Fertility, Family Planning and Population Control in China (Routledge Studies in Asia's Transformations)

Fertility, Family Planning, and Population Policy in China China’s one-child population policy, first initiated in 1979, has had an enormous effect on the country’s development. By reducing its fertility in the past two decades to less than two children per woman, and developing a family planning program focused heavily on sterilization and abortion, China has undergone a significant transition in status to a demographically developed country. Fertility, Family Planning, and Population Policy in China brings together contributions from leading scholars on such developments as family planning policy and contraceptive use, biological and social determinants of fertility, patterns of family and marriage, and China’s future population trends. As such it will be essential reading for academics, researchers, policy-makers, and government officials with an interest in China’s demography, fertility, and population policy. Dudley L.Poston, Jr is Professor of Sociology at Texas A&M University, College Station, Texas. Che-Fu Lee (deceased) was Professor of Sociology at the Catholic University of America, Washington, DC. Chiung-Fang Chang is Assistant Professor of Sociology at Texas A&M University, Kingsville, Texas. Sherry L.McKibben is Assistant Professor of Sociology at the University of Texas of the Permian Basin, Odessa, Texas. Carol S.Walther is a PhD Candidate at Texas A&M University, College Station, Texas.

Routledge Studies in Asia’s Transformations Edited by Mark Selden Binghamton and Cornell Universities, USA

The books in this series explore the political, social, economic, and cultural consequences of Asia’s transformations in the twentieth and twenty-first centuries. The series emphasizes the tumultuous interplay of local, national, regional, and global forces as Asia bids to become the hub of the world economy. While focusing on the contemporary, it also looks back to analyze the antecedents of Asia’s contested rise. This series comprises several strands: Asia’s Transformations aims to address the needs of students and teachers, and the titles will be published in hardback and paperback. Titles include: China in War and Revolution, 1895–1949 Peter Zarrow Confronting the Bush Doctrine Critical views from the Asia-Pacific Edited by Mel Gurtov and Peter Van Ness Japan’s Quiet Transformation Social change and civil society in the 21st century Jeff Kingston State and Society in 21st Century China Edited by Peter Hays Gries and Stanley Rosen The Battle for Asia From decolonization to globalization Mark T.Berger Ethnicity in Asia Edited by Colin Mackerras Chinese Society, 2nd edition Change, conflict and resistance Edited by Elizabeth J.Perry and Mark Selden

The Resurgence of East Asia 500, 150 and 50 year perspectives Edited by Giovanni Arrighi, Takeshi Hamashita, and Mark Selden The Making of Modern Korea Adrian Buzo Korean Society Civil society, democracy and the state Edited by Charles K.Armstrong Remaking the Chinese State Strategies, society and security Edited by Chien-Min Chao and Bruce J.Dickson Mao’s Children in the New China Voices from the Red Guard Generation Yarong Jiang and David Ashley Chinese Society Change, conflict and resistance Edited by Elizabeth J.Perry and Mark Selden Opium, Empire and the Global Political Economy Carl A.Trocki Japan’s Comfort Women Sexual slavery and prostitution during World War II and the US occupation Yuki Tanaka Hong Kong’s History State and society under colonial rule Edited by Tak-Wing Ngo Debating Human Rights Critical essays from the United States and Asia Edited by Peter Van Ness Asia’s Great Cities aims to capture the heartbeat of the contemporary city from multiple perspectives emblematic of the authors’ own deep familiarity with the distinctive faces of the city, its history, society, culture, politics, and economics, and its evolving position in national, regional, and global frameworks. While most volumes emphasize urban developments since the Second World War, some pay close attention to the legacy of the longue durée in shaping the contemporary. Thematic and comparative volumes address such themes as urbanization, economic and financial linkages,

architecture and space, wealth and power, gendered relationships, planning and anarchy, and ethnographies in national and regional perspective. Titles include: Singapore Wealth, power and the culture of control Carl A.Trocki Representing Calcutta Modernity, nationalism and the colonial uncanny Swati Chattopadhyay Hong Kong Global city Stephen Chiu and Tai-Lok Lui Shanghai Global city Jeff Wasserstrom Beijing in the Modern World David Strand and Madeline Yue Dong Bangkok Place, practice and representation Marc Askew Asia.com is a series which focuses on the ways in which new information and communication technologies are influencing politics, society, and culture in Asia. Titles include: The Internet in Indonesia’s New Democracy David T.Hill and Krishna Sen Asia.com Asia encounters the internet Edited by K.C.Ho, Randolph Kluver, and Kenneth C.C.Yang Japanese Cybercultures Edited by Mark McLelland and Nanette Gottlieb Literature and Society is a series that seeks to demonstrate the ways in which Asian Literature is influenced by the politics, society, and culture in which it is produced. Titles include:

Chinese Women Writers and the Feminist Imagination (1905–1945) Haiping Yan The Body in Postwar Japanese Fiction Edited by Douglas N.Slaymaker Routledge Studies in Asia’s Transformations is a forum for innovative new research intended for a high-level specialist readership, and the titles will be available in hardback only. Titles include: 1. Fertility, Family Planning, and Population Policy in China Edited by Dudley L Poston, Jr, Che-Fu Lee, Chiung-Fang Chang, Sherry L.McKibben, and Carol S.Walther 2. Genders, Transgenders and Sexualities in Japan Edited by Mark McLelland and Romit Dasgupta 3. Developmental Dilemmas Land reform and institutional change in China Edited by Peter Ho 4. Japanese Industrial Governance Protectionism and the licensing state Yul Sohn 5. Remaking Citizenship in Hong Kong Community, nation and the global city Edited by Agnes S.Ku and Ngai Pun 6. Chinese Media, Global Contexts Edited by Chin-Chuan Lee 7. Imperialism in South East Asia ‘A fleeting, passing phase’ Nicholas Tarling 8. Internationalizing the Pacific The United States, Japan and the Institute of Pacific Relations in War and Peace, 1919–1945 Tomoko Akami 9. Koreans in Japan Critical voices from the margin Edited by Sonia Ryang

10. The American Occupation of Japan and Okinawa* Literature and memory Michael Molasky * Now available in paperback. Critical Asian Scholarship is a series intended to showcase the most important individual contributions to scholarship in Asian Studies. Each of the volumes presents a leading Asian scholar addressing themes that are central to his or her most significant and lasting contribution to Asian studies. The series is committed to the rich variety of research and writing on Asia, and is not restricted to any particular discipline, theoretical approach, or geographical expertise. China’s Past, China’s Future Energy, food, environment Vaclay Smil China Unbound Evolving perspectives on the Chinese past Paul A.Cohen Women and the Family in Chinese History Patricia Buckley Ebrey Southeast Asia A testament George McT.Kahin

Fertility, Family Planning, and Population Policy in China Edited by

Dudley L.Poston, Jr, Che-Fu Lee, Chiung-Fang Chang, Sherry L.McKibben, and Carol S.Walther

LONDON AND NEW YORK

First published 2006 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 This edition published in the Taylor & Francis e-Library, 2006. “ To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to http://www.ebookstore.tandf.co.uk/.” © 2006 Dudley L.Poston, Jr, Che-Fu Lee, Chiung-Fang Chang, Sherry L.McKibben, and Carol S.Walther, selection and editorial matter; the contributors, their own chapters 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 A catalog record for this book has been requested ISBN 0-203-35644-6 Master e-book ISBN

ISBN 0-203-38955-7 (Adobe e-Reader Format) ISBN 0-415-32330-4 (Print Edition)

To our co-editor, Che-Fu Lee, who played a major role organizing, editing, and writing this book, but who did not live to see it published.

Requiescat in Pace!

Contents List of figures

xiii

List of tables

xv

List of contributors Preface

Prologue DUDLEY L.POSTON, JR AND CAROL S.WALTHER 1 Fertility and population policy: an overview QIUSHENG LIANG AND CHE-FU LEE PART I Family planning policy and contraceptive use 2 Patterns of induced abortion JUAN WU AND CAROL S.WALTHER 3 Patterns of sterilization CAN LIU AND CHIUNG-FANG CHANG PART II Family and marriage patterns 4 The impact of family structure on fertility FEINIAN CHEN 5 The impact of intermarriage on the fertility of minority women CHIUNG-FANG CHANG 6 Emerging patterns of premarital conception CAROL S.WALTHER 7 Changing patterns of desired fertility LI ZHANG, XIAOTIAN FENG, AND QINGSONG ZHANG PART III Biological and social determinants of fertility 8 Age at menarche and the timing of the first birth SHERRY L.MCKIBBEN

xviii xix

1 7 20

21 35 48

49 61 75 86 106

107

9 The effect of floating migration on fertility XIUHONG YOU AND DUDLEY L.POSTON, JR 10 The impact of language dialect on fertility XIAODONG WANG AND XIUHONG YOU PART IV Implications and the future 11 The managed fertility transition in rural China and implications for the future of China’s population CHE-FU LEE AND QIUSHENG LIANG 12 China’s demographic destiny: marriage market implications for the twenty-first century DUDLEY L.POSTON, JR AND KAREN S.GLOVER Index

122 140 152

153

167

182

Figures 1.1

Total fertility rate (average number of children per woman) for urban and rural places and all China, 1949–1999

12

1.2

Crude rates of birth, death, and natural increase in China, 1949– 14 2000

2.1

Probability of surviving the hazard of an abortion after the birth 27 of the first child, stratified by rural and urban residence, China, 1997

3.1

Kaplan-Meier estimates of Chinese women surviving the hazard 41 of sterilization, 1997

3.2

Kaplan-Meier estimates of Chinese women surviving the hazard 42 of sterilization, by residence, 1997

4.1

Cumulative distribution of first birth by duration months since date of marriage, CHNS

53

4.2

Residential patterns of parents-in-law and parents of married women, CHNS, 1993

54

5.1

Comparison of the observed distribution of CEB, with the univariate Poisson distribution with mean of 2.82

65

8.1

Kaplan-Meier survival estimates for the hazard of a first birth: Chinese Han women, 1997

113

8.2

Kaplan-Meier survival estimates for the hazard of a first birth: Chinese minority women, 1997

114

9.1

Frequency distribution of CEB variable, and univariate Poisson 127 distribution based on mean of 2.148

9.2

Generation of the migration status variable

129

11.1 Age-specific fertility rates by urban and rural places: all China

158

11.2 Age-specific fertility rates by urban and rural places: Hebei province

159

11.3 Population growth rates of China and India, 1950–2000

162

12.1 TFRs: China, 1950–2001

168

12.2 SRB: Mainland China and the United States, 1980–2001

171

12.3 Sex ratios by age: the United States, July 1, 1999

173

12.4 Males and females born in Mainland China, 1980–2001

174

12.5 Number of excess males at marriageable age of 22: Mainland China, 2000–2021

174

Tables 1.1

Population, birth, and death rates of China, 1949–2000

10

2.1

Descriptive data: 11,762 women with at least one child, China, 1997

27

2.2

Cox proportional hazards models for the hazard of having abortion for the pregnancy following a woman’s first child, China, 1997

31

3.1

Descriptive data of 10,406 currently married women with at least one child, China, 1997

40

3.2

Hazard ratios of female sterilization in China, 1997

43

4.1

Descriptive statistics of selected sample characteristics (N=218) 56

4.2

Woman month risk of first birth, 1989–1993, CHNS (N=12432) 57

5.1

Descriptive data for ever-married women aged 15–49, by ten ethnic groups, China, 1990

5.2

Mean numbers of CEB for exogamous women and endogamous 68 women, by ten ethnic groups, China, 1990

5.3

Reduced Poisson regression models predicting number of CEB: 69 minority ever-married women, aged 15–49, China, 1990

5.4

Poisson regression models predicting number of CEB, by rural and urban, minority ever-married women, ages 15–49, China, 1990

71

6.1

Dependent and independent variables: means and standard deviations, China, 1997

80

67

6.2

Logits and odd ratios of Chinese women having a premarital conception, 1997

82

7.1

Patterns of the desired number of children in rural areas of China

90

7.2

Pattern of the desired number of children in urban areas of China

94

7.3

Patterns of the desired gender of children in rural areas of China 95

7.4

Patterns of the desired gender of the next child and the number of existing children in rural China

97

7.5

Patterns of the desired gender of children in urban areas of China

99

8.1

Descriptive statistics of Chinese Han women for the hazard of a 112 first birth, 1997

8.2

Descriptive statistics of Chinese minority women for the hazard 112 of a first birth, 1997

8.3

Cox Proportional Hazard analysis of the hazard of a first birth: ever-married Han females, China, 1997

116

8.4

Cox Proportional Hazard analysis of the hazard of a first birth: ever-married minority females, China, 1997

117

9.1

Social demographic variables by migration status: ever-married 126 women in China, aged 15–49, 1990

9.2

Percentage distributions of education status, for nonmigrants 130 and groups of migrants: ever-married Chinese women, aged 15– 49, 1990

9.3

Poisson regressions of number of CEB: ever-married Chinese women, aged 15–49, 1990

131

9.4

Poisson regressions of number of CEB: ever-married Chinese

135

women, aged 15–49, 1990 (rural nonmigrants and rural-tourban floating migrants) 9.5

Poisson regressions of number of CEB: ever-married Chinese women, aged 15–49, 1990 (urban nonmigrants vs. rural-tourban floating migrants and urban permanent migrants)

136

10.1 Chinese dialects and their major locations in China

143

10.2 Lu’s quantified indices for selected cities and the dialects groups

144

10.3 The coding of Chinese dialects by province

145

10.4 Descriptive statistics of independent and dependent variables

147

10.5 Zero-order correlations of dependent and independent variables 147 10.6 Standardized regression coefficients of effects of independent variables on CBR

147

11.1 Crude rates of birth, death and natural increase for urban and rural China, selected years, 1957–1999

155

11.2 Development indices of China and India around the year 1950

161

Contributors Chiung-Fang Chang is Assistant Professor of Sociology at Texas A&M University, Kingsville, Texas. Feinian Chen is Assistant Professor of Sociology at North Carolina State University, Raleigh, North Carolina. Xiaotian Feng is Professor of Sociology at Nanjing University, Nanjing, China. Karen S.Glover is a Doctoral Candidate in Sociology at Texas A&M University, College Station, Texas. Che-Fu Lee (deceased) was Professor of Sociology at the Catholic University of America, Washington, DC. Qiusheng Liang is Research Associate at the Catholic University of America, Washington, DC. Can Lin is a JD Candidate at the School of Law at the University of Texas at Austin. Sherry L.McKibben is Assistant Professor of Sociology at the University of Texas of the Permian Basin, Odessa, Texas. Dudley L.Poston, Jr is Professor of Sociology, and Holder of the George T. and Gladys H.Abell Endowed Professorship in Liberal Arts, at Texas A&M University, College Station, Texas. Carol S.Walther is a Doctoral Candidate in Sociology at Texas A&M University, College Station, Texas. Xiaodong Wang is a Doctoral Candidate in Sociology at Texas A&M University, College Station, Texas. Jnan Wu is a Doctoral Candidate in Finance at Texas A&M University, College Station, Texas. Xiuhong You is a Doctoral Candidate in Sociology at the University of Texas at Austin; and a Research Associate in the Office of the State Demographer. Austin, Texas. Li Zhang is a Doctoral Candidate in Sociology at Texas A&M University, College Station, Texas. Qingsong Zhang is a Human Resources Officer at the Huawei Company, Shenzhen, China.

Preface The idea of a book on fertility, family planning, and policy in China was introduced at the meetings of the North American Chinese Sociologists Association (NACSA) held in Anaheim, California in August of 2001. Earlier versions of several of the chapters in this book were presented at the NACSA conference. Che-Fu Lee of the Catholic University of America served as the NACSA conference organizer. Since the authors of several of the NACSA conference papers were at the time graduate students of Dudley Poston at Texas A&M University, Lee and Poston discussed developing a China fertility book using the NACSA papers as a base. They approached several other scholars who did not make presentations at the NACSA conference to write chapters for the book, to help fill voids and to flesh out theoretical and empirical aspects of Chinese fertility and family planning not covered in the conference presentations. Poston then asked Chiung-Fang Chang, Sherry McKibben, and Carol Walther to assist him and Lee in putting the book together. In the subsequent development of this book, we are very much indebted to Mark Selden of the State University of New York, Binghamton, who serves as a China book series editor for Routledge, for assisting us in the preparation of a book proposal; to Zoë Botterill of Routledge Publishers for working with us in developing the book contract; and to Helen Baker of Routledge Publishers for helping us get the book ready for production. As the book chapters were revised and/or written, the five editors each, in turn, read and edited them. The edited chapters were then returned to the authors for revision. Poston was responsible for the final reading and editing. All the chapters were then copy-edited at Texas A&M University by Chris Lewinski, who then discussed and reviewed the final changes and edits with Poston. We thank her for her dedicated and timely work, and all the others who assisted us. When the book was in the final stages of editing in February of 2005, we were saddened to lose our co-editor Che-Fu Lee, who died at the age of 64. Without Che-Fu’s encouragement and hard work, this book would never have been completed. As already noted, Che-Fu was the organizer of the 2001 annual meeting of the North American Chinese Sociologists Association where several of the chapters in this book were first presented. He was very instrumental in commissioning several of the other chapters, he provided valuable editorial assistance to the authors of all the chapters, and he wrote two of the chapters with his colleague Qiusheng Liang. We dedicate our book to him. Dudley L.Poston, Jr College Station, Texas Chiung-Fang Chang Kingsville, Texas Sherry L.McKibben Odessa, Texas Carol S.Walther College Station, Texas

Prologue Dudley L.Poston, Jr and Carol S.Walther This book focuses on fertility, family planning, and population policy in China—the most populated country in the world. In 2004 China had nearly 1.3 billion inhabitants. After India, the United States is the third most populated country, with a population in 2004 of 293 million. But China has a land mass slightly less than that of the United States (9.6 million square kilometers of surface area compared to 9.8 million of the United States) with a population that is 4.4 times larger. China has reduced its fertility in the past two decades to less than two children per woman, has a family planning program focusing heavily on sterilization and abortion, and a population policy based on one child, and no more than two children, per woman. The country’s fertility transition, family planning programs, and one-child policy have captured the attention of academicians, researchers, policy practitioners, government officials, and laypeople the world over. Indeed, some have seen China’s experiences as providing important lessons for the demographic transitions of many countries in the developing world. Others have denounced the policy’s system of quotas, and particularly forced abortions, as gross violations of human rights. What is certain is that the transition has made it possible for China to achieve in a relatively brief time the status of a demographically developed country. Its fertility, family planning, and policy dynamics that are analyzed in the chapters of this book need to be considered in this context. This Prologue first places these demographic issues in a historical perspective. China today is not demographically a country with an aged population. In 2000, only one-tenth of China’s population was over age 60, compared to 16 percent for the United States. By comparison, in 2000, 10 countries, all in Europe, had more than 20 percent of their populations over age 60 (United Nations 2003). But China is one of the oldest countries in existence. Statistics on the size of China’s population suggest that the country had a population of around 60 million people at the time of Christ (Durand 1960). Of course, Chinese civilization began much earlier than the time of Christ, with the Xia Dynasty, the first dynasty of China, lasting from about the twenty-first century BC to the sixteenth century BC. There are, however, no demographic records of the Chinese population in the centuries before Christ, other than an estimate of about 13 million at the start of the Xia Dynasty (Sun 1988:9), a figure whose accuracy is difficult to establish. During the Han Dynasty, China took a population count in the second century AD, and it showed a population size of just under 60 million people (Banister 1992). The population increases and decreases over the almost 20 centuries since the time of Christ have usually been associated with dynastic growth and decay. Typically, the beginning of a new dynasty was followed by a period of peace and order, cultural development, and population growth. As population density increased, it often exceeded the availability of food, and the Malthusian struggle for existence was intensified. Then there would come a period of pestilence and famine resulting in a reduction in the size of the population. Two thousand years of Chinese records and archives show that for all the centuries prior to the seventeenth century, China’s population size increased to around 50–60

Fertility, family planning, and population policy in china

2

million before declining. Indeed at the start of the Ming Dynasty (in 1368) the size of China’s population was only slightly larger than it was at the time of Christ. For all the dynasties up until China’s last dynasty, the Qing (1644–1911), China’s population swayed roughly with the rise and fall of a dynasty (most dynasties reigned for about 200– 300 years). The population grew in the initial years of the dynasty, then fell, so that onethird or sometimes one-half of the original population was decimated. Mortality then was too high to allow much of an increase in population. To illustrate, from 1400 to 1500, the size of the Chinese population grew by around 25 million. It grew by another 50 million from 1500 to 1600. But since the mid-1700s after the establishment of the Qing Dynasty, slight reductions in mortality enabled the population to continue growing. By 1850 there were roughly 420 million people in the country, 6–8 times the traditional level (of 60–80 million) that was the demographic norm 200 years or so previously. The Qing Dynasty was supremely successful at living up to the Chinese ideal of “numerous descendants.” It is indeed ironic that by achieving this ideal, not only was the Qing Dynasty wiped out, but China’s dynastic system of almost four thousand years was eradicated. Previously, declines in population resulted in the collapse of the dynasties. The Qing fell in 1911, among other reasons, because the population became too large. By the date of the birth of the People’s Republic of China (PRC) on October 1, 1949, the population exceeded 500 million. It is at this point in the country’s demographic narrative that we may turn to this book’s first substantive chapter “Fertility and population Policy” by Qiusheng Liang and Che-Fu Lee. The authors revisit the process of the formation of the Chinese government’s population policies over the past half-century. Their primary aim, through a systematic review of the evolving policies over a period of five decades, is to provide a comprehensive picture of the dynamics between policy decisions and their implementation at different points in time since the 1950s. The population policy considerations are reviewed in the context of the political economy and are demarcated into three stages over the past 50 years: (1) the harbinger of population planning in the 1950s, (2) the chaotic decade of the 1960s and the establishment of the birth-control institution in the 1970s, and (3) a policy experiment followed by a decentralization since the early 1980s. Liang and Lee’s policy reviews serve to frame the discussions of China’s current fertility by presenting and discussing year-to-year baseline fertility data since the early 1950s. They show how the “on-again and off-again” fertility control policies of the 1950s and 1960s interacted with non-fertility related policies and ideologies to keep fertility rates high, especially in the rural areas. Their discussions provide an important perspective for the contemporary analyses of fertility that follow in later chapters of this book. Two key features of China’s family planning policy are abortion and sterilization. These are the subjects of Chapters 2 and 3 and comprise Part I of the book. The chapter by Juan Wu and Carol Walther deals with induced abortion. In 1957, induced abortion was introduced by the Chinese government as part of the first birth control campaign. This method was to be used during the first ten weeks of pregnancy. In later decades, however, for some women, induced abortion became the primary form of birth control, although the government discourages this practice. In this chapter, the authors show that population policy factors play a significant role in affecting a woman’s chances of

Prologue

3

aborting the pregnancy that occurs following the birth of her first child. The authors also show that if the first child is a daughter, the woman has a lower chance of aborting her next pregnancy. Chapter 3 is by Can Liu and Chiung-Fang Chang and deals with patterns of sterilization. Sterilization is the world’s most widespread form of birth control and has contributed significantly to China’s fertility decline. This chapter analyzes women’s knowledge of contraception as a significant predictor of sterilization. Women with greater contraceptive knowledge are shown to be less likely to experience sterilization than women with less contraceptive knowledge, among other reasons. Liu and Chang suggest that improvements in the education of couples about contraception and improved communication between clients and providers can achieve “informed choice.” Part II deals with family and marriage patterns. Feinian Chen’s Chapter 4 on “The impact of family structure on fertility” introduces the concept of a “modified extended family” in her analysis of fertility. Instead of focusing only on co-residence, she extends the boundary of the household by also incorporating parents who live close by, a residential type that is sometimes referred to as “quasi-co-residence.” One of the key mechanisms of how family structure affects fertility is via the reduced opportunity costs of children. In contemporary China, while grandparents do not necessarily bear the direct economic costs of raising children, they act as important alternative childcare givers. Indeed grandmothers are often identified as the most important caregivers other than parents themselves. But caregiving is not just limited to co-residing grandparents. Grandparents who live nearby are often just as likely to help. Thus, with parents in the household or living nearby as future childcare helpers, couples may choose to start having children earlier than if no such choice was available. Moreover, young couples could be under stronger normative pressures from their parents to start the childbearing process earlier for the purpose of carrying on the family line. Chen’s results show that the extended family living arrangements have a clear impact on the transition to the first birth. Co-residence with the husband’s parents substantially increases the likelihood of an early first birth. Moreover, when Chen expands her definition of extended family to include co-residence and quasi-co-residence with parents, a very similar effect is found. Her findings clearly demonstrate the importance and relevance of the family context, and the fact that this is not bounded by the household. Chiung-Fang Chang (Chapter 5) deals with the effects of intermarriage on the fertility of minority women. She measures intermarriage in two ways by coding the ethnicities of the woman and her spouse; if the minority woman married a Han man; or if the minority woman married a man from another minority ethnic group. In her analyses the reference group is the endogamous couple, that is, minority women who marry in the same ethnic group. She shows that both intermarriage variables have negative effects on fertility irrespective of whether the intermarriage was between a minority and a Han or between a minority and another minority. And these effects are independent of an assortment of other socioeconomic, demographic, and policy variables. Exogamous women, that is, women who marry outside their ethnic group, have lower levels of fertility than endogamous women. In particular, for minority women who intermarried with Han, they appear to be more willing to accept the norms and values of the mainstream Han culture.

Fertility, family planning, and population policy in china

4

Chang (Chapter 5) writes that the issue of ethnic assimilation is particularly important when one considers the fertility rates of the minority populations. Since intermarriage is an important consequence of ethnic assimilation, she notes the need to move beyond the research reported in her chapter. For instance, one area requiring attention is the ethnic identity issue of the second generation of intermarried couples in relation to the dynamics of overall population growth and the future population policy. Attention also needs to be directed to the phenomenon of multiple ethnicities and ethnic inequalities, an area just now being addressed in the West, but not yet in China. Chapter 6 by Carol Walther analyzes emerging patterns in China of premarital conception. More than 20 years ago Rindfuss and Morgan (1983) claimed that a silent but profound sexual revolution was occurring throughout Asia in which couples were moving away from arranged marriages to romantic marriages, thus leading to increases in premarital conceptions. In this chapter, Walther finds that among other factors, having friends with more nontraditional attitudes toward premarital sex, and having had sex education, are positively correlated with premarital conceptions. This is a new area of inquiry in China, and further study is needed regarding the impact of sex education on romantic relationships, premarital conceptions, and contraceptive usage. Chapter 7 is by Li Zhang, Xiaotian Feng, and Qingsong Zhang and focuses on desired fertility in number and gender preference. Using a meta-analysis, the authors show that there is a declining pattern in the desired number of children and son preference from the 1970s to the 1990s in both the rural and urban areas of China. A family that consisted of two children was preferred by rural and urban citizens in the 1980s, but in the 1990s, more urban people preferred one child. Generally speaking, urban citizens have lower desired number of children compared to their rural counterparts. Also, Beijing and Shanghai residents desired a small number of children, and Fujian and Guangdong residents desired a higher number of children. Part III moves to a consideration of biological and social determinants of fertility. The biological effect that is analyzed is age at menarche, and the social effects studied are migration and language dialect. Sherry McKibben’s Chapter 8 investigates whether a biological factor such as age at menarche has an independent effect on the likelihood over time of having a first birth. Her central goal is to model the likelihood of giving birth to a first child during the period of time starting with the onset of a woman’s menarche. The analysis is conducted for Han women and for high fertility minority women. She finds that the effect of age at menarche is consistently significant and positive for both minority and Han Chinese women. As a woman’s age at menarche increases, her likelihood of having a first birth also increases. McKibben notes that among the Han women, as they get older when reaching menarche, their hazard of a first birth increases by 3.6 percent and for the minority women, by 2.4 percent. Biologically, women who reach menarche at a later age have a shorter period of subfecundity and are thus more likely to experience a first birth sooner after reaching menarche than those women with an early age at menarche. The findings reported in McKibben’s chapter support this hypothesis. A woman who reaches menarche early is likely to waste many of her most viable follicles before she is ready to conceive, and her ovulation cycles will be spaced further apart. Therefore, if menarche is postponed, the woman’s chances of conception increase and she will experience her first birth soon after reaching menarche.

Prologue

5

Xiuhong You and Dudley Poston (Chapter 9) investigate the effect of floating migration on fertility. Many in China believe that floating (or temporary) migrants have higher fertility than non-migrants. They hold that one of the main reasons why some floaters migrate is to escape family planning controls in order to have more children. You and Poston point out that such an expectation is contrary to the prediction of most migration theories that expect migrants to have lower, not higher, fertility than nonmigrants. To date there has been no comprehensive empirical examination of the fertility of floating migrants and the degree to which their fertility surpasses that of non-migrants and permanent migrants. You and Poston show that after controlling for relevant factors, rural to urban floating migrants have lower, not higher, fertility than those left behind. Floating migrants do not have fertility rates higher than most of the population and they should not be referred to as “Child-bearing Guerillas,” as they often are in the popular press. Chapter 10 in Part III by Xiaodong Wang and Xiuhong You inquires whether language dialect has a significant effect on fertility. Expanding on the diffusion perspective of Susan Watkins (1991), the authors show that speaking different dialects does indeed appear to affect fertility. Those provinces speaking Mandarin and those provinces speaking dialects similar to Mandarin have fewer births than those not speaking Mandarin. Another interesting finding is that among the provinces, there is a significant negative relationship between the percentage of households with television sets and fertility. The researchers suggest that as China becomes more dominated by Mandarin, and as it experiences increased access to mass media, especially television, the crude birth rates overall, and differences in birth rates among provinces should further decrease. Part IV deals with implications and the future. It consists of two chapters. Chapter 11 is by Che-Fu Lee and Qiusheng Liang and evaluates the effects of birth control in China. China’s population of 1.27 billion, as reported by the most recent 2000 census, serves as a baseline reference for their analysis. Various projections extrapolating China’s population and its growth rate in earlier eras, with explicit assumptions of the policy effects (or lack thereof), are compared with China’s population in 2000. Lee and Liang demonstrate that China’s population today is at least 500 million less than it would otherwise have been as a result of changes during the last three decades in government efforts to implement a compulsory population control policy. Of the more than 500 million births averted, the authors show that the administrative and close monitoring of citizens’ reproductive behavior since 1970 in both the urban and rural regions accounted for more than half of the reduced population growth. The spurt of the radical “one-child” policy in the early 1980s was soon adjusted in 1984 in the rural areas to allow for a second birth if the first was a girl. The immediate effect was seen in a rise in period fertility rates in the subsequent years of the 1980s. Allowing for the possibilities of the underreporting of unplanned births, especially in rural places, fertility declined further in rural China in the 1990s to nearly converge with urban China. Chapter 12 by Dudley Poston and Karen Glover inquires into the future marriage market implications of China’s unbalanced sex ratio at birth, another outcome related to the one-child and modified one-child policy. The authors first show that in China in every year, beginning in 1980 through to the year 2001, many more boys have been born than girls. Indeed in the year 2000, almost 120 boys were born for every 100 girls. The authors

Fertility, family planning, and population policy in china

6

estimate that when the number of boys, who have already been born in China, reach marriage age, they will all not be able to find Chinese women to marry. They estimate that there are more than 23 million boys already born, who will not be able to find brides when they reach marriage age. Poston and Glover (Chapter 12) consider a number of possibilities about what these numerous extra boys will do when they are unable to find brides. The migration of foreign mail-order brides, a solution available to some rich countries such as Japan, is deemed unlikely. So too is a major increase in levels of homosexuality. Polyandry too is not thought by the authors to be a likely solution. The likely possibility is that these Chinese bachelors will never marry and will have no choice but to develop their own lives and livelihoods in the absence of nuclear families. Many are likely to live with one another in “bachelor ghettos” either in the countryside or in cities where commercial sex outlets are available. There is thus the real possibility of major increases in China in the incidence of HIV and AIDS owing to their transmission mainly through commercial sex outlets. Research also suggests that these bachelors will be more prone to crime than if they were married. This possibility has alerted some to the potential increases in crime in China’s future, and perhaps political ramifications, resulting from these excess males. The authors conclude that no one knows for sure what this excess number of young Chinese males will do. The only known fact is that there have already been born in China over 23 million more Chinese boys than there will be Chinese girls for them to marry.

References Banister, J. 1992. “A Brief History of China’s Population” pp. 51–57 in Dudley L.Poston, Jr and David Yaukey (eds), The Population of Modern China. New York: Plenum. Durand, J.D. 1960. “The Population Statistics of China, A.D. 2–1953.” Population Studies, 13:209–249, 255–256. Rindfuss, R.R. and S.P.Morgan. 1983. “Marriage, Sex, and the First Birth Interval: The Quiet Revolution in Asia,” Population and Development Review, 9(2):259–278. Sun, J. 1988. “The Chinese Population: Its Size and Growth.” pp. 9–14 in China Financial and Economic Publishing House, New China’s Population. New York: Macmillan. United Nations. 2003. World Population Prospects: The 2002 Revision. Two volumes. New York: United Nations. Watkins, S.C. 1991. From Provinces into Nations: Demographic Integration in Western Europe, 1870–1960. Princeton, NJ: Princeton University Press.

1 Fertility and population policy An overview Qiusheng Liang and Che-Fu Lee China, the most populous nation in the world, reported a population of 1.27 billion in the year 2000. Compared with the population of the United States, 280 million, China had 1 billion more people at the turn of the new millennium. Given 6 billion people over the whole world, Chinese represented a little more than one-fifth of the human race on earth. In late 1979, when the Chinese Government launched an ambitious one-child-percouple population policy, the target was to limit the population of the country within 1.2 billion by the end of the twentieth century. To implement the “one-child” policy the government devised various “carrot and stick” measures to reward couples who pledged to have no more than one child and to penalize those who exceeded the one-child limitation. Anecdotal reports of forced abortions by local officials who were overzealous in meeting reproductive quotas produced international uproars over human rights violations and calls for sanctioning foreign aid to China’s family planning programs. Many were left with the impression that China had only one population policy, that is, a policy permitting only one child per couple. This fragmented picture of China’s attempts to control its population problem tended to obscure the historical processes of social construction of population growth and efforts to control it in the context of broader development strategies. This chapter revisits the process of the formation of the Chinese government’s population policies over the past half-century. The primary aim, through a systematic review of the evolving policies over a period of five decades, is to provide a comprehensive picture of the dynamics between policy decisions and their implementation at different points in time since the 1950s. Developments of population policy in the context of political economy, which are discussed here in some detail, can be demarcated into three stages over the past 50 years: (1) the harbinger of population planning in the 1950s, (2) the chaotic decade of the 1960s and the establishment of the birth-control institution in the 1970s, and (3) a policy experiment followed by a decentralization since the early 1980s. These policy reviews will serve to frame the discussions of China’s current fertility by showing and discussing year-to-year baseline fertility data for the country since the early 1950s. It will be shown how the “on-again and off-again” fertility control policies of the 1950s and 1960s interacted with non-fertility-related policies and ideologies to keep fertility rates high, especially in the rural areas. The discussions in this chapter will provide an important perspective for the more contemporary analyses of fertility that follow in later chapters of this book (also see Poston and Glover’s Chapter 12 in this book for more discussion).

Fertility, family planning, and population policy in china

8

The 1950s—harbinger of population planning The Korean war (1950–1953) delayed China’s post-civil-war reconstruction for a few years after the establishment of the new People’s Republic in 1949. On July 1, 1953, China conducted for the first time a national census of its population. The census was required for the preparation of the official First Five-Year Plan and for the apportioning of political representation, the same political purpose for which the United States took its first census in 1790 (Baumle and Poston 2004). The first Plenary Congress of the People was held in 1954. The results of the first population census revealed that China had a total population of 602 million in 1953, a figure that was one-third larger than 450 million, the number commonly believed to be the size of the population in China at that time. Many concerned scholars and government officials, especially those involved with the drafting of the First Five-year Plan, were alarmed by the unexpectedly large number of people that they had to deal with. The first communication at the upper echelon of the ruling party pertaining to the issue of population may be traced to Deng Xiaoping, the late supreme leader and architect of China’s economic reform from the late 1970s to the 1990s. He was then the Secretary General of the Chinese Communist Party (CCP). In his letter (1954) responding to Madam Zhou Enlai (Mr Zhou was the Premier), he stated, “I think it is necessary and beneficial to promote effectively the use of contraception” (Ca et al. 1999:112–113). At the first meeting of the People’s Congress in 1954, a delegate and a noted scholar in China, Shao Lize proposed, “in our country, we may set aside the issue of abortion, but health knowledge about contraception must be propagated. In addition, practical guides and necessary means and material supplies of contraception must be provided” (People’s Daily 1954). Apparently the top leaders were concerned about population pressure, and this was echoed in the public forum. Shao subsequently published a pamphlet entitled, “Basic Knowledge Regarding the Propagation of Contraception,” and continued to write a series of related articles in the public media. Shao would become the first person in the history of the new Republic to openly advocate the practice of family planning or birth control (Ca et al. 1999:113). Under the leadership of President Liu Shaoqi, a meeting was called at the ministry level in December 1954, to gather inputs on family planning from various government offices. At that meeting, Liu pronounced the support of the Party for planned fertility. In 1955, the Party’s Central Committee approved a report submitted by the Ministry of Health, in which it was said that regulation of birth is of concern with the living of people’s lives. It is an important policy matter. Given the conditions at this point of our history, for the benefits of the nation, family, and future generations, the Party supports appropriate regulation of births. The Party Secretariats at all levels of government (except in the autonomous regions of the ethnic minorities) must work with the masses to spread this policy of the Party. (She 1988:120–121) This became the first official document known as “the Directive on the Problem of Population Control.” This policy decision was then followed by the organization of a

Fertility and population policy: an overview

9

leading team for Planned Births, headed by Chen Muhua, who later became the first female vice-premier. Thereupon, the first official policy on population was announced, and the first government office in charge of implementing population policy was born. In subsequent years, there were other reiterations of the need for population control in relation to economic development. For example, an official document on agricultural development distributed in 1957 included population control as an intrinsic component of the policy. On the other hand, beginning with the 1955 “Directive” followed by other government initiatives in population control, a heated debate was unfolding in academic circles on the pros and cons of the perception of the so-called “population problem.” In the year 1957 alone, it is estimated that more than eighty scholars of all disciplines engaged themselves in writing hundreds of articles on this debate. The most memorable was by an economist and a prominent intellectual, Ma Yinchu, the then president of Beijing University. He treaded a thin line between Marxian ideology and Malthusian theory by publishing a booklet called, “The New Population Theory” (Ma 1997). His major argument was to negate the Marxist doctrine that a socialist society could not have a population problem. He maintained that in a socialist planned economy, population reproduction must be an integral part of the planning. The opposite camp, characterizing themselves as orthodox Marxists, came forth immediately to accuse Ma of adopting bourgeoisie Malthusian population theory. They claimed that the Malthusian view of the population problem blamed the victims, or the oppressed class, for the problem generated by a capitalist system, and hence was antithetical to Marxist doctrine. The misfortune of Ma’s new population theory was, ironically, aggravated by the Great Leap Forward that Chairman Mao launched in 1958 in the midst of the Party’s internal power struggle, an Anti-Rightist political movement. Mao led this political movement to eradicate the socalled capitalist revisionists within the Party. Under fire from this political movement, Professor Ma was labeled as a political rightist and in 1960 was forced to resign the presidency of the most prestigious university in China. The discussions of the population problem and ways to deal with it were muted not only by political pressures but by the economic calamity during the years 1959–1962.

The 1960s—the haunting concern of population in a chaotic decade If the idea of a planned population and regulated reproduction dawned on the horizon of China in the 1950s, its debate and controversy were limited to the elite circle of the Party leaders and concerned academicians. For the general masses, controlling the number of births contradicted the age-old cultural tradition which deemed a large family as a blessing of good fortune. To the Marxist ideologues, this was antithetical to the building of a Utopian socialist society. Aside from cultural and ideological resistance, the Chinese economy and medical technology in the 1950s were not equipped with a public health system capable of implementing a policy of birth control. Even if the camp advocating control of population growth were to prevail, there was no realistic means to provide the public with the necessary knowledge and effective means of contraception on any scale. As if adding salt to the wound of the idea of population policy, there was an economic calamity caused by misguided policy and aggravated by radical natural disasters in 1959– 1961. During that period, China experienced a sharp decline in the birth rate together

Fertility, family planning, and population policy in china

10

with an unprecedented spike in the death rate. An estimated 30 million or more deaths from starvation or diseases related to under-nourishment occurred during the turn of the decade of the 1960s. In this period, China actually witnessed a negative rate of population growth. Obviously, this historical accident or incident did not help in resolving the paradox of population control in China. Immediately following the years of the economic fiasco, a “baby boom” occurred, beginning in 1962. During the years 1962–1966, the crude birth rate hovered at around 40 per thousand, representing total fertility rates between 5 and 7 (see Chapter 12 by Poston and Glover in this book for more discussion). The return to traditional high-level uncontrolled fertility was especially true in the rural areas. The total fertility rate was as high as 6 and 7 during most of the 1960s (Table 1.1 and Figure 1.1). The rise in fertility was due mainly to compensatory births delayed during the prior bad years and a recovery of the normal rate of marriage. The demographic impact of this baby boom in the early 1960s would have an echo in the 1980s, as will be discussed in a later section. Alarmed by the bumper baby crop, the central government returned to the consideration of population control. In 1962, the State Council, the government central executive office, announced a new “Directive.” This time, it was to “seriously” confront the problem of planned reproduction. It commanded the government at all levels not only to promote birth control but also to review the results of its birth-control campaign. To clear up any ideological ambivalence, the new policy document pronounced that planned reproduction was necessary for the construction of a socialist society and was in no way to be confused with reactionary Malthusian thoughts. The seriousness of this official installation of a population policy was indicated by the institution of a Committee on Planned Births at the central government, and similar administrations in charge of family planning at all subordinate levels down to the local authorities. Government health departments and hospitals began to conduct medical research in contraceptive devices, induced abortion, and sterilization. In addition to the dissemination of contraceptive knowledge, the public health agencies provided contraceptives free of charge. Health workers were recruited for training in order to meet the expanded need for birth control. Finally, the central government set a goal to reduce the population growth to

Table 1.1 Population, birth, and death rates of China, 1949–2000 Year

Population (‘000)

Crude birth Crude death rate (0/00) rate (0/00)

Natural increase (0/00)

Total fertility rate National Rural Urban

1949

541,670

36.00

20.00

16.00

6.139

—

—

1950

551,960

37.00

18.00

19.00

5.813 5.963

5.001

1951

563,000

37.80

17.80

20.00

5.699 5.904

4.719

1952

574,820

37.00

17.00

20.00

6.472 6.670

5.521

1953

587,960

37.00

14.00

23.00

6.049 6.183

5.402

1954

602,660

37.97

13.18

24.78

6.278 6.390

5.723

Fertility and population policy: an overview

11

1955

614,650

32.60

12.28

20.32

6.261 6.391

5.665

1956

628,280

31.90

11.40

20.50

5.854 5.974

5.333

1957

646,530

34.07

10.80

23.23

6.405 6.504

5.943

1958

659,940

29.22

11.98

17.24

6.679 5.775

5.253

1959

672,020

24.78

14.59

10.19

4.303 4.323

4.172

1960

662,070

20.86

25.43

−4.57

4.015 3.996

4.057

1961

658,590

18.20

14.24

3.78

3.287 3.349

2.982

1962

672,950

37.01

10.02

26.99

6.023 6.303

4.789

1963

691,720

43.37

10.04

33.33

7.502 7.784

6.207

1964

704,990

39.14

11.50

27.64

6.176 6.567

4.395

1965

725,380

37.88

9.50

28.38

6.076 6.597

3.749

1966

745,420

35.05

8.83

26.22

6.259 6.958

3.104

1967

763,680

33.96

8.43

25.53

5.313 5.847

2.905

1968

785,340

35.59

8.21

27.38

6.448 7.025

3.872

1969

806,710

34.11

8.03

26.08

5.732 6.263

3.299

1970

829,920

33.43

7.60

25.83

5.812 6.379

3.267

1971

852,290

30.65

7.32

23.33

5.442 6.001

2.882

1972

871,770

29.77

7.61

22.16

4.984 5.503

2.637

1973

892,110

27.93

7.04

20.89

4.539 5.008

2.387

1974

908,590

24.82

7.34

17.48

4.170 4.642

1.982

1975

924,200

23.01

7.32

15.69

3.571 3.951

1.782

1976

937,170

19.91

7.25

12.66

3.235 3.582

1.608

1977

949,740

18.93

6.87

12.06

2.844 3.116

1.574

1978

962,590

18.25

6.25

12.00

2.716 2.968

1.551

1979

975,420

17.82

6.21

11.61

2.745 3.045

1.370

1980

987,050

18.21

6.34

11.87

2.238 2.480

1.147

1981

1,000,720

20.91

6.36

14.55

2.631 2.910

1.390

1982

1,016,540

22.28

6.60

15.68

2.860 3.230

1.580

1983

1,030,080

20.19

6.90

13.29

2.420 2.780

1.340

1984

1,043,570

19.90

6.82

13.08

2.350 2.700

1.220

1985

1,058,510

21.04

6.78

14.26

2.200 2.480

1.210

1986

1,075,070

22.43

6.86

15.57

2.420 2.770

1.240

Fertility, family planning, and population policy in china

12

1987

1,093,000

23.33

6.72

16.61

2.590 2.940

1.360

1988

1,110,260

22.37

6.64

15.73

2.520 2.740

1.460

1989

1,127,040

21.58

6.64

15.04

2.350 2.540

1.550

1990

1,143,330

21.06

6.67

14.39

2.310 2.400

1.490

1991

1,158,230

19.68

6.70

12.98

2.200 2.250

1.420

1992

1,171,710

18.24

6.64

11.60

2.000 2.110

1.360

1993

1,178,440

18.09

6.64

11.45

1.810 1.960

1.300

1994

1,192,840

17.70

6.49

1121

1.810 1.890

1.290

1995

1,204,860

17.12

6.57

10.55

1.780 1.810

1.280

1996

1,217,550

16.98

6.56

10.42

1.700 1.700

1.210

1997

1,230,080

16.57

6.51

10.06

1.620 1.610

1.170

1998

1,242,210

16.52

6.55

9.97

1.490 1.520

1.130

1999

1,259,090

15.23

6.46

8.77

—

—

—

2000

1,268,000

—

—

6.94

—

—

—

Figure 1.1 Total fertility rate (average number of children per woman) for urban and rural places and all China, 1949–1999. Source: Table 1.1. 1 percent or less by the end of the twentieth century, and each of the provinces and municipalities was required to project short- and long-term targets for the birth-control campaign. Hebei province, for instance, in its 1963 “ten-year plan of fertility control,” aimed at a reduction of its crude birth rate to 13 per thousand by 1975; and the Shanghai municipality proposed to decrease its birth rate to 15 per thousand by the year 1967.

Fertility and population policy: an overview

13

It was unclear whether these targets of fertility control were supported by any scientific demographic analysis. It was undoubtedly, however, advancement from a stage merely advocating a planned reproduction by the government to another at which the government engages actively in the implementation of birth control. This policy imprint on the precipitous decline during the years 1962–1966 was visible in terms of the period’s total fertility rate (TFR), especially in urbanized areas, as shown in Figure 1.1 in which the rural and urban TFRs are displayed separately. Led by Shanghai, a campaign slogan, “one is not too few, two, just right, and three, too many” (Ca et al. 1999:157–158), was spread quickly throughout the nation. And the effective measure to achieve a small family size was put in three words, “wan, xi, shao,” or “late, sparse, and few.” “Late” referred to a delay in marriage and the first birth to a later age, “Sparse” was to space the birth interval by 4–5 years, and “few” to limit the number of births to 2 or no more than 3 per couple. To the extent that urban areas and the rural regions in the vicinity of major cities were regimented into work-unit organizations under a socialist system, the planned-birth policy was effectively carried out. The neighborhood committee in the city and production brigade in the countryside were formed as the grass-roots unit in which a women’s association was put in charge of regulating the childbearing behavior of all families in their jurisdiction. Technically, more than a dozen contraceptive paraphernalia were distributed free of charge by these local offices. There were clinics providing sterilization without charging a fee, and the patients were given leave from work with pay. The Great Cultural Revolution of 1966 in which Chairman Mao mobilized the masses to eradicate his political foes within the Party, brought the administrative system literally to a halt. The government’s role in the birth-control campaign dissolved quickly and was not revived until 1970. The demographic mark of this political turmoil in the late 1960s was a spike in the TFR in both the urban and rural areas (see Figure 1.1). The Cultural Revolution was not ended politically until the arrest in 1976 of the socalled “gang of four” led by Madam Mao, a group allegedly plotting to usurp the control of the Party central power. Premier Zhou Enlai, who oversaw the economic deterioration while handling a difficult act of balancing the political struggles, was determined to put China’s economic house in order. In 1970 he assembled the various ministries under his command to plan an economic revival. He addressed the issue of planned births and criticized the top officials, saying that “planned population reproduction is not limited to the jurisdiction of the health department, it is an integral part of overall planning. If you cannot plan for population growth, it is futile to talk about national planning” (Sun 1990:158). In 1971 the State Council issued a referendum on the nationwide campaign of fertility planning. The short-lived movement of birth control in the early 1960s was thereby resuscitated and even more vigorously enforced. This time the reinstitution of a population policy was also blessed by Chairman Mao’s endorsement. In approving a government document in 1974 in his later years, he uttered that “population must be controlled by all means” (Ca et al. 1999:166). At that time Mao was still a god-like supreme center of power in China. With his stature, he commanded the eager following of the planning policy not only by party officials but also widely among the masses, especially the city folks who were mostly employed by the state. Unlike the one-child policy in the 1980s, which had compulsory penalties for violators, the vigorous administrative means of watching closely over women of

Fertility, family planning, and population policy in china

14

reproductive ages and regulating their birth behavior in the 1970s was leaning more toward the prevention of excess births. Moreover, the allowance of two children per couple, spaced by 4–5 years, was easily acceptable by city residents, and it did not meet with strong resistance from the people in most of the densely populated rural areas. The results of the 1970s’ planned-birth campaign may be seen in the decline of the crude birth rate nationwide from 33 per thousand in 1970 to 18 per thousand in 1980 (see Figure 1.2). The TFR was reduced from 5.8 in 1970 to 2.2 in 1980 nationwide; and dropped to slightly less than 1.2 in the urban areas (Figure 1.1). This policy-induced fertility transition in a matter of one decade for a nation of 1 billion population was unprecedented in the demographic history of mankind. It is little wonder that the United Nations and other international organizations involved in population planning, in the era of its active promotion in the 1970s, picked China as a model for other developing countries.

Figure 1.2 Crude rates of birth, death, and natural increase in China, 1949– 2000. Source: Table 1.1.

1980 onward: a policy experiment followed by decentralization In 1979, China adopted, at its third plenary meeting of the fifth People’s Congress (or Parliament), an economic reform and open policy. Along with the restoration of the previously abolished social sciences, population research institutes were established in a number of major universities and in the Academy of Social Sciences. The new Premier, Zhao Ziyang, in consultation with experts’ projections of China’s population for the future 20–30 years, proposed a population limit of 1.2 billion by the end of the twentieth century. In order to achieve this goal of population control, the State Council issued an ambitious “one-child policy” in late 1979. This was to limit family size to one child for all newly wed couples except in the autonomous regions where the ethnic population was

Fertility and population policy: an overview

15

under 10 million. In fact, the goal of this new compulsory birth-control program to limit to one child per family, had already been nearly accomplished in urban areas, as reflected in the TFR of less than 1.2 in 1980. For the rural populace it was a different and much more complicated story. The rural economic reforms reverted from collective farming to family farming, for which family laborers, especially male ones, were minimally required assets. Moreover, for all the years of socialist construction, China could barely afford a prototype social security system, including the provision of a retirement pension in the cities in which most workers were state employees. The rural Chinese were left to fend on their own for old age support, depending on their male posterity. It was nearly impossible to convince the majority of Chinese in rural areas that having only one child, especially when the first birth was a girl, meant an advantage to their own families. The local enforcement of the one-child policy by the village Party officials was inevitably translated into forced abortion, the under-registration of female first births, and even bribery for the permission to have a second birth. The tension between the government’s demand on local officials and the wish of the rural people boiled. Reports on persecutions of women and families who failed to conform to the one-child policy spread outside the country, which led to an outcry regarding violation of human rights from the international community, especially in the United States during the Reagan administration of the 1980s (Hao 2003:17–19). Under these pressures inside and outside the country, the Chinese authorities reconsidered their radical one-child policy. In 1984, the central government approved a progress report submitted by the State Committee on Fertility Planning on the “Situations of Implementing Planned Births,” and announced a policy adjustment. In its official language, “the policy of planned births must be built upon the people’s support. It must be sensible and reasonable to the people, and doable to the responsible officials” (National Committee on Planned Births 1992:35– 43). To mitigate the tensions among the people and corruption of the officials thereof, a second birth may be allowed in the rural areas and a third one permitted under special conditions, but no more beyond that. This signaled the end of the “one-child policy” in rural China. In the province of Shandong, a policy allowing “only-girl families” to have a second child, after a minimum interval of four year, was initiated. By 1986, similar policies were soon adopted in other provinces for rural families (see Scharping 2003, for more discussion). By a rough estimate, this meant that one-half of the rural families would proceed to have two children instead of one. Notable in this policy adjustment in the late 1980s was the decentralization of the planning and implementation of population regulations from the central authorities to the provincial and county levels of the government. Regional differences and local variations in economic development and socio-cultural environment dictated realistic targets and appropriate measures of population control tailored by the local governments. In general, the one-child policy continued to be enforced among the state employees and residents in the cities and rural townships, although exceptions were granted case by case. In the wider countryside and regions of ethnic minorities, the population control policy continued to be enforced, but in a more flexible fashion. To correct over-zealous local officials and, at times, erratic compromises of laws, the new policy guideline emphasized “the three keys” to family planning work: the key of prevention (of unplanned pregnancies), the key of education, and the key of consistency (in policy implementation). Together these meant, in other words, that planned births must be

Fertility, family planning, and population policy in china

16

integrated with assistance to the peasant families in economic development, in raising the civility of rural culture, and in guaranteeing the reproductive health of the women. The policy adjustment in the mid-1980s contributed inevitably to a “baby-boomlet” as may be seen in the crude birth rates in the period of 1984–1990 (Figure 1.2). There were other factors besides birth-planning concerns that added to this period increase in the birth rate. As mentioned earlier, one was the “demographic echo,” that is, those who were born during the early 1960s’ baby boom were entering the marriage and reproductive ages by the mid-1980s. Moreover, in connection with the government’s reforms beginning in 1980, a new marriage law was legislated in the People’s Congress, which stipulated the minimum age of marriage at 20 for females and 22 for males. These legal ages for marriage were lower than the practice of 25 for females and 27 for males during the “late” policy of the 1970s, which was an executive order but never made a law. The new marriage law was oblivious of the population control policy. It nonetheless added many younger married couples in the 1980s to increase the momentum of child births in that period (Figure 1.2). The government, in view of the rising fertility rate in the later part of the 1980s, adjusted in the early 1990s its projected target number for the total population from 1.2 billion to nearly 1.3 billion. This latter figure proved to be more realistic and would soon be verified by China’s latest census conducted in November 2000. It is advisable, in order not to be overly alarmed by the upturn in fertility as a consequence of relaxing the onechild policy, to look at the TFR in addition to the crude birth rate, during the late 1980s (Figure 1.1). In spite of a discernible rise in the crude birth rate (Figure 1.2), the TFR was relatively stable over the same period. The TFR in urban China continued to decrease from 1.5 to about 1.2 per woman throughout the 1980s, and that of the rural population remained below 3.0. The breakdown separately for the urban and rural TFR was not yet available for the 1990s. The national average TFR was, however, brought down from 2.3 in 1990 to around 1.8 in the late 1990s. This is below the TFR level of 2.1, generally regarded as required to replace a population. The statistics cited so far suggest that the population control policy may have been stabilized or institutionalized, and the norm of a small family size had taken root in the 1990s in China.

Summary China was among the earliest countries in the world to realize population pressures in planning for economic development. The government’s intent to control population can be traced back to the 1955 document, “Directive on the Issue of Population Control.” Constrained by the postwar conditions of the economy and the lack of modern technology, the government then was not equipped to actually deliver any effective measures to reduce the traditionally high fertility rate. The innovative idea of birth control was even more hampered by the age-old Chinese value that a large family is equated with prosperity, on the one hand, and by the ideological contradiction to Marxist doctrine, on the other. Following the economic fiasco at the turn of the 1960s and the subsequent bumper crop of babies due to compensatory births for the delay during the bad years, the government resolved in 1962 to slow the rapid rate of population growth. This time, the official pronouncement of a population policy was to clear away the

Fertility and population policy: an overview

17

ideological ambivalence regarding whether population planning was contradictory to the Marxist economic planning. The government’s action then was to establish an office in charge of birth planning at the central government and related offices at the provincial and local levels. Research in contraceptive knowledge and technology was widely engaged in by hospitals and health agencies, and public health workers were recruited for special training in methods of averting births, including the use of contraceptive devices, induced abortion, and sterilization. This system of government machinery in birth control campaigns, however, was short lived, although it did make a clear mark on bringing down fertility from its peak in 1962. It was, unfortunately, disrupted by the political upheaval of the Great Cultural Revolution, beginning in 1966. Systematic birth control was, nonetheless, restored and put into more vigorous use in the early 1970s. Throughout the decade of the 1970s, the control policy of “late, sparse, and few” was effectively implemented to decrease precipitously the fertility rates in rural as well as in urban areas. Perhaps, encouraged by the 1970s’ success in carrying out the birth control program, the reform-minded leadership of the central government decided in 1979 to adopt a radical one-child policy to achieve a projected goal of limiting China’s population to a magic number of 1.2 billion in the year 2000. The rigid demand of one child per couple ran into strong resistance, especially in the rural areas. In 1984, the central government was forced to revise the one-child policy, and allow provincial and local authorities the option of permitting a second child if the first child was a girl. Although there appeared to be an upturn in the birth rate associated with this relaxation of the one-child policy in the 1980s, much of that short-term increase may also be attributed to the demographic echo-effect. The baby crop in the early 1960s had come of age in the 1980s to marry and have children. The relatively stable TFR, less than 2.0 per woman through the years of the 1990s, however, suggests that fertility regulation behavior and the norm of a small family have become institutionalized and taken root in the population of China. The imprints of the hesitation in putting a government’s population program in place in the 1950s and early 1960s and the vigorous implementation of birth control throughout the 1970s, on the changing fertility in China can be seen in the year-to-year statistics displayed in Table 1.1. For a quick overview, Figure 1.2 shows in graphic form the fertility transition over the past five decades, and, together with the declining death rate, shows also the declining rate of population growth in China. Figure 1.1 depicts the parallel transition of fertility in terms of total fertility rate between urban and rural China, though the latter remains at a higher level than the former. In a white paper on population control and development, released in December 19, 2000, China’s government set a target to limit its population to 1.33 billion by 2005, with an annual growth rate lower than 1 percent. For the long term, China projects its population to peak at 1.6 billion by the mid-twenty-first century and to gradually decline thereafter (People’s Daily, December 20, 2000). With a population pushing toward 1.2 billion in 1998, the latest available estimate of China’s GDP per capita was around US$ 800. This average level of living was higher than other highly populated developing countries like India, Indonesia, and Pakistan, which had a per capita GDP below US$ 500, but was in no way comparable to other Chinese in Hong Kong and Taiwan where the GDP per capita was between US$ 20,000

Fertility, family planning, and population policy in china

18

and 25,000. Although since the 1980s China has seen an annual rate of GDP growth averaging about 8 percent, the economic gains in the post-reform era were limited to the more developed urban regions on the east coast of China. Three-quarters of China’s population live in rural areas, and an estimated 50 million or more in remote pockets live in abject poverty. China is aiming, as stated in the most recent white paper on population and development, for a universal nine-year education for all its populace by 2005 and a popularized secondary and post-secondary education nationwide by 2010. Basic medical health care and social security are also expected to become generally accessible. In view of the fast-growing proportion and number of older people in China, because of the drastically reduced rate of fertility in the most recent three decades, a secure system providing basic living and health care for the retired elderly will be critical in the near future. Control of population, in view of all these goals of development, is of vital national interest for China. Just as national security is a vital national interest, China will continue to monitor and intervene as necessary in people’s reproductive behavior in the coming decades. The difference, if any, is seen in China’s turn to openness to multilateral international cooperation in defining and solving problems of national interest. This increased openness to international dialogue is even more foreseeable in the post-Cold War era, and especially since China has joined the World Trade Organization with imported intelligence from outside, together with the increasingly advanced social sciences, including demographic studies inside, China will likely help design future population policy in a more comprehensive and less intrusive way than the draconian one-child policy.

References Baumle, A.K. and D.L.Poston, Jr. 2004. “Apportioning the House of Representatives in 2000: The Effects of Alternative Policy Scenarios.” Social Science Quarterly, 85: 578–603. Ca, R., W.Hu, and J.Zei. 1999. Demography of A Century. Beijing: Beijing Arts and Literature Publishers. Hao, L. 2003. “Regarding the United States’ Refusal to Contribute to the U.N. Fund for Population Activities.” Population Research, 1(27):17–19. Ma, Y. 1997. The New Population Theory. Changchun, China: Jilin People’s Publishers. National Committee on Planned Birth. 1992. Collection of Planned Birth Documents, 1981–1991. Beijing: China’s Democratic-Law Publishers. People’s Daily. 1954, September 18. People’s Daily (overseas edition). 2000, December 20. “China’s Population and Development in 21st Century.” A White Paper by the Press Office of the State Council. Scharping, T. 2003. Birth Control in China, 1949–2000. London: RoutledgeCurzon. She, C. 1988. A History of Planned Birth Activities in China. Urumqi, China: Xinjiang People’s Publishers. Sun, M. 1990. A History of China’s Planned Birth Programs. Beijing: Beijing Publishers of Women and Children.

Part I Family planning policy and contraceptive use

2 Patterns of induced abortion Juan Wu and Carol S.Walther

Introduction China has experienced one of the most successful fertility declines in demographic history, an observation also made in other chapters of this book. Its total fertility rate (TFR) fell from about 6.5 in the 1960s to 1.8 in the year 2000 (United Nations Economic and Social Commission for Asia and the Pacific: http://www.unescap.org/). While socioeconomic development has undoubtedly played an important role in this process, the precipitous decline in fertility is more likely due to the implementation of the “one couple, one child” population policy initiated in 1979 (Poston and Gu 1987; Poston 1988). Induced abortion, among other proximate causes, is recognized as having greatly contributed to fertility decline in many developing countries (Bongaarts 1978; Frejka 1985; Knodel et al. 1987). There are three main objectives in this chapter. First, a general discussion about abortion will help the readers’ understanding of abortion issues in China. There has been much publicity about the controversial practice in China of using abortion to end unplanned pregnancies. And some international journalists have focused on the extreme examples rather than on the typical situation in China. Second, although there have been a few studies of abortion in China, the findings have not been conveyed effectively to the general public, especially to the international audience. Research reported in this chapter, based on a nationally representative sample, will present very detailed information about this issue, and will contribute to a better understanding of abortion. Third, the research findings reported here may indicate ways in which the Chinese Government can make family planning programs more effective and may also help understand various other social issues. For instance, an important social issue is China’s increasingly abnormal sex ratio at birth since the mid-1980s and its serious social consequences in years to come (see Poston and Glover’s Chapter 12 in this volume). Zeng and his associates (1993) argue that the increase in the sex ratios at birth is due in part to sex-selective abortion. Examining the effects of son preference on the hazard of having an abortion, as will be done in this chapter, will hopefully shed more light on the issue of rising sex ratios at birth. In fact, the analyses in this chapter will not only be important for China, but for other countries which may encounter similar problems at a later time.

Fertility, family planning, and population policy in china

22

History of abortion in China In China, abortion law may be traced back to the Tang dynasty (618–906) where it was a crime to induce an abortion by assaulting a pregnant woman (Rigdon 1996). In the following dynasties up to the Qing (1644–1911), the legal codes stipulated that the fetus had to have assumed a human form for the abortion-in-assault to apply. During the Qing dynasty, abortion was illegal, though there were some exceptions permitting a pregnant woman to obtain one (Rigdon 1996). In the first four years after the founding of the People’s Republic of China (PRC) in 1949, abortion was still illegal. Even though family planning as an official policy was initiated as early as 1953, abortion was available only when there was a risk to the mother’s health, or in cases when the birth interval since the last child was too short and the mother experienced difficulty in breastfeeding the previous child. Abortions were conducted in hospitals with the consent of the doctor, parents, and their work units (Li et al. 1990; Rigdon 1996). The regulations also stated that doctors were not authorized to perform abortions unless the couple seeking it already had four children (Rigdon 1996). In 1957, China legalized induced abortion in the first 10 weeks of pregnancy as part of the government’s first birth control campaign. However, due to a lack of medical facilities and personnel, and the unwillingness of many physicians to accept the new law, there was not a large increase in the number of abortions performed (Tu and Smith 1995; Rigdon 1996). Family planning was not a priority during the mobilization for the Great Leap Forward that began in 1958 when programs were initiated for the rapid communalization of agriculture and industry. In the following years of widespread famine in China, fertility control was not a priority. Only after the recovery from the consequences of the Great Leap Forward and the severe famine of the early 1960s did birth control become a priority. However it was not until the late 1970s when the onechild policy was initiated that abortion became a significant complement to the family planning program (Tien 1987; Henshaw 1990; Li et al. 1990; Tu and Smith 1995). The increases in the number of abortions in the 1970s were probably caused by both increasingly rapid socioeconomic development, especially in urban areas, and the implementation of the wan, xi, shao campaign (Tien 1987; see Chapter 1 of this volume for more discussion). As mentioned in the previous paragraph, with the implementation of the one-child policy in 1979, the number of abortions performed has risen rapidly. And the number has varied with changes in population policies. Generally speaking, China’s population policies have changed from a strict and centrally enforced policy implementation in the late 1970s and early 1980s to a more decentralized policy implementation characterized by local family planning regulations since the mid-1980s. These changes have affected the incidence of abortion in China (Greenhalgh 1986). For instance, according to HardeeCleveland and Banister (1988), in 1979, 7.8 million abortions were performed in China. In 1982, it was 12 million, and in 1983 when the policy was the strictest, it reached a peak of 14.4 million; the number dropped to about 9 million in 1984 when rural couples with one daughter were allowed to have a second birth; the number rebounded to about 12 million in 1986. Over the 1970s and 1980s, 200 million abortions are estimated to have been performed in China (Zeng 1991).

Patterns of induced abortion

23

Reasons for and patterns of abortion in China This section of the chapter focuses on abortion reasons that are specific for Chinese women in the special context of China’s population policy. One major reason for a Chinese woman to have an abortion is compliance with China’s family planning policy. The one-child population policy in China regulates not only the family size that a couple can have but also the spacing between children (in most cases, a four-year period between the first and the second child is required). When couples fail to meet those regulations, they are often encouraged to seek an abortion (Tu and Smith 1995). For some women, abortion has been used as a primary means of birth control, even though the state strongly discouraged the reliance on abortion because of the risks to women’s health and its high costs relative to contraception (Greenhalgh 1986). The state preferred couples to use sterilization or reliable contraception, but women with unauthorized pregnancies were sometimes encouraged to have abortions, especially during those periods when the enforcement of family planning policy was very strict. Cadres in some provinces encouraged women to have abortions because their work units would be fined if the birth quota assigned to them was exceeded (Greenhalgh 1986; Hardee-Cleveland and Banister 1988). As Tu and Smith note, under the family planning responsibility system, abortion serves as the main alternative that families can use to avoid penalties for an unplanned birth in cases of contraceptive failure, a last protection for local authorities against penalties for the failure to meet their preset demographic target, and a sensitive indicator of shifts in official family planning policy and its implementation at national and local levels. (1995:279) But Rigdon (1996) points out that the extent to which coercion was used varied widely from place to place; some villagers said that they never experienced it while others have reported a variety of coercive practices. Son preference in China is another major reason for ending a pregnancy with an abortion. Son preference is a product of the Confucian tradition that emphasizes the value of sons. There are a variety of historical, moral and ethical, and economic factors underlying the existence of son preference in China (Arnold and Liu 1986; Zeng et al. 1993; Poston et al. 1997; Zeng and George 2000). In the past, family structure in China was patriarchal. This patriarchal structure and the resulting preference for sons were so strong that it was even written in the Book of Rites that “A woman is to obey her father before marriage, her husband during married life, and her son in widowhood” (Arnold and Liu 1986:226). Chinese traditions also stress the importance of carrying on the family line through male progeny. Sons are desired mainly because of their economic value to the family, old-age support to their parents, and the provision of labor in farming or family business (Arnold and Liu 1986). With the implementation of the one-child policy, many couples could not have as many children as they desired. However, the ideology of son preference and the economic value of sons urged many couples to have at least one son in the family, especially in rural areas (Arnold and Liu 1986; Poston et al. 1997; Zeng and George 2000).

Fertility, family planning, and population policy in china

24

The existence of sex preference by itself will not influence fertility unless technologies that translate the gender preference to fertility modification are applied. Modern technologies like ultrasound and amniocentesis have enabled many Chinese parents to determine the sex of the fetus. Consequently, there have been reports of sex-selective abortions in China, that is, couples would undergo an abortion if they found out the fetus is female (Hull 1990; Zeng et al. 1993). The practice of sex-selective abortion is thought to be a reason for the rising sex ratios at birth in China since the mid-1980s (Zeng et al. 1993). The true sex ratio at birth does not vary much in most human societies. Countries with reliable data show that the sex ratio at birth is around 104–107 male births for every 100 female births (Chahnazarian 1991). Nevertheless, prenatal sex identification followed by gender-specific induced abortion is a main reason for the abnormally high sex ratio at birth and will have serious social consequences (Zeng et al. 1993; Gu and Xu 1994; Gu and Roy 1995; see Chapter 12 in this volume by Poston and Glover for more discussion). The third reason for Chinese women to seek abortion is social norms. There are various social taboos against single or widowed women bearing children. Not only do social norms pressure those pregnant women to have an abortion, but they also are officially sanctioned because family planning regulations do not allow births to single women (Rigdon 1996). With increasing exposure to Western lifestyles, the loosening of restrictions on social behavior, especially in large urban areas, and inadequate sex education among young Chinese women have contributed to increases in the number of pregnancies outside marriage that often end in abortion. Although abortion has increased across China since the implementation of the onechild policy in 1979, some differences may be observed at the regional level. Specifically, the inland areas often have lower abortion rates than the wealthier coastal provinces. This is partly due to the large concentration of minorities in the inland areas for whom the Chinese government has relaxed its population control policies (Chhabra 1984). Also urban areas are found to have higher abortion rates than rural areas (Kang 1991; Tien et al. 1992). In urban areas, the reported percentage of all pregnancies terminated by abortion rose from less than 3 percent in 1960 to about 30 percent in 1987; in rural areas, it rose from virtually zero to more than 15 percent (Kang 1991). Some studies show that education, age, marriage duration, and residence have apparent effects on abortion order (Li et al. 1990). Based on data from the 1982 One-Per-Thousand Fertility Survey, Arnold and Liu (1986) studied pregnancy outcomes by family composition and found that couples without a son were much less likely to have an abortion than couples with at least one son (Tu and Smith 1995).

Data and methods In this chapter, data are used from the 1997 Sample Survey of Population and Reproductive Health in China conducted by the China Population Information and Research Center and the State Family Planning Commission of China. The survey randomly interviewed 15,213 women between the ages of 15 and 49. Basic demographic data such as age, ethnicity, residence, marital status, and education were collected for each woman. In addition, each woman’s pregnancy histories were recorded in great

Patterns of induced abortion

25

detail, including information on the ending time of each pregnancy, the outcome of each pregnancy, the health status of each live birth, and birth parity. Every pregnancy was coded with six outcome categories: son, daughter, miscarriage, stillbirth, abortion, and currently pregnant. This survey also includes information about a woman’s attitude towards the ideal number of children and the sex composition of this desired number of children. In this chapter, for ever-married Chinese women with at least one child, their hazard of having an abortion is modeled for the pregnancy following their first child. Several restrictions have been imposed on the sample. Specifically, the woman’s hazard of having an abortion is examined for the pregnancy following her first child. Women who had only one child and were pregnant at the time of the survey were excluded from the study. Furthermore, the number of months between the birth of the first child and the second birth should usually not equal zero. When this occurred, it is assumed that the woman had multiple births, and they too are excluded from the analysis. The last restriction is that women under the age of 20 are excluded. This constraint was introduced because on one hand, women’s educational status, one of the independent variables, is not settled until they reach 20 years of age (reasons are specified later in the discussion of the education independent variable), and on the other hand, China’s Marriage Law requires that the minimum legal marriage age is 20 for women (China Population Information and Research Center: http://www.cpirc.org.cn/). The effects of individual characteristics and population policy on a woman’s hazard of having an abortion are analyzed. The dataset consists of ever-married women with at least one child, a total of 11,762 married women. Cox proportional hazard models are estimated to predict the hazard of having an abortion for the pregnancy following the first child.

Hazard models The most satisfactory way statistically to model either the occurrence or non-occurrence of an event, and its timing, is via hazard analysis. The term hazard rate, which refers to the rate at which the event of interest occurs, comes from the biomedical sciences’ use of survival analysis, that is, the surviving the hazard of a death (Yamaguchi 1991; Kleinbaum 1996). Hazard models are concerned with the patterns and correlates of the occurrence of events. Death, disease incidence, or any designated experience of interest that may happen to an individual, such as an abortion, would be an event. And “the occurrence of an event assumes a preceding time interval that represents its nonoccurrence” (Yamaguchi 1991:1). Specifically, there must exist a certain time period or duration of nonoccurrence in order for an occurrence to be recognized as an “event” (Yamaguchi 1991:1). Hazard analysis is used to examine duration data, which represent the nonoccurrence of a given event. Yamaguchi (1991:2) points out that “in conceptualizing the duration of the nonoccurrence of a given event, another important concept is the risk period.” Generally, the time period that represents the nonoccurrence of a given event can be divided into two parts: the period at risk and the period not at risk of experiencing the event. So specifically, hazard analysis is the analysis of the duration for the nonoccurrence of an

Fertility, family planning, and population policy in china

26

event during the risk period, or the analysis of the rates of the occurrence of the event during the risk period. One reason why Cox models are so widely used is that the Cox proportional hazard model has advantages over other hazard models. The most significant advantage is that Cox models can assume time dependence without having to specify its form (Yamaguchi 1991). In the situation when there are only time-independent variables, that is, when the values of the independent variables do not change over time, the form that the Cox model assumes is represented as follows: log h(t)=log h0(t)+b1X1+b2X2+…+BiXi where log h(t) is the hazard rate (or hazard function) which is an unobserved value gauging the instantaneous probability that a subject will have the event of interest during the interval. It is a rate, rather than a probability. Thus the values of the hazard function range between 0 and infinity; h0(t) is any function of time t; and X1 to Xi are independent variables. This function of time (h0(t)) does not have to be specified, and is the characteristic that makes the Cox model so attractive. Another advantage of Cox’s method is that it has the ability to estimate stratified models, which permit the investigator to control for a categorical independent variable that may have a complicated form of interaction effects with time, without specifying the form of the interaction effects (Yamaguchi 1991). One assumption of the Cox proportional hazards model is that the hazard rate for any one individual is proportional to the hazard for any other individual (Yamaguchi 1991; Kleinbaum 1996). However, the importance of meeting this assumption has been debated in the literature. Allison states that “as models go, the proportional hazards model is extraordinarily general and nonrestrictive—the main reason for its popularity. Even when the proportional hazards assumption is violated, it is often a satisfactory approximation” (Allison 1985:38). One way to describe the survival-time data for the subjects is to graph their Kaplan– Meier (K–M) survivor function curve. The K–M survivor curve is an empirical plot showing the probabilities of surviving for each unit of time. It starts with a survival probability of 1 and then steps down to the other survival probabilities, as one moves from one ordered failure time to another (in these analyses, from one month to the next). Let nj represent the number of observations that have not failed and are not censored at the beginning of time period t; and dj represent the number of failures that occur to the observations during the time period t. The following formula is the K–M estimator of surviving beyond time t, and is the product of survival probabilities in t and the preceding periods:

Figure 2.1 graphs S(t) against the number of months between the birth of the first child and the occurrence of the event of interest, abortion, in separate curves for all rural and urban women with at least one child. The K–M survivor curves show the probabilities of surviving the hazard of having an abortion following a woman’s first child for each group of women. For urban women, the curve steps down gradually from a probability of near 1.0 of surviving the hazard of having an abortion, to a probability of 0.6 by about the

Patterns of induced abortion

27

180th month, leveling off for the remainder of the risk period. The analysis to follow focuses on all women. Dependent variable The survival-time data for ever-married Chinese women consist of two variables. Abortion1 is a dummy variable indicating for each woman whether or not the event of interest, abortion for the pregnancy following the first child, occurred during the observation period. This is coded 1 if the pregnancy was aborted and 0 if otherwise. The data show that among women with at least one child, 22 percent

Figure 2.1 Probability of surviving the hazard of an abortion after the birth of the first child, stratified by rural and urban residence, China, 1997. Table 2.1 Descriptive data: 11,762 women with at least one child, China, 1997 Variable Abortion1

Mean

Standard deviation

Minimum value

Maximum value

0.22

0.41

0

1

Time1

37.40

40.15

0

351

Rural

0.78

0.42

0

1

Han

0.91

0.29

0

1

After80

0.57

0.50

0

1

Fertility, family planning, and population policy in china

28

Girl

0.49

0.50

0

1

Illiterate

0.25

0.43

0

1

Elementary

0 32

0.47

0

1

Junior

0.29

0.45

0

1

Senior

0.14

0.35

0

1

Number_ preference

1 86

0.74

0

1

35 10

7.79

20

49

North

0 14

0.35

0

1

Northeast

0.09

0.28

0

1

East

0.29

0.46

0

1

Mid-South

0.26

0.44

0

1

Southwest

0 15

0.36

0

1

Northwest

0.06

0.24

0

1

Age

Source: The 1997 sample survey of population and reproductive health in China.

had an abortion following the birth of their first child (see Table 2.1). The second variable, Time1, refers to the amount of time since the birth of the first child and the abortion or the censoring event. Independent variables The variable, Girl, represents the gender of the first child. This variable is intended to measure a woman’s sex preference for her children. If there is no sex preference in the society, there should be no association between the sex of an already born child and the likelihood of having another child. Since various studies point out the existence of son preference in China, the general hypothesis in this chapter is that controlling for several other independent variables, having a boy as the first child will increase the hazard of ending the following pregnancy with an abortion (see Chapter 3 by Liu and Chang in this volume). If there was no or only minimal sex preference in China, the sex of the first child should not be associated with the hazard of having an abortion. The Girl variable is a dummy variable coded 1 if the first child is a girl and 0 if the first child is a boy. The education level of a woman has been found to be closely associated with her fertility attitudes and behavior. Previous studies find that in most societies a woman’s educational attainment is positively associated with her probability of having an abortion (Powell-Griner and Trent 1987; Ahiadeke 2001). In this analysis, four dummy variables are used representing a woman’s educational attainment, namely, illiterate or semiilliterate (Illiterate), elementary (Elementary), junior high school (Junior), and senior high school (Senior). Each variable is coded 1 if yes, 0 if no, and Senior is the reference group. It is hypothesized that other things being equal, illiterate women, women with

Patterns of induced abortion

29

elementary education, and women with junior high school education are less likely to have an abortion, compared to the reference group, that is, women with senior high school education or above. Also, as already mentioned, the data contain detailed information about a woman’s attitudes toward the ideal number and sex composition of her children. One question asks about the number of children a woman deems to be ideal. The responses are categorized into no child, 1 boy, 1 girl, 1 child regardless of sex, 1 boy and 1 girl, 2 boys, 2 girls, 2 children regardless of sex, at least 1 boy, at least 1 girl, at least 1 boy and 1 girl, the more the better, do not care, others, and no idea. Using these 15 categories, one variable has been constructed to estimate a woman’s attitude about the ideal number of children, named Number_preference. It is coded 0 if a woman prefers no children, 1 if she prefers 1 child, 2 if she prefers 2 children, 3 if she feels the more, the better, and 4 if she does not care. It is hypothesized that other things being equal, the more children a woman desires, the lower the hazard of having an abortion. This is because if women desire more children or do not care, they are less likely to terminate a pregnancy with abortion, even if they are faced with fines or other punishments. Three independent variables are used that are related to China’s population policy. The first is a woman’s residence. It is termed Rural, a dummy variable, coded 1 if a woman lives in rural areas and 0 if she lives in urban areas. The hypothesis is that other things being equal, living in a rural area decreases the hazard of having an abortion for the pregnancy following the first child, compared to living in an urban area. Another policy-related variable is the ethnicity of a woman. In China, population policies for the most part have been relaxed for the minority nationalities. In most cases, minorities are allowed to have 2 children (China Population Information and Research center: http://www.cpirc.org.cn/). Minority couples may have 2 children and under special circumstances, they may even have 3 (Heberer 1989). Members of the Han nationality are subject to a stricter policy. A dummy variable, Han, is used that is coded 1 if a woman is Han and 0 if not. It is hypothesized that other things being equal, being a Han increases the hazard of having an abortion for the pregnancy following the first child, compared to being a non-Han woman. Another important policy-related factor is the time of the pregnancy following the first child. As mentioned earlier, China started its one-child policy in 1979; therefore pregnancies after 1979 were under the influence of the policy. Allowing some time for the policy to be diffused, the time of January 1980 is used. A dummy variable named After80 is introduced which is coded 1 if a woman’s first child was born after January 1980, and 0 if otherwise. It is hypothesized that other things being equal, a pregnancy following the birth of the first child that occurred after the policy is more likely to be terminated with an abortion than the pregnancy that occurred before the population policy. Women’s age is also one of the factors among the intermediate variables affecting fertility (Bongaarts 1978). It is known that in most societies, the age-specific fertility rate first rises and then falls, with the age of the mother. Therefore, age of the women at the time of the survey is also used as a control variable. Six regional dummy variables are also introduced as control variables. China has six very large regions, namely, Dongbei (Northeast) which includes Liaoning, Jilin, and Heilongjiang provinces, Huadong (East) which includes Shanghai city, Jiangsu, Zhejiang,

Fertility, family planning, and population policy in china

30

Anhui, Fujian, Jiangxi, and Shandong provinces, Huabei (North) which includes Beijing city, Tianjin city, Hebei, Shanxi, and Inner Mongolia provinces, Xinan (Southwest) which includes Sichuan, Guizhou, Yunnan, and Tibet provinces, Xibei (Northwest) which includes Shaanxi, Gansu, Qinghai, Ningxia and Xinjiang provinces, and Zhongnan (Mid-South) which includes Henan, Hubei, Hunan, Guangdong, Guangxi, and Hainan provinces. There are considerable differences in socioeconomic and demographic development among the six regions. The East is more developed, particularly along the east coast, and the Northwest and Southwest regions are less developed (Poston and Jia 1990). Also the eastern region is more rigorous in implementing the population policy. Each of the six region variables indicating where a woman resides is scored 1 if she resides in that region, and 0 if not. The East region will be used as the reference group. Table 2.1 lists the means, standard deviations, minimum, and maximum values of each variable in the dataset containing 11,762 women with at least one child. Rural women account for 78 percent of the sample. Also Han women dominate the sample. Among the first-born children, 49 percent were girls, and 57 percent of them were born after the one-child policy (After80). One quarter of the women are illiterate, 32 percent have an elementary education, 29 percent have a junior high school education, and 14 percent have a senior high school education and above. On average, most women prefer less than two children. Also it can be seen that the average age of women with at least one child is 35, and women are unevenly distributed among the six different regions of China.

Results Table 2.2 reports the results of eight nested hazard models. From Model 1, the most basic model, to Model 8, the complete model, are added increasing numbers of control variables. The results reported in the eight models are consistent and support the hypotheses. Other things being equal, having a daughter as a woman’s first child significantly lowers her likelihood of having the following pregnancy aborted; an illiterate woman has a significantly lower hazard of having an abortion for the pregnancy following a woman’s first child, compared to a woman with senior high school education or above; the more children a woman prefers to have, the lower her hazard of having an abortion for the pregnancy following her first child; living in rural areas reduces a woman’s hazard of having an abortion for the pregnancy following her first child; a Han woman is significantly more likely to end the pregnancy following her first child with abortion; and a pregnancy occurring after the implementation of the one-child policy has a much higher hazard of being terminated with abortion.

Patterns of induced abortion

31

Table 2.2 Cox proportional hazards models for the hazard of having abortion for the pregnancy following a woman’s first child, China, 1997 Model Model Model Model Model Model Model 7 Model 8 1 2 3 4 5 6 Girl

Illiterate

Elem entary Junior Senior Number_ prefe rence

Age

Rural

Han

After80

North

Northeast East

Mid-South Sout

−0.08* −0.10* −0.09* −0.09* −0.09* −0.09* −0.23*** −0.21*** (0.92) (0.90) (0.91) (0.91) (0.91) (0.91) (0.80) (0.81) −1.46 −1.39 −1.39 −0.99 −0.97 −0.90 −0.9 *** *** *** *** *** *** 9*** (0.23) (0.25) (0.25) (0.37) (0.38) (0.40) (0.37) −0.84 −0.79 −0.79 −0.44 −0.43 −0.38 −0.43 *** *** *** *** *** *** *** (0.65) (0.69) (0.43) (0.46) (0.46) (0.64) (0.65) −0.1 −0.17** −0.41 −0.40 −0.40 −0.19 −0.19 4** *** *** *** *** *** (0.66) (0.67) (0.67) (0.82) (0.82) (0.87) (0.84) Refer Refer Refer Refer Refe Refer Refere ence ence ence ence rence ence nce −0.19 −0.19 −0.15* −0.15 −0.29 −0.27 *** *** ** *** *** *** (0.82) (0.82) (0.86) (0.86) (0.75) (0.76) 0.00 −0.01* −0.01* −0.03 −0.03 *** *** (1.00) (0.99) (0.99) (0.97) (0.97) −0.53 −0.52 −0.85 −0.86 *** ** *** *** (0.59) (0.59) (0.43) (0.42) 0.33 0.45 0.37 *** *** *** (1.39) (1.57) (1.45) 2.82 2.82 *** *** (16.77) (16.86) −0.45 *** (0.64) −0.23 ** (0.79) Refer ence −0.38 *** (0.69) −0.09 (0.91)

Semistand ardized 0.90

0.65

0.82 0.92

0.82

0.79

0.69

1.11

4.11

0.86 0.94

0.85 0.97

Fertility, family planning, and population policy in china hwest North west Pseudo R2 Final log

32 −0.56* ** (0.57)

0.00 0.01 0.01 0.01 −22,26 −21,99 −21,97 −21,97 2.18 3.74 4.47 4.45

0.02 0.02 −21,9 −21,91 21.63 3.93

0.08 −20,54 4.03

0.87

0.08 −20,49 7.31

Note ***p