of why such disparities have occurred are not valid in the Indian context. Unlike what we have observed in China, coastal regions have not necessarily benefited since the reforms, nor have regions with initial capabilities in manufacturing. What has perhaps been more significant in the evolution of regional disparities has been the ability of some state governments to provide a conducive environment for the manufacturing sector to expand – for example, this is true of the governments of the states of Andhra Pradesh and Tamil Nadu. Thus, the relationship of the governments of these states to the business sector has been more critical in explaining manufacturing performance in the states in question, rather than innate advantages resulting from geography or history. This augurs well for states like Orissa and Madhya Pradesh, with the former not having a history of industrial experience and the latter not having access to a coastline and to ports. The clear implication here is that policy-makers in these states can expect to bring about a stronger performance in manufacturing if they provide a more conducive environment for the business sector to expand, with greater infrastructural development and a credible investment climate that will create incentives for industrial firms to locate in these states. On the widening disparity in wages between unskilled and skilled workers, clearly the solution to this problem lies in the increased supply of skilled workers to India’s manufacturing industries, so that the skill premium is driven down. Given the state’s fiscal constraint in expanding places in

140

Conclusion

publicly funded universities and institutions, this can come about with a more liberal attitude towards the involvement of the private sector in the provision of vocational skills and in higher education. The final policy implication is to do with India’s future economic growth and the role of the manufacturing sector in this projected growth scenario. One surprising feature of India’s recent growth experience has been that the manufacturing sector has played less of a role in India’s economic growth as compared to the services sector. This is clear from the statistics that we have surveyed in Chapters 2 and 4 – the service sector has contributed more to GDP growth and exports in the 1990s relative to the manufacturing sector. It has been conjectured that India may leapfrog from an agriculture-based economy to a service-based economy, and that India’s future economic growth will be fuelled by service sector growth. While this may be true, there are, however, reasons to be cautious about such an optimistic scenario. First, the component of the service sector that is growth oriented is the exportoriented information technology sector, which is significantly skilled labour intensive. Given the large amounts of surplus labour that exists in India’s rural areas, clearly the prospects for inclusive growth rest very much on the growth of the labour-intensive manufacturing sector. Second, the service sector, like the manufacturing sector, comprises two distinct parts – the organised and unorganised sectors, and about two-thirds of service sector output is in the unorganised sector – which comprises small enterprises unlikely to be affected by trade reforms, and suffering from low productivity similar to their manufacturing sector counterparts (Bardhan 2007). If the technology-enabled service sector is to be the engine of India’s future economic growth, the pattern of such growth may not be different from the ‘jobless growth’ phenomenon that has been observed in Indian manufacturing. Therefore, it seems unavoidable that the unskilled labour-intensive component of the manufacturing sector may still provide the best route out of poverty for India’s rural masses. For the skilled labour- and technologyintensive component of Indian manufacturing, past government policies and private investments in building technological capabilities in these industries have made India a major exporter among developing countries of the commodities produced by these industries, and this will continue to be the case in the future. The preferred strategy for Indian policy-makers should be to encourage the growth of India’s manufacturing sector, both in its skilled and unskilled intensive components, and that leapfrogging into a service sector economy may not be in India’s interests, from the viewpoints of either efficiency or equity.

Appendix: data sources, list of industries and matching of industry codes

Data sources An important feature of this study (and the underlying strength of the empirical analysis) is that we have used the same data for several of the core empirical chapters. This is the detailed industry level, compiled at the disaggregated NIC 3-digit level from the Annual Survey of Industries (ASI) by the Central Statistical Organisation (CSO), Government of India. The NIC 3-digit level corresponds to the International Standard Industrial Classification (ISIC) 4-digit industry level, and is the maximum level of disaggregation that is available for the period of the study. Our largest sample will be a set of 139 industries over the period 1973/74 to 1997/98. A full list of the industry codes along with a description of the industries is provided in Table A1. For different measures of trade policy, we will work with smaller samples, but rarely less than 50 industries. This implies that the data we use has considerable cross-sectional and time-series variation so we are able to quantitatively assess the impact of trade policy both across industries and for the same industries, over time. We do not use the NIC 3-digit (ISIC 4-digit) level data that we describe above in three cases. First, when we need data for the entire manufacturing sector to assess the patterns and trends in Indian manufacturing, as we do in Chapter 3, and to examine the effect of international trade on total manufacturing employment, as we do in Chapter 6, we use ISIC 3-digit industry data obtained the United Nations Industrial Development Organisation (UNIDO) and ISIC 3-digit matched trade data from the International Economic Databank (IEDB) of the Australian National University, for the period 1975–99. It should be noted that the UNIDO industry data is itself obtained from the Annual Survey of Industries, CSO, India. However, the advantage for us in using UNIDO data is that the matching and reclassification of the NIC 3-digit data to ISIC 3-digit has already been undertaken by the UNIDO. The list of ISIC 3-digit industries is provided in Table A2. The NIC 3-digit data-set is incomplete in that while it includes most industries, some industries have to be excluded due to missing observations for some years. This is not the case with ISIC 3-digit data of IEDB. The second case

142

Appendix

when we deviate from the NIC 3-digit data in the empirical analysis is when we examine the impact of trade policy on regional inequality in Chapter 8. In this case, we use state-level NIC 2-digit industry data, as state-level NIC 3-digit data is not easily available for the period of analysis. In Chapter 4, where we examine the relationship between trade policy and economic growth, and estimate growth regressions which are at the national or sectoral levels, we use the National Accounts Statistics (NAS) of the CSO for data on the dependent and explanatory variables. At several points in the empirical analysis, we have to use trade data which are available in the SITC classification matched to the ISIC or NIC classification. In Chapters 3 and 6, the IEDB has already provided us with the matched data at the ISIC 3-digit level. For other chapters, we do the matching ourselves – with the details provided in Table A3. We do not use any other data source apart from the CSO, UNIDO and the IEDB throughout this book.

Table A1 Complete list of industries used for Chapters 5 and 7 Industry code

Description of industries

200 201 202 203 204 205 206 207 209

Slaughtering, preparation and preservation of meat Manufacture of dairy products Canning and preservation of fruits and vegetables Processing, canning and preservation of fish, crustacean and similar foods Grain milling Manufacture of bakery products Manufacture and refining of sugar (vacuum pan sugar factories) Production of indigenous sugar, boora, khandsari, gur, etc. Manufacture of cocoa products and sugar confectionery (incl. sweetmeats) Manufacture of hydrogenated oils and vanaspati, ghee, etc. Manufacture of vegetable oils and fats/other than hydrogenated Processing and blending of tea including manufacture of instant tea Coffee curing, roasting, grinding and blending, etc. Processing of edible nuts Manufacture of ice Manufacture of prepared animal and bird feed Manufacture of starch Distilling, rectifying and blending of spirits, ethyl alcohol production Manufacture of wine Manufacture of malt liquors and malt Manufacture of soft drinks and syrups

210 211 213 214 215 216 217 218 220 221 222 224

Appendix 225 226 227 228 230 231 232 233 234 235 236 240 241 242 243 244 245 246 248 250 254 255 260 261 262 263 264 265 266 267 268 270 271 272 273 274 275 276

143

Tobacco stemming, redrying and all other operations Manufacture of bidi Manufacture of cigars, cigarette, cheroots and cigarette tobacco Manufacture of snuff, zarda, chewing tobacco and other tobacco products n.e.c. [not elsewhere classified] Cotton ginning, cleaning and bailing Cotton spinning other than in mills (charkha) Weaving and finishing of cotton khadi Weaving and finishing of cotton textiles on handlooms Weaving and finishing of cotton textiles on powerloom Cotton spinning, weaving and processing in mills Bleaching, dyeing and printing of cotton textiles Preparation of raw wool, silk and artificial/synthetic textile fibres for spinning Wool spinning, weaving and finishing other than in mills Wool spinning, weaving and processing in mills Bleaching and dyeing of woollen textiles Spinning, weaving and finishing of silk textiles other than in mills Spinning, weaving and processing of silk textiles in mills Bleaching, dyeing and printing of silk textiles Bleaching, dyeing and printing of artificial/synthetic textile fabrics Jute and mesta pressing and baling Spinning, weaving and finishing of jute and mesta textiles Spinning, weaving and finishing of coir textiles Manufacture of knitted or crocheted textile products Manufacture of all types of threads, cordage, ropes, twines and nets, etc. Embroidery work, zari work and making ornamental trimmings Making of blankets, shawls, carpets, rugs and other similar textile products Manufacture of floor covering of jute, mesta, sannhemp, etc. Manufacture of all types of textile garments and clothing accessories n.e.c. [not elsewhere classified] Manufacture of rain coats, hats, caps and school bags, etc. Manufacture of made up textile articles; except apparel Manufacture of waterproof textiles fabrics Sawing and planing of wood (other than plywood) Manufacture of veneer sheets, plywood and their products Manufacture of structural wooden goods Manufacture of wooden and cane boxes, crates, drums, barrels, etc. Manufacture of wooden industrial goods n.e.c. [not elsewhere classified] Manufacture of cork and cork products Manufacture of wooden furniture and fixtures (Continued Overleaf )

144

Appendix

Table A1 Continued Industry code


279

Manufacture of products of wood, bamboo cane, reed and grass n.e.c. [not elsewhere classified] Manufacture of pulp, paper and paper board incl. manufacture of newsprint Manufacture of containers and boxes of paper and paper board Manufacture of paper and paper board articles and pulp products n.e.c. [not elsewhere classified] Printing and publishing newspapers Printing and publishing of periodicals, books, journals, directories, etc. Printing of bank notes, currency notes, stamps, security passes, etc. Engraving, etching and block making, etc. Book binding on account of others Tanning, curing, finishing, embossing and japanning of leather Manufacture of footwear except vulcanised or moulded rubber or plastic Manufacture of wearing apparel of leather and substitutes of leather Manufacture of consumer goods of leather and substitutes of leather Manufacture of industrial organic and inorganic chemicals Manufacture of fertilisers and pesticides Manufacture of plastics in primary forms; and, synthetic rubber Manufacture of paints, varnishes and related products; artists’ colours and inks Manufacture of drugs, medicines and allied products Manufacture of perfumes, cosmetics, lotions, hair dressings, etc. Manufacture of matches Manufacture of explosives, ammunition and fireworks Manufacture of chemical products n.e.c. [not elsewhere classified] Tyre and tube industries Manufacture of footwear made primarily of vulcanised or moulded rubber and plastics Manufacture of rubber products n.e.c. [not elsewhere classified] Manufacture of plastic products n.e.c. [not elsewhere classified] Manufacture of refractory products and structural clay products Manufacture of glass and glass products Manufacture of earthen and plaster products Manufacture of non-structural ceramic-ware Manufacture of cement, lime and plaster Manufacture of mica products Stone dressing and crushing; manufacture of structural stone goods and stoneware

280 281 282 284 285 286 287 288 290 291 292 293 300 301 302 303 304 305 307 308 309 310 311 312 313 320 321 322 323 324 325 326

Appendix 327 329 330 331 332 333 334 335 336 340 341 342 343 345 346 349 350 351 352 353 355 356 357 358 359 360 361 362 364 366 367 368 369 370 371

145

Manufacture of asbestos cement and other cement products Manufacture of miscellaneous non-metallic mineral products n.e.c. [not elsewhere classified] Manufacture of iron and steel in primary/self-finished forms Manufacture of semi-finished iron and steel products Manufacture of ferro alloys Copper manufacturing Brass manufacturing Aluminium manufacturing Zinc manufacturing Manufacture of fabricated structural metal products Manufacture of fabricated metal products n.e.c. [not elsewhere classified] Manufacture of furniture and fixtures primarily of metal Manufacture of hand tools, weights and measures and general hardware Treatment or coating of metals such as plating, polishing, etc. Manufacture of metal cutlery, utensils and kitchenware Manufacture of metal products (except machinery and equipment) n.e.c. [not elsewhere classified] Manufacture of agricultural machinery and equipment and parts thereof Manufacture of machinery / equipment used by construction and mining industries Manufacture of prime movers, boilers, steam generating plants and nuclear reactors Manufacture of industrial machinery for food and textile industry Manufacture of refrigerators, air conditioners and fire-fighting equipment Manufacture of general purpose non-electrical machinery/equipment Manufacture of machine tools, their parts and accessories Manufacture of office computing and accounting machinery and parts Manufacture of special-purpose machinery/equipment Manufacture of electrical industrial machinery, apparatus and parts thereof Manufacture of insulated wires and cables, incl. manufacture of optical fibre cables Manufacture of accumulators, primary cells, and primary batteries Manufacture of electric fans and electric/electro-thermic domestic appliances Manufacture of television receivers, apparatus, etc. Manufacture of computers and computer based systems Manufacture of electronic valves and tubes and other electronic components n.e.c. [not elsewhere classified] Manufacture of radiographic X-ray apparatus, X-ray tubes and parts, etc. Ship and boat building Manufacture of locomotive and parts (Continued Overleaf )

146

Appendix

Table A1 Continued Industry code


372 374 375 376 378 379

Manufacture of railway/tramway wagons and coaches Manufacture of motor cars and other motor vehicles Manufacture of motor cycles and scooter and parts (incl. three wheels) Manufacture of bicycles, cycle-rickshaws and parts thereof Manufacture of bullock-carts, push-carts and hand-carts, etc. Manufacture of transport equipment and parts n.e.c. [not elsewhere classified] Manufacture of medical, surgical, scientific and measuring equipment Manufacture of photographic, cinematographic and optical goods and equipment Manufacture of watches and clocks Manufacture of jewellery and related articles Manufacture of sports and athletic goods Manufacture of musical instruments (excl. toys)

380 381 382 383 385 386

Note: Industry Code as in NIC 1987 Classification.

Table A2 Complete list of industries used for Chapters 3 and 6 ISIC 3-DIGIT CODE 311-FOOD MANUFACTURING 312-OTHER FOOD MANUFACTURING 313-BEVERAGE INDUSTRIES 314-TOBACCO MANUFACTURES 321-MANUFACTURE OF TEXTILES 322-MANUF WEARG APP EXCL FTWR 323-MANUF PROD LTHR EXCL FTWR, APP 324-MANUF FTWR EXCL RUBBR, PLSTC 331-MANUF WOOD CORK PRD EXCL FURN 332-MANUF FURN, FIXT EXCL PRIM MTL 341-MANUF PAPER + PRODUCTS 342-PRNTNG, PUBLNG, ALLD IND 351-MANUF OF INDUST CHEMICALS 352-MANUF OTH CHEMICAL PRODS 353-PETROLEUM REFINERIES 354-MISC PRODS OF PETR, COAL 355-RUBBER PRODUCTS

Appendix

147

356-PLASTIC PRODUCTS NEC 361-POTTERY,CHINA,EARTHWARE 362-GLASS + PRODUCTS 369-OTHER NON-MET MINL PRODS 371-IRON AND STEEL BASIC INDS 372-NON-FER METAL BASIC IND 380-MET SCRAP FOR MANUF OF MET PRDS 381-FAB MET PRDS, EXCL MACH, EQP 382-MANUF OF MACH EXCL ELECTRICAL 383-ELEC MACH, APP, APPL + SUPP 384-TRANSPORT EQUIPMENT 385-PROF, SCIEN, MSRNG, CNTL EQU 390-OTHER MANUF INDUSTRIES

Table A3 Matching ISIC 4-digit codes to NIC 1987 codes ISIC

NIC (1987) classification

3111-SLGHTRING, PREP, PRESERV MEAT 3112-MANUF OF DAIRY PRODUCTS 3113-CANNING, PRESERV FRUITS VEGS 3114-CANNING, PRESERV, PROCSSNG OF FISH, CRUSTACEANS 3115-MANUF VEG, ANIMAL OILS + FATS 3116-GRAIN MILL PRODUCTS 3117-MANUF OF BAKERY PRODUCTS 3118-SUGAR FACTORIES REFINERS 3119-MANUF COCOA,CHOC + SUG CONF 3122-MANUF OF PREPD ANIMAL FEEDS 3131-DISTILLNG, BLENDING SPIRITS 3132-WINE INDUSTRIES 3133-MALT LIQUORS AND MALT 3134-SFT DRNKS + CARB WATERS IND 3140-TOBACCO MANUFACTURES 3212-MANUF MDUP TXT GDS EXCL WEARG APP 3213-KNITTING MILLS 3214-CARPETS 3215-CORDAGE ROPE,TWINE INDS 3220-MANUF WEARG APP EXCL FTWR 3231-TANNERIES, LTHER FINISHNG

200 201 202 203 211 204 205 206 209 217 220 221 222 224 227 267 260 263 261 265 294 (Continued Overleaf )

148

Appendix

Table A3 Continued ISIC

NIC (1987) classification

3233-MANUF PRODS LTHER EXCL FTWR, APP 3240-MANUF FTWR EXCL RUBBR, PLASTC 3311-SAWMLS, PLNG OTH WD MILLS 3312-MANUF WOOD CONTAINERS 3319-MANUF WOOD CORK PRODS n.e.c. [not elsewhere classified] 3320-MANUF FURN, FIXT EXCL PRIM MTL 3411-MANUF PULP, PAPER, PAPERBOARD 3412-MANUF CONTS,BXES PPR, P/BRD 3419-MANUF ART PULP, PPR, P/BRD n.e.c. [not elsewhere classified] 3420-PRNTNG, PUBLNG ALLIED IND 3511-BASIC IND CHEMS EXCL FERT 3512-FERTILISERS PESTICIDES 3521-PAINTS,VARNISH LACQUERS 3522-DRUGS + MEDICINES 3523-SOAP, CLN PRPS, PERF, COSM 3529-CHEMICAL PRODUCTS n.e.c. [not elsewhere classified] 3530-PETROLEUM REFINERIES 3551-TIRE + TUBE INDUSTRIES 3559-MANUF OF RUBBER PRODS n.e.c. [not elsewhere classified] 3560-PLASTICS PRODUCTS n.e.c. [not elsewhere classified] 3610-POTTERY, CHINA, EARTHWARE 3620-GLASS + GLASS PRODUCTS 3691-STRUCTURAL CLAY PRODUCTS 3692-CEMENT, LIME AND PLASTER 3699-NON-MET MINL PRODS n.e.c. [not elsewhere classified] 3710-IRON + STEEL BAS INDS 3811-CUTLY, HAND TLS, GEN HDWRE 3812-FURN + FIXT PRIM OF METAL 3813-STRUCTURAL METAL PRODUCTS 3821-ENGINES + TURBINES 3822-AGRIC MACHINERY AND EQUIP 3824-SPEC IND MACH + EQP EXCL 3823 3825-OFFICE, COMPUTG, ACCOUNTG MACH 3829-MACH, EQUIP EXCL ELECT n.e.c. [not elsewhere classified] 3832-RADIO, TELE, COMM EQP, APPAR 3841-SHIPBUILDING + REPAIRING

293 291 270 273 275 276 280 281 282 285 300 301 303 304 305 309 314 310 312 313 323 321 320 324 329 330 343 342 340 352 350 353 358 356 366 370

Appendix 3842-RAILROAD EQUIPMENT 3843-MOTOR VEHICLES 3844-MOTOR CYCLES + BICYCLES 3845-AIRCRAFT 3851-PROF, SCIEN, MSRG, CNTRL EQU 3853-WATCHES + CLOCKS 3901-JEWELRY + RELATED ARTICLES 3902-MUSICAL INSTRUMENTS 3903-SPORTING + ATHLETIC GOODS

149

372 374 375 377 380 382 383 386 385

Table A4 US producer price index matched to NIC (1987) classification US code

Matched NIC code

US code

Matched NIC code

3113 3115 31161 3121 31222 3131 3151 3161 3251 3252 3253 3254 3256 32621 3332 3336 3371 3399

206–210 201 200 220–224 225–229 230–259 260–269 290–299 300, 303, 306–309 312 301 304 305 310 35–36, except as below 352 271–279 380–389

332211 332212 331111 327310 3331111 333131 333294 333313 333314 333415 333512 3342 334310 335110 335222 336510 336611 336991

346 343 330, 331 324 350 351 353 358 381 355 357 365 366 363 355 371 370 375

Notes

1 Trade policy, inequality and performance – the key analytical isssues 1 2 3 4

5 6 7

8

9

See Caves (1985) and Schmalensee (1989) for excellent overviews of the literature. See Domowitz et al. (1986, 1988), Levinsohn (1993), Harrison (1994) and Katics and Petersen (1994) for recent examples of such studies. See Rama (2003) for a recent review of the limited research in this area. See also Ghose (2003: esp. Ch. 4). We use dummy variables when measuring trade policy changes at the aggregate level, as in Chapter 4 of this book. At the level of the economy, there are no reasonable alternatives to dummy variables to capture trade policy changes. The organised manufacturing sector comprises all factories that fall under the purview of the Indian Factories Act of 1948, and are registered under sections 2m(i) and 2m(ii) of this Act. These are factories that employ ten or more workers and use power, and those employing 20 or more workers and do not use power. The sector also includes manufacturing establishments registered under the Bidi and Cigar Workers Act 1966, with a similar definition of the numbers of workers employed with and without power. According to Saha et al. (2004), 39.2 per cent of manufacturing output originated in the informal manufacturing sector, which also accounted for 94.9 per cent of total manufacturing employment. See the Appendix for a more detailed discussion of the data used, including the sources of the data. To take two examples, consider the 3-digit ISIC industries – Electrical Machinery (383) and Transport Equipment (384). In the first case, audio and video equipment (ISIC 3832) – a consumer goods industry – is clubbed together with several capital goods industries, such as electrical industrial machinery (ISIC 3831) and insulated wires and cables (ISIC 3839). Similarly, in the second case, another consumer goods industry, motor vehicles (ISIC 3843), is lumped together with other capital goods industries, e.g. railroad equipment (ISIC 3842). As we have already noted, Indian trade policy with respect to capital goods has differed significantly from that with respect to consumer goods. This is particularly true when we compute the price wedge, which is the difference between domestic and foreign prices. The ‘price wedge’ as a measure of protection is feasible only for products for which the quality adjustments are not overwhelming and where the products compared are fairly homogeneous in terms of their characteristics (Pritchett 1996). The greater the level of disaggregation, the more similar is the industry’s product between the two countries. An appendix provides a description of the data used, and the sources. The appendix also lists the NIC 3-digit and ISIC 3-digit industries, and provides the scheme

Notes

151

used to match the NIC 3-digit to ISIC 4-digit codes wherever we have attempted to combine data from the two different industrial classifications. 2 Trade policy in India 1 In a detailed study of sixty appraisal reports prepared by two of India’s leading term-lending institutions for the projects they financed in 1988 and 1989, Ettori (1990) found a marked bias of the effective protection structure and incentives in favour of the domestic market and import substitution and against export orientation. 2 For a detailed study of India’s trade regime prior to 1991, see World Bank (1990), Aksoy and Ettori (1992) and Bhagwati (1993). 3 As Joshi and Little (1997) argue, the concentration on emphasising an early reduction in tariffs on capital goods in the reform process was probably intended to avoid discouraging investment because of the expectation of a later reduction in tariffs. 4 For a critical evaluation of India’s trade policy changes since 1991, see Pursell (1996) and Panagariya (2004). 5 This section draws significantly from Ganesh Kumar et al. (2003). 6 Therefore, ERPj = (VA*j − VAj ) / VAj, where VA*j is value added of the final product j at free trade prices, and VAj is the value added of the final product j at tariff-distorted prices. 7 Therefore, ICRj = Σ Di Mij / Σ Mij, where D is a dummy variable which takes the value one if the product is subject to a NTB – that is, is banned/restricted/limited permissible or canalised, zero otherwise; j stands for industry and i for product category; M is the value of imports for the product which is subject to a NTB, and Σ Mi is the value of imports of all product lines within the industry. 8 As Harrison (1996: 421) observes, ‘price comparisons between goods sold in domestic and international markets could provide an ideal measure of the impact of trade policy’. 9 See Appendix, Table A1 for list of industries with their NIC 1987 codes. 10 From the methodological viewpoint, this will imply that we will not attempt to use regression methods that control for the possibility of simultaneity bias if the trade policy variables – which will figure in the right-hand side of the equations we estimate – are assumed to be independent of economic factors, and especially the dependent variables in our empirical analysis, such as productivity growth, employment and relative wages. 3 Growth and structural change in Indian manufacturing, 1975–99 1 The Grubel–Lloyd (1975) measure for a particular industry is defined as: [(X + M) − abs(X − M)] / abs(X − M), where X and M are exports and imports of the industry respectively, and abs is the absolute value. 2 The import penetration ratio is defined as imports / (imports + domestic production − exports). 3 The export orientation ratio is defined as exports/domestic production. 4 Trade policy and economic growth 1 The rate of GDP growth for the period 1955–1979 was 3.6 per cent per annum. 2 In a recent study, using panel data methods and a cross-country data-set, Greenaway et al. (2002) find that trade liberalisation does appear to impact on growth, albeit with a lag. Harrison (1996) finds that there is a positive association between economic growth and different measures of openness for a large panel of

152

3 4

5

6 7 8

Notes

countries. Frankel and Romer (1999) and Noguer and Siscart (2005) also find that international trade promotes income, using sophisticated econometric methods that treat trade as endogenous and use geographical factors to instrument for trade. This section draws from Sen (2002). India’s low equipment investment rate in the 1960s and 1970s can be attributed to the strong restrictions on capital goods imports during this period, which led a high relative price of equipment. As De Long and Summers (1993: 399) point out, ‘India’s policies have managed to enrich industrialists instead of encouraging industry.’ Unfortunately, we do not have disaggregated investment data in PPP terms from the Summers and Heston (1991) database to see if India’s rate of equipment investment relative to other countries was different in the late 1990s as compared to the early 1980s. We omit the constant term as, in the standard growth models (both in the neoclassical and endogenous growth tradition), growth of output will be zero if the investment rate is zero. Real domestic credit to the private sector is obtained by deflating nominal domestic credit to the private sector by the price deflator for fixed investment. For a fuller development of this argument, see Sen (2007) who argues that the effect of the attitudinal shift of the state towards the private sector on India’s growth acceleration was second order, and if it did have an effect on growth at all, it was only through changes in economic policies, rather than independently of the latter. The increase in economic growth occurred, according to this view, due to three policy-influenced fundamentals – financial deepening, a rise in public investment and a fall in the relative price of equipment.

5 Trade policy, productivity and prices 1 This section draws from Chand and Sen (2002). 2 This section draws from Chand and Sen (1999). 3 See Feinberg (1986, 1989) and Dornbusch (1986) for a fuller treatment of these issues. 4 A survey of the theoretical and empirical literature is provided in Tybout (1992). 5 More precisely, total factor productivity growth (TFPG) is as follows: TFPGt = yt − 0.5(1−slt − sLt−1)kt − 0.5(slt + sLt−1)lt; where yt is growth of value added at time t, kt is growth of capital stock at time t, lt is growth of labour – both production and non-production – at time t, and sLt is share of wages in value added in time t. We assume a constant returns to scale production function in our calculations. 6 By including sectoral foreign prices, along with the nominal exchange rate and the aggregate domestic price level, as independent variables, we are, in essence, estimating the effect of the sectoral real exchange rate on domestic real prices. We also substituted the aggregate foreign price level (the CPI for the United States) for sectoral foreign prices without any change in the results. 6 International trade and employment 1 This and the next four sections draw heavily from Jenkins and Sen (2006). 2 This is the assumption made by the individual country studies in the NBER project on trade and employment led by Krueger, and is a fairly plausible assumption for most low-income developing countries. 3 Strictly speaking intra-industry trade would not lead to any change in the composition of output because imports and exports are in the same industry. In practice, however, if intra-industry is defined in terms of relatively broad industrial

Notes

4 5 6 7

8

9 10 11

12

153

categories while a more detailed classification of production is used, then it is possible for intra-industry trade to have some impact on the composition of output and hence on employment. In their model the factors of production are unskilled and skilled labour, and they show trade increasing the demand for skilled labour in the less developed country (Feenstra and Hanson, 1996). For reviews of such studies, see Wood (1994: Ch. 3), Lawrence (1996: Ch. 2). The six developing countries in Ghose’s sample are: China, India, Indonesia, Malaysia, the Philippines and Taiwan. It should be noted that the paper uses the International Standard Industrial Classificiation (ISIC) definition of manufacturing, which is broader than the Standard International Trade Classification (SITC) and includes processing of many primary products. The trade data comes from the International Economic Databank of the Australian National University and has been reclassified from COMTRADE data according to the International Standard Industrial Classification (ISIC Rev. 2). Because the trade data is only available at the 4-digit level and, in a small number of cases, Krause uses a 5-digit classification, we have had to slightly modify his groupings. A more conventional trade theory approach could regard both these categories as capital intensive. To compute the employment coefficients, we use employment per constant price rupee of output rather than per rupee of value-added as the export and import figures are in gross terms. One problem with the data that we encounter when implementing the growth accounting methodology in relation to India is the large growth in exports and imports that has occurred in the same industry – which is jewellery and related industries (ISIC Code 3901). This is because the Indian jewellery industry imports raw gems to be processed into jewellery for export. Including this industry will overstate the effect of trade on employment since the growth accounting methodology treats all imports as necessarily substituting for domestic production in the same industry. Including Industry 3901 will give the inaccurate result that the increase in imports that occurred in this industry has led to a large employment decline in the industry, which is not the case. Thus, we exclude industry 3901 from our calculations. We use the package STATA, which does not report the Sargan test for overidentifying restrictions, when we compute the estimates with robust standard errors.

7 Trade policy and wage inequality 1 It should be noted that that the H–O model has one set of factor market equilibrium equations that links factor supply conditions to trade quantities, and an entirely separate set of zero profit conditions that link product prices to factor prices. As Leamer argues, in this model: the news of changes in external product markets is carried to US labour markets by changes in product prices not by trade volumes, whether measured directly or indirectly through factor contents. [. . .] Trade volumes are entirely irrelevant. It doesn’t matter if apparel is exported or imported or if trade is exactly balanced. What matters is whether or not apparel is produced in the US, competing with products made in China. (2000: 20) Thus, there is no necessary relationship between changes in factor contents and the response of domestic labour markets to changes in the external product markets.

154

Notes

2 One should be careful to note that the relaxing of one or more of these assumptions does not necessarily lead to a weakening (or reversal) of the S–S result. For example, perfect competition is a sufficient, but not necessary, assumption of the S–S theorem. What is really essential in the latter is the existence of a systematic link between product prices and factor prices. Thus, it is possible to have imperfect competition, as long as there is a positive but unchanging price-cost mark-up, or monopolistic competition in which sufficient entry by new firms ensures zero profits in equilibrium. Moreover, as argued by Slaughter (1998), the imperfect mobility of factors of production in the short run may imply that the H–O model may have greater predictive power as a theory of changes in inequality in the long run. 3 A selective set of references is Bound and Johnson (1992), Katz and Murphy (1992), Berman et al. (1994), Autor et al. (1998) and Berman et al. (1998). 4 Thoenig and Verdier (2003) construct a dynamic general equilibrium model to show that when international trade triggers an increased threat of technological leapfrogging or imitation, firms would tend to respond to the threat by biasing the direction of their innovations towards skilled labour-intensive technologies. 5 Again, more complex versions of the H–O that allow skill-biased technological change in the North and the South predict that wage inequality will rise in both developed and developing countries with international trade (Panagariya 2000). 6 There is no finite profit maximum at constant prices in the case of constant or increasing returns to scale. 7 The reason is that the larger is labour’s share in the firm’s total cost and revenues, the larger the pass-through from ηj to ηLLj; but at the same time, the more difficult it is for the firm to substitute away from labour. 8 As Panagariya argues, ‘while it is possible for an appropriately defined labour demand curve to be more elastic in an open than a closed economy, under standard assumptions, it is equally possible for the opposite to be true’ (2000b: 1). 9 It should be noted that there is no necessary link between labour demand elasticities at the firm and the national levels (Slaughter 1997). The national labour demand elasticity is derived from a country’s optimal output mix, given exogenous factor endowments, product prices and production technology. 10 The countries that exhibit an unambiguous negative relationship between wage inequality and openness are Hong Kong, Korea, Singapore and Taiwan, while changes in wage inequality in the Philippines show no clear trend. In the case of Malaysia, there was decreasing wage inequality in the 1970s and 1980s, but this was possibly because of changes in domestic factor supply. 8 Trade policy and regional inequality 1 Two exceptions are Lall et al. (2003) and Lall and Chakravorty (2005). The first paper finds significant effects of agglomeration economies on productivity in Indian industry, and the second finds that structural reforms may have led to increased spatial inequality in industrialisation, with private sector industrial investments biased towards existing industrial and coastal districts. 2 Aghion et al. (2005) have argued that following the dismantling of the licensing regime, industries located in states with pro-employer labour institutions grew more quickly than those in pro-worker environments. The descriptive statistics presented in this chapter are consistent with this argument. States like Maharashtra and West Bengal, with pro-worker labour market institutions, have seen industrial declines, and states like Andhra Pradesh and Tamil Nadu, with pro-employer labour market institutions, have seen industrial expansion.

Notes

155

9 Conclusion 1

Under the Industrial Disputes Act 1946, firms in the organised sector are not permitted to retrench any worker who has been employed with the firm for at least one year without permission from the government. If such permission is granted, the firm must give three months’ notice to the worker as well as retrenchment compensation as in the case of smaller firms. However, such permission is rarely granted (Rudoph and Rudolph 1987, Ganesh-Kumar et al. 2003).

References

Aghion, P., Burgess, R. and Redding, S. (2005) ‘The Unequal Effects of Liberalization: Evidence from Dismantling the License Raj in India’, mimeo, London: LSE. Agrawal, P., Gokarn, S., Mishra, V., Parikh, K.S. and Sen, K. (1995) Economic Restructuring in East Asia and India: Perspectives on Policy Reform, Basingstoke: Macmillan. Ahluwalia, I.J. (1991) Industrial Growth in India, New Delhi: Oxford University Press. Ahluwalia, I.J. (1999) ‘India’s Economic Reforms: An Appraisal’, in J.D. Sachs, A. Varshney and N. Bajpai (eds) India in the Era of Economic Reforms, New Delhi: Oxford University Press. Aksoy, M.A. and Ettori, F. (1992) ‘Protection and Industrial Structure in India’, Working Paper No. 990, Washington, DC: The World Bank. Amadeo, E.J. and Pero, V. (2000) ‘Adjustment, Stabilisation and the Structure of Employment in Brazil’, Journal of Development Studies, 36(4): 120–48. Anderson, J. and Wincoop, E.V. (2004) ‘Trade Costs’, Journal of Economic Literature, 62: 691–751. Arellano, M. and Bond, S. (1991) ‘Some Tests of Specification for Panel Data: Monte Carlo Evidence and an Application to Employment Equations’, Review of Economic Studies, 58: 277–97. Athukorala, P. and Sen, K. (2002) Saving, Investment and Growth in India, Delhi: Oxford University Press. Attanasio, O., Goldberg, P. and Pavcnik, N. (2004) ‘Trade Reforms and Wage Inequality in Colombia’, Journal of Development Economics, 74: 331–66. Autor, D.H., Katz, L.F. and Krueger, A.B. (1998) ‘Computing Inequality: Have Computers Changed the Labour Market?’, Quarterly Journal of Economics, 113: 1169–213. Balakrishnan, P. and Pushpangadan, K. (1994) ‘Total Factor Productivity Growth in Manufacturing Industry: A Fresh Look’, Economic and Political Weekly, 30 July: 2028–35. Balakrishnan, P., Pushpangadan, K. and Suresh Babu, M. (2002) ‘Trade Liberalisation, Market Power and Scale Efficiency in Indian Industry’, Working Paper No. 336, Trivandrum, India: Centre for Development Studies. Balasubramanyam, V.N. (2003) ‘India: Trade Policy Review’, The World Economy, 26(9): 1357–68. Bardhan, P. (2007) ‘Poverty and Inequality in China and India: Elusive Link with Globalisation’, Economic and Political Weekly, 22 September: 3849–52.

References

157

Bell, C. and Rousseau, P. (2001) ‘Post-Independence India: A Case of Finance-Led Industrialization’, Journal of Development Economics, 62(1): 153–75. Bencivenga, V.R. and Smith, B.D. (1991) ‘Financial Intermediation and Endogenous Growth’, Review of Economic Studies, 58(2): 195–209. Berman, E., Bound, J. and Griliches, Z. (1994) ‘Changes in the Demand for Skilled Labour: Evidence from the Annual Survey of Industries’, Quarterly Journal of Economics, 109: 367–97. Berman, E., Bound, J. and Machin, S. (1998) ‘Implications of Skill-Biased Technological Change’, Quarterly Journal of Economics, 113(4): 1245–79. Besley, T. and Burgess, R. (2004) ‘Can Labour Regulations Hinder Economic Performance? Evidence from India’, Quarterly Journal of Economics, 19(1): 91–134. Beyer, H., Rojas, P. and Vergara, R. (1999) ‘Trade Liberalization and Wage Inequality’, Journal of Development Economics, 59: 103–23. Bhagwati, J.N. (1993) India in Transition: Freeing the Economy, Oxford: Clarendon Press. Bhagwati, J. and Desai, P. (1970) India: Planning for Industrialization, London: Oxford University Press. Bhagwati, J. and Srinivasan, T.N. (1975) Foreign Trade Regimes and Economic Development: India, New York: Columbia University Press. Bhalotra, S. (1998) ‘The Puzzle of Jobless Growth in Indian Manufacturing’, Oxford Bulletin of Economics and Statistics, 60(1): 1–32. Bleaney, M.F. and Greenaway, D. (1993) ‘Long-Run Trends in the Relative Price of Primary Commodities and in the Terms of Trade of Developing Countries’, Oxford Economic Papers, 45(3): 349–63. Blejer, M. and Khan, M.S. (1984) ‘Government Policy and Private Investment in Developing Countries’, IMF Staff Papers, 31: 379–403. Borjas, G., Freeman, R. and Katz, L. (1997) ‘How Much do Immigration and Trade Affect Labour Market Outcomes?’, Brookings Papers on Economic Activity, 1: 1–90. Bound, J. and Johnson, G. (1992) ‘Changes in the Structure of Wages in the 1980s: An Evaluation of Alternative Explanations’, American Economic Review, 82: 371–92. Bruno, M. (1978) ‘Duality, Intermediate Inputs and Value Added’, Production Economics: A Dual Approach to Theory and Applications, Amsterdam: North Holland. Bruno, M. (1984) ‘Raw Materials, Profits and the Productivity Slowdown’, Quarterly Journal of Economics, 1: 1–29. Caves, R. (1985) ‘International Trade and Industrial Organisation: Problems, Solved and Unsolved’, European Economic Review, 28: 377–95. Chakravarty, S. (1987) Development Planning: The Indian Experience, Delhi: Oxford University Press. Chamarbagwala, R. (2006) ‘Economic Liberalization and Wage Inequality in India’, World Development, 34(12): 1997–2015. Chand, S. and Sen, K. (1999) ‘Competitive Pressures from Trade Exposure: Evidence from Indian Manufacturing’, Indian Economic Review, 34(2): 113–26. Chand, S. and Sen, K. (2002) ‘Trade Liberalisation and Productivity Growth: Evidence from Indian Manufacturing’, Review of Development Economics, 6(1): 120–32. Currie, J. and Harrison, A. (1997) ‘Sharing the Costs: The Impact of Trade Reform on Capital and Labour in Morocco’, Journal of Labour Economics, 15(3): S44–S71.

158

References

Das, D.K. (2003) Quantifying Trade Barriers: Has Protection Declined Substantially in Indian Manufacturing?, Working Paper No. 105, New Delhi: ICRIER. De Long, J.B. (2003) ‘India since Independence: An Analytical Growth Narrative’, in D. Rodrik (ed.) In Search of Prosperity: Analytical Narratives on Economic Growth, Princeton, NJ: Princeton University Press, pp. 184–204. De Long, J.B. and Summers, L.H. (1991) ‘Equipment Investment and Economic Growth’, Quarterly Journal of Economics, 106: 455–502. De Long, J.B. and Summers, L.H. (1992) ‘Equipment Investment and Economic Growth: How Strong is the Nexus? Brookings Papers on Economic Activity, 2: 157–99. De Long, J.B. and Summers, L.H. (1993) ‘How Strongly do Developing Countries Benefit from Equipment Investment?’ Journal of Monetary Economics, 32: 395–415. Demurger, S., Sachs, J., Woo, W., Bao, S., Change, G. and Mellinger, A. (2002) ‘Geography, Economic Policy and Regional Development in China’, mimeo. Cambridge, MA: Harvard University. Dixit, A. and Stiglitz, J.E. (1977) ‘Monopolistic Competition and Optimum Product Diversity’, American Economic Review, 67: 297–308. Domowitz, I., Hubbard, R.G. and Petersen, B.C. (1986) ‘Business Cycles and the Relationship between Concentration and Price-Cost Margins’, Rand Journal of Economics, 17: 1–17. Domowitz, I., Hubbard, R.G. and Petersen, B.C. (1988) ‘Market Structure and Cyclical Fluctuations in US Manufacturing’, Review of Economics and Statistics, 70(1): 55–65. Dornbusch, R. (1986) ‘Exchange Rates and Prices’, American Economic Review, 77(1): 93–106. Driffield, N. and Kambhampati, U.S. (2003) ‘Trade Liberalization and the Efficiency of Firms in Indian Manufacturing’, Review of Development Economics, 7(3): 409–30. Edwards, S. (1993) ‘Openness, Trade Liberalization and Growth in Developing Countries’, Journal of Economic Literature, 31: 1358–93. Ettori, F. (1990) ‘The Pervasive Effects of High Taxation of Capital Goods in India’, Working Paper No. 433, Washington, DC: World Bank. Fajnzylber, P. and Maloney, W.F. (2005) ‘Labour Demand and Trade Reform in Latin America’, Journal of Development Economics, 66: 423–46. Fanelli, J.M. and Medhora, R. (eds) (2002) Finance and Competitiveness in Developing Countries, London: Routledge. Feenstra, R. and Hanson, G. (1996) ‘Foreign Investment, Out-Sourcing and Relative Wages’, in R. Feenstra, G. Grossman and D. Irwin (eds) Political Economy of Trade Policy: Essays in Honour of Jagdish Bhagwati, Cambridge, MA: MIT Press. Feenstra, R. and Hanson, G. (1997) ‘Foreign Direct Investment and Relative Wages: Evidence from Mexico’s Maquiladoras’, Journal of International Economics, 42: 371–94. Feinberg, R.M. (1986) ‘The Interaction of Foreign Exchange and Market Power Effects on German Domestic Prices’, Journal of Industrial Economics, 35(1): 61–9. Feinberg, R.M. (1989) ‘The Effects of Foreign Exchange Movements on US Domestic Prices’, Review of Economics and Statistics, 71: 505–11. Frankel, J. and Romer, D. (1999) ‘Does Trade Cause Growth?’, American Economic Review, 89(3): 379–99.

References

159

Fujita, M. (1988) ‘Monopolistic Competition and Urban Systems’, European Economic Review, 37: 308–15. Galiani, S. and Sanguinetti, P. (2003) ‘The Impact of Trade Liberalization on Wage Inequality: Evidence from Argentina’, Journal of Development Economics, 72: 497–513. Ganesh Kumar, A., Sen, K. and Vaidya, R.R. (2003) International Competitiveness, Investment and Finance: A Case-Study of India, London: Routledge. Gang, I. and Pandey, M. (1996) ‘Trade Protection in India: Economic or Politics?’, Departmental Working Papers 199616, Rutgers University, Department of Economics. Ghose, A. (2000) ‘Trade Liberalization, Employment and Global Inequality’, International Labour Review, 139(3): 281–305. Ghose, A. (2003) Jobs and Incomes in a Globalizing World, Geneva: ILO. Goldar, B. and Renganathan, V.S. (1990) ‘Liberalisation of Capital Goods Imports in India’, Working Paper No. 8, National Institute of Public Finance and Policy, New Delhi. Goldberg, P.K. and Pavcnik N. (2007) ‘Distributional Effects of Globalization in Developing Countries’, Journal of Economic Literature, 45: 39–82. Greenaway, D., Hine, R. and Wright, P. (1999) ‘An Empirical Assessment of Impact of Trade on Employment in the United Kingdom’, European Journal of Political Economy, 15: 485–500. Greenaway, D., Hine, R. and Wright, P. (2002) ‘Trade Liberalisation and Growth in Developing Countries’, Journal of Development Economics, 67: 229–44. Grossman, G. and Helpman, E. (1991) Innovation and Growth in the Global Economy, Cambridge, MA: MIT Press. Grubel, H. and Lloyd, P.J. (1975) Intra-Industry Trade, London: Macmillan. Gurley, J. and Shaw, E. (1955) ‘Financial Aspects of Economic Development’, American Economic Review, 45(4): 515–38. Hanson, G.H. (1998) ‘Regional Adjustment to Trade Liberalization’, Regional Science and Urban Economics’, 28(4): 419–44. Hanson, G.H. and Harrison, A.E. (1999) ‘Trade Liberalization and Wage Inequality in Mexico’, International and Labour Relations Review, 52(2): 271–88. Harrison, A.E. (1994) ‘Productivity, Imperfect Competition and Trade Reform: Theory and Evidence’, Journal of International Economics, 36: 53–73. Harrison, A.E. (1996) ‘Openness and Growth: A Time-Series, Cross-Country Analysis for Developing Countries’, Journal of Development Economics, 48: 419–47. Harrison, A.E. and Hanson, G.H. (1999) ‘Who Gains from Trade Reform? Some Remaining Puzzles’, Journal of Development Economics, 59: 125–54. Hasan, R., Mitra, D. and Ramaswamy, K.V. (2003) ‘Trade Reforms, Labour Regulations and Labour-Demand Elasticities: Empirical Evidence from India’, NBER Working Paper No. 9879, Washinton, DC: NBER. Havrylyshyn, O. (1990) ‘Trade Policy and Productivity Gains in Developing Countries: A Survey of the Literature’, World Bank Research Observer, 5(1): 1–24. Helpman, E. and Krugman, P. (1989) Trade Policy and Market Structure, Cambridge, MA: MIT Press. Henderson, J.V. (1974) ‘The Size and Types of Cities’, American Economic Review, 64: 308–15. Hine, R. and Wright, P. (1998) ‘Trade with Low-Wage Economies, Employment and Productivity in UK Manufacturing’, Economic Journal, 108: 1500–10.

160

References

Horn, H., Lang, H. and Lundgren, S. (1996) ‘Managerial Effort Incentives, X-inefficiency and International Trade’, European Economic Review, 39(1): 117–38. Howes, C. and Singh, A. (eds) (2000) Competitiveness Matters: Industry and Economic Performance in the US, Ann Arbor: University of Michigan Press. Jenkins, R.O. and Sen, K. (2006) ‘International Trade and Manufacturing Employment in the South: Four Country Case-Studies’, Oxford Development Studies, 34(3): 299–322. Jones, C.I. (1994) ‘Economic Growth and the Relative Price of Capital’, Journal of Monetary Economics, 34: 359–82. Jorgenson, D.W. (1967) ‘The Theory of Investment Behaviour’, in R. Ferber (ed.) Determinants of Investment Behavior. New York: Columbia University Press, pp. 129–55. Joshi, V. and Little, I.M.D. (1994) India,: Macroeconomics and Political Economy, 1964–1991. Delhi: Oxford University Press. Joshi, V. and Little, I.M.D. (1996) India’s Economic Reforms 1991–2001, Oxford: Clarendon Press. Joshi, V. and Little, I.M.D. (1997) ‘India: Reform on Hold’, Asian Development Review, 15(2): 1–42. Kanbur, R. and X. Zhang (2005) ‘Fifty Years of Regional Inequality in China: A Journey Through Central Planning, Reform and Openness’, Review of Development Economics, 9(1): 87–106. Katics, M.M. and Petersen, B.C. (1994) ‘The Effect of Rising Import Competition on Market Power: A Panel Data Study of US Manufacturing’, Journal of Industrial Economics, 52: 277–86. Katz, L.F. and Murphy, K.M. (1992) ‘Changes in Relative Wages 1963–1987: Supply and Demand Factors’, Quarterly Journal of Economics, 107(1): 35–78. Khambhampati, U. (2003) ‘Trade Reforms and the Efficiency of Firms in India’, Oxford Development Studies, 31(2): 219–33. Khambhampati, U. and Parikh, A. (2003) ‘Disciplining Firms: The Impact of Trade Reform on Profit Margins in Indian Industry’, Applied Economics, 35: 461–70. Khambhampati, U. and Parikh, A. (2005) ‘Has Liberalisation Affected Profit Margins in Indian Industry?’, Bulletin of Economic Research, 57: 273–304. Khambhampati, U., Krishna, P. and Mitra D. (1997) ‘The Effects of Trade Policy Reform on Labour Markets: Evidence from India’, Journal of International Trade and Economic Development, 6(2): 287–97. King, R. and Levine, R. (1993) ‘Finance and Growth: Schumpeter May Be Right’, Quarterly Journal of Economics, 108(3): 717–37. Kohli, A. (1989) The State and Poverty in India, Cambridge: Cambridge University Press. Kohli, A. (2006) ‘Politics of Economic Growth in India: 1980–2005’, Economic and Political Weekly, 1 April: 1251–9. Krause, L.B. (1982) US Economic Policy Towards the Association of Southeast Asian Nations: Meeting the Japanese Challenge, Washington, DC: The Brookings Institution. Krishna, P. and Mitra, D. (1998) ‘Trade Liberalisation, Market Discipline and Productivity Growth: New Evidence from India’, Journal of Development Economics, 56: 447–62. Krueger, A.O. (1977) ‘Growth, Distortions and Patterns of Trade Among Many Countries’, Princeton Studies in International Finance, 40: 1–50.

References

161

Krueger, A.O. (1997) ‘Trade Policy and Economic Development’, American Economic Review, 87: 1–22. Krueger, A.O. and Tuncer, B. (1982) ‘Growth of Factor Productivity in Turkish Manufacturing Industries’, Journal of Development Economics, 11: 307–26. Krueger, A.O., Lary, H.B., Monson, T. and Akrasanee, N. (1981) Trade and Employment in Development Countries, Chicago: University of Chicago Press. Krugman, P. (1991) Geography and Trade, Cambridge, MA: MIT Press. Krugman, P. (1995) ‘Growing World Trade: Causes and Consequences’, Brookings Papers on Economic Activity, 1: 327–62. Krugman, P. (2000) ‘And Now for Something Completely Different: An Alternative Model of Trade, Education and Inequality’, in R. Feenstra (ed.) The International Trade on Wages, Chicago: University of Chicago Press. Krugman, P. and Livas, R. (1992) ‘Trade Policy and the Third World Metropolis’, NBER Working Paper No. 4238. Lall, S. (1998) ‘Technological Capabilities in Emerging Asia’, Oxford Development Studies, 26(2): 213–43. Lall, S. (2001) ‘Competitiveness Indices and Developing Countries: An Economic Evaluation of the Global Competitiveness Report’, World Development, 29(9): 1501–25. Lall, S. and Chakravorty, S. (2005) ‘Industrial Location and Spatial Inequality: Theory and Evidence from India’, Review of Development Economics, 9(1): 47–68. Lall, S., Shalizi, Z. and Deichmann, U. (2003) ‘Agglomeration Economies and Productivity in Indian Industry’, Journal of Development Economies, 73: 643–73. Lawrence, R. (1996) Single World, Divided Nations?, Paris: OECD. Leamer, E.E. (1987) ‘Patterns of Development in the Three Factor n-Good General Equilibrium Model’, Journal of Political Economy, 95(5): 961–99. Leamer, E.E. (2000) ‘What’s the Use of Factor Contents?’ Journal of International Economics, 50: 17–49. Leamer, E.E., Maul, H., Rodriguez, S. and Schott, P. (1999) ‘Does Natural Resource Abundance Increase Latin American Income Inequality?’, Journal of Development Economics, 59: 3–42. Levine, R. (1997) ‘Financial Development and Economic Growth: Views and Agenda’, Journal of Economic Literature, 35: 688–726. Levine, R. and Renelt, D. (1992) ‘A Sensitivity Analysis of Cross-Country Growth Regressions’, American Economic Review, 82(4): 942–63. Levinsohn, J. (1993) ‘Testing the Imports-as-Market-Discipline Hypothesis’, Journal of International Economics, 35: 1–22. Levinsohn, J. (1999) ‘Employment Responses to International Liberalization in Chile’, Journal of International Economics, 47: 321–44. Lucas, R.E. Jr (1988) ‘On the Mechanics of Economic Development’, Journal of Monetary Economics, 22: 3–42. Macdonald, J.M. (1994) ‘Does Import Competition Force Efficient Production’, Review of Economics and Statistics, 76(4): 722–7. Magee, S. (1994) ‘The Political Economy of Trade Policy’, in D. Greenaway and L.A. Winters (eds) Surveys in International Trade, London: Basil Blackwell. Majumdar, S. and Sen, K. (2007) ‘The Debt Wish: Rent-Seeking by Business Groups and the Structure of Borrowing in Indian Industry’, Public Choice, 130: 209–23. Marjit, S. and Kar, S. (2007) ‘Regional Trade Openness Index, Income Disparity, and

162

References

Poverty – An Experiment with Poverty Data’, mimeo. Centre for Studies in Social Sciences, Kolkata, India. Marjit, S. and S. Mitra (1996) ‘Convergence in Regional Growth Rates, Indian Research Agenda’, Economic and Political Weekly, 31(33): 2239–42. Martin, J. and Evans, J. (1981) ‘Notes on Measuring the Employment Displacement Effects of Trade by the Accounting Procedure’, Oxford Economic Papers, 33(1): 154–64. Mayer, J. and Wood, A. (2001) ‘South Asia’s Export Structure in a Comparative Perspective’, Oxford Development Studies, 29(1): 5–29. Michaely, M., Papageorgiou, D. and Choksi, A.M. (1991) Liberalizing Foreign Trade – Lessons of Experience in the Developing World, Cambridge, MA: Basil Blackwell. Milner, C. and Wright, P. (1998) ‘Modelling Labour Market Adjustment to Trade Liberalisation in an Industrialising Economy’, The Economic Journal, 108: 509–28. Milner, C., Vencappa, D. and Wright, P. (2007) ‘Trade Policy and Productivity Growth in Indian Manufacturing’, The World Economy, 30(2): 249–65. Mishra, P. and Kumar, U. (2005) Trade Liberalization and Wage Inequality: Evidence from India, IMF Working Paper WP/05/20, Washington, DC: IMF. Mohan, R. (2003) ‘Small-Scale Industrial Policy in India: A Critical Evaluation’, in A. Krueger (ed.) Economic Policy Reforms and the Indian Economy, Chicago: University of Chicago Press, pp. 213–302. Mookherjee, D. (1995) ‘Introduction’, in D. Mookherjee (ed.) Indian Industry: Policies and Performance, Oxford: Oxford University Press, pp. 1–43. Moreira, M.M. and Najberg, S. (2000) ‘Trade Liberalisation in Brazil: Creating or Exporting Jobs?’, Journal of Development Studies, 36(3): 78–99. Nayyar, D. (2006) ‘India’s Unfinished Journey: Transforming Growth into Development’, Modern Asian Studies, 40: 797–832. Nishimizu, M. and Page, J. (1990) ‘Trade Policy, Market Orientation and Productivity Change in Industry’, in J. de Melo and A. Sapir (eds) Trade Theory and Economic Reform: Essays in Honor of Bela Balassa, Cambridge, MA: Basil Blackwell. Noguer, M. and M. Siscart (2005) ‘Trade Raises Income: A Precise and Robust Result’, Journal of International Economics, 65: 447–60. Palmer-Jones, R. and Sen, K. (2003) ‘What has Luck Got to Do With It? A Regional Analysis of Poverty and Agricultural Growth in India’, Journal of Development Studies, 40(1): 1–32. Panagariya, A. (1999) ‘Trade Openness: Consequences for the Elasticity of Demand for Labour and Wage Outcomes’, mimeo, College Park, University of Maryland. Panagariya, A. (2000) Technology and Wage Inequality in North and South, Department of Economics, University of Maryland, Center for International Economics. Panagariya, A. (2004) ‘India’s Trade Reform’, India Policy Forum 2004, 1: 1–69. Panagariya, A. (2007a) ‘Why India Lags Behind China and How It Can Bridge the Gap’, The World Economy, 229–248. Panagariya, A. (2007b) ‘Transforming India’, mimeo, Columbia University, New York. Pritchett, L. (1996) ‘Measuring Outward Orientation in LDCs: Can It Be Done?’, Journal of Development Economics, 49: 307–35. Pursell, G. (1992) ‘Trade Policies in India’, in D. Salvatore (ed.) National Trade Policies, New York: Greenwood Press. Pursell, G. (1996) Indian Trade Policies since the 1991/1992 Reforms, Washington, DC: The World Bank.

References

163

Rama, M. (2003) ‘Globalisation and Workers in Developing Countries’, in R. Hasan and D. Mitra (eds) The Impact of Trade and Labour, Amsterdam: North Holland. Revenga, A. (1997) ‘Employment and Wage Effects of Trade Liberalization: The Case of Mexican Manufacturing’, Journal of Labour Economics, 15(3): S20–S43. Robbins, D. (1995) ‘Trade, Trade Liberalization and Inequality in Latin America and East Asia: Synthesis of Seven Country Studies’, Harvard Institute for International Development, Cambridge, MA. Robbins, D. (1996) ‘HOS Hits Facts: Facts Win: Evidence on Trade and Wages in the Developing World’, Development Discussion Paper No. 557, Harvard Institute of International Development. Rodriguez, F. and Rodrik, D. (1999) ‘Trade Policy and Economic Growth: A Skeptic’s Guide to the Cross-National Evidence’, NBER Working Paper 7081, Washington, DC: NBER. Rodrik, D. (1988) ‘The Limits of Trade Policy Reforms in Developing Countries’, Journal of Economic Perspectives, 6(1): 87–105. Rodrik, D. (1992) ‘Closing the Productivity Gap: Does Trade Liberalization Really Help?’, in G.K. Helleiner (ed.) Trade Policy, Industrialization and Development, New York: Oxford University Press, pp. 155–75. Rodrik, D. (1995) ‘Trade and Industrial Policy Reform’, in J. Behrman and T.N. Srinivasan (eds) Handbook of Development Economics, vol. III, Amsterdam: North Holland. Rodrik, D. (1997) Has Globalisation Gone too Far? Washington DC, Institute for International Economics. Rodrik, D. and Subramanian, A. (2004) ‘From Hindu Growth to Productivity Surge: The Myth of the Indian Growth Transition’, NBER Working Paper No. W10376. Romer, P.M. (1986) ‘Increasing Returns and Long-Run Growth’, Journal of Political Economy, 94: 1002–37. Romer, P.M. (1987) ‘Growth Based on Increasing Returns due to Specialisation’, American Economic Review, 77(2): 56–62. Rudolph, L. and Rudolph, S.H. (1987) The Political Economy of the Indian State, Chicago: University of Chicago Press. Sachs, J. (2003) ‘Institutions Don’t Rule: Direct Effects of Geography on Per Capita Income’, NBER Working Paper 9490. Sachs, J.D., Bajpai N. and Ramiah, A. (2002) ‘Understanding Regional Economic Growth in India’, Working Papers No. 88, Center for International Development at Harvard University. Saha, V., Kar, A. and Baskaran, T. (2004) ‘Contribution of Informal Sector and Informal Employment in Indian Economy’, paper presented at the 7th meeting of the Expert Group on Informal Sector Statistics, New Delhi. Sapsford, D. and Chen, J.R. (1998) ‘The Prebisch-Singer Terms of Trade Hypothesis: Some (Very) New Evidence’, in D. Sapsford and J.R. Chen (eds) Development Economics and Policy: The Conference Volume to Celebrate the 85th Birthday of Sir Hans Singer, Basingstoke: Macmillan. Sarkar, P. and Singer, H.W. (1991) ‘Manufacturing Exports of Developing Countries and their Terms of Trade Since 1965’, World Development, 19(4): 333–40. Schmalensee, R. (1989) ‘Inter-Industry Studies of Structure and Economic Performance’, in R. Schmalensee and R. Willig (eds) Handbook of Industrial Organisation, Amsterdam: North Holland, pp. 951–1009.

164

References

Sen, K. (2002) ‘Trade Policy, Equipment Investment and Growth in India’, Oxford Development Studies, 30(3): 317–31. Sen, K. (2007) ‘Why Did the Elephant Start to Trot? India’s Growth Acceleration Re-Examined’, Economic and Political Weekly, 43: 37–49. Slaughter, M. (1997) ‘International Trade and Labor Demand Elasticities’, Journal of International Economics, 54(1): 27–56. Slaughter, M. (1998) ‘International Trade and Labour-Market Outcomes: Results, Questions and Policy Options’, The Economic Journal, 108: 1452–62. Srivastava, V. (1996) Liberalization, Productivity and Competition: A Panel Study of Indian Manufacturing, New Delhi: Oxford University Press. Summers, R. and Heston, A. (1991) ‘The Penn World Table (Mark V): An Expanded Set of International Comparisons, 1950–1988’, Quarterly Journal of Economics, 106(2): 327–68. Temple, J. (1998) ‘Equipment Investment and the Solow Model’, Oxford Economic Papers, 50(1): 39–62. Thoenig, M. and Verdier, T. (2003) ‘A Theory of Defensive Skill-Biased Innovation and Globalization’, American Economic Review, 93(3): 709–28. Topalova, P. (2004) ‘Trade Liberalisation and Firm Productivity: The Case of India’, IMF Working Paper No. 04/28, Washington, DC: IMF. Tybout, J. (1992) ‘Linking Trade and Productivity: New Research Directions’, The World Bank Research Observer, 6(1): 189–211. Varshney, A. (1999) ‘Mass Politics or Elite Politics: India’s Reforms in Comparative Perspective’, in J.D. Sachs, A. Varshney and N. Bajpai (eds) India in the Era of Economic Reforms, Delhi: Oxford University Press, pp. 222–60. Venables, T. (2003) ‘Spatial Disparities in Developing Countries: Cities, Regions and International Trade’, WIDER Project Meeting on Spatial Inequality in Development, Helsinki. Vousden, N. (1993) ‘Variable Specific Factors and the X-efficiency Cost of Protection’, Review of International Economics, 1(3): 234–42. Winters, L.A. (2000) ‘Trade, Trade Policy and Poverty: What Are the Links?’, School of Social Sciences, University of Sussex. . Wood, A. (1994) North–South Trade, Employment and Inequality: Changing Fortunes in a Skill-Driven World, Oxford: Clarendon Press. Wood, A. (1995) ‘How Trade Hurt Unskilled Workers’, Journal of Economic Perspectives, 9(3): 57–80. Wood, A. (1997) ‘Openness and Wage Inequality in Developing Countries: The Latin American Challenge to East Asian Conventional Wisdom’, World Bank Research Observer, 11: 33–57. Wood, A. (2003) ‘Could Africa Be Like America?’, Annual Bank Conference on Development Economics, Washington DC: World Bank. Wood, A. and Calandrino, M. (2000) ‘When the Other Giant Awakens: Trade and Human Resources in India’, Economic and Political Weekly, 30 December: 4677–94. World Bank (1990) India: Strategy for Trade Reform, Washington, DC: World Bank.

Index

Note: italic page numbers denote references to Figures/Tables.

Africa 4–5, 80, 85 agglomeration 121, 122, 125, 133, 154n1 Aghion, P. 154n2 agriculture 9; agricultural-intensive goods 87, 88, 89, 90, 91; GDP ratio 48, 49, 50 Amadeo, E.J. 86 Annual Survey of Industries 122, 141 Arellano, M. 95 Argentina 109 Asia 4–5, 80 Balakrishnan, P. 69 Bangladesh 4, 5, 87, 91–3, 91, 94, 96, 98, 136 Bell, C. 58 Bencivenga, V.R. 58 Beyer, H. 106 Bhagwati, J. 9, 20 Bond, S. 95 Brazil 86 business groups 29 capital 69, 72, 152n5 capital controls 17 capital goods: Effective Rate of Protection 21, 22; Import Coverage Ratio 23, 25; import licensing 14, 15; intra-industry trade 42, 43, 73, 77; liberalisation 3, 47, 66, 88, 112; manufacturing imports 41–2; pricecost margins 71; price ratio 24, 28; protection 28, 29; restrictions on 152n4; tariff reductions 16; Total Factor Productivity 70, 71, 73, 135–6; trade reforms 90, 134, 136; wage inequality 111

Central Statistical Organisation (CSO) 11, 12, 141, 142 Chakravorty, S. 154n1 Chamarbagwala, R. 107, 108 Chand, S. 68 chemical products 38, 41, 132 Chile 86, 106 China 10, 107, 120, 138 coastal regions 121, 125, 129, 130, 133, 137–8, 139 Colombia 109 comparative advantage 5, 80–1, 100, 112, 129, 138 competition: imperfect 3–4, 67, 154n2; perfect 100, 101; price 67, 74, 75, 76, 77–8, 136; quantity 67, 68, 74, 75–6, 77, 136; see also import competition composition effect 80, 81, 83 congestion costs 121 consumer goods: Effective Rate of Protection 21, 22, 24; import ban 15, 16; Import Coverage Ratio 23, 24, 25, 26; liberalisation 47; price-cost margins 71; price ratio 24, 29; protection 28; Total Factor Productivity 70, 71; trade reforms 134 credit 57, 58 CSO see Central Statistical Organisation Currie, J. 85 De Long, J.B. 56, 57, 61, 152n4 ‘debt accumulation’ hypothesis 58 ‘defensive innovation’ 6, 7, 102 Desai, P. 9, 20 Dixit, A. 64 Driffield, N. 68

166

Index

East Asia 107 economic growth 1, 2–4, 47–63, 135, 136, 140; equipment investment 55–7, 152n8; investment behaviour 48–55; ‘jobless growth’ 138, 140; poverty reduction 139; regional disparities 7; relative price of equipment 62–3 education 139–40 Effective Rate of Protection (ERP) 8, 20–1, 22, 23, 24, 24; relative skill intensity 114, 116; Total Factor Productivity 66, 72, 73, 75, 76; wage inequality 109, 110, 117, 118 efficiency 68, 69, 78, 84 electrical machinery 38, 40 employment 1, 79–98, 135, 136–7; Africa 85; decomposition of employment changes 92–5; factor content approach 82, 87–92, 98; female 36–7, 36; globalisation impact on 4; growth 31, 34, 35–6; growth accounting approach 82–3, 92–5, 98; labour demand modelling 83–4, 95–7, 98; Latin America 85–7; manufacturing 10, 30, 31, 34–7, 44–5, 79, 80–1; regional disparities 122–9, 124, 126, 127, 128, 130, 131; relative wages 108–11; service sector 9–10; stability of employment shares 113; structure of 33; trade reforms 106–8; see also labour; wages endogenous growth theories 2–3 equipment investment 48, 51–5, 51, 52, 53, 54, 135; economic growth 55–7, 59–61, 62–3; household investment 55; relative price of equipment 55, 56, 57–8, 59, 61–3, 62, 136, 152n8 ERP see Effective Rate of Protection Ettori, F. 151n1 Europe 6 exchange rate 16–18, 19, 68, 72; domestic prices 67, 74, 76, 77, 78, 136, 152n6; price wedge 24 export orientation 8, 42–4, 46, 81, 86; information technology 140; labour demand modelling 97; regional disparities 121, 129, 132, 133, 138; relative skill intensity 114, 116; wage inequality 118 export promotion 15, 17 exports 18, 41–4, 41, 135; Africa 5, 80; factor-intensity 87–92, 88, 90, 91, 112–13; GDP ratio 16, 17; impact on employment 80, 82–3, 89–92, 92, 93, 94–5, 96, 98, 136; increase in 34;

jewellery industry 153n11; labour demand modelling 84; regional disparities 129–32, 132; structure of 42; see also export orientation factor content approach 82, 87–92, 98, 101 factor endowments 5, 80, 112 factories 31, 39, 40–1, 40, 72, 74, 77 Fajnzylber, P. 106 FDI see foreign direct investment Feenstra, R. 81, 106 financial deepening 57, 58, 61, 152n8 financial intermediaries 58 financial services 48, 49, 50 food manufacturing 34, 36, 125, 132 foreign direct investment (FDI) 5, 9, 20, 106 Frankel, J. 152n2 Gandhi, Indira 61 Gandhi, Rajiv 15, 61, 136 Gang, I. 29 GDP see gross domestic product gender inequalities 108 Generalised Method of Movements (GMM) 95–6 GFCF see gross fixed capital formation Ghose, A. 86, 106 globalisation 4, 5, 6, 7, 107 GMM see Generalised Method of Movements Greenaway, D. 84, 151n2 gross domestic product (GDP): equipment investment rate 56, 57; exports and imports 16, 17; growth 48, 49, 59, 60, 62; investment ratio 48–52, 51, 53, 57, 58 gross fixed capital formation (GFCF) 48, 51 growth accounting approach 82–3, 92–5, 98, 153n11 Grubel-Lloyd measure 42, 66, 73, 75, 76, 151n1 Gurley, J. 58 Hanson, G. 81, 106, 108, 113 Harrison, A. 85, 106, 108, 113, 151n2 Hasan, R. 108 Havrylyshyn, O. 68 Heckscher-Ohlin (H-O) model 7, 120–1, 153n1; elasticity of labour demand 105; employment 80, 81, 85, 86, 137;

Index wage inequality 6, 7, 100–2, 103, 109, 111–12, 119, 154n2 Hine, R. 84 hotels and restaurants 48, 49, 50 household investment 52, 54, 55 human capital 87, 88, 88, 89, 90, 91, 112 ICR see Import Coverage Ratio import competition 9, 86, 92, 94; exchange rate depreciation 136; regional disparities 121, 129, 132, 133, 138 Import Coverage Ratio (ICR) 8, 20, 23–4, 25, 26, 27; domestic prices 77; quantity competition 68, 74, 75–6; relative skill intensity 114, 116; Total Factor Productivity 66, 72–3, 72, 75, 76; wage inequality 117, 118 import penetration 8, 74, 75–6, 77; capital goods 15; impact on employment 80, 81, 83, 92, 93, 95, 96; labour demand modelling 84, 97; relative skill intensity 114, 116; trends 42, 45; wage inequality 118 import substitution 1, 82 imports 18, 41–2, 43; factor-intensity 88, 89, 90, 91, 112–13; GDP ratio 16, 17; impact on employment 80, 82–3, 89, 92, 92, 94–5, 96, 98, 136; jewellery industry 153n11; labour demand modelling 84; licensing 14–16, 21; regional disparities 129–32, 133; structure of 44; see also import competition; Import Coverage Ratio; import penetration industrial policy 18–20, 61 industrialisation 29, 125, 129, 132, 133, 137, 154n1 industry classification 11, 68, 87, 95, 141–9, 153n7; employment shares 113, 114; regional disparities 122, 123, 132; relative employment 116–17, 117; wage inequality 111 inequality 1, 4–7; Heckscher-Ohlin model 100–2, 119; own-price elasticity of labour demand 103–5, 106, 108; regional 120–33, 137–8, 139; relative employment 108–11; skill-biased technological change 116–18, 119; Stolper-Samuelson effects 111–16; trade-induced technological progress 102–3, 119; wages 6–7, 99–119, 137 infant industry protection 28, 138 informal sector 10, 79, 86, 135

167

information technology 16, 140 interest rate 57, 58, 61 intermediate goods: Effective Rate of Protection 21, 22, 23, 109; Import Coverage Ratio 23, 25, 26; import licensing 14–15; intra-industry trade 42, 43, 73, 77; liberalisation 3, 47, 66, 88, 112; manufacturing imports 41–2; price-cost margins 71; price ratio 24, 28; protection 28, 29; Total Factor Productivity 70, 71, 73, 135–6; trade reforms 90, 134, 136; wage inequality 111 inventory investment 57, 61 investment 47, 48–55, 50, 51; economic growth 55–7, 59–63; household 52, 54, 55; inventory 57, 61; private 50–3, 53, 54, 57–8, 61, 132, 135, 136; public 50–3, 52, 53, 57–8, 61, 135; regional development 139; relative price of equipment 57–8, 59, 61–3, 136; technology-intensive sectors 138; see also equipment investment; foreign direct investment Jenkins, R.O. 4, 5 jewellery industry 153n11 job creation 10, 79, 86, 139 ‘jobless growth’ 138, 140 Jones, C.I. 62 Joshi, V. 151n3 Kenya 4, 5, 87, 91–3, 91, 94, 96, 136 Khambhampati, U. 68, 69, 87 Kohli, A. 61, 139 Krause, L.B. 87 Krishna, P. 69 Krueger, A.O. 80, 152n2 Krugman, P. 6 Kumar, U. 107, 108, 109, 154n11 labour: costs 32, 33, 135; demand 83–4, 87, 95–7, 98, 103–5, 106, 108, 118, 154n9; factor content approach 82; land/labour ratio 5, 80; manufacturing goods 94; productivity 31, 32, 33, 38, 44–5, 135, 136; Total Factor Productivity growth 65, 69, 72, 152n5; wearing apparel 93; see also employment; skilled workers; unskilled workers; wages labour demand modelling 83–4, 95–7, 98 Lall, S. 154n1 land/labour ratio 5, 80

168

Index

Latin America 80, 85–7, 106, 107, 109 Leamer, E.E. 80, 107, 153n1 Levine, R. 55–6 Levinsohn, J. 86 liberalisation 1–2, 7, 47, 112; economic growth 3, 151n2; efficiency-enhancing effects 78; elasticity of labour demand 105; impact on employment 85, 86, 93; imports 15, 88; industrial controls 18–20; market power 3–4; profitability 69; relative price of equipment 62; South Africa 5; Total Factor Productivity 7–8, 64, 65, 66, 74; wage inequality 106, 107–8, 111–12 Little, I.M.D. 151n3 Lucas, R.E. Jr. 2 machinery 56, 59 Magee, S. 28 Maloney, W.F. 106 manufacturing 2, 8–10, 13, 30–46, 135; Africa 85; comparative advantage 80–1; employment 31–7, 44, 45, 79, 80–1, 93, 94–5; equipment investment 61; export and import trends 16, 18, 41–4; factor-intensity 87–92; GDP growth 48, 49, 50, 140; industry classification 141–9; international trade impact on wages 86–7; ‘jobless growth’ 138; labour-intensive 5, 35, 45, 80–1, 87–90, 112–13, 135, 137, 138–9, 140; Latin America 85–6; price-cost margins 71; production patterns and trends 37–41, 45; protection 72; regional disparities 7, 122–9, 132, 133, 137–8, 139; skilled workers 110, 137; Stolper-Samuelson effects 116; Total Factor Productivity 70, 77, 136; trade reforms 106; wage inequality 109, 113, 119, 137 market power 3–4, 69, 70; Africa 85; incumbent firms 104; price competition 67, 74, 78; quantity competition 67, 74, 76, 77, 136 market structure 74 Mauritius 85 metropolises 125, 129, 130, 133, 137–8 Mexico 85–6, 106, 109 Milner, C. 68–9, 84, 85 Mishra, P. 107, 108, 109, 154n11 Mitra, D. 69 modernisation 15, 20, 29 Moreira, M.M. 86 Morocco 85

Najberg, S. 86 National Industrial Classification (NIC) 11 National Sample Survey Organisation (NSSO) 10, 108 natural resources 80, 87, 88, 89, 90, 91, 107, 137 ‘new economic geography’ 7, 120, 122, 129 NIC see National Industrial Classification Noguer, M. 152n2 non ferrous metals 35 non-tariff barriers (NTBs) 20, 21–3 NSSO see National Sample Survey Organisation NTBs see non-tariff barriers Open General Licence (OGL) 14–15, 42 openness 1, 4, 7, 16, 81; balanced city structures 121; impact on employment 92; labour demand modelling 84; productivity improvements 103; wage inequality 107 output effect 104 output growth 122–9, 124, 126, 127, 128, 130, 131; see also gross domestic product outward-orientation 1, 9, 82 Panagariya, A. 105, 154n8 Pandey, M. 29 Parikh, A. 69 Pero, V. 86 petroleum refineries 35, 40 plastic products 35, 40 political factors 29, 139 poverty 4, 9, 139, 140 price ratio 24, 28, 29 price wedge 8, 20–1, 24, 150n8; Total Factor Productivity 66, 73, 75, 76 prices 66–8, 72, 74, 77–8, 77; foreign 67, 68, 76, 101–2, 152n6; non-traded goods 102; price-cost margins 69–70, 71, 72, 75; relative factor 81; StolperSamuelson theorem 100–1, 112, 113–14 process effect 80, 81–2 productivity 2, 64–78, 135–6; economic growth 3, 135; impact on employment 81–2, 83, 84, 85, 92, 95, 96; labour 31, 32, 33, 38, 44–5, 135, 136; trade openness 103; trends and patterns

Index 69–70; see also Total Factor Productivity professional and scientific equipment 41 profitability 69, 100 protection 3, 8, 28–9, 134; exchange rate depreciation 78; infant industry 28, 138; Latin America 86; Total Factor Productivity 65, 72, 74; ‘true level’ of 11; wage inequality 109, 117–18, 119, 137; see also Effective Rate of Protection; Import Coverage Ratio; tariffs PROWESS database 12 Pushpangadan, K. 69 quotas 14, 23–4, 134, 136 RBI see Reserve Bank of India regional inequality 120–33, 137–8, 139 regulation 138–9 Renelt, D. 55–6 Reserve Bank of India (RBI) 12, 17–18 Revenga, A. 85–6 Robbins, D. 102, 106, 107 Rodriguez, F. 1 Rodrik, D. 1, 3, 61, 85, 103, 104–5 Romer, D. 2, 152n2 Rousseau, P. 58 rupee 17, 18, 19, 68 SBTC see skill-biased technological change scale economies 3, 20, 81, 121 scale effect 80, 81, 83, 104 Sen, K. 4, 5, 68, 152n8 service sector 9–10, 79, 140 Shaw, E. 58 shocks 73, 74, 84 Siscart, M. 152n2 skill-biased technological change (SBTC) 6, 109, 114, 116–18, 119, 137, 154n5 skilled workers: education 139–40; employment growth 115; information technology sector 140; relative employment 108–9, 110, 116–17, 117, 119, 137; wage inequality 6, 99, 100–3, 106–7, 108–11, 112–18, 119, 137 Slaughter, M. 6, 105, 154n2 small-scale sector reservation policies 138–9 Smith, B.D. 58

169

South Africa 4, 5, 87, 91–3, 91, 94, 96, 136 Srinivasan, T.N. 9 Stiglitz, J.E. 64 Stolper-Samuelson (S-S) theorem 81, 100–1, 111–16, 117, 154n2 structures investment 48, 51–3, 51, 52, 53, 54, 135; economic growth 61; household investment 55; relative price of 56 Subramanian, A. 61 substitution effect 104 Summers, L.H. 56, 57, 152n4 tariffs 14, 16, 78, 134, 136, 151n3; Effective Rate of Protection 20, 21; impact on wages 86; increase in 15; unskilled labour industries 109, 154n11; see also protection technology: comparative advantage 80–1; information technology 16, 140; skill-biased technological change 6, 109, 114, 116–18, 119, 137, 154n5; technology-intensive goods 87–8, 88, 89, 90, 91, 112, 138; trade-induced technological progress 3, 102–3, 119; trade reforms in Africa 85 textiles 34, 35–6, 37–8, 41, 135; contraction of textile industries 45; regional disparities 122, 125, 132; see also wearing apparel TFP see Total Factor Productivity Thoenig, M. 154n4 tobacco 36–7 Topalova, P. 68 Total Factor Productivity (TFP) 2, 3, 7–8, 64–6, 70–7, 135–6; growth 152n5; manufacturing 70, 72; protection 72; trade reforms 68; trends and patterns 69–70; see also productivity trade: employment 79–98, 136–7; Heckscher-Ohlin model 153n1; intra-industry 42, 43, 66, 72, 73, 75, 76, 77, 81, 152n3; manufacturing trade ratio 34; positive and negative impacts of 1; wage inequality 6–7, 100–5; see also exports; imports; openness trade balance 16, 19 trade policy 1–2, 14–29, 134; economic growth 2–4, 47–63, 135; industry level analysis 12; inequality 4–7, 137; measuring 7–8, 11; overview 14–16; political economy of protection 28–9;

170

Index

prices 66–8; productivity 64–78, 135–6; regional disparities 120–2, 129–33, 137; relative price of equipment 59; restrictiveness 20–6 trade unions 29 transport costs 121 transport equipment 34, 41 unemployment 6 United States (US) 6, 21, 24, 68, 101, 105, 153n1 unskilled workers: elasticity of labour demand 105; employment growth 115; labour-intensive goods 87, 88, 89, 90, 91, 98, 112, 138, 140; relative employment 108–9, 110; tariff reductions 154n11; wage inequality 6, 99, 100–3, 106–7, 108–11, 112–16, 119, 137 value added 69, 72; growth 31, 37, 38, 40, 44, 135, 152n5; regional disparities 124, 126, 127, 128, 130, 131 Varshney, Ashutosh 139 Venables, T. 7

Verdier, T. 154n4 Vietnam 4, 5, 87, 91–3, 91, 94, 96, 98, 136 wages 6–7, 99–119, 135, 137; Africa 85; agglomeration 121; growth 31–3, 31, 35, 36, 44, 45; Heckscher-Ohlin model 100–2, 119; international trade impact on 86–7; labour demand modelling 84, 97; Latin America 85–6; own-price elasticity of labour demand 103–5, 106, 108; relative employment 108–11; skill-biased technological change 116–18, 119; Stolper-Samuelson effects 111–16; Total Factor Productivity growth 152n5; tradeinduced technological progress 102–3, 119; trade reforms 106–8, 119 wearing apparel 35, 37, 40, 42–4, 89–90, 93, 113, 132 women 36–7, 36, 108 Wood, A. 4–5, 80, 83, 85, 102, 103, 107 Wright, P. 84, 85 X-inefficiency 64, 82, 135

Trade Policy, Inequality and Performance in Indian Manufacturing (Routledge Advances in South Asian Studies)

Starvation and India's Democracy (Routledge Advances Ibn South Asian Studies)

Studies in South Asian Film and Media

Political Mobilisation and Democracy in India: States of Emergency (Routledge Advances in South Asian Studies)

Sikh Nationalism and Identity in a Global Age (Routledge Advances in South Asian Studies)

Manufacturing Citizenship: Education and Nationalism in Europe, South Asian and China (Routledge Research in Education)

Trade Policy Issues in Asian Development (Routledge Studies in the Growth Economies of Asia, 20)

British Policy in India 1858-1905 (Cambridge South Asian Studies)

British Policy in India 1858-1905 (Cambridge South Asian Studies)

Military Control in Pakistan: The Parallel State (Routledge Advances in South Asian Studies)

Advances in Sustainable Manufacturing

Parsis in India and the Diaspora (Routledge South Asian Religion)

Health Inequality and Development (Studies in Development Economics and Policy)

South Asian Religions on Display: Religious Processions in South Asia and in the Diaspora (Routledge South Asian Religion)

Trade, Growth, and Inequality

Boats of South Asia (Routledge Studies in South Asia)

Social Policy in East and South East Asia: Hong Kong, Korea, Singapore and Taiwan (Routledge Advances in Asia-Pacific Studies)

Memory, Allegory, and Testimony in South American Theater: Upstaging Dictatorship (Routledge Advances in Theatre and Performance Studies)

Inequality and Public Policy in China

Deconstruction and the Ethical in Asian Thought (Routledge Studies in Asian Religion and Philosophy)

Parking Policy in Asian Cities

Trade Policy in Developing Countries

Trade Policy in Developing Countries

Desire and Motivation in Indian Philosophy (Routledge Hindu Studies Series)

Intra-Asian Trade and the World Market (Routledge Studies in the Modern History of Asia)

South Indian Factory Workers: Their Life and their World (Cambridge South Asian Studies)

Perception, Politics and Security in South Asia: The Compound Crisis of 1990 (Routledgecurzon Advances in South Asian Studies, 1)

Tibetan Exile Politics: An Analysis of the Tibetan Government-in-Exile (Routledge Advances in South Asian Studies)

Trade Policy, Inequality and Performance in Indian Manufacturing (Routledge Advances in South Asian Studies)

Starvation and India's Democracy (Routledge Advances Ibn South Asian Studies)

Studies in South Asian Film and Media

Political Mobilisation and Democracy in India: States of Emergency (Routledge Advances in South Asian Studies)

Sikh Nationalism and Identity in a Global Age (Routledge Advances in South Asian Studies)

Manufacturing Citizenship: Education and Nationalism in Europe, South Asian and China (Routledge Research in Education)

Trade Policy Issues in Asian Development (Routledge Studies in the Growth Economies of Asia, 20)

British Policy in India 1858-1905 (Cambridge South Asian Studies)

British Policy in India 1858-1905 (Cambridge South Asian Studies)

Military Control in Pakistan: The Parallel State (Routledge Advances in South Asian Studies)

Advances in Sustainable Manufacturing

Parsis in India and the Diaspora (Routledge South Asian Religion)

Health Inequality and Development (Studies in Development Economics and Policy)

South Asian Religions on Display: Religious Processions in South Asia and in the Diaspora (Routledge South Asian Religion)

Trade, Growth, and Inequality

Boats of South Asia (Routledge Studies in South Asia)

Social Policy in East and South East Asia: Hong Kong, Korea, Singapore and Taiwan (Routledge Advances in Asia-Pacific Studies)

Memory, Allegory, and Testimony in South American Theater: Upstaging Dictatorship (Routledge Advances in Theatre and Performance Studies)

Inequality and Public Policy in China

Deconstruction and the Ethical in Asian Thought (Routledge Studies in Asian Religion and Philosophy)

Parking Policy in Asian Cities

Trade Policy in Developing Countries

Trade Policy in Developing Countries

Desire and Motivation in Indian Philosophy (Routledge Hindu Studies Series)

Intra-Asian Trade and the World Market (Routledge Studies in the Modern History of Asia)

South Indian Factory Workers: Their Life and their World (Cambridge South Asian Studies)

Perception, Politics and Security in South Asia: The Compound Crisis of 1990 (Routledgecurzon Advances in South Asian Studies, 1)

Tibetan Exile Politics: An Analysis of the Tibetan Government-in-Exile (Routledge Advances in South Asian Studies)

Recommend Documents