Ultra-low Fertility in Pacific Asia
Japan, Singapore, Taiwan, South Korea and Hong Kong SAR are among the very lowest-...
60 downloads
632 Views
3MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Ultra-low Fertility in Pacific Asia
Japan, Singapore, Taiwan, South Korea and Hong Kong SAR are among the very lowest-fertility countries in the whole world; even China has reached fertility levels lower than those in many European countries. If these levels continue over long periods, East Asia will soon face accelerating population decline, and in addition the changes in age distributions in such populations raise major new questions for planning of economic and social welfare. This book brings together work by noted experts on the low-fertility countries of East Asia with an up-to-date analysis of trends in fertility, what we know about their determinants and consequences, the policy issues and how these are being addressed in the various countries. Its role in bringing together information on policy trends and initiatives of a pro-natalist kind adopted over recent years in these countries is extremely important, as is the fact that the discussion of these pro-natalist policies is set in the context of a thorough analysis of what has driven fertility so low in these countries. Ultra-low Fertility in Pacific Asia is invaluable to students and scholars of East Asian public and social policy, as well as fertility studies more generally. Gavin Jones is Professor in the Asia Research Institute, National University of Singapore, where he is research team leader on the changing family in Asia. Paulin Tay Straughan is an Associate Professor and Deputy Head of the Department of Sociology, and Vice-Dean at the Faculty of Arts & Social Sciences, National University of Singapore. Angelique Chan is an Associate Professor of Sociology at the National University of Singapore.
Routledge Research on Public and Social Policy in Asia Edited by M. Ramesh National University of Singapore
1. Ultra-low Fertility in Pacific Asia Trends, causes and policy issues Edited by Gavin Jones, Paulin Tay Straughan and Angelique Chan
Ultra-low Fertility in Pacific Asia Trends, causes and policy issues
Edited by Gavin Jones, Paulin Tay Straughan and Angelique Chan
First published 2009 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 5RN Simultaneously published in the USA and Canada by Routledge 270 Madison Avenue, New York, NY 10016 Routledge is an imprint of the Taylor & Francis Group, an informa business
This edition published in the Taylor & Francis e-Library, 2008. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” © 2009 Editorial selection and matter, Gavin Jones, Paulin Tay Straughan and Angelique Chan. Individual chapters, the contributors. 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 Ultra-low fertility in Pacific Asia: trends, causes and policy dilemmas / editors: Gavin Jones, Paulin Tay Straughan and Angelique Chan. p. cm. – (Routledge research on public and social policy in Asia; 1) 1. East Asia—Population policy. 2. Pacific Area—Population policy. 3. Fertility, Human—East Asia. 4. Fertility, Human—Pacific Area. I. Jones, Gavin W. II. Straughan, Paulin Tay. III. Chan, Angelique Wei Ming. HB3650.5.A3U48 2008 304.6'32095–dc22 2008014894 ISBN 0-203-89006-X Master e-book ISBN ISBN 13: 978-0-415-46884-8 (hbk) ISBN 13: 978-0-203-89006-6 (ebk) ISBN 10: 0-415-46884-1 (hbk) ISBN 10: 0-203-89006-X (ebk)
Contents
List of figures and maps List of tables and appendices Contributors’ biographies Definition of terms Preface 1 Very low fertility in Pacific Asian countries: causes and policy responses
vii x xii xvi xvii
1
GAVIN JONES, PAULIN Tay Straughan, and ANGELIQUE CHAN
2 Explanations of low fertility in East Asia: a comparative perspective
23
PETER MCDONALD
3 Japan’s declining fertility and policy responses
40
NAOHIRO OGAWA, ROBERT D. RETHERFORD, AND RIKIYA MATSUKURA
4 The arrival of low fertility in China
73
BAOCHANG GU
5 Socioeconomic development and fertility in the People’s Republic of China and the Republic of Korea
96
DUDLEY L. POSTON, JR., HEATHER TERRELL KINCANNON, AND JUNGWON YOON
6 The 1997 Asian economic crisis and changes in the pattern of socioeconomic differentials in Korean fertility DOO-SUB KIM
110
vi Contents
7 Ultra-low fertility in Hong Kong: a review of related demographic transitions, social issues, and policies to encourage childbirth
132
PAUL YIP, C. K. LAW, and KAREN CHEUNG
8 Ultra-low fertility in Singapore: some observations
160
YAP MUI TENG
9 From population control to fertility promotion – a case study of family policies and fertility trends in Singapore
181
PAULIN Tay Straughan, ANGELIQUE CHAN, AND GAVIN JONES
10 Fertility in Pacific Asia: looking to the future
204
GAVIN JONES, PAULIN Tay Straughan, AND ANGELIQUE CHAN
Index
215
Figures and maps
Figures 3.1 Trends in the Total Fertility Rate (TFR), the Total Marital Fertility Rate (TMFR), and ideal family size (number of children), Japan, 1947–2005 3.2 Trends in Mean Age at Marriage (SMAM) by sex, Japan, 1950–2005 3.3 Trends in the synthetic proportion still single at age 50, by sex, 1951–2005 3.4 Trends in Period Parity Progression Ratios (PPPRs), Japanese women, 1950–2005 3.5 Mean Age at First Marriage (MAM), by sex, 1955–2002 3.6 Average age at first marriage for men, by education, 1955–2002 3.7 Average age at first marriage for women, by education, 1955–2002 3.8 Trends in the proportions of currently married women who worked before marriage and who worked for pay before marriage, Japan, 1955–2002 3.9 Trends in the proportion working in part-time jobs among paid employees age 20–29, by sex, 1972–2005 3.10 Trends in the proportion of firms reporting that they follow the lifetime-employment system, by firm size (number of employees), Japan, 1988–2002 3.11 Change in score for mate-selection criteria among single women, 1988–1998 3.12 Trends in the proportion of newly married couples who coresided with parents at the time of marriage, 1955–2002 3.13 Among currently married women who have had at least one birth and who worked before marriage, trends in the proportions who quit when they got married, who quit when they had a first birth, and who quit either when they got married or when they had a first birth, Japan, 1965–2002 3.14 Among single women not currently enrolled in school, trends in the proportions working part-time and full-time, 1990–2004
42 44 44 45 46 51 51
52 54
55 56 60
61 62
viii Figures and maps 3.15 Trends in proportion of women who were pregnant at the time of first marriage 3.16 Trends in mean age at marriage for those who were pregnant and those who were not at time of first marriage 3.17 Proportion who agree that husband should be breadwinner, wife should stay at home 3.18 Trends in the proportion of married women of reproductive age who took advantage of childcare leave, by year of child’s birth, 1980–2003 3.19 Among currently married women below age 50, trends in the proportions working full-time and part-time 4.1 Total fertility rate in China, 1950–2005 4.2 Population structure of China, 2000 (shaded) and 2050 4.3 Population structure of Shanghai, 2000 (shaded) and 2050 4.4 Population structure of Beijing, 2000 (shaded) and 2050 4.5 Infant mortality rate by sex and sex ratio of infant deaths in China for selected years, 1975–2002 5.1 Total fertility rates: China and South Korea, 1960–2005 6.1 Trends in the number of marriages and divorces, crude marriage rates, and crude divorce rates, 1981–2005 6.2 Trends in age-specific fertility rates, 1993–2005 6.3 CEB and post-1998 CEB by education of the couple, 2003 6.4 CEB by occupation of the couple, 2003 6.5 Post-1998 CEB by occupation of the couple, 2003 6.6 CEB by working status of the couple, 2003 6.7 Post-1998 CEB by working status of the couple, 2003 6.8 Post-1998 CEB by place of work of the couple, 2003 6.9 CEB by household income and family assets, 2003 6.10 Post-1998 CEB by household income and family assets, 2003 6.11 CEB of the two groups by wife’s place of work, 2003 6.12 CEB by the duration of work after marriage for the wife, 2003 6.13 CEB of the two groups by family assets, 2003 6.14 CEB of the two groups by status of housing ownership, 2003 7.1 Hong Kong: number of live births and TFR, 1961–2004 7.2 Age-specific fertility rates of Hong Kong, 1971–2004 7.3 Hong Kong: percentage distribution of live births by live-birth order, 1981–2003 7.4 Hong Kong: mean age of mothers by live-birth order, 1981–2003 7.5 Number of live births in Hong Kong to Mainland women whose spouses are Hong Kong residents (BSHKR), and live births with both parents being Chinese nationals but not Hong Kong residents (BnotHKR), 2001–2004 7.6 Proportion of live births in Hong Kong by parentage, 2001–2004 7.7 Average number of children ever born by the mother’s year of birth and age, Hong Kong
62 63 68
68 70 76 87 88 88 90 96 112 113 117 118 118 119 120 121 122 122 124 124 125 126 135 136 137 137
138 139 140
Figures and maps ix 7.8 Age-specific marital fertility rates of Hong Kong, 1971–2001 7.9 Proportions of never-married population by age and sex, Hong Kong, 1981–2001 7.10 General marriage rate and number of marriages, Hong Kong, 1971–2004 7.11 Median age of first marriage by sex, Hong Kong, 1981–2004 7.12 Median age for first marriage among women and interval between marriage and first-order live birth, Hong Kong, 1981–2001 7.13 Number of successful applicants for certificate of no-marriage registration (CAMR), Hong Kong, 1990–2003 7.14 Proportion of live births in Hong Kong by cohabitating parents, 1981–2003 7.15 Number of divorce decrees granted and the general divorce rate, Hong Kong, 1981–2001 7.16 Abortion to live-birth ratio, Hong Kong, 1992–2003 8.1 Live births and total fertility rate, 1957–2005 8.2 Age-specific marriage rates, male and female, 1980–2005 8.3 Age-specific fertility rates, 1980–2005 8.4 Proportion married among resident females, 2000–2005 8.5 Resident marital fertility rates, 2000–2005 8.6 TFR and GDP growth/unemployment rates, 1980–2005 10.1 Trends in percentage of population in two broad age groups
141 143 144 145 146 147 148 149 151 161 162 166 167 167 172 207
Maps 4.1 Geographic distribution of policy fertility, prefecture level, China, late 1990s 4.2 Policy-required fertility level by broad economic regions, China, late 1990s
77 81
Tables and appendices
Tables 1.1 Trends in Total Fertility Rates and projected population growth, selected East Asian countries 1.2 Delays in reversing anti-natalist policies, selected East Asian countries 1.3 Labor-force participation rates for females in Japan, Singapore, and South Korea, certain age groups, 1960–2000 2.1 Total Fertility Rates, 2005 2.2 Women aged 30–34, percentages never married and percentages never married and not cohabiting, selected countries 3.1 Model estimates of proportion of women still single at age 40, by education, childhood residence, and year of survey, Japan, 1990–2000 3.2 Men’s income and women’s expectations of prospective husband’s income, never-married persons age 25–34 in Hirosaki-shi in Aomori prefecture and Nerima-ku ward in Tokyo, 2003 3.3 Proportion never married among men age 25–34 by work status and annual income in Hirosaki-shi in Aomori prefecture and Nerima-ku ward in Tokyo, 2003 3.4 Major Japanese government actions aimed at raising fertility 4.1 Population changes in China, 2000–2006 4.2 Age-specific fertility rates and total fertility rate in China, 2000–2005 4.3 Demographic distribution of fertility policy, China, late 1990s 4.4 Policy fertility and observed fertility of China’s provinces, c. 2000 4.5 Respondents’ answers to attitude questions on family and having children, Jiangsu survey, end 2006 4.6 Reproductive ideal and reproductive desire, Jiangsu survey, end 2006 4.7 Factors to consider for having a second child, women aged 18–40 who are qualified to have two children, Jiangsu survey, end 2006
4 7 13 25 30
49
57
57 64 74 75 78 79 84 85
86
Tables and appendices xi 4.8 5.1
Life expectancy by sex in China for selected years, 1981–2002 Descriptive statistics, fertility rates: 2,870 counties of China in 2000, and 231 counties of South Korea in 2004 5.2 Multiple regression coefficients for the TFR on DTT independent variables: 2,432 counties and county equivalents of China, 2000 5.3 Multiple regression coefficients for the GFR on socioeconomic status, female status, and cultural values variables: 231 counties of Korea, 2004 6.1 Demographic profiles of the study population, 2003 6.2 Distribution of educational attainment and occupation of the two study groups, 2003 8.1 Total Marriage Rate (per 1,000 unmarried singles) 8.2 Proportions single by age and sex (percent) 8.3 Proportions single by age, sex, and educational attainment (percent) 8.4 Proportions single by age, sex, and ethnic group (percent) 8.5 Decomposition of TFR change, 2000–2005 8.6 Total Fertility Rates by ethnic group 8.7 Mean number of children born among ever-married resident females 8.8 Mean number of children born by resident ever-married females aged 40–49 by ethnic group 8.9 Mean number of children born among ever-married resident females aged 40–49 by education 8.10 Some trends in Singapore’s human resources 9.1 Summary of Singapore family policies and TFR, 1980–present 9.2 Macro data on economic health and employment 9.3 Bivariate correlations between TFR and economic-health indicators 9.4 Regression model predicting TFR – macro factor 9.5 Awareness of pro-family policies announced by the Government at the end of 2004 9.6 Perception of and attitudes towards pro-family policies 9.7 Items in scale measuring extrinsic expectations of child 9.8 Marital satisfaction scale 9.9 A Logistic regression model predicting future procreation decision, performance of model 9.9 B Logistic regression model predicting likelihood of having child in future 10.1 United Nations population projections: world and selected regions and countries, 2005–2050
89 98 102
104 114 115 163 163 164 165 168 168 169 169 170 170 188 191 191 192 193 194 194 195 196 196 205
Appendices 8
Chronological development of pro-natalist measures in Singapore 176
Contributors’ biographies
Angelique Chan is an Associate Professor of Sociology at the National University of Singapore. She obtained her PhD from the University of California, Los Angeles, and completed a post-doctoral fellowship at the University of Michigan, Population Studies Center. Dr Chan has been researching aging issues since 1990 when she began studying living arrangements of older adults in Malaysia and subsequently Singapore. Recently, she has been focusing on health issues among the aged, specifically, gender and socioeconomic differentials in health status among older adults in Singapore and the region. Her published works include articles and chapters focusing on aging in Singapore and comparative analyses of Singapore with other South-East Asian countries. Karen S. L. Cheung is a visiting scholar in the Division of Social Science at the Hong Kong University of Science and Technology. She is a demographer specializing in studies of mortality, population health, and longevity. She was a research associate at the French National Institute of Health and Medical Research (Institut national de la santé et de la recherche médicale, INSERM)/ the Health and Demography team (L’équipe démographie et santé). She earned her post-doctoral fellowship at the Department of Demography, University of Rome “La Sapienza,” Italy, after her PhD at the Hong Kong University of Science and Technology. Baochang Gu received his PhD in Sociology and Demography at the University of Texas in 1986. He is a Professor of Demography at Renmin University of China, and holds adjunct professor positions in a number of other universities. His committee memberships include WHO’s Specialist Panel for Social Science and Operations Research on Reproductive Health, the IUSSP Panel on low fertility, the Population Advisory Committee of China’s National Population and Family Planning Commission (NPFPC), the Specialist Panel on Reproductive Health, Ministry of Health (MOH), and the standing Council of the China Population Association. He has also served as Deputy Executive Director and Director of International Cooperation at China Family Planning Association (2001–2004), as Senior Associate at the Population Council, New York (1998–2000), and Associate Director of China Population Information and Research Center (1990–1998).
Contributors’ biographies xiii Gavin Jones is Professor in the Asia Research Institute, National University of Singapore, where he is research team leader on the changing family in Asia. After completing his PhD degree at the ANU in 1966, he joined the Population Council, where he worked first in New York, then in Thailand and Indonesia, before returning to Australia. He was then with the Demography and Sociology Program at the Australian National University for 28 years, serving as head of program for an 8-year period. Professor Jones has conducted joint research with many colleagues in South-East Asia. He has a long-standing interest in the formulation and evolution of population policy in the region. His research interest in recent years has focused especially on very low fertility regimes in Asia, delayed marriage, non-marriage, and cross-boundary marriage, and the dynamics of population and employment change in mega-urban regions of Asia. He has served as consultant to many international agencies, and has published about 20 books and monographs and some 130 refereed journal articles and book chapters. Doo-Sub Kim is Professor of Sociology, Hanyang University, Korea. He is also President, Population Association of Korea and Director, Institute of Population and Aging Research, Hanyang University. His current research focuses on low fertility, population aging, and marriage migration in Asia. He received his MA in Sociology from the Seoul National University and PhD from Brown University. He also has research and teaching experience at the East-West Center, Seoul National University, University of Oxford, University of Iowa, and University of British Columbia. He has published 26 books and monographs and some 80 journal articles and book chapters. His latest publication includes The 1997 Asian Economic Crisis and Changes in Korean Fertility. In recognition of his active consultation to the Korean government as well as academic contributions, he was recently awarded the Order of Service Merit from the President of Korea. Heather Terrell Kincannon is a doctoral student in the Department of Sociology at Texas A&M University. She holds an MS degree in sociology and demography from Texas A&M University. Her dissertation research deals with the patterns and dynamics of inter-racial marriage. Her prior research focused on socioeconomic development and fertility in China. Chi Kin Law has a post-doctoral appointment in the Faculty of Social Sciences of the University of Hong Kong. His areas of expertise are in demography, public health, and health economics. His PhD thesis at the University of Hong Kong was about Hong Kong’s demographic transition. Subsequently, he obtained further training at the World Health Organization (Geneva). Rikiya Matsukura is a staff researcher at the Nihon University Population Research Institute and a lecturer at the Advanced Research Institute for Science and Humanities, Nihon University. He has been a guest researcher and lecturer at the Statistical Research and Training Institute of the Ministry of Internal Affairs and Communications, Japan, from 2002 to the present. He also spent a
xiv Contributors’ biographies year (2005–2006) as a collaborative researcher at the Japanese Government’s Institute of Statistical Mathematics. He has more than 20 years of experience in statistical research and his research areas include the development of statistical methods for complicated models and the application of these methodologies to social science. He has published articles in journals such as Population and Development Review, The Japanese Economy, Asian Population Studies, and Asia-Pacific Population Journal. Peter McDonald is Professor of Demography at the Australian National University. He has been elected Vice President (2006–2009) and President (2010–2013) of the International Union for the Scientific Study of Population, the international association of professional demographers. He is frequently consulted on the issue of population futures (causes, consequences, and policies) by governments around the world, especially in Australia, Europe, and East Asia. He is a leading expert on policies, including labor-supply policies, for countries facing very low fertility rates. His work has changed the nature of the population debate in Australia by bringing demographic realities into a debate that had relied mainly on rhetoric. He has been highly influential in work and family policy and labor-supply futures policy. His theoretical work on the causes of very low fertility rates is widely cited in academic papers and meetings and acknowledged by governments around the world, which are addressing the problems associated with very low fertility rates. Naohiro Ogawa is Director of the Population Research Institute of Nihon University, Tokyo. He is also Professor of Population Economics at the Advanced Research Institute for the Sciences and Humanities of Nihon University. Over the last 25 years, he has written numerous academic papers and books on population and development in Japan and other Asian nations. His many papers have been published in internationally respected journals including Population and Development Review, Population Studies, American Economic Review, and Journal of Labor Economics. Dudley L. Poston, Jr. is Professor of Sociology, Director of Asian Studies, and the George T. and Gladys H. Abell Endowed Professor of Liberal Arts at Texas A&M University. He holds Adjunct Professorships at Renmin University, Fuzhou University, and Nanjing Normal University. His research focuses on the demography of China and South Korea, the demography of homosexuality, and the social demography of gender. Robert D. Retherford is Coordinator of Population and Health Studies at the East-West Center and Affiliate Graduate Faculty in Sociology at the University of Hawaii. He joined the East-West Center in 1970 after receiving his PhD in Sociology at the University of California at Berkeley. His current research focuses on below-replacement fertility, population aging, and a variety of maternal and child-health issues in Asia. His current work on below-replacement fertility and population aging focuses on Japan and China. His current work on maternal and child health focuses mainly on India. He has also done
Contributors’ biographies xv previous research on the Philippines, Thailand, Pakistan, Nepal, and Sri Lanka. His work at the University of Hawaii involves teaching graduate courses and supervising students in the University’s Population Studies Program. Paulin Tay Straughan is an Associate Professor and Deputy Head of the Department of Sociology, and Vice-Dean at the Faculty of Arts & Social Sciences, National University of Singapore. Her research interests include medical sociology and health-care systems, with a special focus on preventive health (especially cancer prevention), the importance of social-support networks in health-related issues, sociology of the family, work–family relations and fertility, and aging. She is currently researching Singapore’s fertility trends; and divorce in Singapore. She is also a Co-principal Consultant for the National Survey of Elderly in Singapore. Yap Mui Teng is a Senior Research Fellow at the Institute of Policy Studies, where she leads research in the area of demography and family. Her current research interests include policy responses to low fertility in Singapore and elsewhere, policies on the aged, migration policies, and poverty alleviation policies. She worked as a statistician at the Population Planning Unit, Ministry of Health, before joining IPS in 1989. She was a Research Fellow at the Population Institute, East-West Center, Honolulu, Hawaii, after graduating with a PhD in Sociology from the University of Hawaii. Prior to that, she worked as a statistician at the Research and Evaluation Unit, Singapore Family Planning and Population Board. Paul Yip is a Professor in social work and social administration at the University of Hong Kong (HKU). His areas of expertise are in biostatistics, demography, and population health. He is also the director of the Center for Suicide Research and Prevention at HKU. He is a member of the working group of the population policy of the Hong Kong Government and a Council and Executive member of the Family Planning Association of Hong Kong. Jungwon Yoon is currently a PhD candidate in the history and sociology of technology and science program at Georgia Institute of Technology. She earned a Master’s degree in sociology from Texas A&M University. Her recent research focuses on social studies of science and technology, including human resources in science and technology, the digital divide, and the development of information technology in Asian developing countries. Her prior research focused on the pattern of low fertility in South Korea.
Definition of terms
This book is meant to be accessible to the general reader, but there are necessarily some demographic terms and analysis of a more specialized nature. We have tried to keep these to a minimum. In order to facilitate the general reader’s navigation of the book, we provide here definitions of the key demographic measures that are widely used throughout the book. Other specialized measures used less widely in the book are defined where they occur. Age-specific fertility rate (ASFR) The ASFR in a particular calendar year is calculated as the number of births that occurred during the year to women of a given age, divided by the mid-year number of women of the same age. Total fertility rate (TFR) The TFR is widely used throughout this book, and is also the measure of fertility most commonly used by policymakers. It can be understood as the number of births a woman would have by age 50 if she lived through her reproductive period experiencing the age-specific fertility rates that prevailed in the population in the particular calendar year. In other words, it is a hypothetical cohort measure based on the fertility experience of women in a particular year. The TFR is calculated by summing the ASFRs (births per woman per year at each age) between the ages of 15 and 50. Replacement-level fertility This term refers to the level of fertility that, if continued over time, would result in a population (closed to migration) reproducing itself exactly, resulting in a steady-state population, neither growing nor declining. The level of TFR corresponding to replacement level depends on the level of mortality in the population, but in the populations being studied in this book, a TFR of 2.1 approximates the replacement level. It is important to be aware that in the short to medium term (which can be as long as several decades), low-mortality populations with TFR of 2.1 can be growing or declining, sometimes quite substantially, as a result of wide differences in their age and sex structures. But if this fertility level is continued long enough, age-structure differences between populations will gradually disappear, and in the long run they will be exactly replacing themselves.
Preface
The issue of ultra-low fertility in Pacific Asia appears to have crept up on the world’s consciousness surreptitiously. Within the region it receives considerable attention from the media and from policymakers. But outside of the region there does not appear to be much awareness of it. No doubt this can be partly explained by the recency of ultra-low fertility in Pacific Asian countries. In Europe, very low fertility was in place in a number of countries well before the end of the twentieth century. By contrast, the movement to very low levels of fertility in Pacific Asian countries – with the exception of Japan and Hong Kong SAR – has been very recent, in the first five years of the twenty-first century. Japan, Singapore, Taiwan, South Korea, and Hong Kong SAR are now among the lowest-fertility countries in the whole world, and even China has reached fertility levels lower than those in many European countries. Fertility has sunk so low in many East Asian countries that if these levels continue over long periods, populations will face accelerating population decline not very far into the future. Aside from this, changes in age distributions in such populations raise major new questions for planning of economic and social welfare. The best-recognized prospect raised by ultra-low fertility is population aging, which brings with it an entirely new set of issues, for example, increasing old-age dependency ratios, financing old age and old-age health care, continuing familial support of the elderly, and elderly political participation. Existing pro-natalist policies do not appear to be having much effect in these countries, and although governments realize that more needs to be done to encourage fertility, exactly what should be done remains elusive. Part of this elusiveness stems from a lack of information on what motivates people in Asia to marry in the first place, and once married, to have, or not have, children. The three editors have been studying determinants of low fertility in Singapore for some time, and had the opportunity in February 2007 to bring together some of the world experts on low fertility in Asian countries at a small conference, funded by the National University of Singapore. The decision was then taken to publish a book comprising revised versions of some of the papers presented at the conference, together with some additional commissioned papers. The editors have prepared a concluding chapter to complete the book. The need for such a book is very clear. It is indeed surprising, in view of the
xviii Preface extreme concern governments of the ultra-low-fertility countries of Pacific Asia are showing about this situation, that there has been no recent book summarizing the trends and dealing with their causes and the policy issues arising from them. Two useful earlier studies (Richard Leete and Iqbal Alam, eds, The Revolution in Asian Fertility: Dimensions, Causes and Implications, Oxford: Clarendon Press, 1993; and Korea Institute for Health and Social Affairs and UNFPA, Low Fertility and Policy Responses to Issues of Ageing and Welfare, Seoul: Korea Institute for Health and Social Affairs, 2000) dealt with some of the issues, but both publications examined Asian countries as a whole, thus failing to focus specifically on the issues faced in common by the ultra-low-fertility countries. Both books also pre-dated the sharp further fall in fertility that has taken place in these countries over the first five years of the twenty-first century. The present volume fills this need by bringing together papers by noted experts on the low-fertility countries of East Asia with an up-to-date analysis of trends in fertility, what we know about their determinants and consequences, the policy issues, and how these are being addressed in the various countries. It is thus both comprehensive and focused. Not only does it bring together information on policy trends and initiatives of a pro-natalist kind adopted over recent years in these countries, but it sets these in the context of a thorough analysis of what has driven fertility so low in these countries. In preparing this book for publication, we are indebted to many people. First of all, the National University of Singapore has provided the resources and supportive environment in which our own research has been able to take place. The Faculty of Arts and Social Sciences and the Asia Research Institute contributed the funds for the conference which provided the starting point for this book. We have received excellent support from our research assistant, Windel Lacson, and Farheen Mukri has given editorial assistance to prepare the papers for publication. Finally, the editors at Routledge, Stephanie Rogers and Leanne Hinves, have been enthusiastic about the book and have given us every assistance to arrange its timely publication. Gavin W. Jones Paulin Tay Straughan Angelique Chan Singapore, February 2008
1
Very low fertility in Pacific Asian countries Causes and policy responses Gavin Jones, Paulin Tay Straughan, and Angelique Chan
Introduction Only 40 years ago, population experts were still worried about a population explosion that would threaten the future of humanity. Fortunately, while population growth is currently largely under control, sub-Saharan Africa and parts of South Asia still face massive increases with very serious potential consequences.1 Paradoxically, however, a new problem is emerging, with its key locus in Pacific Asia (the term used in this book to refer to Asian countries with a Pacific littoral). This problem is ultra-low fertility. Japan, Singapore, Taiwan, South Korea, and Hong Kong SAR are among the very lowest-fertility countries in the whole world, and even China has reached fertility levels lower than those in many European countries. Fertility has sunk so low in many East Asian countries that if these levels continue over long periods, populations will face accelerating population decline not very far into the future. Not only this, but changes in age distributions in such populations raise major new questions for planning of economic and social welfare. The best-recognized prospect raised by ultra-low fertility is population aging, which brings with it an entirely new set of issues, for example, increasing old-age dependency ratios, financing old age and old age health care, continuing familial support of the elderly and elderly political participation. But there are many others, for example, the decline in size and changing age structure of the workforce, and the declining visibility of and perhaps attention paid to the needs and interests of children and young people. Existing pro-natalist policies do not appear to be having much effect in these countries, and although governments realize that more needs to be done to encourage fertility, exactly what should be done remains elusive. Part of this elusiveness stems from a lack of information on what motivates people in Asia to marry in the first place, and once married, to have, or not have, children. (The route of bearing children without marrying is a little-travelled one in Asian countries, and not socially sanctioned.) Though the shift in emphasis in population policy from anti-natalism to pro-natalism in some Asian countries over the course of just a decade or two is striking, the underlying issue is the same: whether governments have a legitimate interest in nudging fertility rates in desired directions, and if so, how policies designed to
2 Gavin Jones, Paulin Tay Straughan, and Angelique Chan accomplish this relate to family and social policy. Thus, it is of some relevance to recapitulate briefly the history of anti-natalist policies.
The case for fertility reduction in high-fertility countries The general consensus among economists has been that there are social and economic benefits to be achieved by reducing fertility rates in high-fertility conditions. The literature on this is vast, but three references summarizing the received wisdom are National Research Council, 1986 (which drew very cautious conclusions), Cassen, 1994, and Birdsall, Kelley and Sinding, 2001. Thus, although there have always been dissenters (for example, Simon, 1981), the consensus among most economists over the most active period of family planning efforts was that governments are justified in taking active steps to bring birth rates down – in the interest of the welfare of future generations. The economic case for family planning was bolstered by the arguments of environmentalists, and those dealing with issues of food and water availability (see Pimentel, et al., 1999; Alexandratos, 2005; Falkenmark, 1997). Family planning programs were seen as the most direct route to achieving the goal of fertility reduction. They were designed to make information on contraception more readily available, and to facilitate the adoption of contraception by couples wishing to do so (Seltzer, 2002). Studies that estimated “unmet need” for contraception indicated that many couples at risk of pregnancy and who did not want any more children were not doing anything to protect themselves from pregnancy. Satisfying this unmet need was seen as appropriate in both meeting people’s expressed needs and in lowering the overall level of fertility (see, for example, Sinding, Ross, and Rosenfield, 1994; Casterline and Sinding, 2000). Many family planning programs went well beyond the mere supply of information and services, engaging in campaigns to persuade people of the advantages of delayed marriage and small family size. But there are other routes as well – arguably more fundamental ones – to low fertility. Studies dealing with the determinants of fertility show consistently that fertility is inversely related to education of women, for example. In general terms, fertility is inversely related to levels of economic development and (a somewhat different indicator) human development (see UNFPA, 2003: 4), and an ongoing debate is therefore about whether “development is the best contraceptive” (the slogan adopted by many at the World Population Conference in 1974), or whether the urgency of reducing birth rates calls for more direct measures. Finally, to confuse the situation even more, there are countries (Myanmar a notable example) where fertility has fallen to fairly low levels without either much evidence of development or the assistance of an official family planning program. Two key problems with the family planning approach were:
•• In its implementation by governments that perceived (and were pressed by donors to perceive) the lowering of population growth as an overriding goal, the basic rights of the population were often given second place.2
Very low fertility in Pacific Asian countries 3
•• Family planning programs had limited success in settings in which religious opposition was not effectively counteracted, and where economic development was sluggish and not effectively felt by the masses (examples include Pakistan, the Philippines, and parts of Latin America). Over the past decade, demographic trends have taken the wind out of the sails of the population-control movement. Fertility rates have fallen consistently across the world, though Africa and parts of South Asia are regions where fertility remains high, massive population increases are still in prospect, and the effect of this on human well-being requires ongoing attention. Still, population projections by the United Nations now show world population peaking at below 10 billion, whereas figures closer to 15 billion had earlier been thought likely.
The collapse of fertility in East Asia and the delayed switch to pro-natalist policies There is now an interesting mix of situations in the world: in some countries, lowered fertility is arguably crucial to sustainability and the well-being of populations; in others, increased fertility is arguably equally critical to future wellbeing; in others, a laissez-faire approach to fertility seems justified. In the first two groups of countries, the issue of whether, and if so how, governments should become involved in matters that are frequently argued to be private and confined to the bedroom continues to exercise the minds of government planners. The second group – countries needing to increase fertility – includes countries such as Japan, South Korea, Singapore, and Taiwan. Over the past five years or so, fertility in the first four of these countries has fallen to levels below those of almost all countries in Europe. There is controversy about where China and Thailand fit: needing to reduce fertility further or needing to raise it. Table 1.1 gives a brief summary of the population prospects faced by some of these countries. It shows that Japan, South Korea, Taiwan, and Singapore have now reached such low fertility levels that population declines have begun or are in prospect. The momentum towards a shrinking of population, inherent in the age structure and fertility levels, is well documented for these countries. In particular, they face sharp contractions in numbers in the adolescent and young-adult age groups, the key age groups affecting the size and dynamism of the workforce. Only a sharp rise in fertility or very high levels of immigration (the latter to some extent built into United Nations projections for Singapore and Hong Kong SAR) could prevent population decline. Politicians and planners in these countries are viewing ultra-low fertility as a “crisis”. It is important to note that some four or five decades ago, four of these East Asian countries with ultra-low fertility (South Korea, Taiwan, Singapore, and Hong Kong) were facing high fertility and rapid population growth; all of them were considered densely populated, and their governments were very concerned at the prospect of rapid population growth. It is not surprising, then, that they were pioneers in adopting policies to reduce fertility, including family planning
4 Gavin Jones, Paulin Tay Straughan, and Angelique Chan Table 1.1 Trends in Total Fertility Rates and projected population growth, selected East Asian countries Total Fertility Rates 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 Projected % growth, UN medium projection, 2005–30* Population 15–64 age group 15–24 age group Projected % growth, UN low projection, 2005–30* Population 15–64 age group 15–24 age group
Japan 1.42 1.43 1.39 1.38 1.34 1.36 1.33 1.32 1.29 1.29 1.25 1.32
South Korea 1.64 1.70 1.54 1.47 1.42 1.47 1.30 1.17 1.17 1.16 1.08 1.13
Taiwan 1.78 1.76 1.77 1.47 1.56 1.68 1.40 1.34 1.24 1.18 1.12 1.12
Singapore 1.67 1.66 1.61 1.47 1.47 1.60 1.41 1.37 1.25 1.24 1.25 1.26
Hong Kong SAR n.a. 1.19 1.12 1.02 0.98 1.04 0.93 0.94 0.90 0.93 0.97 0.98
–4.3 –17.0 –20.0
2.8 –8.7 –36.9
n.a. n.a. n.a.
20.2 –0.3 –25.2
22.3 4.2 –15.3
–9.9 –19.8 –37.7
–4.3 –12.2 –53.1
n.a. n.a. n.a.
12.7 –3.3 –42.3
13.0 0.9 –38.2
Source: Japan: Ministry of Health and Welfare, Vital Statistics, various years; Korea: June 2004, Table 3.6; Hong Kong: Census and Statistics Department, The Government of the Hong Kong Special Administrative Region (using resident population approach); Taiwan Province of China: Tsay, 2003. Figures for 2003–2006 from Department of Household Registration Affairs, Ministry of Interior; Singapore: Singapore Department of Statistics, various years. * United Nations Population Division, 2006.
programs (Robinson and Ross (eds), 2007). Japan’s situation was different, in that it had already reached replacement-level fertility at that time. The history of population policies in the other four countries, adopted to deal with a situation contrasting so sharply with the current demographic prospects they face, should be kept in mind in understanding delays in altering policies to respond to changing circumstances, as discussed below. Governments in ultra-low-fertility countries are now arguing that raising birth rates is crucial for national survival and welfare. But the change from anti-natalist to pro-natalist policies did not come easily. Singapore was the first government in the region to reverse anti-natalist policies. The long lag in recognizing the need for policy change meant that the reorganization of policy was abrupt, and somewhat shocking to the general public, who had long grown used to anti-natalist messages. Singapore’s fertility fell to replacement level in 1975, but it was not until 1986 that the first signs of a change in policy were seen, with a government
Very low fertility in Pacific Asian countries 5 announcement of plans to review population policies and dialogue sessions with the public. Pro-natalist measures were finally introduced in 1987, 12 years after replacement fertility had been reached and with fertility close to 25 percent below replacement level.3 The curious result of the haste in which the policy changes were made was that almost overnight anti-natalist messages were replaced by pro-natalist measures on the nation’s bus stops and other places, presumably to the bemusement of a populace well primed to follow the exhortations of a seemingly omniscient government. The Korean government showed a similar reluctance to modify policies after fertility fell below replacement level in 1984. By 1990, fertility was well below replacement level, prompting a debate on population policies. Those supporting continuation of fertility control argued that the current level of low fertility is mostly due to the strong population control policies and the change of policies would bring the fertility level up resulting in rapid population growth again, slowing down economic growth and effecting heavy burdens on environment and resources. Those supporting the relaxation of fertility control policies argued that the socioeconomic conditions of Korea have changed greatly resulting in changes in attitudes and values towards preference for small size families. They also argued that further decline in fertility would result in rapid population aging and increasing burden of support for the elderly. (Choe and Park, 2005: 8) It was not until 1996 that the emphasis of population policy was shifted from population control to quality and welfare of the population, one of the aims being to prevent fertility from declining further from its level of 1.7 at that time. Taiwan’s fertility had been below replacement level for eight years before a new population policy statement was issued by the Executive Yuan, calling for raising fertility to replacement levels. However, there were no explicit proposals as to how to raise the marriage rate or the birth rate. In an assessment of this policy, Freedman and Freedman (1993: 28) wrote: “We see nothing in the new policy statement that is likely to have an effect on fertility levels, one way or the other. Therefore, the balance of the existing pro-natalist and anti-natalist influences in the society are likely to determine what actually happens”. It was not until 2006, 22 years after fertility fell below replacement level, that any national-level pronatalist policies were introduced, although a few county-level administrative units had introduced small-scale pro-natalist measures. Japan followed a different fertility trajectory, having experienced belowreplacement-level fertility over the 1970s and 1980s. Although, unlike the other countries, it did not have anti-natalist policies to reverse, it was not until it reached a record low TFR of 1.57 in 1990 that the government started looking into possible measures to reverse the downward trend. As for China, the implementation of the “one-child policy” after 1979 has been greatly modified over time, and the “one-child policy” these days applies
6 Gavin Jones, Paulin Tay Straughan, and Angelique Chan to only 35 percent of China’s population (Gu, Chapter 4, this volume). Nevertheless, China’s population policy remains clearly anti-natalist. There is controversy over China’s current fertility levels, but it is clear that fertility is so far below replacement level that China is actually in much the same position that Singapore was in 1975: namely, of having succeeded in the policy of fertility reduction, but being slow to “take off the brakes” when fertility slid well below replacement level. In China, many interpret the slide in fertility to well below replacement level as a sign of success. On the other hand, the population structure, aging and other issues arising from recent low fertility levels (compounded by the distorted age structures inherited as a legacy of past events – notably the planning disaster of the “Great Leap Forward” – and changing population policy) have been well publicized in the literature. Many demographers and economists are now arguing that pro-natalist policies are needed (Wang Feng, 2005; Zeng Yi, 2007). But government leaders and the population-control establishment have been slow to change their mindset. In February 2008, the Vice Minister of the National Population and Family Planning Commission did announce that China is considering “incrementally” lifting limits on the number of children a couple can have – an announcement that appeared to be subsequently rebutted by Premier Wen Jiabao (Straits Times, March 6, 2008). But even the initial announcement indicated a weakening of anti-natalist measures, rather than foreshadowing a switch to pro-natalism. There is therefore a fascinating history with respect to the change of government policy in East Asian countries in response to changing demographic circumstances. South Korea, Taiwan, Singapore, and China all had long-standing policies to lower fertility, centered on family planning programs. Their delays in reversing policy can be summarized by noting the number of years that their TFR had fallen below replacement level, and the percentage below replacement reached at the time policy was altered, as shown in Table 1.2. It should be noted that in the cases of South Korea and Taiwan, though policies were modified at earlier points in time, it was more than 20 years after the replacement fertility level was breached that more serious pro-natalist measures were introduced. The same may well turn out to be the case in China. Why the delay? There were probably three main reasons:
•• Demographic momentum meant that population kept increasing despite
•• ••
below-replacement fertility, thus seemingly obviating the need to modify or reverse policy. This is certainly a major reason for the continuation of a strongly anti-natalist policy in China. Inertia and the entrenched bureaucratic interests and mindsets of agencies entrusted with anti-natalist policies and leaders who had been promoting anti-natalist policies. A deficiency in the theory of demographic transition. Little attention was given to what happens after fertility reaches replacement level, and the assumption prevailed that fertility would not fall far below replacement (Demeny, 1997). The United Nations Population Division population projections for a
Very low fertility in Pacific Asian countries 7 Table 1.2 Delays in reversing anti-natalist policies, selected East Asian countries Country
Year in which replacement fertility was reached
Year in which Number % below Comments anti-natalist of years replacement policy was elapsed when policy reversed reversed
Singapore 1975 South Korea 1984
1987 1996
12 12
25 20
South Korea 1984
2004
20
50
Taiwan
1984
1992
8
20
Taiwan
1984
2006
22
47
Japan
19731
1990
17
25
China
1992
No reversal
16 (+)
—
— Very mildly pro-natalist policies More serious pro-natalist measures Pro-natalist statement but no measures Specific pro-natalist measures under consideration Mildly pro-natalist measures 25% below replacement in 2007; policy not altered
1
Actually, Japan’s TFR was slightly below replacement level as early as the 1950s, but it hovered around that level for two decades, and did not fall definitively below replacement level until 1973.
long time assumed, against all the evidence, that TFRs that had sunk below replacement level, even well below this level, would gradually rebound to replacement level. This added official weight to the idea that there was no need for a policy response to very low fertility.
Pro-natalist policies, family policy and social policy in the very low-fertility countries Before giving some details about pro-natalist policies in Asian countries, we will set these in a broader context by looking first at some aspects of family policy in general, and considering how it has been used in European low-fertility countries. One thing to keep in mind is that, unlike in East Asia, “in most European countries overt population-policy measures would meet resistance rather than acclamation among the population” (Neyer, 2003: 49). But family policy measures are an acceptable means of encouraging child-bearing. Of course, family policies are difficult to conceptualize and measure, and represent a diverse range of policy objectives. The primary purpose is not always connected to child-bearing and child-raising as such. Because of this diversity, “family policies may encompass inconsistent or even divergent aims” (Neyer, 2003: 51). McDonald (2002: 435) classifies policies directed toward the reversal of low fertility into three broad categories:
8 Gavin Jones, Paulin Tay Straughan, and Angelique Chan
•• financial incentives •• support for parents to combine work and family •• broad social change supportive of children and parenting. The first category includes child benefits (public transfers paid for children). The second includes such items as maternity-leave policies, parental-leave policies (leaves of absence from employment granted parents by law in order to take care of their child during the first few years of life), and childcare services (offered by the state, the market, employers or non-profit institutions). The extent to which support of these kinds is provided to parents varies greatly between European countries, largely according to the kind of welfare-state regime they follow. Common classifications of welfare-state regimes in Europe distinguish between universalistic welfare states (the Nordic countries), conservative welfare states (continental European countries), liberal welfare states (Anglo-Saxon countries), and Southern European welfare states (see, for example, Gauthier, 2002, Table 1). Conservative welfare states rely heavily on familialism – that is, on the family as a provider of welfare. Southern European countries display an even higher degree of familialism. In this respect, Southern European countries clearly have an important element in common with the East Asian countries currently facing the issue of how to raise birth rates, to match the other element they share in common: that of having the lowest levels of fertility in the world. It has been argued persuasively by McDonald (2000a; 2000b) that the sharing of these two common elements is not accidental; it is precisely their familialism, in the context of widened educational and employment opportunities for women, that poses strong conflicts of interest for women and leads to delayed marriage and low levels of child-bearing. To return to the Asian countries that are the focus of this paper, we will summarize briefly the kinds of pro-natalist policies some of these countries have been introducing. Japan Japan4 has gradually been cranking up its pro-natalist policies as the level of concern with very low fertility rises. Japanese policy has followed two main approaches: direct subsidies for child-bearing and child-raising; and changing the institutional framework to facilitate marriage and child-raising. As in many other countries, child allowances (first introduced in 1972) were a family policy measure to assist low-income families, rather than a pro-natalist measure. But after 1990, pro-natalist concerns led to large increases in the allowances, though an eligibility criterion remains. In 1991, unpaid leave for childcare was introduced, though part-time workers were excluded. The “Angel Plan,” introduced in 1994, called for major expansion of childcare centres, with eligibility criteria varying by locality. Later, part payment during childcare leave was introduced. In 1999 the “New Angel Plan” called for further expansion of the heavily subsidized day-care centres, after-school programs and family support centres. In 2001,
Very low fertility in Pacific Asian countries 9 the proportion of salary received by an employee on childcare leave was raised from 25 percent to 40 percent, and this was raised to 50 percent in late 2007. It seems that many employees, especially women, were not taking childcare leave because of social disapproval from fellow workers. Thus the government introduced measures aimed at creating an atmosphere within firms that would encourage parents to take the childcare leave to which they were entitled. Complying firms could use a logo saying “we support child-bearing among our employees”. Finally, in 2005, the government extended the right to childcare leave to part-time workers, under certain circumstances. Workers on short-term (for example, three-month or six-month) contracts whose contracts were not renewed were not eligible for such leave, however. The gradual gearing up of pro-natalist policies in Japan is clear, but it is noteworthy that large tax deductions for children are not yet entertained. Moreover, the typically very small contribution of husbands to housework and child-rearing tasks (see Tsuya et al., 2005) means that the burden borne by the working wife remains very high. Without change in broader gender relations in Japan, the task of raising fertility appears formidable. South Korea5 In 1996, the South Korean government adopted new population-policy goals, in the face of continuing declines in fertility (the TFR had been around 1.6 or 1.7 for a decade). However, these goals were hardly pro-natalist, but rather emphasized maintaining the level of fertility, better reproductive health, redressing the imbalance in sex ratios at birth and reducing the incidence of induced abortion, tackling the sex-related problems of youth and adolescence, and empowering women by expanding employment opportunities and welfare services for them (Cho and Lee, 2000: 151–161). A decade later, with fertility dipping even lower, in 2006 the Korean government was developing policies (“Vision 2020”) in response to low fertility and an aging society. It is attempting to create a favorable environment for childbirth and child-rearing by transferring some of the burden of child-raising from family to society. More specifically, subsidies for the costs of child-rearing and education, currently available for low-income groups, are to be available for the middle class as well. Day-care for children up to age 4 would be subsidized according to the family’s income level. After-school programs would be expanded, particularly for lower-grade primary school children, as an alternative to expensive private tutoring institutions. Taxes would also be lowered for households with young dependent children or large families, and the tax system altered to reduce the costs of health insurance for such families. The introduction of a child-allowance system, which Korea as yet does not have, is being seriously considered. Childcare facilities are to be expanded in number and quality, with the government playing a central role. This is part of an effort to create a family-friendly and gender-equal social culture, making work and family more compatible. Companies providing maternity leave will be supported. Women workers at small
10 Gavin Jones, Paulin Tay Straughan, and Angelique Chan to mid-sized firms who take maternity leave can receive up to three months of employment insurance protection. Starting in 2008, male partners will automatically receive three days off to help after childbirth. Childcare leave that hitherto applied only to parents of children under the age of one will now be extended to children up to three years of age. Singapore6 Tracing the history of anti-natalist policy in Singapore is complicated by the fact that for a period of time, beginning in 1984, Singapore adopted policies that were seemingly unique in Asia. These policies were selectively pro-natalist for the well-educated but anti-natalist for the poorly educated. They were adopted in the interests of improving the genetic quality of the population, but not directed at a general increase in fertility rates. Not very long after that, in 1987 however, Singapore reversed its fertility objectives. Since then, it has experimented with a wide range of measures designed to raise fertility. The general trend has been towards a strengthening of these measures over time, and especially since the further downward trend in fertility since 1996.7 The first task was to loosen or abandon the old anti-natalist policies, and this was gradually done (Saw, 2005: Chapter 11). A number of limited pro-natalist measures were also introduced in 1987. For example, couples having a third child were given priority in getting access to a larger HDB apartment through sale of their smaller apartment, and tax rebates were granted for third or fourth children. Subsequently, the involvement of the government in encouraging marriage was widened from a concern with ensuring that well-educated women married (through the Social Development Unit) to a broader program targeting also the non-tertiary educated. Further pro-natalist policies were announced in 2000, with the introduction of the baby bonus scheme for second and third children. The scheme consists of a two-tier payment given annually by the government for a period of six years after the birth of the child. The first tier is an outright cash gift (paid in five instalments over five years), totalling S$3,000 for the second child and S$5,000 for the third child, while in the second tier both parents and government contribute to a co-savings account. The scheme has been structured so that the funds must be used solely for the benefit of the children. Other pro-natalist provisions announced in 2000 included limited arrangments for paid leave in the public sector to marry and to attend to sick children, flexible working hours, and childcare subsidy for enrolment in childcare centres. In 2004, it was decided that these measures had been insufficient, and a raft of new schemes were introduced, including a Medisave maternity package and extra paid maternity leave (extending paid maternity leave from the 8 weeks in 2001 to 12 weeks). In addition there were: further modification of the provisions for getting an HDB apartment to encourage marriage; extensions to the baby bonus from the second and third children to include the first and fourth child; an increase in the subsidy paid by the government for enrolment of an infant
Very low fertility in Pacific Asian countries 11 in childcare; a more liberal parenthood tax rebate; and a streamlined working mother’s child-relief scheme. Further measures announced in 2004 included: provision of the statutory two-day paid childcare leave for a parent of a child under 7 years of age; a lower maid levy for parents with children under 12 years of age; tax relief for working mothers where the grandparent serves as care-giver; introduction of a five-day working week in the civil service (though the week’s total working hours remain the same); equal medical benefits for male and female civil servants; and incentives for firms to seek better “work-life harmony” for their employees. Taiwan As noted earlier, the new population policy announced in Taiwan in 1992, though advocating the maintenance of reproduction at the replacement level, and essentially ending the family planning program, included no specific pro-natalist measures. It seems that even up to the present no financial incentives have been provided for childbearing, except for limited tax deductions and, in 2007, free kindergarten schooling for low income families. Policies have included the exhortation for more single Taiwanese to marry, but the rapid rise in marriages between Taiwanese men and foreign women (Tsay, 2004) appears to have caught the government by surprise. Assessment of policies in these East Asian countries How do the pro-natalist policies in Japan, Korea, Singapore, and Taiwan compare with those in European countries? A systematic assessment of this would require much more time and resources than we are able to apply to it, but instead we will try to put forward a few generalizations. First, these Asian countries follow a familialist approach that places much higher expectations on the family to provide the kinds of support needed for children and the elderly than is the case in the European countries, with the possible exception of the Mediterranean countries. Second, the amount of resources put into the first two kinds of family support programs listed above (financial incentives, and support for parents to combine work and family) is far less in these Asian countries than in the Scandinavian countries or indeed in European countries in general, again with the possible exception of the Mediterranean countries. Third, the range of policies in the ultra-low fertility Asian countries (except Taiwan) has widened considerably in recent years and the financial resources provided for these policies are growing rapidly. Fourth, there is a long way to go in bringing about broad social change supportive of children and parenting in these countries. Thus, for example, in Japan and Korea there is evidence of only a glacially slow change in the amount of time husbands put into household tasks and child-rearing in families where both spouses are working (Tsuya and Bumpass (eds), 2004; Tsuya et al., 2005). The policies recently introduced by South Korea appear to illustrate the “too little, too late” problem – Korean fertility is probably now the lowest in the world
12 Gavin Jones, Paulin Tay Straughan, and Angelique Chan among countries with populations over 30 million, yet the range of policies introduced in 2006 appears modest compared with that in the European lowfertility countries.
What has been the effect of pro-natalist programs on fertility? There is a substantial literature on the effect of family planning programs on fertility (see, for example, Ross and Forrest, 1978; United Nations, 1979; Mauldin, 1983; Phillips and Ross (eds), 1992) and an almost equally substantial literature on the effect on fertility of pro-natalist measures adopted by the low-fertility countries (summarized in Neyer, 2003: 78–81; see also Gauthier and Hatzius, 1997). In both cases, the problem for measurement is the uncertainty about the “counter-factual”: what would have happened to fertility in the absence of the programs? There are, of course, many ways to deal with this problem, none of them yielding absolutely trustworthy conclusions, hence the continuing spate of writings on the subject. The most sophisticated studies on the anti-natalist effects of family planning programs have dealt with population-based methods measuring the net program effects upon fertility for the population as a whole. These can be grouped as areal regression techniques, multilevel regression and experimental design (for details, see Ross and Lloyd, 1992: 36–43; Phillips et al., 1988). In this book we are more interested in assessing the impact of pro-natalist programs on fertility. But before attempting such an exercise, it is necessary to have some idea of the underlying causes of the decline in fertility to very low levels in these East Asian countries. Only when the causes are reasonably well understood is it likely to be possible to assess the impact of policy. Background factors influencing the decline to ultra-low fertility in East Asia The context in which delayed marriage and sharp fertility declines have occurred in East Asia has been one in which women’s educational levels have been rising and their workforce participation generally increasing. Concurrent with the rapid rise in numbers of women with secondary and tertiary education, there has been a very sharp rise in the proportion of females working. Trends in labor-force participation rates (LFPRs) for women aged 25–39 in Japan, Singapore and South Korea are shown in Table 1.3. In both Japan and Singapore, there was a sharp increase from 1975 onwards for all three age groups. In South Korea, by contrast, it was only after 1985 that female participation rates rose sharply, and even then to levels well below those in Japan or Singapore. The sharp rise in LFPRs in Japan and Singapore coincided with sharp increases in delayed marriage (Jones, 2007). The delayed increase in LFPRs in South Korea was consistent with the much later onset of delayed marriage for females in that country. Causation, of course, remains difficult to determine, because of the “chicken and egg” issue: did women remain single longer because they were
Very low fertility in Pacific Asian countries 13 Table 1.3 Labor-force participation rates for females in Japan, Singapore, and South Korea, certain age groups, 1960–2000
1970 1975 1980 1985 1990 1995 2000 2004
Japan
Singapore
South Korea
25–29 30–34 35–39
25–29 30–34 35–39
25–29
30–34
35–39
46 43 49 54 61 66 70 74
31 47 59 67 76 79 84** 86
31.7 35.1 34.1 35.9 42.8 47.8 55.9 63.9
36.3 40.7 46.1 43.2 49.6 47.5 48.5 50.4
43.1 49.3+ 55.5 55.8 58.0 59.2 59.1 58.9
48 44 48 51 52 54 57 61
46 43 49 54 61 66 70 74
23 32 44 49 63 64 74** 77
19 29 37 45 55 58 63** 68
Source: Jones, 2007: Table 5. ** 2001.
in the workforce, or were they in the workforce because they were still single? In Japan, which has the best data for examining these issues, the fact that the labor-force participation rate (LFPR) increased faster after 1975 for women aged 25–29 than for women in other age groups indicates that some reverse causality was occurring at age 25–29 (Retherford et al., 2001: 77). There was a major increase in the proportion single in this age group, and because the LFPR was much higher for single women, this drove up the LFPR independently of other factors that were also driving up this rate. But the LFPR for single women was also rising over this period (from 81 percent in 1972 to 92 percent in 1999). Retherford et al. (2001: 82) claim that there is evidence of a substantial increase in the opportunity cost for women of quitting their job to marry and have children, which would have driven up the age at marriage, though this pressure may be decreasing because a declining proportion of women are quitting the labor force after marriage and first birth, especially in the period between 1995 and 1998. In theory, young people may be avoiding marriage for reasons that differ from those that influence married couples to avoid having children. For example, marriage itself could be seen by young women intent on pursuing a career as a distraction from their goal, or they may be delaying it until, inadvertently, they find that they are faced by a lack of suitable partners (the “good man is hard to find” syndrome). In reality, however, marriage is a package, as it is not simply about the relationship between two people, but tightly linked with child-bearing, childrearing, and other family obligations. Thus the delay in marriage in the region is undoubtedly linked to a considerable – though not easily measurable – extent with the same kinds of concerns that are influencing married couples to delay childbearing or to have only one or two children. To marry without the intention to have children is still considered aberrant behavior throughout the region. There is no doubt that after marrying young couples are subjected to considerable pressure to have a baby, from family and to a lesser extent from friends. The easiest way to
14 Gavin Jones, Paulin Tay Straughan, and Angelique Chan avoid this is to remain single. Although single people are also pressured to marry, this may be less than that exerted on married couples to produce their first child. The arguments against having children, especially for upwardly mobile women throughout the region, appear compelling. Not only are the opportunity costs of having children alarmingly high, but the actual financial costs of raising and educating children are also very substantial. These financial costs have been estimated at US$253,000 for an average family in Korea, based on a survey conducted by the Korea Institute for Health and Social Affairs,8 and about the same level for Singapore. When opportunity costs are added in as well, it has been estimated that for female university graduates in Japan, the costs of raising and educating a child could well exceed 1 million US dollars. (For the data on which this estimate is based, see Retherford and Ogawa, 2005: 15–16.) The role of women in these South-East and East Asian societies may provide additional reasons to avoid bearing children. Patriarchal attitudes among employers and in government, resulting in poor workplace provision for the needs of working mothers (including lack of flexible working hours, childcare leave and provision of crèches), and reluctance of husbands to take much of the housework and childrearing tasks from working wives, adds considerable stress to the lives of women with children. The extreme pressure in countries such as Japan, South Korea and Singapore to raise high-quality children, and the unequal role given to mothers in achieving this goal, is another strong deterrent to beginning the process of family building (see, for example, Hirao, 2004). There are, then, clearly substantial obstacles to raising fertility rates in the ultra-low fertility countries. But this is not a reason to despair about the possible efficacy of policy. Not only in these East Asian countries but also in the lowfertility countries of Europe, women report family size ideals that are very close to the levels needed for population replacement (Quah, 2003: 71–73; Prachuabmoh and Mithranon, 2003: 39–40). There is an argument that expressed fertility desires do not prove very much about “real” fertility desires in low-fertility situations. In the European countries with well-below-replacement fertility, expressed desired fertility is much above the actual levels of fertility and rarely falls below two, nor does it differ much by cohort or social group (Van de Kaa, 2001; Bachrach, 2001). Indeed, international evidence indicates that there tends to be a crossover in actual and expressed desired family size when actual family size falls to levels a little above replacement level. Above this level, desired fertility tends to be below actual, but below this level the reverse holds (Bongaarts, 2001: 263–266). As Livi Bacci (2001: 284) notes, “the suspicion is that stated preferences are heavily influenced by stereotypes and particularly by the model of the two-child family (a boy and a girl). This stereotype is pervasive and many surveyed individuals are ‘prisoners’ of it”. It would be inappropriate, however, to push such an argument to the extreme of concluding that expressed preferences give no cause for optimism about the possible efficacy of pro-natalist policy. As McDonald (2006: 485) argues, “in expressing higher ‘ideal preferences’ on average, women are effectively commenting upon the nature of the social-institutional setting in which they
Very low fertility in Pacific Asian countries 15 consider having children”. The interest of society is in assisting them to realize their stated desires, in two key ways:
•• by designing policies that make it more desirable for individual women to have children within the given social-institutional setting;
•• by altering aspects of the social-institutional setting that influence their fertility decisions. Assessment of the impact of pro-natalist programs on fertility The preceding paragraphs summarize the East Asian setting in which pro-natalist policies are being practiced. Western countries, especially in Europe, have a longer history of pro-natalism, and the impact of pro-natalist programs on fertility in this setting has been discussed by various authors (for example, McDonald, 2002, 2006; Castles, 2003; Grant et al., 2004; Adkins, 2003; Neyer, 2003). Assessment of the impact of these policies on fertility has often been pessimistic, but, by and large, the references cited support the argument that direct child-support payments have a positive impact on fertility, as does greater childcare availability, with the evidence on the effect of family-friendly workplaces more mixed. McDonald (2002: 442) argues that the impact of individual pro-natalist measures can rarely be treated in isolation, because the effect of any policy will depend on the broader setting: A range of brilliant gender equity policies will be ineffective if unemployment rates are high for young people of childbearing age. Work and family policies can only work if there is work. Likewise, these same gender equity policies would be put under strain if the direct financial costs of children were very high or if the general tenor of the social setting was child-unfriendly. As Sleebos (2003: 5) argues, “what is required is coherent application of a range of well-designed interventions, applied consistently over time”. The general consensus about pro-natalist policies in East Asian countries seems to be that they have failed, because there is no evidence that fertility has risen as a result of their introduction. A reasonable response to this is that fertility may have fallen even further if these policies had not been introduced. Another is that the policies have not gone as far as they might, and in many cases, strengthening of policies has occurred only recently, so it is therefore premature to judge their apparent lack of success as proof that pro-natalist policies will always fail. But it is also true that the methodology of assessing the impact of pro-natalist policies on fertility is complex, and that for all countries, the likely efficacy of pro-natalist policies is difficult to predict. For example, as noted by Ogawa, Retherford and Matsukura (Chapter 3, this volume), programs making it easier to combine work and child-rearing can have both fertility reducing and fertility increasing effects. On average, for women who are already in the paid workforce, the effects of such policies should clearly be to raise their fertility. However, again
16 Gavin Jones, Paulin Tay Straughan, and Angelique Chan on average, by pulling some housewives into the labor force, this is likely to lower their fertility, since working women have lower fertility than housewives: Which of these two effects predominates depends partly on: • the magnitude of pre-existing fertility differences between workers and housewives • how much the fertility of working women rises as a result of the government’s policies and programs, • the magnitude of the probability that a housewife will be pulled into the labor force as a result of the government’s policies and programs, • the relative numbers of workers and housewives to begin with. If almost all women are already working, the overall effect will be to raise the TFR. If almost all women are housewives, the overall effect will be to lower the TFR. (Ogawa, Retherford, and Matsukura, Chapter 3, this volume: p. 67).
Some implications The World Population Plan of Action, 1974, Article 14(f) stated: All couples and individuals have the basic right to decide freely and responsibly the number and spacing of their children and to have the information and means to do so; the responsibility of couples and individuals in the exercise of this right takes into account the needs of their living and future children, and their responsibilities toward the community. This statement is careful in balancing rights and responsibilities. But there is frequently a tension between “individual good” and “common good” when it comes to matters of optimal family size. Many governments have taken the position that couples, acting on their individual right, are collectively causing population to develop in ways that have negative consequences for the community and nation, and that governments therefore have the responsibility to nudge population trends in ways that are collectively desirable.9 This was the argument for population control policies in high-fertility countries, and it is the argument for pro-natalist policies in low-fertility countries. In most lowfertility countries, under present arrangements, the fact is that the market will provide lower benefits to those with children than to those without children, and this affects fertility decision-making. “The collapse of birth rates in most industrialized countries is telling evidence of the failure of the market approach to allow social reproduction to proceed” (McDonald, 2002: 432). Governments, then, must act to redress the situation, but the corporate sector needs to be supportive, given the likely impact of many pro-natalist policies on its “bottom line”. The efforts to reduce fertility in high-fertility countries undoubtedly paid off in terms of slowing population growth, providing a context more favorable to poverty alleviation and providing a greater chance that humankind can reach a population equilibrium without putting intolerable strain on the earth’s resources.
Very low fertility in Pacific Asian countries 17 It is striking, though, that governments have proven to be far from omniscient in determining appropriate population policies. As we have seen, they have frequently been very slow to react to changing circumstances, for example, in failing to release the policy-imposed brakes on population growth early enough after their fertility level has fallen below replacement level. This raises the following issue: at the present time, which countries should be following a policy that is neither anti-natalist nor pro-natalist? We would argue that countries in which total fertility rates have fallen below about 2.7 do not need to follow specifically anti-natalist policies any longer, and those whose fertility remains above about 1.8 do not need to follow specifically pro-natalist policies. This is because international evidence suggests that once TFR declines to 2.7 or so, the decline tends to continue, and that countries where TFR is as high as 1.8 or so seem to have their family and other policies about right in maintaining their long-term population (and could make up any deficit through migration). For countries in this key range of one child in the TFR – from about 1.8 to about 2.7 – attention would be better directed at:
•• good-quality reproductive health services directed at the welfare of mothers •• ••
(and fathers) and children, one element of which would be the encouragement of delay in child-bearing and better spacing; development of sound family policies directed towards the welfare of children; provision of appropriate options to combine work and child-rearing.
In terms of the effects of these policies on fertility, planners can afford to “let the chips fall where they may,” of course keeping a watching brief on where they do indeed fall. Available evidence suggests that by concentrating on the quality and ready availability of reproductive health services and on family policy providing financial and other support for child-raising and an appropriate “work-life balance” the chips will fall appropriately in the sense of keeping fertility from dropping too low. Finally, there is need for introspection about many issues of social and family policy in the new ultra-low fertility situation. For example, the earlier concern about the treatment of women’s reproductive rights in population-control programs has surfaced again recently in relation to pro-natalist policy. Women must never be viewed as mere pawns in the game of either lowering or raising birth rates. In a speech to local members of Japan’s ruling Liberal Democratic Party (LDP) in January 2007, the Japanese Minister of Health, Hakuo Yanagisawa, said, “Because the number of birth-giving machines and devices is fixed, all we can ask for is for them to do their best per head”. Although after the subsequent furore, Mr Yanagisawa apologized, critics claimed that his words revealed a more widespread attitude among male government officials. Another issue is that where those who give birth to and raise children are seen to be “national heroes” or rather “national heroines,” the kudos given to such parents seems to imply an inevitable downplaying of the recognition given to those who do not bear and raise babies – the single, the childless couples, the gay couples.
18 Gavin Jones, Paulin Tay Straughan, and Angelique Chan This seems particularly unfortunate in cases where failure to produce children is by circumstance rather than choice – infertile couples; those who would like to have children but who have not found a suitable partner; those who choose to remain single to care for aging parents. But even in the case of those who choose not to have children for what some would judge to be “selfish” reasons, the need for a varied, vibrant and inclusive society would seem to dictate that people’s contributions to society be judged on a much broader set of criteria than whether they produce and raise children to ensure society’s replacement by the next generation. This is not to argue that income redistribution from those who do not produce children to those who do is inappropriate. Indeed, given the enormous direct financial and indirect opportunity costs – to well-educated women in particular – of having babies, much more substantial monetary transfers than are presently contemplated by countries of East Asia would appear to be necessary.
Notes 1 For example, the poor, semi-arid country of Niger in West Africa faces a trebling (or more) of population with very bleak prospects for development (Cleland et al., 2006: 1812). 2 There was a long-running debate over how women’s rights fared under population control policies. The debate tended to pit feminists against family planners (see Hartmann, 1987; Dixon-Mueller, 1993; Sen, 1994). There were awkward elements to the debate on both sides. Feminists had to deal with the evidence of the great need for family planning information and methods by women seeking to assert control over their fertility, and the potential contribution of appropriately conducted family planning programs to meeting this need. The family planners had to confront the evidence that in many countries the goal of reducing fertility was given priority over the need to provide women with appropriate choices. The argument that women were being treated simply as objects producing excess babies was irrefutable when applied to fertility control programs in China and India at particular times. The taking up of positions – the need for population control on one hand and the need to empower women to take control of their own lives on the other – tended to leave little common ground between family planning advocates and those arguing for reproductive rights, although, arguably, there was a large area of common ground to be occupied. 3 Actually, important modification in Singapore’s population policy had been made in 1983, when Prime Minister Lee Kuan Yew’s National Day speech discussed the differences in fertility levels according to the education of the mother. The following year, the first pro-natalist programs were launched, but these were only directed to the issue of fertility differentials between well-educated and less-educated women, not at raising the overall level of fertility. 4 For more detail, see the chapter by Ogawa, Retherford and Matsukura in this book. 5 The following paragraphs draw heavily on Eun, 2007. 6 The following paragraphs draw heavily on Saw, 2005, Chapters 11–13. 7 This downward trend was interrupted by a rise in a single year – 2000 – which was a Dragon Year (Saw, 2005: 207–210). This was only a blip in the downward trend. 8 This survey covered 11,816 children across Korea (Straits Times, 28/10/2007). 9 As Demeny (1986: 476) has argued, “When socially advantageous modification of demographic behavior is beyond the capacity of private markets to provide, it assumes the character of a public good that must be acquired, if at all, through the political marketplace”.
Very low fertility in Pacific Asian countries 19
References Adkins, D., 2003, “The role of institutional context in European regional fertility patterns,” paper presented at the Annual Meeting of the Population Association of America. Alexandratos, Nikos, 2005, “Countries with rapid population growth and resource constraints: issues of food, agriculture and development,” Population and Development Review, 31(2): 237–258. Bachrach, Christine, 2001, “Comment: the puzzling persistence of postmodern fertility preferences,” in Rodolfo A. Bulatao and John B. Casterline (eds), Global Fertility Transition, supplement to Vol. 27, Population and Development Review: 332–336. Birdsall, N., A.C. Kelley, and S.W. Sinding, 2001, Population Matters: Demographic Change, Economic Growth and Poverty in the Developing World. Oxford: Oxford University Press. Bongaarts, John, 1997, “The role of family planning programmes in contemporary fertility transition,” in Gavin W. Jones, Robert Douglas, John C. Caldwell, and Rennie d’Souza (eds), The Continuing Demographic Transition. Oxford: Clarendon Press. Bongaarts, John, 2001, “Comment: desired family size and the future course of fertility,” in Rodolfo A. Bulatao and John B. Casterline (eds), Global Fertility Transition, supplement to Vol. 27, Population and Development Review: 332–336. Cassen, Robert, 1994, Population and Development: Old Debates, New Conclusions. Oxford: Transaction Publishers. Casterline, John B. and Steven W. Sinding, 2000, “Unmet need for family planning and implications for population policy,” Population and Development Review, 26(4): 691–724. Castles, F., 2003, “The world turned upside down: below replacement fertility, changing preferences and family-friendly public policy in 21 OECD countries,” Journal of European Social Policy, 13(3): 209–227. Cho Nam-Hoon and Lee Sam-Sik, 2000, “Low fertility and population policy development in the Republic of Korea,” in Korea Institute for Health and Social Affairs, Low Fertility and Policy Responses to Issues of Ageing and Welfare. Seoul: KIHASA. Choe, Minja Kim and Kyung-Ae Park, 2005, “Fertility decline in South Korea: Forty years of policy-behavior dialogue,” paper presented at the XXV IUSSP International Population Conference, Tours, France. Cleland, John, Stan Bernstein, Alex Ezeh, Anibal Faundes, Anna Glasier, and Jolene Innes, 2006, “Family planning: the unfinished agenda,” Lancet, 368(9625): 1637–1722. Demeny, Paul, 1986, “Pronatalist policies in low fertility countries: patterns, performance and prospects,” in Kingsley Davis et al. (eds), Below-Replacement Fertility in Industrial Societies: Causes, Consequences, Policies, supplement to Vol. 12, Population and Development Review: 335–358. Demeny, Paul, 1997, “Replacement-level fertility: the implausible endpoint of the demographic transition,” in Gavin W. Jones, Robert Douglas, John C. Caldwell, and Rennie d’Souza (eds), The Continuing Demographic Transition. Oxford: Clarendon Press. Dixon-Mueller, Ruth, 1993, Population Policy and Women’s Rights: Transforming Reproductive Choice. Westport, CT: Praeger. Eun, Ki-Soo, 2007, “Lowest-low fertility in the Republic of Korea: causes, consequences and policy responses,” Asia-Pacific Population Journal, 22(2): 51–72. Falkenmark, M., 1997, “Meeting water requirements of an expanding world population,” Philosophical Transactions of the Royal Society B: Biological Sciences, 352(1356): 929–936.
20 Gavin Jones, Paulin Tay Straughan, and Angelique Chan Feyisetan, Bamikale and John B. Casterline, 2000, “Fertility preferences and contraceptive change in developing countries,” International Family Planning Perspectives, 26(3): 100–109. Freedman, Ronald and Deborah Freedman, 1993, “Policy options at the end of the demographic transition: Taiwan, 1993,” Industry of Free China 80(6): 25–36. Gauthier, A., 2002, “Family policies in industrialized countries: is there convergence?” Population (English Edition), 57(3): 447–474. Gauthier, Anne H. and Jan Hatzius, 1997, “Family benefits and fertility. An econometric analysis,” Population Studies, 51(3): 295–306. Grant, J., S. Hoorens, S. Sivadasan, M. van het Loo, J. DaVanzo, L. Hale, S. Gibson, and W. Butz, 2004, Low Fertility and Population Ageing: Causes, Consequences and Policy Options. Santa Monica, CA: RAND. Gwatkin, Davidson, R., 1979, “Political will and family planning: the implications of India’s Emergency experience,” Population and Development Review, 5(1): 29–59. Hartmann, Betsy, 1987, Reproductive Rights and Wrongs: The Global Politics of Population Control and Contraceptive Choice. New York, NY: Harper and Row. Hirao, Keiko, 2004, “Contradictions in maternal roles in contemporary Japan,” paper presented at International Workshop on Working and Mothering: Asian Women Negotiating Work Challenges and Family Commitments, National University of Singapore, January 29–30. Jones, Gavin W., 2007, “Delayed marriage and very low fertility in Pacific Asia,” Population and Development Review, 33(3): 453–478. Jones, Gavin W. and Richard Leete, 2002, “Asia’s family planning programs as low fertility is attained,” Studies in Family Planning, 33(1): 114–126. Livi Bacci, Massimo, 2001, “Comment: desired family size and the future course of fertility,” in Rodolfo A. Bulatao and John B. Casterline (eds), Global Fertility Transition, Supplement to Vol. 27, Population and Development Review: 282–289. McDonald, Peter, 2000a, “Gender equity in theories of fertility transition,” Population and Development Review, 26(3): 427–439. —— 2000b, “Gender equity, social institutions and the future of fertility,” Journal of Population Research, 17(1): 1–16. —— 2002, “Sustaining fertility through public policy: The range of options,” Population (English Edition), 57(3): 417–446. —— 2006, “Low fertility and the state: the efficacy of policy,” Population and Development Review, 32(3): 485–510. McIntosh, C. Alison and Jason L. Finkle, 1995, “The Cairo Conference on Population and Development: a new paradigm?” Population and Development Review, 21(2): 223–260. Mauldin, W.P., 1983, “Population programs and fertility regulation,” in R.A. Bulatao and R.D. Lee (eds), Determinants of Fertility in Developing Countries, Vol. 2, Fertility Regulation and Institutional Influences. New York, NY: Academic Press. National Research Council, 1986, Population Growth and Economic Development: Policy Questions. Washington DC: National Academy Press. Neyer, Gerda, 2003, “Family policies and low fertility in Western Europe,” Journal of Population and Social Security (Population), Supplement to Volume 1: 46–93. Phillips, J.F., M.A. Koenig, and J. Chakraborty, 1988, “Determinants of reproductive change in a traditional society: evidence from Matlab, Bangladesh,” Studies in Family Planning, 15(2): 62–73. Phillips, J.F. and J.A. Ross (eds), 1992, Family Planning Programmes and Fertility. Oxford: Clarendon Press. Pimentel, D., O. Bailey, P. Kim, E. Mullaney, J. Calabrese, L. Walman, F. Nelson, and X.
Very low fertility in Pacific Asian countries 21 Yao, 1999, “Will limits of the earth’s resources control human numbers?” Environment, Development and Sustainability, 1(1): 19–39. Prachuabmoh, Vipan and Preeya Mithranon, 2003, “Below-replacement fertility in Thailand and its policy implications,” Journal of Population Research, 20(1): 35–50. Quah, Stella R., 2003, Home and Kin: Families in Asia. Singapore: Eastern Universities Press. Retherford, Robert D. and Naohiro Ogawa, 2005, “Japan’s baby bust: causes, implications and policy responses,” East-West Center Working Papers, Population and Health Series, No. 118. Retherford, Robert D., Naohiro Ogawa, and Rikiya Matsukura, 2001, “Late marriage and less marriage in Japan,” Population and Development Review, 27(1): 65–102. Robinson, Warren C. and John A. Ross (eds), 2007, The Global Family Planning Revolution: Three Decades of Population Policies and Programs. Washington DC: The World Bank. Ross, J.A. and J.D. Forrest, 1978, “The demographic assessment of family planning programs: a bibliographic essay,” Population Index, 44(1): 8–27. Ross, J.A. and Cynthia B. Lloyd, 1992, “Measures for measuring the fertility impact of family planning programmes: the experience of the last decade,” in J.F. Phillips, and J.A. Ross (eds), 1992, Family Planning Programmes and Fertility. Oxford: Clarendon Press. Saw, Swee-Hock, 2005, Population Policies and Programmes in Singapore. Singapore: Institute of Southeast Asian Studies. Seltzer, Judith R., 2002, The Origins and Evolution of Family Planning Programs in Developing Countries. Santa Monica, CA: RAND. Sen, Gita, 1994, “Reproduction: The Feminist Challenge to Social Policy,” in Gita Sen and Rachel Snow (eds), Power and Decision: The Social Control of Reproduction. Harvard, MA: Harvard University Press. Simon, Julian L., 1981, The Ultimate Resource. Princeton, NJ: Princeton University Press. Sinding, Stephen W., John A. Ross, and Allan G. Rosenfield, 1994, “Seeking common ground: unmet need and demographic goals,” International Family Planning Perspectives, 20(1): 23–29. Singapore Department of Statistics, various years, Population Trends. Singapore: Ministry of Trade and Industry. Sleebos, J., 2003, Low Fertility Rates in OECD Countries: Facts and Policy Responses, OECD Social, Employment and Migration Working Papers No. 15, Paris: OECD. Tsay, Ching-lung, 2003, “Below-replacement fertility and prospects for labor force growth in Taiwan,” Journal of Population Research, 20(1): 67–87. Tsay, Ching-lung, 2004, “Marriage migration of women from China and Southeast Asia to Taiwan,” in Gavin W. Jones and Kamalini Ramdas (eds), (Un)tying the Knot: Ideal and Reality in Asian Marriage. Singapore: Asia Research Institute, National University of Singapore. Tsui, Amy Ong, 2001, “Population policies, family planning programs and fertility: the record,” in Rodolfo A. Bulatao and John B. Casterline (eds), Global Fertility Transition, supplement to Vol. 27, Population and Development Review: 184–204. Tsuya, Noriko and Larry Bumpass (eds), 2004, Marriage, Work and Family life in Comparative Perspective: Japan, South Korea and the United States. Honolulu, HI: University of Hawaii Press. Tsuya, Noriko, Larry L. Bumpass, Minja Kim Choe, and Ronald R. Rindfuss, 2005, “Is the gender division of labour changing in Japan?” Asian Population Studies, 1(1): 47–68. UNFPA, 2003, “Achieving the Millennium Development Goals: Population and Reproduc-
22 Gavin Jones, Paulin Tay Straughan, and Angelique Chan tive Health as Critical Determinants,” Population and Development Strategies Series, No. 10. New York, NY: UNFPA. United Nations, 1979, Manual IX: The Methodology of Measuring the Impact of Family Planning Programmes on Fertility (Sales No. E.78.XIII.8). United Nations, Department of Economic and Social Affairs, Population Division, 2006, World Population Prospects: The 2006 Revision. New York, NY: United Nations. Van de Kaa, Dirk J., 2001, “Postmodern fertility preferences: from changing value orientation to new behaviour,” in Rodolfo A. Bulatao and John B. Casterline (eds), Global Fertility Transition, supplement to Vol. 27, Population and Development Review: 60–92. Wang, Feng, 2005, “Can China afford to continue its one-child policy?” Asia Pacific Issues, Analysis from the East-West Center, No. 77. Zeng, Yi, 2007, “Options for fertility policy transition in China,” Population and Development Review 33(2): 215–246.
2
Explanations of low fertility in East Asia A comparative perspective Peter McDonald
Based mainly on experience and research in Western European and Englishspeaking countries, I have argued that low fertility in advanced societies today has been an unintended outcome of two major waves of social and economic change, social liberalism and economic restructuring (McDonald 2006a). Both these waves have enhanced individual aspirations in relation to the quality of personal and economic lives. However, in differing cultural and welfare environments, both have brought pressure to bear upon the capacity to form and maintain families. Social liberalism and economic restructuring have given rise to two important changes for individuals: the provision of gender equity through an opening up of opportunities for women beyond the household, and growing levels of risk aversion among young people of both sexes in an increasingly competitive labor market. These changes influence fertility in the following ways.
Gender equity and fertility Most advanced societies have a recent history of differentiated family roles for men and women where men specialized in wage-earning while women specialized in home-making and caring for relatives, especially children. Rigidly differentiated roles for men and women were questioned as part of the reaction in the 1960s and 1970s to socially prescribed roles for men and women and the movement toward greater freedoms for the individual. As a result, education levels for women increased dramatically, and opportunities in paid employment were opened to women to the extent that, in the institutions of education and market employment, considerable gender equity was afforded to women as individuals. The movement to gender equity has been focused, however, upon individual-oriented social institutions. In contrast, family related institutions, especially the family itself, have continued to be characterized by gender inequity. By the time women begin to consider family formation, they have experienced considerable freedom and gender equality as individuals, but they are keenly aware that these gains will be distinctly compromised once they have a baby (McDonald 2000a, 2000b). This is especially the case in labor markets where little or no provision is made for the combination of work and family. There is a considerable economic dimension to
24 Peter McDonald the gender argument, the mechanism being the lifetime earnings lost to women through having children. In these circumstances, women exercise careful control over their own fertility, delay their family formation, and have fewer children to such an extent that fertility falls to very low levels. In some settings where child-bearing is a strongly expected, rapid consequence of marriage – in East Asia, for example – women may remain unmarried. The central problem is that family formation involves greater risks for women than for men. Accordingly, women are wary about embarking upon marriage and child-bearing if they do not feel confident about their ability to combine family with the other opportunities that have opened up for them, especially through paid employment.
Economic restructuring, risk aversion and fertility Globalization and sharply rising education levels have created high economic aspirations among young people. At the same time, the competitive nature of labor-market deregulation has led to a wider variation in their earnings and career stability and progression. Engagement in the deregulated labor market is now seen as involving greatly increased risk. Under these conditions, young people tend to become risk-averse, that is, they follow pathways that have lower risk. Any recent economic shock, such as the 1997 Asian financial crisis, adds both to the sense of risk among young workers and to the determination of employers to protect their enterprises through reduced employment security of workers. While, at an individual level, early labor-market success can promote earlier family formation, the societal balance is toward later achievement of economic security, in a context where security is defined within an environment of greatly enhanced economic aspirations. Investment in one’s human capital (education and labor-market experience) is seen as the essential hedge against these risks, the optimal path of risk aversion. This investment involves considerable commitment to self and to one’s employer, especially through long work hours, in opposition to a commitment to more altruistic endeavors such as service to family members and family formation. As a consequence, family formation is put on hold while human capital is accumulated. While family formation remains the goal of most people, within the context of the opportunities and risks of the economic restructuring it can be delayed to an extent that achieved fertility falls short of individual preferences.
Cross-country variations in fertility While all advanced countries have been subject in varying degrees to these two major waves of social and economic change, the outcomes in terms of fertility rates have been different. One group of countries has very low fertility rates (defined as being under 1.5 births per woman) while another group has fertility rates that are moderate (defined as being above 1.5 births per woman). These groups of countries are shown in Table 2.1. The table displays an interesting cultural divide.
Explanations of low fertility in East Asia 25 Table 2.1 Total Fertility Rates, 2005 Group 1 Countries
TFR
Group 2 Countries/Regions
TFR
United States (2004) Iceland New Zealand France Ireland Norway Australia Finland Denmark United Kingdom Sweden Netherlands Belgium Luxembourg Canada (2004)
2.05 2.05 2.00 1.94 1.88 1.84 1.82 1.80 1.80 1.80 1.77 1.73 1.72 1.70 1.53
Switzerland Austria Portugal Malta Germany Italy Spain Greece Japan Singapore Taiwan Republic of Korea Hong Kong SAR
1.42 1.41 1.40 1.37 1.34 1.34 1.34 1.28 1.26 1.24 1.12 1.08 0.97
Source: Eurostat and national statistical offices.
The Group 1 countries consist of all of the English-speaking countries, all Frenchand Dutch-speaking Western European countries and all of the Nordic countries. The Group 2 countries consist of all of the Southern European countries, all of the German-speaking Western European countries and all of the Asian countries/ regions with advanced economies. Indeed, with the exception of Canada, there is a fairly substantial gap (around 0.3 births per woman) between the fertility rates of the two groups. Furthermore, the fertility rates of the Group 1 countries have never fallen below 1.5 births per woman (except for Denmark and Canada dipping slightly below a fertility rate of 1.5 for very brief periods). And, once having fallen below a fertility rate of 1.5 births per woman, none of the Group 2 countries have yet risen again above the 1.5 level. This last observation has led some demographers to argue that there may be “a low-fertility trap”. Their hypothesis is that sustained very low fertility may lead to demographic, social and economic changes that institutionalize very low fertility (Lutz et al. 2006). While the lowfertility trap is based upon observations that I have made, my own view is more optimistic, that very low fertility rates are reversible through urgent, concerted policy effort.
The effects of very low fertility While seemingly small, I have argued that the gap of 0.3–0.4 births per woman between the Group 1 and Group 2 countries is the difference between future demographic sustainability and future demographic crisis (McDonald 2006b). At the macro level, sustained low fertility rates have a devastating impact on future labor supply at the same time as the number of older people is increasing. If recent demographic trends were to continue, by 2040, the labor force in Japan would
26 Peter McDonald fall by 20 million workers and in both Italy and Germany by 11 million workers (McDonald and Kippen 2001). These losses are entirely losses of younger workers and in an era of rapidly changing technology, economies that are blasé about their future supply of young technologically skilled workers are taking a considerable risk (McDonald and Temple 2006, Skirbekk 2003). Using several economic indicators, Ogawa et al. (2005) demonstrate that, in Japan, the demographic bonus conferred by the earlier decline in fertility was transformed to a “demographic onus” from 2000 onwards due to the impacts on the age structure of sustained low fertility. In Europe as well, this message is receiving more prominence as very low fertility rates persist for longer periods and as aging societies face the certain knowledge that their social insurance systems are not viable in a future with a much older population and a sharply reduced labor supply (Commission of the European Communities 2005). In European countries, there is evidence also that very low fertility, on average, is counter to the preferences of individuals experiencing it (van Peer 2002, d’Addio and d’Ercole 2005). While individual-level economic analysis attributes meaning only to revealed preferences, actual fertility behavior (a revealed preference) is achieved within a given social-institutional setting. In expressing higher “ideal preferences” on average, women are effectively commenting upon the nature of the social-institutional setting in which they consider having children. They are saying that, in a different institutional setting, they believe they would have had more children. Benefits accrue both to the individuals concerned and to the society as a whole if fertility ideals can come closer to being fulfilled. There are severe questions about the nature of social organization if citizens fail to have the number of children they would prefer to have when that number is as low as one, two, or three children. The argument is a subtle one because the societies under discussion do not remove the right to have children as some societies remove other personal freedoms. Rather, the argument is that ideals go unrealized because of countervailing forces ensuing from the nature of modern society. Low fertility is an unintended rather than a deliberate outcome of changing social and economic institutions. The argument here is that very low fertility is explained not by individual behavior but by institutional constraints upon individual behavior.
The cultural divide between Group 1 and Group 2 countries The sharp cultural divide shown in Table 1 is very unlikely to be the outcome of mere chance and, as such, demands explanation. In broad terms, the Group 2 countries share a strong tradition in which family and state are separate entities and families are expected to support their own members without intervention from the state. Accordingly, these states have been slow to implement family assistance measures or to “intervene” in family life. With some variation, the opposite is the case in the Group 1 countries; in general, they are notable for the family-friendly institutional arrangements that their governments and employers have implemented in the past 20 years and for relatively higher
Explanations of low fertility in East Asia 27 levels of gender equity within the family.1, 2 The responsibility for family caring and maintenance (beyond income) in Group 2 countries falls almost exclusively upon women, that is, the male-breadwinner model of the family remains largely intact. Because women are expected to provide caring and maintenance work, in Group 2 countries, the service and public sectors are generally smaller than in Group 1 countries (Bettio and Villa 1998). In Group 1 countries, these are the sectors that are more likely to employ women and to have family friendly work environments. It is no surprise then that both fertility and labor-force participation rates for women are lower in Group 2 countries. Ironically, Group 2 countries see themselves as having strong traditional “family values”. This image of themselves makes change from the traditional family organization politically more difficult. The cultural divide indicates that differences between countries with very low fertility and those with moderately low fertility are due to institutional factors rather than to individual-level factors. Very low fertility in Group 2 countries is not the result of young people in Group 2 countries being more selfish or materialistic than young people in Group 1 countries. They simply do not receive the level of family support from governments and employers that young people in Group 1 countries receive. Individual preferences are formed in the shadow of the social and economic institutions of the society that, in turn, are within the realm of political choice. Governments and employers have played a leading role in the social and economic changes that have brought pressure to bear on all forms of altruistic behavior but, most especially, upon family life. Because of this, it is incumbent upon governments and employers to restore the balance through family support programs.
Very low fertility in the advanced economies of Asia As shown in Table 2.1, the fertility rates in all of the advanced economies of Asia are very low. Indeed, fertility is lower in all of these Asian societies than it is in the other Group 2 countries. Many countries in East and Central Europe also have fertility rates under 1.3 but there are special circumstances applying in these countries related to the massive restructuring of their economies. Why do advanced Asian societies have the lowest fertility rates in the world? The explanations of very low fertility that I have provided above are based largely on research in European contexts and not so much in Asian contexts. Thus there is some possibility that explanations of very low fertility in advanced Asian contexts may be different. If the explanations that I have provided are applicable to Asian advanced societies then we would expect that:
•• The incoherence between gender equity in individually oriented institutions and family oriented institutions would be very large in these societies.
•• There would be a strong sense of economic risk and economic pressure upon young people in these societies.
•• There would be a relative absence of support from governments and employers for the family needs of young people.
28 Peter McDonald If there are other explanations specific to advanced Asian economies, what might they be? There are three main alternatives: 1
2
3
Much attention has been given to delayed marriage and non-marriage in the East and Southeast Asian context (Jones 2005). Individual country studies for Japan, South Korea, Singapore, and Hong Kong have attributed very low fertility mainly to delay of marriage or non-marriage (Takahashi 2004; Atoh et al. 2004; Choe et al. 2004; Yip et al. 2006). These studies are based upon decompositions of the change in total fertility over time in one country into changes in proportions married and changes in marital fertility. These studies tend to show that the effect of changes in marriage is dominant, although Koh’s study of Singapore shows an equal effect of the two components. The conclusion often drawn from this type of analysis is that the correct policy direction is to get people married rather than to promote fertility within marriage. In Japan, Yamada (1996) has referred to those delaying marriage as “parasite singles” (Yamada 1996; Takahashi and Voss 2000). I shall refer to this as the marriage explanation. There has been some emphasis in studies in Asia on high direct costs of children (especially the costs of their future education) as explanations of low fertility. The hypothesis here is that young people, themselves having experienced acute competition in the education and labor markets, consider that the only way that they can have their own children achieve success is to have fewer of them. If they have few children, they figure that they are more likely to be able to meet the very steep education costs (“cram” schools, foreign degrees, etc.) that will apply to their children across their future lifetimes. I shall refer to this as the direct-cost explanation. There is an argument that most people in advanced Asian economies live in large, densely populated cities with high housing costs and housing and urban environments that are unfriendly to children. I shall refer to this as the housing explanation.
Individual preferences Before discussing the potential explanations of very low fertility in East Asia, it is worth pointing out that young people in advanced Asian countries also express fertility ideals that are above replacement-level fertility. For Japan, Suzuki (2006) reports that the average ideal number of children (for wives aged less than 50) has never fallen below 2.5 in the past 25 years and that the average number of children expected by the same women was around 2.1 in 2005. Furthermore, very few single Japanese women expressed a preference to remain single (under 8 percent). Suzuki concludes, consistently with the argument above, that “recent fertility decline should be explained not from demand itself but from obstacles to fulfilling the demand” (Suzuki 2006: 8). Choe et al. (2004) present data showing that the ideal number of children has not fallen below 2 in the Republic of Korea at any time in the past 20 years.
Explanations of low fertility in East Asia 29 However, they also consider “planned fertility,” the final number of children that women plan to have. This fell to 1.4 in 2000 and only 50 percent of women with one child planned to have another. Choe et al. (2004: 10) conclude: currently, the TFR and the planned number of children among married women are both well below the mean ideal number of children, suggesting that women would have more children if their socio-economic circumstances were also ideal”. In the case of Singapore, Yap (2003: 651) reports that the average number of children preferred by women in their 20s was 2.5 in 1997 and that “the desired family size among women in their twenties has remained fairly stable over the past two decades or so”. If these studies and their interpretations are correct, advanced Asian countries are not yet in the situation of the “low-fertility trap” where individual ideals fall to a level that corresponds with the number of children that people are actually having.
The marriage explanation Decomposition of the decline in fertility to changes in proportions married and changes in age-specific marital fertility rates is a peculiarly Asian approach. While often applied in Asia, this approach is rarely applied to investigations of fertility decline in European countries, although reference is made to delay of marriage as an explanatory factor in Italy (Dalla Zuanna et al. 1998). This probably stems from the fact that it has been assumed generally that cohabitation and births outside of marriage are negligible in advanced Asian countries whereas this is not the case in Europe.3 European analyses usually focus on cumulated fertility of cohorts by given ages or cumulated proportions by age having a first, second and third birth (Frejka and Sardon 2004; Lesthaeghe 2001). These European studies indicate that age at first birth has increased sharply in all countries and this leads to falling cumulated fertility at younger ages. The difference between the Group 1 and Group 2 European countries is that as women age through their thirties, cumulated fertility rises relative to a standard in the Group 1 countries but fails to rise or rises only slightly in the Group 2 countries. Lesthaeghe (2001) refers to this process as “recuperation,” meaning that women in Group 1 countries having had relatively low fertility in their younger ages “recuperate” much of the loss at older ages. Suzuki (2006) has presented data in this form for Japanese birth cohorts. The results show the increasing onset of a failure to recuperate for successive Japanese cohorts. In the context of the decomposition methodology, recuperation implies that marital fertility rates at older ages would rise as proportions married at younger ages fell. From the policy perspective, the important feature of very low fertility is not so much that marriage rates are low in the twenties but that marital fertility rates fail to rise (or even fall) in the thirties. While much attention is focused on low marriage rates in advanced Asian countries, Table 2.2 indicates that, by age group 30–34, most Asian women have married and that the proportions “partnered” in most European countries (both Group 1 and Group 2) are very similar to the proportions married in the Asian countries in the table.
30 Peter McDonald Table 2.2 Women aged 30–34, percentages never married and percentages never married and not cohabiting, selected countries Group 1 Country
Females Aged 30–34 % Never % Never Married Married and Not Cohabiting
Ireland France Norway Finland Denmark Sweden United Kingdom Netherlands Belgium
39.1 37.3 43.0 40.5 38.4 51.1 34.5 33.6 27.0
30.3 19.4 20.9 20.9 17.6 — 21.9 16.9 18.9
Group 2 Country or Region
Females Aged 30–34
Portugal Switzerland Austria Germany Italy Spain Greece Japan Singapore Chinese South Korea Hong Kong
16.2 28.7 32.1 29.6 30.0 31.1 20.3 26.6 21.6
— 20.0 22.1 19.2 28.0 26.3 18.6 — —
10.7 26.5
— —
% Never % Never Married Married and Not Cohabiting
Sources: European countries: Eurostat; Asian countries: Jones 2005: Table 1.
While there is a strong argument that the commencement of child-bearing has been excessively delayed in contemporary advanced societies, if most women are married by their early thirties, there is still adequate time for them to have two children. In Group 1 countries, they do; in Group 2 countries, they don’t. To repeat the Choe et al. (2004) finding, only 50 percent of South Korean women with one child plan to have a second child. In Singapore, Koh (2006) shows that, rather than rising over time, age-specific marital fertility rates at ages 30 and over have continued to fall to 2005. Of all the countries shown in Table 2.2, South Korea has by far the lowest percentage of 30–34-year-old women that are not married yet and, aside from Hong Kong, it has the lowest fertility rate among the countries and regions shown in Table 2.1. In behavioral terms, low marital fertility among women in their thirties is the essential difference between Group 1 and Group 2 countries, not differences in their marriage/relationship patterns. Nevertheless, in policy terms, encouragement of child-bearing at somewhat younger ages than have become the norm is worthwhile in both Group 1 and Group 2 countries. A further important point to be made is that decisions about marriage and child-bearing tend to be concurrently determined in advanced Asian societies: a decision to marry is equivalent to a decision to have a child (Shirahase 2000). There is some recent evidence in Japan of a rising trend to delay the first birth within marriage which may change this picture, but it remains true for most Japanese women that marriage means pregnancy and withdrawal from the labor force. Thus, any observed reluctance to marry must be considered in this context.
Explanations of low fertility in East Asia 31
The direct-cost explanation Japan is the country where high direct costs of children, especially the costs of their education, are most often cited as a reason for low fertility. Suzuki (2006: 8) claims: “the rising cost of children including public and private educational costs is thought to be the main reason of the recent low fertility in Japan”. Suzuki points out that envisaged high direct costs is the most frequently mentioned reason that people give when asked why they plan to have fewer children than their ideal number, especially respondents aged less than 30 (see also Oyama 2001: reporting on a data collection in 1992). High direct costs are also considered to be part of the explanation of Japan’s fertility decline by Retherford and Ogawa (2006). They cite estimates of the direct costs of children that range from US$286,000 (not including the cost of “cram” schooling) to US$630,100 for a child that attends private schools through medical school. It seems a little far-fetched to suggest that potential Japanese parents make estimates of the costs of their children’s education across their lifetime, compare this with their own estimated earnings and the expected earnings of their children and then discount all these estimations back to the present. Few potential Japanese parents are skilled econometricians. It is much more likely that their attitudes stem from their own experiences with their own parents. The current generation of potential parents in Japan is a large generation who have struggled in a very competitive educational environment. They have witnessed their own parents sacrificing their living standard in the interests of higher educational achievements for their children, today’s potential parents. Thus, while the next generation of Japanese children will be smaller in number than their parents’ generation and while competition for educational places is already falling, today’s potential parents are influenced more by the past than by projections of the future. The argument that the assessment of the costs of children by potential Japanese parents is more subjective than objective is supported by the research of Oyama (2003a) on costs of children. He finds that using an objective measure, the Rothbarth model based on consumption of adult goods, the costs of children in Japan are low. However, using a subjective measure based on levels of satisfaction with income, the costs are high. In a subsequent paper, Oyama (2003b) concludes that factors associated with low fertility in Japan are low income of the husband and lack of housing support from the older generation. He concludes that these results are “probably caused by the recent higher cost of raising and educating children in Japan”. Again, this implies a subjective assessment of the costs of children. There is also an issue about the question: why do you plan to have fewer children than your ideal? The answer is contingent on the number that is considered ideal. Using the 2000 Korean data provided by Choe et al. (2004), an extremely low percentage of Koreans with two children planned to have a third (about 4 percent). It is easy to envisage that parents would consider the direct costs of three children to be too expensive. A high proportion of today’s potential parents would have come from families with three or more children and, again, their own experience as children may have an important bearing on their thinking. However, Korea’s
32 Peter McDonald fertility appears to be very low because a high proportion of those with one child do not plan to have a second, although fully 92 percent plan to have at least one child. Lifting the progression from the first to the second child in Korea from 0.5 to 0.7 would increase the fertility of a cohort by about 0.2; lifting the progression from the second to the third child from 4 percent to 10 percent would raise it by only 0.05 of a child. The progression from the first to the second child is the phenomenon that needs close investigation in the very low-fertility societies of Asia. In summary, there seems to be some evidence that young people in advanced East Asian countries may carry perceptions that the direct costs of children are high and therefore their number should be limited to as low as one child. Objectively, there does not seem to be strength to the argument that the direct costs of children are high. In policy terms, this suggests two directions: financial support attached to the first and the second child (as recently proposed in Russia) and education campaigns that carry the message that the costs of two children will be able to be managed by most families. Governments might also consider providing education guarantees subject to adequate performance of the child.
The housing explanation While we know that fertility rates are lowest in the large metropolitan cities of the advanced Asian countries/regions (Tokyo, Seoul, Taipei, Hong Kong and Singapore), little attention has been focused on the argument that city life and housing costs and conditions may not be conducive to child-rearing in these cities. Most residents of these cities live in high-rise apartment blocks. Generally, apartments are small. This is not an ideal environment for young children. There is also an impression that Group 1 countries are more likely to be “suburban” in nature and the suburbs are the home of the child. In policy terms, these massive cities and their housing form cannot be changed. In future developments, larger apartments would mean higher costs and perhaps further delay in the commencement of child-bearing. It may be possible that architects can consider ways in which existing high-rise apartment blocks and their immediate environment can be made more child-friendly. This is an area in which more research is required. However, in speaking to colleagues from advanced Asian economies (anecdotal evidence only), most consider that the form of housing is not a major determinant of low fertility because the forms available are consistent with standard expectations.
Alternative explanations: a summary The above discussion indicates that the key to interpreting very low fertility rates in advanced East Asian countries/regions lies in low marital fertility rates for women in their thirties, manifested primarily in a low progression from the first to the second child. While more research is needed, the marriage explanation, the direct-cost explanation and the housing explanation could not be said, on the available evidence, to provide satisfactory explanations for the low fertility in
Explanations of low fertility in East Asia 33 these settings. However, there is sufficient evidence to suggest that some policy initiatives related to these explanations could have some effect:
•• •• •• •• ••
encouragement of an early age at first birth financial supports for the first and second child public awareness campaigns on the direct costs of children government educational guarantees refurbishment of apartment blocks to be more child-friendly.
As will be discussed below, there is more evidence that the standard “European” explanations of low fertility are also highly relevant in the advanced East Asian context. These can be described under three headings: work-family balance and gender equity, a sense of economic risk among young people, and a relative absence of family support provided by governments and employers stemming from the pervasive cultural value that families support their own. Work-family balance and gender equity Empirical work by Oyama (2001: 24) has shown that, for women in Japan, decisions about work and children are jointly determined and “the effects of wife’s work choice on number of children was mostly significant and negative”. As Atoh et al. (2004) argue, opportunities for young Japanese women in the labor market have mushroomed through higher levels of education and strong demand from employers. However, the capacity to combine work and family is limited and women are faced with stark choices regarding career or family. Tsuya et al. (2000) attribute low fertility rates in Japan and South Korea to lack of support for working women both outside and inside the household. They conclude that Korean women are faced with the starkest choice between work and family: “In Korea, married women seem to have only two options: work for long hours or not at all. This, in turn, demonstrates the limitations that the market and society impose on women’s ability to reconcile employment and family responsibilities” (Tsuya et al. 2000: 216). In like manner, Suzuki concludes: Thus, many Japanese women have the ability and opportunity to work but they have to give up their career on child-bearing. Such incompatibility is attributed to the remaining gender role attitude, low participation of the husband in housework, characteristics of the labour market, and underdevelopment of family policy. (Suzuki 2006: 11) For Korea, Chang observed: the very family-centeredness of South Koreans has been responsible for their internationally unparalleled fertility decline. … women’s high education has inevitably made their fertility, and sometimes even marriage, a serious
34 Peter McDonald personal sacrifice … Under these circumstances [increased participation of young women in the labor market], developed a growing awareness of the need for gender-equitable role and status arrangements both at home and in society. Furthermore, more and more women consider marriage merely one of the compromisable options in life and postpone it very late or even avoid it altogether. … more and more women [are] hesitant about marriage, in particular when they acquire a more than sufficient economic capacity for self-support mainly thanks to the expansion of service industries. Their high educational levels enable them to undertake whatever jobs newly created if arbitrary barriers of gender discrimination are checked. (Chang 2003: 596–603) Awareness of this explanation for low fertility is extending to more popular levels as indicated by the following quotation from the Korea Times: A new social contract guaranteeing that those who have children are not severely disadvantaged in the workplace is needed. The government is asked to review a wide range of policy options to effectively counter our disastrous situation. The cooperation of business is also important to increase the number of births in our country. Many business corporations here don’t provide parental leave for the birth of children of employees. People also need to know about the calamity that a low fertility rate can bring. (Korea Times, January 25, 2007) South Korea’s new policy direction, Vision2020, recognizes explicitly that “society should take on more responsibility as a whole for raising the next generation” (Eun 2006: 26). Gender inequity arguments have had the greatest force in South Korea followed by Japan. South Korea is active in addressing the situation through the creation of a Ministry of Gender Equality and discussion of a range of family support policies (Lee 2005; Eun 2006). The National Institute of Population and Social Security Research, Tokyo, has been at the forefront of making these arguments in Japan especially through its former Director-General, Makoto Atoh. The argument seems to be less well developed in Singapore, Taiwan and Hong Kong, all countries in which there has been a central focus on non-marriage as the explanation of low fertility. However, Singapore has made some recent tentative steps in this direction including a five-day working week for public-sector workers, although the impact of this policy change has been reduced considerably by requiring workers to work longer on the five working days. The list of work and family policy initiatives is well known: paid parental leave; flexible work hours; availability of part-time work that is not inferior in level, hourly wage rate and promotion prospects; a five-day working week; and affordable, quality childcare. These policy directions remain severely under-developed in the developed countries of Asia. In all advanced Asian countries/regions, governments have failed to confront employers in order to achieve workplaces that are more cognizant of the needs of parents, especially mothers. Expected work hours remain in sharp
Explanations of low fertility in East Asia 35 conflict with family responsibilities. Women aged less than 30 in Singapore, for example, work an average of 52 hours per week. Sense of economic risk among young people The impacts on fertility of rising job insecurity (or the sense of it) and the competitive labor market among young people are much under-researched in the Asian context. Generally, levels of direct unemployment have been low in the advanced economies of Asia but all have faced economic downturn in the recent past: the burst of the bubble economy in Japan and the 1997 Asian financial crisis in the other countries. All these countries have a strong reliance upon manufacturing and in the past decade have faced increasing competition, especially from China. The crisis and the increased competition have led to companies taking a leaner and meaner approach to job security. Ki-Soo Eun (2003) asserts strongly that the above economic arguments explain the fall in fertility in South Korea since the mid-1990s. He argues that labor-market deregulation has been accelerated since the Asian economic crisis in 1997: The core principle of the reconstruction of the labor market in accordance with the spirit of “new liberalism” was “flexibility”. According to this principle, workers should be … laid off whenever necessary. Rather than regular full-time workers, part-time or temporary workers have been preferred in various places by owners. … the barrier to the entrance into the labor market for younger people is still so high that they always have to undergo a kind of “war” to find a job in the labor market. On the other hand, the condition of labor or job status has become deteriorated and unstable. The concept of lifetime job has suddenly gone away in Korea since the economic crisis. (Eun 2003: 584) Eun has continued this analysis and, in 2006, published a detailed account of the ways in which the economic crisis of 1997 led to the changes in the Korean psyche that led to low fertility: The crisis fundamentally dismantled the belief in lifetime job security which had been maintained since widespread economic growth in the 1970s. With the introduction of the flexible labor market, much as happened in Japan, lifetime job security dissolved into a mirage. Ordinary people were forced to recognize that unemployment could happen to anyone, anytime or anywhere. No compensation or unemployment benefits were available and there were few opportunities to be re-hired in the formal sector after unemployment. The shock of the crisis weighed more heavily on the Korean psyche and society when compared to other Asian societies hard hit by the 1997 financial crisis because of the stark contrast between the undeniable expectations of economic prosperity and the subsequent economic privations. (Eun 2006: 11–12)
36 Peter McDonald Between 1997 and 1998, the unemployment rate among Korean men aged 25–29 years, the prime ages of career formation, rose from 4.9 to 10.9 percent. Although the rate fell after the crisis, it has plateaued at a new higher level of 7–8 percent. Thus, while indeed there was an unemployment shock around the years of the crisis, the sharp falls in fertility from 2000 to 2005 were associated with somewhat improved levels of unemployment for young men compared to the crisis years. There was also little change through the crisis in the proportion of 25–29 year olds who were working part-time. Thus, direct experience of unemployment or part-time work during the crisis or since the crisis is not sufficient to explain the fall in fertility in Korea. Indeed, a shift to low fertility had been underway for many years before the 1997 economic crisis, and fertility had already reached the low level of 1.54 births per woman before the impact of the crisis. Increases in human capital associated with protection against risk had also been underway for many years as indicated by the educational achievement data for successive age cohorts. Thus, while Eun (2003) may have placed undue emphasis on the impact of the relatively short-term economic crisis and insufficient emphasis on economic and labor-market changes that had been building for two decades, this does not negate his argument that employment has become much less secure for young people and that this has had flow-on effects to fertility. His argument that the crisis provided both a stark demonstration to young people of the risks inherent in the labor market and a stimulus for government and employers to pursue labor-market deregulation with greater vigor is likely to be highly relevant to any attempt to reverse the fall in Korea’s fertility. Suzuki (2006) makes a similar argument for Japan. He says that young people who grew up in the period of rapid economic growth tend to have high aspirations. However, the poor state of the economy of Japan throughout the 1990s discouraged the career achievement aspirations of young Japanese. Their incomes dropped relative to those of older men. Suzuki refers to the Cabinet Office Opinion Survey on people’s expectation of their future life circumstances. The percentage expecting a worse future rocketed upwards from 1995 and has remained at a highly pessimistic peak from 2001 to 2005. He concludes: “It is thought that such uncertainty about the future is one of the major sources of lowest-low fertility in Japan” (Suzuki 2006: 10). In general, governments in advanced Asian countries, like employers fearful of competition from China, have not been prepared to attempt to modify the worsening employment conditions of young workers.
Conclusion The preceding discussion indicates that delay of family formation is based not so much on experienced economic outcomes but, like any other investment, on the degree of confidence that potential parents have about their capacity to undertake family formation while not placing themselves at economic risk or at risk of falling short of their individual aspirations. Does the solution lie in lowering the
Explanations of low fertility in East Asia 37 level of aspirations or in providing a higher level of security? It is often suggested that young people’s values are “too materialistic” and this is why their fertility is low. However, rising economic aspirations and materialism have a long history, and, at present, they are encouraged by the new capitalism. It is a gross contradiction to support new capitalism and at the same time describe young people as materialistic. While young people are aware that almost inevitably they will reduce their material outcomes if they have children, most are willing to accept the loss so long as it is not overly destructive of their aspirations. In particular, they would like to have confidence that they will have adequate financial resources during the period when children are very young and that the period of loss will be temporary. The solution to low fertility therefore lies in providing a greater sense of assurance to young women and young men that, if they marry and have children, they will be supported by the society in this socially and individually important decision. If instead they look ahead to societal arrangements that severely disadvantage those who have children, they will delay their family formation until they feel they have reached a secure enough position to assume its costs. Individual delay means very low fertility for the society. Having been instrumental in the rise of the conditions leading to low fertility, governments should take the lead in providing this greater sense of security. Effectively, this implies large public transfers from those who do not have the care of young children to those who do. There is a role for other institutions as well, especially the institutions of employment. And there is scope for increased awareness among young people of the countervailing risk of delay, that is, the risk that couples will not be able to have the child they want to have. Many other risks and fears are associated with having children, and to the extent that a society is able to address these risks, it increases the chances it will have a higher fertility rate.
Notes 1 The United States is often cited as the exception to this rule and, more strongly, that the rule is invalid because it does not apply to the United States. The first point to be made is that the US occupies first position in Table 1 because of its diversity. Ethnic diversity has been shown to add about 0.2 births per woman to the US Total Fertility Rate (Lesthaeghe and Neidert 2006). Fertility for non-Hispanic whites in the US is around 1.8 and, as such, is similar to many other countries in Table 1. Religious and regional diversity contribute further to the relatively high fertility among US nonHispanic whites. The function of age-specific fertility rates in the US is bi-polar. A large section of US non-Hispanic whites, regionally and religiously concentrated, have very high fertility rates at young ages, unlike any other advanced Western society. Thus, it is difficult to talk about the US as a coherent whole. Beyond this, government assistance to families in the US through the tax system is substantial but largely unseen (for example, tax allowances for childcare expenses, child costs and mortgage payments). Childcare is readily available in the US, although there may be questions about its quality. Many US employers, through informal means, recognize the family demands of their workers. Finally, gender equity within the family is probably at a more advanced level (especially as the education level of the wife rises) than in any other advanced society (Tsuya et al. 2000).
38 Peter McDonald 2 Note: fertility in New Zealand is high because of the high fertility of the Maori population. Non-Maori fertility is also around 1.8, as in other countries in Table 2.1. 3 Cohabitation is now on the rise in some advanced East Asian countries (Iwasawa et al. 2005)
References Atoh, M., Kandiah, V., and Ivanov, S. 2004. “The second demographic transition in Asia? Comparative analysis of the low fertility situation in East and South-East Asian Countries”, Japanese Journal of Population 2(1): 42–75. Bettio, F. and Villa, P. 1998. “A Mediterranean perspective on the breakdown of the relationship between participation and fertility,” Cambridge Journal of Economics 22(2): 137–171. Chang, K-S. 2003. “The state and families in South Korea’s compressed fertility transition: a time for policy reversal?,” Journal of Population and Social Security (Population), supplement to Volume 1: 596–610. Choe, M.K., Retherford, R. and Kim, S.-K. 2004. “Very low fertility in South Korea: Patterns and prospects”. Paper presented to the 2004 Annual Meeting of the Population Association of America, April 1–3, Boston. Commission of the European Communities. 2005. “Confronting demographic change: a new solidarity between the generations,” Green Paper. Brussels: European Commission. D’Addio, A. and d’Ercole, M.. 2005. Trends and Determinants of Fertility Rates in OECD Countries: The Role of Policies, OECD Social, Employment and Migration Working Papers No. 27. Paris: OECD. Dalla Zuanna, G., Atoh, M., Castiglioni, M. and Kojima, K. 1998. “Late marriage among young people: the case of Italy and Japan,” Genus 54: 187–232. Eun, K-S. 2003. “Understanding recent fertility decline in Korea,” Journal of Population and Social Security (Population), supplement to Volume 1: 574–595. —— 2006. “Lowest-low fertility in the Republic of Korea: causes, consequences and policy responses,” Seminar on Fertility Transition in Asia: Opportunities and Challenges, December 18–20 2006, Bangkok, United Nations Economic and Social Commission for Asia and the Pacific. Frejka, T. and Sardon, J-P. 2004. Childbearing Trends and Prospects in Low-fertility Countries: A Cohort Analysis. Dordrecht: Kluwer Academic Publishers. Jones, G. 2005. “The ‘Flight from Marriage’ in South-East and East Asia,” Journal of Comparative Family Studies, Winter 2005, 36(1): 93–119. Lee, S-S. 2005. “Lowest fertility and policy directions in Korea,” International Workshop on Low Fertility and Population Policies. Seoul: Korea Institute for Health and Social Affairs. Lesthaeghe, R. 2001. “Postponement and recuperation: recent fertility trends and forecasts in six Western European countries,” paper presented at the International Perspectives on Low Fertility Seminar, International Union for the Scientific Study of Population, Tokyo, March 21–23. Lesthaeghe, R. and Neidert, L. 2006. “The second demographic transition in the United States: exception or textbook example?” Population and Development Review, 32(4): 669–698. Lutz, W., Skirbekk, V. and Testa, M. 2006. “The low fertility trap hypothesis; forces that may lead to further postponement and fewer births in Europe,” Vienna Yearbook of Population Research 2006: 167–192. McDonald, P. 2000a. “Gender equity, social institutions and the future of fertility,” Journal of Population Research 17(1): 1–16.
Explanations of low fertility in East Asia 39 McDonald, P. 2000b. “Gender equity in theories of fertility transition,” Population and Development Review 26(3): 427–439. —— 2006a. “Low fertility and the state; the efficacy of policy,” Population and Development Review 32(3). —— 2006b. “An assessment of policies that support having children from the perspectives of equity, efficiency and efficacy’, Vienna Demographic Yearbook 2006: 213–234. McDonald, P. and Kippen, R. 2001. “Labor supply prospects in 16 developed countries, 2000–2050,” Population and Development Review 27(1): 1–32. McDonald, P. and Temple, J. 2006. Immigration and the Supply of Complex Problem Solvers in the Australian Economy. Canberra: Australian Government, Department of Immigration and Multicultural Affairs. Ogawa, N., Kondo, M. and Matsukura, R. 2005. “Japan’s transition from the demographic bonus to the demographic onus,” Asian Population Studies, 1(2): 207–226. Oyama, M. 2001. “Fertility decline and female labour force participation in Japan,” Graduate School of Economics, University of Chicago, IL. —— 2003a. “Measuring costs of children using equivalence scale on Japanese panel data,” Faculty of Economics, Hitotsubashi University. —— 2003b. “On the factors which caused the recent fertility decline in Japan,” The Economic Review, 54(2). Retherford, R. and Ogawa, N. 2006. “Japan’s baby bust: causes, implications, and policy responses,” in F.R. Harris (ed.), The Baby Bust: Who Will Do the Work? Who Will Pay the Taxes? Langham, MD: Rowman & Littlefield Publishers: pp. 5–44. Shirahase, S. 2000. “Women’s increased higher education and the declining fertility rate in Japan,” Review of Population and Social Policy 9: 47–64. Skirbekk, V. 2003. “Age and productivity: a literature survey,” MPIDR Working Paper WP 2003-028, Max Planck Institute for Demographic Research, Rostock. Suzuki, T. 2006. “Lowest-low fertility and governmental ctions in Japan,” National Institute of Population and Social Security Research, Tokyo. Takahashi, H. and Voss, J. 2000. “Parasite singles – a uniquely Japanese phenomenon?,” Japan Economic Institute Report No. 31 A. Washington, DC: Japan Economic Institute, Takahashi, S. 2004. “Demographic investigation of the declining fertility process in Japan”, Japanese Journal of Population 2(1): 93–116. Tsuya, N., Bumpass, L. and Choe, M.K. 2000. “Gender, employment, and housework in Japan, South Korea and the United States,” Review of Population and Social Policy 9: 195–220. Tsuya, N. and Choe, M.K. 2004. “Investments in children’s education, desired fertility, and women’s employment,” in Noriko O. Tsuya and Larry L. Bumpass (eds), Marriage, Work, and Family Life in Comparative Perspective: Japan, South Korea, and the United States. Honolulu: University of Hawaii Press: pp. 76–94. van Peer, C. 2002. “Desired and achieved fertility,” in E. Klijzing and M. Corijn (eds), Dynamics of Fertility and Partnership in Europe: Insights and Lessons from Comparative Research, Volume II. New York, and Geneva: United Nations: pp. 117–142. Yamada, M. 1996. The Sociology of Marriage; Will Marriage Continue to Be Delayed? Tokyo: Maruzen Library (in Japanese). Yap, M-T. 2003. “Fertility and population policy: the Singapore experience,” Journal of Population and Social Security (Population), supplement to Volume 1: 643–658. Yip, P., Li, B., Xie, K. and Lam, E.. 2006. “An analysis of the lowest Total Fertility Rate in Hong Kong SAR”, Discussion Paper No. 289, Institute of Economic Research, Hitotsubashi University, Tokyo.
3
Japan’s declining fertility and policy responses Naohiro Ogawa, Robert D. Retherford, and Rikiya Matsukura
At the end of 2006, world population was approximately 6.6 billion persons (US Census Bureau, 2006). Currently the annual rate of growth of world population is about 1.3 percent, considerably lower than the peak value of 2.1 percent in the latter half of the 1960s. Growth has slowed because of reductions in fertility, which in many countries is now below the replacement level of approximately 2.1 births per woman, as measured by the total fertility rate (TFR). Between 1970 and 1975 and 2000 and 2005, the number of countries with below-replacement fertility increased from 19 to 65. Most of these countries are in the world’s more-developed regions, but the number in less-developed regions also increased, from 0 to 19. Worldwide over this same period, the number of countries with “lowest-low” fertility (TFR below 1.3) increased from 0 to 17. A large majority of the world’s low-fertility countries are in Europe. European countries can be divided into two groups (McDonald, 2005): Group 1 includes the Nordic countries, the English-speaking countries, and the French- and Dutchspeaking countries. In this group of countries the TFR is at or above 1.5 births per woman. In most of these countries, fertility is below replacement, but in a few it is at replacement or slightly higher. Group 2 includes countries with a TFR below 1.5, including the Southern European countries and the German-speaking countries. One can expand Group 2 to include most countries of East Asia, where some of the lowest fertility in the world is found. Japan is one of those countries. Surprisingly, the Group 2 countries, including those in East Asia, tend to be characterized by strong family values that emphasize the responsibility of the family for taking care of its own members. It appears that, in modern circumstances, meeting this responsibility poses many difficulties, especially for women who work or want to work outside the home – the result being later marriage, less marriage, and fewer children within marriage. In East Asia, Japan was the first country in which the transition from high to low fertility occurred. The TFR fell rapidly during the 1950s, reaching 2.04 in 1957. It then leveled off for the most part until 1973, after which it gradually fell to 1.26 in 2005. Japan’s labor force began shrinking in 1998, and overall population, which continued to grow after 1973 because of temporary age-structure effects known as “population momentum,” finally began shrinking in 2006. Japan’s fertility decline over the past half century is now contributing substantially to rapid population
Japan’s declining fertility and policy responses 41 aging, as measured by the proportion of the population who are age 65 and over. Already by 2005, Japan had the oldest population in the world, and currently the pace of population aging is accelerating as the post-World War 2 baby boomers enter the retirement ages. In 1990, when it was announced that Japan’s TFR in the previous year had fallen to a record low of 1.57, the Japanese government became sufficiently alarmed about the country’s declining fertility to look into possible measures to reverse the downward trend. Major governmental concerns have included a possible loss of economic dynamism and future difficulties in maintaining the solvency of the country’s social security system, which has provided universal pension and medical coverage since 1961 (Ogawa and Retherford, 1997; Mason and Ogawa, 2001). Businesses are additionally concerned about shrinking numbers of consumers on the demand side and workers on the supply side (Clark and Ogawa, 1997). And with only one or two or even no children (and in many cases no spouse either), adults worry about who will take care of them in their old age (Ogawa and Retherford, 1997; Ogawa, 2005). Since 1990, both the central and local governments have been introducing, implementing, and strengthening a series of policies aimed at reversing the decline of fertility (Retherford, Ogawa, and Matsukura, 2001; Retherford and Ogawa, 2006). Largely as a result of government mandates put into effect after 2003, a majority of large and medium-sized employers have also initiated a variety of programs for relieving child-rearing burdens on their employees and their families. This chapter describes and analyzes these trends. The focus is on the causes of fertility decline and the actual and potential effects of the government’s efforts to raise fertility.
Trends in marriage and fertility in Japan We begin by presenting in more detail the trends in marriage and fertility that we wish to explain. The trends considered pertain to the TFR, mean age at marriage, proportion never marrying, and parity progression ratios. The TFR is the measure of fertility most commonly used by policy-makers, and it is also the principal fertility measure used in this chapter. The TFR for a particular calendar year is conventionally defined as the number of births that a woman would have by age 50 if, hypothetically, she lived through her reproductive years experiencing the age-specific fertility rates (ASFRs) that prevailed in the population in the particular calendar year. An ASFR in a particular calendar year is calculated as the number of births that occurred during the year to women of a given age, divided by the mid-year number of women of the same age. An ASFR therefore has units of births per woman per year. The TFR is calculated by summing the ASFRs (births per woman per year at each age) between the ages of 15 and 50. The term “parity” is a characteristic of a woman, referring to the number of children that she has ever borne. A parity progression ratio (PPR) is simply the fraction of women of a given parity who go on to have at least one more birth – i.e.,
42 Naohiro Ogawa, Robert D. Retherford, and Rikiya Matsukura the fraction who eventually progress to the next parity. In this chapter, parity 0 is subdivided into two parity states, namely never-married with no births and evermarried with no births. The first parity transition is therefore from a woman’s own birth to her first marriage, so that the “event” in this case is a first marriage instead of a birth. As in the case of the TFR, parity progression ratios are period measures calculated for a particular calendar year. Thus a PPR for a particular calendar year is defined as the probability that a woman of a given parity will have at least one more birth (or first marriage, in the case of the first parity transition) if, hypothetically, she lives through her reproductive years experiencing the duration-in-parity-specific birth rates (or first-marriage rates) that prevailed in the population in the particular calendar year. PPRs are calculated from duration-in-parity-specific birth rates (or first-marriage rates) using life table methods (Feeney, 1986; Feeney and Saito, 1985). A TFR can be calculated from PPRs as well as from ASFRs (Feeney, 1986). Usually the two TFRs differ somewhat from each other. Depending on the aims of analysis, we use one or the other of these two TFRs at different points in this chapter. A total marital fertility rate (TMFR) can also be calculated from PPRs, starting with the PPR from first marriage to first birth, and we also make use of this measure. The TMFR is actually a total ever-marital fertility rate, but for simplicity we refer to it as a total marital fertility rate. Trends in Japan’s TFR (calculated from ASFRs) and TMFR (calculated from PPRs) are graphed in Figure 3.1. The trend in overall fertility, as measured by 5.00 4.50
Number of children
4.00 3.50 3.00
Ideal family size
2.50 TMF
2.00
R
1.50
TFR
1.00 0.50 0.00
1947 1952 1957 1962 1967 1972 1977 1982 1987 1992 1997 2002 Year
Figure 3.1 Trends in the Total Fertility Rate (TFR), the Total Marital Fertility Rate (TMFR), and ideal family size (number of children), Japan, 1947–2005. Note Ideal family size is for currently married women of reproductive age. TFR is calculated from agespecific fertility rates, and TMFR is calculated from parity progression ratios.
Japan’s declining fertility and policy responses 43 TFR, can be divided into three stages. In the first stage between 1947 and 1957, when Japan’s economy was recovering from the devastation of World War 2, the TFR fell from 4.54 to 2.04. These were difficult times. During this period, contraception was backed up by abortion, which was legalized in 1948 and widely used because of the limited availability of effective contraception. In the second stage between 1957 and 1973, the TFR stabilized at about 2.1 births per woman. During this period, economic growth was very rapid, averaging about 10 percent a year, and most people were in a better financial position to marry and have children than they were during the previous period. The result of this increasing prosperity was that age at marriage stopped rising and marital fertility stopped falling. In the third stage from 1973 to the present, the TFR fell gradually from 2.14 in 1973 to an all-time low of 1.26 in 2005. In the long run, a TFR as low as 1.26 implies rapid population decline. If the TFR were to remain constant at this level in the future, and if mortality were to remain constant at its present low level, population would eventually start declining at a rate of approximately 40 percent per generation. The length of a generation is approximately equal to the mean age at child-bearing, which in Japan is 30 years. For a while, however, the population would decline more slowly than this, because of the phenomenon of population momentum referred to earlier. The catalyst for the resumption of fertility decline after 1973 was the oil shock of 1973, caused by major oil-price increases imposed by the Organization of the Petroleum Exporting Countries (OPEC). Because Japan imports virtually all of its oil, the steep rise in the price of oil had an especially large impact. The economy, which had been growing at a rate of about 10 percent a year, fell into recession, followed three years later by a rebound to a lower growth rate of 3−4 percent per year. Overall price inflation (for all goods, not just oil) during those three years was 53 percent. As a consequence of this inflation, unions fought for and won large wage increases for regular full-time workers, causing struggling companies to start hiring part-time workers in large numbers at much lower wages. These jobs were filled mainly by women. Previously many of these women did piece-work at home, but after 1973 they increasingly engaged in production work outside the home, which was less compatible with child-rearing and therefore part of the explanation of the resumption of fertility decline. About half of the decline in the TFR (calculated from PPRs) between 1973 and the present has occurred because of later marriage and less marriage, and that is why, in Figure 3.1, TMFR has fallen more gradually than TFR. TMFR hardly changed after 1998, indicating that later marriage and less marriage account for virtually all of the decline in TFR after that year. The effect of marriage on fertility is especially strong in Japan because only about 2 percent of births occur out of wedlock, a percentage that is much lower than in Europe, North America, or Oceania. Between 1970 and 2005 the singulate mean age at marriage (SMAM, calculated from age-specific proportions still single in a census) in Japan rose from 27.5 to 31.1 years for men and from 24.6 to 29.4 years for women, as shown in Figure 3.2. Over the same period, the synthetic proportion still single at age 50, S50 (calculated from age-specific first-marriage rates in a given year), rose from
44 Naohiro Ogawa, Robert D. Retherford, and Rikiya Matsukura 32 30
Age
28
Men
26 Women 24 22 20 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 Year
Figure 3.2 Trends in Mean Age at Marriage (SMAM) by sex, Japan, 1950−2005.
9 to 27 percent for men and from 4 to 24 percent for women, as shown in Figure 3.3. By 2005, Japan was far from the “universal marriage society” that it was in 1970. Both SMAM and S50 continued to increase rapidly between 2000 and 2005. The increase in SMAM was especially rapid for women, reflecting a large jump in proportion single at age 25−29, from 54 to 59 percent. 30 25
Percent
20 Male 15 Female
10 5 0
1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 Year
Figure 3.3 Trends in the synthetic proportion still single at age 50, by sex, 1951−2005. Note: The synthetic proportion still single at age 50 is calculated from age-specific first-marriage rates in the specified calendar year using life table methods. Age-specific first-marriage rates are calculated from census and vital registration data.
Japan’s declining fertility and policy responses 45 Figure 3.4 shows trends in PPRs for successive parity transitions, labeled B−M (birth to first marriage), M−1 (first marriage to first birth), 1−2, 2−3, and so on. Before 1973, PPR declines were concentrated at 2−3 and higher-order transitions. The downward spikes in PPRs in 1966 occurred because 1966 was the Year of the Fire Horse. According to superstition, girls born in that year will be unlucky in life. After 1973, PPR declines were concentrated at B−M, M−1, and 1−2. Declines in these latter three PPRs were somewhat steeper during the bubble economy that emerged during the latter half of the 1980s, when housing prices skyrocketed, and during the “lost decade” of economic stagnation after the bubble burst in 1990. Only after early 2002 did the economy finally enter a period of slow but steady growth that has persisted until the present time. This economic recovery has been accompanied by an increase in PPR(M−1), which, however, commenced in 1999 before the recovery began. The likely reasons for the increase in PPR(M−1) but not other PPRs will be discussed later. Although PPR(M−1) increased, PPR(B−M) continued to decline. (Note that the S50 values in Figure 3.3 are calculated as 1−PPR(B−M) in Figure 3.4.) Figure 3.5 shows that, although the proportion ever marrying, PPR(B−M), continued to decline, the upward trend in mean age at marriage (MAM) finally started to level off after 1997 and even to decline slightly, for both women and their husbands. MAM is calculated as the mean age at marriage for marriages that actually occurred in the particular calendar year. By contrast, SMAM is calculated from age-specific proportions single, which are influenced not only by marriages 100
B to M
90
M to 1 1 to 2
80
PPPR (percent)
70 60 50 40 3 to 4
30 20
2 to 3
4+ to 5+
10 0 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 Year
Figure 3.4 Trends in Period Parity Progression Ratios (PPPRs), Japanese women, 1950−2005. Note: Computed from census and vital registration data.
46 Naohiro Ogawa, Robert D. Retherford, and Rikiya Matsukura 34.0 32.0
Age
30.0 Men
28.0 26.0
Women
24.0 22.0 20.0 1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
Year of marriage
Figure 3.5 Mean Age at First Marriage (MAM), by sex, 1955–2002. Note: Computed from pooled data from various rounds of the National Survey on Family Planning and the 2004 National Survey of Population, Families, and Generations.
in the particular calendar year under consideration but also by marriages that occurred further in the past. Another difference between the two measures is that MAM is influenced by the age structure of the population, whereas SMAM is not. Given these differences, it is not surprising that the trends MAM and SMAM in Figures 3.2 and 3.5 differ from each other. Later we shall discuss some possible reasons why MAM leveled off and even declined slightly after 1998. The earlier graph in Figure 3.1 also shows that, among currently married women, ideal family size (based on survey questions that ask the respondent about her own ideal number of children) has hardly changed over time. Ideal family size remains at about 2.5 children, more than one child higher than the TFR. The existence of this large gap suggests that if ideal family size were eventually to shift downward to 2.0 children, as seems likely, the TFR could decline to a level even lower than its 2005 value of 1.26.
Causes of fertility decline after 1973 There are two major proximate causes of fertility decline in Japan:
•• later marriage and less marriage •• fertility decline within marriage.
Japan’s declining fertility and policy responses 47 Each of these two proximate causes has its own set of causes. The two sets overlap but are not quite the same. Causes of later marriage and less marriage after 1973 As discussed in more detail elsewhere (Retherford, Ogawa, and Matsukura, 2001; Retherford and Ogawa, 2006; Yamada, 1996, 2001, 2004, 2006), the main reasons for later marriage and less marriage after 1973 in Japan include the following:
•• •• •• •• •• •• •• ••
remarkable educational gains by women a rising proportion of single women who work for pay outside the home the near-disappearance of arranged marriage a major decline in coresidence with parents for many so-called “parasite singles,” the prospect of a substantial drop in standard of living if they get married major increases in the prevalence and social acceptability of premarital sex emergence of a “new-single concept” that it is socially acceptable to enjoy single life without pressure to get married women’s increasing desire for more help from husbands and a more egalitarian marital relationship.
A brief summary of key evidence relating to the above points follows (for a more detailed discussion, see Retherford and Ogawa, 2006). Numerous studies throughout the world have shown that education is a key determinant of both marriage timing and permanent non-marriage, mainly because it expands women’s horizons, including new opportunities for employment outside the home (Kohler et al., 2002; United Nations, 1995). In Japan, the proportion of women in the relevant age range who were enrolled in colleges and universities increased from 5 to 50 percent between 1955 and 2005. Work for pay also influences marriage, in large part because a paycheck reduces the financial pressure on single women to get married. In the case of Japan, the proportion of currently married women who worked before marriage increased from 49 to 99 percent between 1955 and 2003 (these are the calendar years in which the marriages occurred) (Retherford and Ogawa, 2006). The decline of arranged marriage has also contributed to later marriage and less marriage. Between 1955 and 2002 the proportion of marriages that were arranged fell from 63 to 2 percent (Retherford and Ogawa, 2006). As a result, young singles have increasingly had to find their own spouses, which is not so easy in Japan, because the decline of arranged marriage has not been fully compensated by the emergence of a well-developed marriage market. For many women, meeting and getting to know potential spouses is a problem. Contact with potential spouses tends to be limited to a small circle of colleagues at work and former schoolmates. Among single women age 20 and over, the proportion reporting they have no male friend (neither a boyfriend nor any other male friend) was 34 percent in the 1990 round of the National Survey on Family Planning, 38 percent in the 1994 round,
48 Naohiro Ogawa, Robert D. Retherford, and Rikiya Matsukura 41 percent in both the 1996 and 1998 rounds, and 39 percent in the 2004 National Survey of Population, Families and Generations (Retherford and Ogawa, 2006). All of these surveys were conducted by the Mainichi Newspapers of Japan. Also between 1955 and 2002, the proportion of young couples who coresided with parents when they got married fell from 64 to 29 percent, making it financially more difficult for the average couple to get married because of the costs to non-coresiding couples of setting up an independent household (Retherford and Ogawa, 2006). The proportion of young single adults who live with parents is still high, however, relative to comparable proportions in other industrial countries. A related development is the growth in the number of “parasite singles,” a term coined by Yamada (1996) to describe young single persons who work for pay, live with parents, contribute little or nothing to household finances, and therefore have ample money to lead the good life. The term “parasite single” is somewhat of a misnomer, however, because in many cases these young single persons would like to get married soon but fail to do so because of difficulties in finding a suitable spouse, as will be discussed in more detail later. In 2004, the proportion of single employed women who were living with their parents was 73 percent at age 20−24, 67 percent at age 25−29, and 69 percent at age 30−34 (computed from the 2004 National Survey of Population, Families and Generations). Increase in the prevalence and social acceptability of premarital sex in Japan has meant that it is increasingly unnecessary to get married in order to have sex, thereby also contributing to later marriage and less marriage. Between 1974 and 1999, the proportion of junior college and university students who reported having had sexual intercourse increased from 23 to 63 percent among men and from 11 to 51 percent among women (Retherford et al., 1996; Japanese Association for Sex Education, 2001). Between 1990 and 2004, the proportion of single women age 20 and over reporting current use of contraception rose from 39 to 57 percent (Retherford et al., 2001). Many of the value changes contributing to later marriage and less marriage are embodied in the rise of the so-called “new-single concept,” which received considerable publicity in the Japanese press in the late 1980s and which is closely associated with the rise of the “parasite single” lifestyle. The new-single concept is that it is socially acceptable to enjoy single life without pressure to get married. Survey data show that the proportion favoring the new-single concept was 78 percent in 1988 and 76 percent in 1993 among single women and 59 percent in 1988 and 62 percent in 1993 among single men, indicating that the values embodied in the new-single concept were already accepted by a majority of single persons before the mass media publicized the new-single concept (Retherford et al., 1996). Collectively, all the above changes affecting marriage, but especially the increases in women’s educational attainment and work for pay outside the home, have led to a rise in values of individualism among young single women. The rise of individualism has also promoted later marriage and less marriage. Women increasingly do not want to live with a mother-in-law, although this is rather difficult to avoid because nowadays in low-fertility Japan about three-quarters of
Japan’s declining fertility and policy responses 49 young men are eldest sons, who traditionally coreside with the parents. Women also increasingly want a more egalitarian marriage with more help from husbands with child-rearing and housework, although this is difficult to achieve because of men’s long working hours and because men’s views about helping with childrearing and housework, though changing, lag behind women’s views (Tsuya and Bumpass, 2004). Table 3.1 provides additional evidence pertaining to the effect of women’s rising educational attainment on proportion never marrying. The table is based on pooled data for women age 22 and over in each of six surveys between 1990 and 2004. A proportional hazards model (which can be viewed as a multivariate life table) is used to model the effects of education, childhood residence, and year of survey on progression to first marriage – or, equivalently, as in Table 3.1, not Table 3.1 Model estimates of proportion of women still single at age 40, by education, childhood residence, and year of survey, Japan, 1990–2000 Predictor variable Education Jr. high or less† Sr. high Jr. college University Childhood residence Urban Rural† Year of survey 1990† 1994 1996 1998 2000 2004
Model including education and childhood residence
Model without education and childhood residence
6 7 12* 21* 10* 9 5 8* 9* 11* 11* 20*
5 8* 9* 11* 12* 20*
Source: Computed from pooled data from the 1990, 1994, 1996, 1998, 2000, and 2004 rounds of the National Survey on Family Planning and the 2004 round of the National Survey of Population, Families and Generations, conducted by the Mainichi Newspapers of Japan. Notes In the full model, the predicted percentages of single 16-year-old women who will marry by age 40 were computed from a proportional hazard model with education, childhood residence, and year of survey as predictor variables. All of the predictor variables are represented by sets of dummy variables in the underlying hazard regressions. The percentages in any given panel of the table (such as the panel for education) control for all the other predictor variables in the model by setting them at their mean values. The second model is the same as the full model, except that education and childhood residence are omitted as predictor variables. The models were estimated only for women who were at least 22 years old (pooled over the six surveys) when they were interviewed, so that their education was completed in most cases. †
Indicates a reference category in the underlying hazard regression. * Indicates that the variable coefficient in the underlying hazard regression differs significantly from zero at the 5 percent level.
50 Naohiro Ogawa, Robert D. Retherford, and Rikiya Matsukura progressing to first marriage. Ideally, a statistical model of progression to first marriage should allow for non-proportionality, because in reality the effects of these three predictor variables are to lower first-marriage risks at younger ages and raise them at older ages, to the extent that marriage is delayed. Thus the effects are not constant (i.e., not proportional) over age. Because of this, the model does not yield good estimates of the effects of the three predictor variables on mean age at first marriage, and these effects are therefore not shown in the table. As Allison (1995) points out, however, the assumption of proportional hazards is usually defensible if all one is interested in is the probability of “failing” or “not failing” by the end of the multivariate life table. In Table 3.1, “not failing” is indicated by the model-predicted probability of still being single at age 40. The specification of an average effect that is constant rather than varying over age usually makes little difference in the proportion who fail or do not fail by the end of the life table. Table 3.1 shows that even though the proportion of women who never marry increases with education (with childhood residence and year of survey controlled), improvements in women’s educational attainment between 1990 and 2004 do not explain the large increase in proportion never marrying that occurred over the same period. This is indicated by the finding that the model in which education and childhood residence are controlled and the model in which they are not controlled yield almost identical estimates of the trend in the proportion never marrying. The likely reasons for this unexpected finding are:
•• Age-specific first marriage risks have declined substantially and to roughly the same extent in all childhood residence and education groups.
•• Compositional changes in the population by educational attainment were fairly small between 1990 and 2004, with the result that the across-the-board changes in age-specific first-marriage risks dominate the trend in the proportion who never marry, and that changes in population composition by childhood residence and education account for almost nothing. The inference that across-the-board effects dominate is supported by Figures 3.6 and 3.7, which show, for women and men separately, trends in mean age at first marriage by education. The figures show that mean age at first marriage has been increasing rapidly and at roughly the same rate in all education categories for both women and men. (The same is true of the trends in mean age at marriage in each of the two childhood residence categories – results not shown.) Figure 3.7 for women also shows that the trends for “junior high or senior high” and “junior college” are almost the same, so that changes in population composition pertaining to these two education categories have little or no effect on the trend age at first marriage. In this regard, it should be noted that it is still true in Japan that most female high-school graduates who continue their education go to junior college rather than university. The findings in Table 3.1 and Figures 3.6 and 3.7 do not necessarily mean, however, that rising educational attainment and the shift toward urban residence have had no effect on the trends in mean age at marriage and proportion never
34.0 32.0
Age
30.0
Junior college or university
28.0 26.0
Junior high or senior high
24.0 22.0 20.0 1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
Year of marriage
Figure 3.6 Average age at first marriage for men, by education, 1955–2002. Note: Computed from pooled data from various rounds of the National Survey on Family Planning and the 2004 National Survey of Population, Families, and Generations.
34 32 30 Age
University 28 26
Jr. college
Jr. high or Sr. high
24 22 20 1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
Year of marriage
Figure 3.7 Average age at first marriage for women, by education, 1955–2002. Note: Computed from pooled data from various rounds of the National Survey on Family Planning and the 2004 National Survey of Population, Families, and Generations.
52 Naohiro Ogawa, Robert D. Retherford, and Rikiya Matsukura marrying among women over the time period in question. It is highly likely that they have had an effect. Our findings suggest, however, that this effect, to the extent that it exists, may operate mainly through changes in marriage-related attitudes and values (often occurring after a time lag) that tend to be shared throughout the population. Figure 3.8 provides further information relating to the effect of women’s work for pay before marriage on the mean age at marriage and the proportion never marrying. The figure shows that almost all of the increase in the proportion of currently married women who worked before marriage occurred before 1980, suggesting that change in the proportion of women who worked before marriage does not explain much of the increases in mean age at marriage and proportion never marrying that occurred after 1980 – and especially after 1993, when the proportion who worked for pay before marriage actually declined. Of course, there is much that is not revealed in Figure 3.8, because changes in the proportion who worked before marriage tell us nothing about changes in type of work (parttime or full-time), occupational structure, rates of pay, or job tenure (how long a woman worked before marriage). Women’s work before marriage plays an important but rather subtle role in Yamada’s (1996, 2006) argument that marriage in contemporary Japan is increasingly delayed or forgone because, for working “parasite single” women, marriage is likely to entail a substantial drop in standard of living. The essence of Yamada’s argument (somewhat modified and elaborated here), is the following: 100 90
Worked before marriage Worked for pay before marriage
Proportion (percent)
80 70 60 50 40 30 20 10 0 1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
Year of marriage
Figure 3.8 Trends in the proportions of currently married women who worked before marriage and who worked for pay before marriage, Japan, 1955–2002. Note Computed from pooled data from various rounds of the National Survey on Family Planning and the 2004 National Survey of Population, Families, and Generations.
Japan’s declining fertility and policy responses 53 In the years after World War 2 until the 1970s, the parents of young adults tended to be relatively poor, having entered the labor force some 20 to 30 years earlier, either before or during the war or soon after the war when the economy was still recovering from wartime devastation. By the mid-1950s, however, the young unmarried male adult children of these parents were finding it relatively easy to get regular full-time jobs with seniority-based promotions in the rapidly expanding economy, a situation that persisted until the oil shock of 1973. Of necessity, these young men helped their parents financially, often turning over their earnings and receiving an allowance in return, especially if they lived with their parents. Reflecting higher fertility in the past, these young men tended to have several siblings. In conformity with Confucian tradition, eldest sons tended to marry first and then coreside with the parents. Younger siblings (both sons and daughters) were likely to have to move out of the parental household due to over-crowded living conditions. Typically, the parents arranged the marriages of these younger children (especially daughters) when they moved out. The contemporary situation is quite different. Today’s parents in their 40s and 50s are financially much better off than their counterparts were several decades ago, and they typically have only two children, so none of their adult children has to move out of the household as a result of over-crowding. On the contrary, most parents consider themselves lucky if one of their children continues to stay on and live with them. Virtually all of their unmarried adult children work, but in most cases the parents no longer need financial support from them, so these “parasite singles” typically contribute little or nothing to household expenses and therefore have ample money to spend on expensive clothes, restaurants, vacations, and other aspects of the good life. Another aspect of the contemporary situation is that, for young adults entering the labor force, good jobs are harder to find than they were during the pre-1973 years when the economy was rapidly expanding. As a result of the 1973 oil shock, the collapse of the bubble economy in 1989, and the lost decade of the 1990s, the economy has gone through two periods of major restructuring, each resulting in an increase in the proportion of jobs that are not regular full-time jobs. These non-regular jobs are part-time or other temporary jobs that are poorly paid with little scope for advancement. Many part-time jobs are actually full-time jobs but classified as part-time jobs, which are temporary in the sense that they can be terminated at short notice. In the remainder of this chapter, we shall refer to regular full-time jobs simply as “full-time jobs” and to part-time and other temporary jobs simply as “part-time jobs”. Employees are in one or the other of these two categories. There is very little mobility between the two categories, and that is why Japan’s labor market is often characterized as “highly segmented”. What little mobility does occur mostly happens when a woman quits a full-time job in order to care for young children for a few years, after which case she usually has to come back to a part-time job. Upward mobility from part-time to full-time status is rare. At the present time, annual income averages about US$10,000 for a part-time job and about US$40,000 for a full-time job.
54 Naohiro Ogawa, Robert D. Retherford, and Rikiya Matsukura (Here and elsewhere in this chapter, we assume a conversion rate of 100 yen to the US dollar.) During the first period of restructuring following the 1973 oil shock, the parttime jobs that were being created in large numbers were filled mainly by women. During the second period of restructuring during the 1990s, many more such jobs were created, and this time they were being filled not only by women but also by young men who were not able to find full-time jobs. This is shown in Figure 3.9, which graphs the trend in the proportion of employed men and women age 20−29 (regardless of marital status) working in part-time jobs. Even full-time jobs are becoming less secure as the proportion of firms that offer lifetime employment and seniority-based promotion declines, as shown in Figure 3.10. The various changes just described have resulted in a growing gap between the average standard of living enjoyed by young single working women living with their parents (the “parasite singles”) and the average standard of living that these women are likely to have if they get married. Not surprisingly in this situation, the income and occupation of prospective husbands are important mate-selection criteria for Japanese single women. Data from the 1988 National Family Survey and the 1998 round of the National Survey on Family Planning, both conducted by the Mainichi Newspapers of Japan, demonstrate this point. These surveys asked 20 18 16
Percent
14 12 10
Women
8 6 Men
4 2 0 1972
1977
1982
1987
1992
1997
2002
Year
Figure 3.9 Trend in the proportion working in part-time jobs among paid employees age 20–29, by sex, 1972–2005. Note Computed from published data from annual rounds of the Labour Force Survey (Statistics Bureau, various years).
Japan’s declining fertility and policy responses 55 60
Proportion (percent)
50
40
5000+
1000–4999
30
300–999 30–99
20 100–299
10
0 1988
1990
1993
1996
1999
2002
Year
Figure 3.10 Trend in the proportion of firms reporting that they follow the lifetime-employment system, by firm size (number of employees), Japan, 1988–2002. Source: Survey on Employment Management (Ministry of Health, Labour and Welfare, various years).
single women to rank each of the following mate-selection criteria, relating to characteristics of prospective husbands, from 1 to 4 according to their degree of importance, with 4 indicating “very important”: income, occupation, coresidence with parents, education, age, and parents’ property. Average scores for these criteria are shown in Figure 3.11, which shows that income and occupation are the most important mate-selection criteria among those considered. Although young single women prefer men with good incomes, the hard reality is that it is difficult for these women to find husbands who meet their income criteria. Because of this, many young single women end up delaying or forgoing marriage. Men also tend to delay marriage until they feel able to support a family at the expected standard of living. To investigate more precisely how difficult it is for young single women to find a suitable husband, Yamada (2004) undertook a small survey on “Consciousness of Risk Involved in Childrearing in the Context of Lifetime Planning Among Young Adults,” which was conducted in the city of Hirosaki (Hirosaki-shi) in Aomori prefecture (sample size: 444 respondents age 25−34; response rate of 44 percent) and Nerima ward (Nerima-ku) in Tokyo (sample size: 609 respondents age 25−34; response rate of 50 percent). At the time the survey was taken, Hirosaki-shi had relatively low incomes and high unemployment, and Nerima-ku had relatively high incomes and low unemployment. Data from these surveys are analyzed in Tables 3.2 and 3.3. The tables show:
56 Naohiro Ogawa, Robert D. Retherford, and Rikiya Matsukura 3.3 Income 3.1 Occupation
2.9
Score
2.7 Coresiding with parent(s)
2.5 2.3
Education
Age
2.1 1.9 Parent’s property 1.7 1.5 1988
1998 Year
Figure 3.11 Change in score for mate-selection criteria among single women between 1988–1998.
•• On average, single women’s expected incomes of prospective husbands far exceed the actual incomes of single men.
•• The proportion single among men age 25−34 falls steeply as income increases. Considered together, Tables 3.2 and 3.3 suggest that the trends toward later marriage and less marriage in Japan have been more pronounced among those with lower incomes – a category that has been growing as a consequence of the general trend, not confined to Japan, of rising income inequalities. Yamada did not control for educational attainment, but his finding of a negative effect of income on proportion single among young men would probably be even stronger if he had done so, because education and income are positively correlated and, as seen earlier in Figure 3.6, age at marriage rises with education. On the other hand, if he had controlled for single years of age within the 25−34 age group, the negative effect of income on proportion single would have been weaker. Further research based on data for the whole country, with higher survey response rates and more controls, is needed to test the validity and generality of Yamada’s findings.
Japan’s declining fertility and policy responses 57 Table 3.2 Men’s income and women’s expectations of prospective husband’s income, never-married persons age 25–34 in Hirosaki-shi in Aomori prefecture and Nerima-ku ward in Tokyo, 2003 (percent in income category) Annual income
Hirosaki-shi Men’s income Women’s expectations Nerima-ku Men’s income Women’s expectations
¥ 2 million (men) Income not important (women)
> ¥ 2 million > ¥ 4 million > ¥ 6 million
48 31
52 69
2 54
1 14
34 30
66 70
24 66
4 39
Source: Yamada (2004).
Table 3.3 Proportion never married among men age 25–34 by work status and annual income in Hirosaki-shi in Aomori prefecture and Nerima-ku ward in Tokyo, 2003 (percent)
Work status Full-time Part-time Self-employed Annual income ¥ 2 million ¥ 2–4 million > ¥ 4 million
Hirosaki-shi
Nerima-ku
44 (114) 76 (59) 71 (31)
48 (151) 89 (70) 41 (34)
70 55 15
85 64 35
Source: Yamada (2004). Note In the case of work status, numbers in parentheses are the numbers of men age 25–34 on which the percentages are based.
Causes of falling marital fertility after 1973 As seen earlier in Figure 3.4, virtually all of the decline in marital fertility after 1973 is due to declines in PPR(M−1) and PPR(1−2) – i.e., in progression from marriage to first birth and progression from first to second birth. After 1997, however, PPR(M−1) began to increase, while PPR(1−2) continued to decline. The aim of this section is to explain not only the overall declines in PPR(M−1) and PPR(1−2) but also the increase in PPR(M−1) after 1997.
58 Naohiro Ogawa, Robert D. Retherford, and Rikiya Matsukura The main reasons for the overall decline in marital fertility after 1973 are:
•• The direct costs of children have risen, involving a substitution of quality for quantity of children.
•• The opportunity costs of children, in terms of lost income for women, have risen.
•• Preferences have shifted away from children toward “other goods,” involving a decline in the “consumption utility” of children.
•• In many respects, families are less secure, so that the wife’s job has come to play a more important role in family finances.
Evidence in support of these points (summarized, for the most part, from Retherford and Ogawa, 2006) will now be examined. Estimates of the direct costs of children under different scenarios around the year 2000 are available from the research division of the AIU Life Insurance Company of Japan (2001). The least expensive scenario assumes that the child will be educated through university in government schools, in which case the cost is US$286,000. This includes food, shelter, and education but not other expenses such as juku (expensive cram schools for passing entrance examinations to get into the best schools). The most expensive scenario assumes attendance at private schools all the way though medical school, in which case the cost is US$630,100. The indirect opportunity cost of children, in terms of a woman’s lost income when she drops out of the labor force temporarily to have children, is often much larger than the direct cost of children, especially for women with higher education. A recent government White Paper (Cabinet Office, 2005) estimates this opportunity cost under two different scenarios. In the first scenario, the woman graduates university, works full-time for six years, drops out for six years to have children, and then comes back to another full-time job. Income lost by age 60 (including pension benefits after age 60) as a result of temporarily dropping out is US$993,600. The second scenario is the same as the first except that the woman returns to a part-time job. Income lost by age 60 is US$2,272,000. The “consumption utility” of children has fallen for several reasons, some of which are demographic in nature. People today typically grow up with only one sibling who is close in age. Teenagers and young adults rarely interact with little children, who are simply not part of their lives. As a consequence, young adults increasingly do not know how to enjoy children. Later marriage magnifies this effect, because young adults increasingly settle into a lifestyle in which children play no part. Accompanying the rise of this new lifestyle is the emergence of new values of such as “finding oneself,” “realizing one’s potential,” and (a fairly new development for women) pursuing a career. Some European demographers refer to these new values as “post-modern fertility preferences” (e.g., van de Kaa, 2006). The consumption utility of children has undoubtedly fallen in Japan, but, given the constancy of ideal family size at about 2.5 children (Figure 3.1), there appears to be considerable potential for further decline. In some ways, the evolving nature of the modern economy and society also
Japan’s declining fertility and policy responses 59 means less security for families. The decline of the extended family means that elderly parents and other relatives are increasingly not there to help a young family during hard times. Economic globalization and market-oriented reforms cause more job competition and increasingly less job security as the shift away from the lifetime-employment and seniority-based wage system continues (already mentioned in the discussion of Figures 3.9 and 3.10). Couples perceive that the husband’s future stream of income is increasingly uncertain, and that they need two incomes so that they do not have to sell a home or other assets if one spouse is unemployed for a while. Thus, a woman’s job becomes more precious, and this also contributes to later child-bearing and fewer children. The rise of divorce also means less security for families. Wives increasingly perceive that they need a good job as a hedge against divorce, especially since they are the ones who are likely to have custody of the children in case of divorce. As a consequence, women increasingly tend to view their jobs as relatively more important and having children as relatively less important. Husbands are also increasingly less keen about having children (or having another one), because in the event of divorce, husbands are likely to have to pay child support without the benefit of the day-to-day joys of children. Japan’s crude divorce rate (divorces per 1,000 population per year) rose from 0.74 to 2.15 between 1960 and 2005. By way of comparison, the rate was 2.09 in France and 2.59 in Germany in 2004. As Ogawa and Ermisch (1994) have noted, the rise of divorce in Japan is related to women’s increasing participation in paid employment, especially regular fulltime employment. A new development is that, according to the 2004 pension reform, a divorcee will be qualified to receive not only her own basic old-age pension but also up to half of her former husband’s earnings-related pension, starting in 2007. Recently, numerous popular press articles have reported that many Japanese wives are currently postponing divorce until 2007 in order to take advantage of this reform. Indeed, the upward trend in the crude divorce rate has recently leveled off in Japan. If the press reports are correct, the trend may resume its upward climb after 2007, especially if Japan’s economy continues to expand so that the proportion of currently married women employed in full-time jobs also resumes its long-term upward climb. Another factor that helps account for continuing marital fertility decline during the 1990s is the stagnating economy during that period. Micro-level data gathered in the 2000 round of the National Survey on Family Planning asked a question: “Has the recent growth of economic insecurity in Japan resulting from the bursting of the bubble economy and business-restructuring due to globalization affected your personal decision about when to have children and how many children to have?” The responses were: 11 percent said “largely affected,” 18 percent said “somewhat affected,” 35 percent said “not affected very much,” and 35 percent said “not affected at all”. A statistical analysis showed that, as expected, women in lower-income households were more affected (Ogawa, 2003). It remains to explain the recent rise in PPR(M−1) in Figure 3.4. This rise first became apparent in 1998, the year when the economy, which had been showing signs of recovery, slid back into recession. By 1998, Japan was eight years into
60 Naohiro Ogawa, Robert D. Retherford, and Rikiya Matsukura its “lost decade,” and many couples had been delaying marriage, waiting for the good times to return. When their hopes of an upturn were dashed in 1998, many of them gave up waiting any longer, got married, and had their first child soon thereafter. This also helps explain why mean age at marriage (MAM) leveled off and even declined slightly after 1998 in Figure 3.5. But times were tough, so many with lower incomes had to move in with parents. This is reflected in the reversal of the downward trend in coresidence with parents at the time of marriage that also occurred after 1998, as shown in Figure 3.12. One might have expected that newly married coresiding women would take advantage of built-in baby-sitting by a mother or mother-in-law and not quit their jobs when they married or when they had their first birth, but this did not happen, as shown in Figure 3.13. The proportion who quit when they got married continued to decline, as expected, but this decline was more than compensated by a large increase in the proportion who quit when they had their first birth. It is likely that at least part of the explanation of the rise in the proportion who quit when they had their first birth is that a large proportion of these women were working in part-time jobs and therefore were not eligible for childcare leave and had to quit under pressure from employers during a period of economic hard times. This interpretation is consistent with Yamada’s earlier-mentioned survey finding of greater delays in marriage among men with lower incomes, who, when they ultimately marry, are more likely to marry a woman who is working part-time. It is also consistent with the rise in the proportion of single women who work part-time, as shown in Figure 3.14. 70 60
Percent
50 40 30 20 10 0 1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
Year of marriage
Figure 3.12 Trends in the proportion of newly married couples who coresided with parents at the time of marriage: 1955–2002. Note Computed from pooled data from various rounds of the National Survey on Family Planning and the 2004 National Survey of Population, Families, and Generations.
Japan’s declining fertility and policy responses 61 10 90 80
Quit working at marriage or first birth
Proportion
70 60 50
Quit working at marriage
40 30 20
Quit working at first birth
10 0 1965
1970
1975
1980
1985
1990
1995
2000
Year of marriage
Figure 3.13 Among currently married women who have had at least one birth and who worked before marriage, trends in the proportions who quit when they got married, who quit when they had a first birth, and who quit either when they got married or when they had a first birth, Japan, 1965–2002. Note Computed from pooled data from various rounds of the National Survey on Family Planning and the 2004 National Survey of Population, Families, and Generations.
Also contributing to the rise in PPR(M−1) has been the rise in the proportion of newly married women who were pregnant when they got married, as shown in Figure 3.15. This is a long-term trend that did not accelerate after 1998, so it cannot by itself explain the upturn in PPR(M−1) after 1998. A related development, however, is that during the latter half of the 1990s the trend in mean age at marriage among those who were pregnant when they got married diverged from the trend in mean age at marriage among those who were not pregnant when they got married, as shown in Figure 3.16. We interpret this to mean that the couples who had delayed marriage during the last decade but finally gave up and got married after 1998 were older on average, and that the immediate impetus for them to get married was often a pregnancy, which, if accidental, would likely have been aborted had they been younger. Figure 3.16 suggests that this mechanism contributed not only to the increase in PPR(M−1) but also to the increase in coresidence after 1998 (Figure 3.12). Another mechanism that may have contributed to the increase in PPR(M−1) is the leveling off of the trend in mean age at marriage after 1998 (Figure 3.5), which apparently occurred because older unmarried couples finally gave up on waiting for the good times to return before getting married. The leveling off of the trend may have led to a temporary bunching of marriages (Bongaarts and Feeney, 1998), and if so, there may have been some bunching of first births as well. But this bunching effect (or tempo effect, as it is usually called) appears to have been small, because although PPR(M−1) increased, PPR(B−M) did not.
10 Part-time or full-time
9 8
Full-time
Percent
7 6 5 4 3 2
Part-time
1 0
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Year
Figure 3.14 Among single women not currently enrolled in school, trends in the proportions working part-time and full-time, 1990–2004. Note Computed from various rounds of the National Survey on Family Planning and the 2004 National Survey of Population, Families, and Generations. 40 35 30 25 20 15 10 5 0 1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
Year of marriage
Figure 3.15 Trends in proportion of women who were pregnant at the time of first marriage. Note Computed from pooled data from various rounds of the National Survey on Family Planning and the 2004 National Survey of Population, Families, and Generations.
Japan’s declining fertility and policy responses 63 30.0 29.0
Pregnant at marriage
28.0 27.0
Age
26.0 25.0 24.0 23.0 22.0 Not pregnant at marriage
21.0 20.0 1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
Year of marriage
Figure 3.16 Trends in mean age at marriage for those who were pregnant and those who were not at time of first marriage. Note Computed from pooled data from various rounds of the National Survey on Family Planning and the 2004 National Survey of Population, Families, and Generations.
Government efforts to raise fertility The Japanese government became actively concerned about low fertility in 1990, when it was announced that the TFR for 1989 was 1.57 births per woman, the lowest it had ever been. The media picked up the story and “1.57 shock” made headlines throughout Japan. In short order, also in 1990, the government established an interministry committee on “creating a sound environment for bearing and rearing children” and over the next 15 years introduced a series of measures aimed at raising fertility, as shown in Table 3.4. These measures have been described in some detail by Retherford and Ogawa (2006). What follows are a few highlights. The Japanese government has employed two approaches to raising fertility:
•• direct subsidies for marriage, child-bearing, and child-rearing •• changing the institutional environment to facilitate marriage, child-bearing, and child-rearing. The government introduced direct subsidies in the form of child allowances in 1972. The intent was not pro-natalist but rather to help low-income families who were being left behind by swift economic growth and the rapid social changes that went with it. After 1990, new pro-natalist concerns led to substantial increases in the allowances. Currently a couple receives an allowance of US$100 per month
64 Naohiro Ogawa, Robert D. Retherford, and Rikiya Matsukura Table 3.4 Major Japanese government actions aimed at raising fertility TFR
Year
Action
2.14 1.54
1972 1990
1.53 1.50 1.42 1.34 1.33
1991 1994 1995 1999 2001
1.32 1.29
2002 2003 2003
1.28 1.26
2004 2005
Establishment of child allowances (no pro-natalist intent at first) Establishment of inter-ministry committee on “Creating a sound environment for bearing and rearing children” Enactment of Childcare Leave Act Announcement of Angel Plan for 1995–1999 Enactment of Childcare and Family Care Leave Act Announcement of New Angel Plan for 2000–2004 Amendment to the Employment Insurance Law, specifying 40% of salary to be paid to regular full-time employees during childcare leave Announcement of Plus One plan Enactment of Next Generation law Enactment of law on “Basic Measures to Cope with a Declining Fertility Society” Announcement of New Angel Plan for 2005–2009 Extension of childcare leave to part-time workers, with some limitations
Sources: NIPSSR (2003) and other government sources.
per child under 3 years of age, regardless of the child’s birth order. In the case of children older than 3 years who have not yet finished elementary school (approximately age 12), the family receives US$50 per month for first and second children and US$100 per month for third and higher-order children. The eligibility criterion for a worker with, for example, a dependent spouse and two dependent children is that the worker’s taxable income must be less than US$57,400 if the worker is self-employed and US£64,600 if the worker is a salaried employee. There is some regional variability in these income thresholds, however. In 1991, the parliament passed the Childcare Leave Act. The intent of this law was to make it easier for working women to have children. The law provided up to one year of unpaid childcare leave for either the mother or the father for a child less than one year of age. Benefits were restricted to full-time employees in firms with more than 30 employees. Part-time workers were excluded. In 1994, the government announced its “Angel Plan” for 1995 to 1999. Again, the intent was to make it easier for working women to have children. The Angel Plan called for major expansion of the number of day-care centers throughout the country. It also called for expansion of after-school sports and other afterschool programs and for family-support centers. The family-support centers provided various additional services, such as picking up a child after school if the parents were working and could not do so. Eligibility criteria (i.e., income criteria) for these services varied by locality. Private-sector day-care services also expanded during this period to serve those not eligible for public-sector day-care services. The 1995 Childcare and Family Leave Act superseded the 1991 Childcare Leave Act. Under the 1995 law, full-time employees were again eligible for one
Japan’s declining fertility and policy responses 65 year of leave for a child under one year of age. Part-time workers were again excluded. (As already noted, about half of married women workers work parttime. About three-quarters of part-time workers are women.) An employee now received 25 percent salary during the leave, paid by the Employment Insurance Fund (originally established to pay unemployment benefits). The employee also accumulated seniority while on leave, and the employer had to continue paying the employer’s share of social security contributions, while the government paid the employee’s share. In 1999, the government announced a New Angel Plan for 2000 to 2004. The new plan called for further expansion of day-care centers, after-school programs, and family-support centers. It should be noted that government-provided daycare centers are heavily subsidized, especially in major urban areas. In Tokyo, the monthly cost incurred by government to fund day-care services for one infant currently exceeds the average monthly wage of a male worker in Tokyo, which is about US$4,500. In 2000, the government started paying the employer’s share as well as the employee’s share of social security contributions while the employee was on childcare leave. And in 2001, the Employment Insurance Law was amended to increase the proportion of salary received by an employee on childcare leave from 25 percent to 40 percent. In 2007, the proportion was increased again to 50 percent. In 2002, the government announced its “Plus One Plan”. Plus One means that the effort to raise marital fertility should be strengthened. The plan exhorted husbands to do more to help with child-rearing and housework in various ways. The plan also called for a “no queue” policy for day-care centers, implying further expansion of the number of day-care centers. In 2003, the parliament passed the “Next Generation” law, the intent of which was to implement the Plus One Plan. The law addressed the problem that many parents (mainly women) were not taking childcare leave because of social disapproval from co-workers and employers. The law therefore included measures aimed at creating an atmosphere within firms that would encourage parents to take the childcare leave to which they were entitled. To accomplish this, the law required employers with more than 300 employees to come up with a plan to raise fertility among their employees and to submit the plan to the government (i.e., to the prefectural Labour Bureau) by April 1 2005. Approval of the plan would allow the firm to obtain a stamp of approval in the form of a logo that could appear on its products and advertisements. Only 36 percent of employers submitted their plans on time, but the proportion with submitted plans increased rapidly to 97 percent by December 2005 and 100 percent by September 2006. So far, however, employers do not appear to be making much use of the logo, suggesting that they are not enthusiastic about the plans that they have been required to submit. The Next Generation law also requires local governments, as employers, to submit their own pro-natalist plans. In 2005 the government extended the right to childcare leave to part-time workers (i.e., persons employed for a fixed period of time), with some limitations
66 Naohiro Ogawa, Robert D. Retherford, and Rikiya Matsukura (Japan Institute of Workers’ Evolution, 2007). Part-time workers are eligible for children-care leave if both of the following criteria are met:
•• the worker has been employed by the employer for a continuous period of at least one year
•• the worker is likely to continue to be employed after the child reaches one year of age. Workers whose labor contracts expire during the year after the child reaches one year of age do not have the right to childcare leave when it is clear that the labor contract will not be renewed. A great many part-time workers are on three- or six-month renewable contracts, so they are not eligible for childcare leave unless the employer decides to renew the contract. Thus there is ample opportunity for employers to avoid granting childcare leave to part-time workers should employers wish to do so. As a substitute for the near-disappearance of arranged marriages, dating services have been playing an increasingly important role in the marriage market. So far, however, the government’s efforts to raise fertility have focused on measures aimed at increasing marital fertility rather than facilitating marriage, even though later marriage and less marriage account for about half of the decline in the TFR since 1973. All the big keiretsu (families of allied industries) already provide dating services for their employees. These services are contracted out to privatesector dating services firms, of which there are approximately 3,100 in Japan. It is likely that the plans submitted by the other employers covered by the Next Generation law often include subsidies for dating services, but so far we have no information on whether this is so. Despite all the government’s efforts to raise fertility, the TFR has continued to decline (Table 3.4). But without these measures, it probably would have declined even more. To understand why the TFR has not recovered, it is useful to begin by considering Japan in global context. To date, more than 30 countries around the world have experienced a TFR lower than 1.5 births per woman. Very few of these countries have been able to bring the TFR back up to 1.5, much less to the replacement level of 2.1. This lack of success has led Lutz et al. (2006) to hypothesize a “low-fertility trap”. Japan’s difficulties in reversing its TFR decline are not unusual. A possible reason why the Japanese government’s programs aimed at making it easier for women to both work and have children have not succeeded in reversing the downward trend in the TFR is that these programs have fertility reducing effects as well as fertility increasing effects. To see why this is so, it is useful to consider the effects of these programs on two groups of women:
•• those who work for pay outside the home •• those who do not work for pay outside the home. For simplicity, we shall refer to these two groups as workers and housewives.
Japan’s declining fertility and policy responses 67 The effect of the government’s pro-natalist programs is clearly to raise the fertility of workers. In the case of housewives, however, the effect is to pull some housewives into the labor force, thereby lowering their fertility inasmuch as working women have lower fertility than housewives. Which of these two effects predominates depends partly on:
•• the magnitude of pre-existing fertility differences between workers and housewives
•• how much the fertility of working women rises as a result of the government’s policies and programs
•• the magnitude of the probability that a housewife will be pulled into the labor force as a result of the government’s policies and programs
•• the relative numbers of workers and housewives to begin with. If almost all women are already working, the overall effect will be to raise the TFR. If almost all women are housewives, the overall effect will be to lower the TFR. In Japan, there is still a great deal of potential for housewives to be pulled into the labor force. Only about half of currently married women work, and among those who work, about half work full-time and half work part-time. Figure 3.17 shows that, as late as 2004 among women age 20 and over (regardless of marital status), the proportion agreeing with the statement, “The husband should be the breadwinner, and the wife should stay at home,” was still as high as 41 percent for all women and 35 percent for women age 20−29, leaving considerable room for further declines in these percentages. Corresponding proportions for men were about ten percentage points higher. It appears that most of the women who are currently not working prefer to be full-time housewives. Figure 3.18 shows additionally that, as late as 2003, the proportion of currently married women who gave birth in that year and took childcare leave was only about 20 percent. There is also considerable room for increase in this percentage. Figure 3.18 also shows that the proportion of women who were working before they gave birth rose somewhat after 2001 but by 2003 was still lower than it was in 1992. It also shows that the proportion of workers for whom childcare was not available has changed little since 1998. Another striking finding in this figure is that the 1991 Childcare Leave Act of 1991 and subsequent revisions of it seem to have had little effect on the proportion of women who take childcare leave. Indeed, the proportion taking childcare leave was only slightly higher after 1992 than it was between 1989 and 1992, when evidently many employers were already offering childcare leave before the government mandated them to do so. On the whole, Figure 3.18 indicates that the government’s efforts to make it easier for women to both work and have children have not had much effect on the proportion who work. If this is indeed true, then the effect of the government’s programs to make it easier for women to both work and rear children has probably been to raise fertility (i.e., to slow the decline in fertility). One must be careful in interpreting the trends in Figure 3.18, however, because the group of women considered in each year is increasingly truncated by age as
90 80 Men total
Percent
70 Men 20–29
Women total
60 50 40
Women 20–29 30 1972
1979
1992
1997
2002
2004
Year
Figure 3.17 Proportion who agree that husband should be breadwinner, wife should stay at home. Source: National Survey of Gender Equality and National Survey on Women, various rounds (from the published reports).
1.0
Proportion
0.8 0.6 0.4 0.2 0.0 1980
1983
1986
1989
1992
1995
1998
2001
Year of child-bearing Used
Did not use
Scheme not available
Not working
Figure 3.18 Trends in the proportion of married women of reproductive age who took advantage of childcare leave, by year of child’s birth, 1980–2003. Note Computed from data from the 2004 National Survey of Population, Families and Generations.
Japan’s declining fertility and policy responses 69 one goes back in time from the 2004 survey, which sampled women only up to age 50. This means, for example, that the maximum age of a woman was 36 in 1990 and only 26 in 1980. Another difficulty is that the figure pertains to all currently married women. It would also be of interest to graph the trend in the proportions taking childcare leave for currently married full-time workers and currently married part-time workers separately, but this is not possible using the 2004 survey. Another complication is that many women who take childcare leave take it more than once, so that the proportion of women who are covered is less than the proportion of their children who are covered, as shown by the following additional information from the 2004 survey on which Figure 3.18 is based. There were 1,604 first births in the birth histories collected in this survey – 15 percent of the mothers of these first births took childcare leave, and of these mothers, 55 percent also took childcare leave for a second birth by the time of the survey. These “repeaters” tended to have above-average education and to work full-time for either large corporations or government. The picture that emerges is that the women who take advantage of childcare leave are a rather select group. An apparent additional effect of more generous childcare-leave provisions since 2001 has been to decrease the proportion of currently married women who work full-time and increase the proportion who work part-time, as shown in Figure 3.19. This is contrary to the expectation that more generous childcareleave provisions, which in Japan until very recently have applied exclusively to full-time workers, would draw more women into full-time work. Increasingly generous childcare-leave provisions probably have caused many women who were already working full-time to continue on the job (while on childcare leave) instead of quitting shortly before childbirth. But it also appears from Figure 3.19 that the additional burden on employers has made employers more likely to hire new women workers as part-time workers instead of full-time workers. Many married part-time workers would prefer to have full-time jobs, and their fertility may be lower than what it would have been had they actually been able to get full-time jobs. Overall, the evidence, which is far from conclusive, suggests that the effect of the government’s efforts to make it easier for women to both work and have children (including not only increasingly generous childcare leave but also the major expansion of government-subsidized day-care facilities throughout the country) has been to slow the decline of marital fertility. Indeed, the TMFR has leveled off in recent years (Figure 3.1), with most of the continuing decline in the TFR a result of later marriage and less marriage.
Concluding remarks The danger of placing too much of the burden of pro-natalist programs on business is that it may reduce the competitiveness of Japanese firms in global markets and slow Japan’s economic recovery. Economic recovery, however, is essential for maintaining the government’s capacity to fund pro-natalist programs, which, if they are to be effective, are very expensive. Large tax deductions for children that
70 Naohiro Ogawa, Robert D. Retherford, and Rikiya Matsukura 40 35 Part-time
Proportion
30 25 Full-time
20 15 10 5 0 1979
1984
1989
1994
1999
2004
Year
Figure 3.19 Among currently married women below age 50, trends in the proportions working full-time and part-time.
spread the burden more evenly across the population might be a more economically efficient way to promote higher fertility, although some of the burden will inevitably fall directly on employers. As Retherford and Ogawa (2006) have noted, the big picture is that the government must:
•• restructure the economy so that Japan can restore the health of its economy and compete more effectively in the global economy
•• restructure society to make it more marriage-friendly and family-friendly in order to raise fertility. The trick is accomplishing the second task without jeopardizing the first, and without jeopardizing women’s hard-won gains in education and employment. As Japan’s experience since 1990 amply demonstrates, this will be both difficult and expensive.
Acknowledgements Research for this paper was funded by two grants from the National Institute of Health, NIA R01-AG025488 and AG025247. This work was also supported by a grant obtained by the Nihon University Population Research Institute from the “Academic Frontier” Project for Private Universities: matching fund subsidy from MEXT (Ministry of Education, Culture, Sports, Science and Technology), 2006–2010.
Japan’s declining fertility and policy responses 71
References AIU Insurance Company. 2001. Cost of Children. Tokyo: AIU Insurance Company (in Japanese). Allison, P. 1995. Survival Analysis Using SAS: A Practical Guide. Cary, NC: SAS Institute Inc. Bongaarts, J., and G. Feeney. 1998. “On the quantum and tempo of fertility,” Population and Development Review 24: 271−291. Cabinet Office. 2005. White Paper on the National Life Style. Tokyo: Government of Japan (in Japanese). Clark, R. and N. Ogawa. 1997. “Transition from career jobs to retirement in Japan,” Industrial Relations 36: 255−270. Feeney, G. 1986. “Period parity progression measures of fertility in Japan,” NUPRI Research Paper Series no. 35. Tokyo: Nihon University Population Research Institute. Feeney, G. and Y. Saito. 1985. “Progression to first marriage in Japan: 1870–1980,” NUPRI Research Paper Series no. 24. Tokyo: Nihon University Population Research Institute. Japan Institute of Workers’ Evolution. 2007. Law concerning the welfare of workers who take care of children or other family members, including child care and family care leave. Downloaded from the Institute’s website on February 16, 2007: http://www.jiwe. or.jp/english/law/law4_1_2.html. Japanese Association for Sex Education. 2001. Sexual Behavior of Youth. Tokyo: Shogakukan (in Japanese). Kohler, H.P., F.C. Billari, and J.A. Ortega. 2002. “The emergence of lowest-low fertility in Europe during the 1990s,” Population and Development Review 28: 641−680. Lutz, W., V. Skirbekk, and M. Testa. 2006. “The low fertility trap hypothesis: forces that may lead to further postponement and fewer births in Europe,” in D. Philipov, A.C. Liefboer, and F.C. Billari (eds), Vienna Yearbook of Population Research. Vienna: Austrian Academy of Sciences Press. McDonald, P. 2005. “Low fertility in Singapore: causes, consequences and policies,” paper presented at the Forum on Population and Development in East Asia, Beijing, May 16−17. Mason, A. and N. Ogawa. 2001. “Population, labor force, saving and Japan’s future,” in M. Bloomstrom, B. Gangnes, and S. Lacroix (eds), Japan’s New Economy: Continuity and Change in the Twenty First Century. Oxford: Oxford University Press. Ministry of Health, Labour and Welfare. Various years. Survey on Employment Management. Tokyo: Government of Japan (in Japanese). NIPSSR (National Institute of Population and Social Security Research). 2003. Child Related Policies in Japan. Tokyo: NIPSSR. Ogawa, N. 2003. “Japan’s changing fertility mechanisms and its policy responses,” Journal of Population Research 20: 89−106. —— 2005. “Population aging and policy options for a sustainable future: the case of Japan,” Genus 61: 369−410. Ogawa, N. and J. Ermisch. 1994. “Women’s career development and divorce risk in Japan,” Labour 8: 193−219. Ogawa, N. and R.D. Retherford. 1997. “Shifting costs of caring for the elderly back to families in Japan: will it work?,” Population and Development Review 23: 59–94. Retherford, R.D. and N. Ogawa. 2006. “Japan’s baby bust: causes, implications, and policy responses,” in F.R. Harris (ed.), The Baby Bust: Who Will Do the Work? Who Will Pay the
72 Naohiro Ogawa, Robert D. Retherford, and Rikiya Matsukura Taxes? Lanham: Rowman & Littlefield Publishers: pp.5–44. A pre-publication version is downloadable at http://www.eastwestcenter.org/stored/pdfs/POPwp118.pdf. Retherford, R.D., N. Ogawa, and R. Matsukura. 2001. “Late marriage and less marriage in Japan,” Population and Development Review 27: 65–102. Retherford, R.D., Naohiro Ogawa, and Satomi Sakamoto. 1996. “Values and fertility change in Japan,” Population Studies 50: 5–25. Statistics Bureau. Various years. Annual Report of the Labour Force Survey. Tokyo: Statistics Bureau, Government of Japan. Tsuya, Noriko O., and Larry L. Bumpass. 2004. Marriage, Work and Family Life in Comparative Perspective: Japan, South Korea and the United States. Honolulu, HI: University of Hawaii Press. United Nations. 1995. Women’s Education and Fertility Behaviour: Recent Evidence from the Demographic and Health Surveys. Department of Economic and Social Affairs, Population Division. New York, NY: United Nations. United Nations. 2005. World Population Projection: the 2004 Revision. New York, NY: United Nations. U.S. Census Bureau. 2006. World POPclock Projection. Downloaded from website on December 14, 2006: http://www.census.gov/ipc/www/popclockworld.html. Van de Kaa, D. 2006. “Temporarily new: on low fertility and the prospect of pro-natal policies,” revised version of a paper presented at the conference on the Postponement of Childbearing in Europe, Vienna Institute of Demography, December 1−3, 2005. Yamada, M. 1996. The Sociology of Marriage: Will Marriage Continue to Be Delayed? Tokyo: Maruzen Library (in Japanese). —— 2001. “Parasite singles feed on family system,” Japan Quarterly, March−April: 10−16. —— 2004. “Consciousness of Risk Involved in Childrearing in the Context of Lifetime Planning Among Young Adults,” report submitted to Ministry of Health, Labour, and Welfare (in Japanese). Downloadable at http:/mhlw-grants.niph.go.jp/niph/search/ NISR00.do. —— 2006. “The real story behind Japan’s marriage crisis,” Japan Echo, February: 20−24.
4
The arrival of low fertility in China Baochang Gu
Introduction The demographic profile of China has often been seen as that of the most populous country in the world, with traditionally high fertility, rapidly growing, mainly rural population, and a young age structure. But suddenly it is found that such a description is no longer true. Fertility has dropped to below replacement level, and the population is rapidly aging, sex unbalance among the new-born has been an issue for more than two decades, and the volume of internal migration has become the largest in world history during peace time. China will soon lose its position as the world’s most populous country to India. Demographic transformation in China has been so swift that it is necessary to update our understanding of population dynamics in China, particularly with the arrival of low fertility. This chapter will first provide a review of fertility trends and fertility policy in China followed by an explanation of low fertility based on the evidence from an on-going survey in Jiangsu province. The impact of low fertility with respect to population aging, gender equality, sex ratio at birth and labor supply will then be presented, and the chapter will end with a discussion suggesting that more work needs to be done to understand the population dynamics along with the arrival of low fertility in China.
Low fertility and population growth China’s population policy and program have, for decades, been focused almost exclusively on demographic outcomes given the overriding concern of the government and society alike with the impact of the excessive growth of the population on socioeconomic development. Since the 1990 census reported a TFR of 2.31, the National Bureau of Statistics of China (NBS) has been reporting that China’s fertility has fallen to a level below replacement, and the 2000 census even reported a TFR way below replacement (1.4). Despite the historical transition of reproduction from a typical high-fertility to a typical low-fertility pattern in half a century (Chen and Guo 2006), it was still argued that “while its fertility has declined to below replacement at the level of 1.8, the annual births remain at 20 million and population increment at 10 million. … The demographic profile of China is
74 Baochang Gu characterized by low fertility level with rapid population growth” (Gu 2003). It was even predicted that the number of annual births would remain at 20 million and population increment at 10 million. Nevertheless, this assertion has been questioned in recent years. Table 4.1 is compiled from data published by the NBS with results from the annual population- change sample surveys and the 2005 1 percent national-population sample survey. It shows that while the total population has continued to increase since 2000, the annual growth of the population has been decreasing, from more than 9.5 million in 2000 to less than 7 million in 2006. The declining trend is quite remarkable. The number of deaths has remained virtually unchanged at an annual level of more than 8 million, while the crude birth rate and the number of births have both been decreasing over the years. The birth rate has declined from 14 per thousand in 2000 to 12.1 per thousand in 2006, and the number of births has declined from close to 18 million to less than 16 million, a decrement of about 2 million in 7 years. As a result, the population growth rate has declined to below 0.6 percent. In fact all the 31 provinces of the country have a population growth rate below 1 percent except for the three remote provinces of Tibet (1.12%), Ningxia (1.12%), and Xinjiang (1.09%). It is therefore no longer tenable for China to claim to be a country with “rapid population growth”. The slowing trend of population growth in China can also be seen from the predictions for the timing of the arrival of peak population and the population size at the peak. A decade ago, it was widely accepted that China would not achieve zero population growth until the mid-twenty-first century with a peak population of 1.6 billion; by the turn of the century, the prediction was brought forward to the 2040s with a population of 1.5 billion. The growing consensus has recently switched to the early 2030s with a population of less than 1.5 billion. The most updated projection from the United Nations suggests that China’s population will reach a peak in 2030 with a population of 1.458 billion, while India will have a population of 1.506 billion, surpassing China as the most populous country in the world (United Nations 2007). Table 4.1 Population changes in China, 2000–2006 Year Year-end Total Crude Birth Births Crude Deaths Natural Population Population Rate (‰) (million) Death (million) Growth Growth (million) Rate (‰) Rate (%) (million) 2000 2001 2002 2003 2004 2005 2006
1267.43 1276.27 1284.53 1292.27 1299.88 1307.56 1314.48
14.03 13.38 12.86 12.41 12.29 12.40 12.09
Sources: NBS 2005, 2006, 2007.
17.71 17.02 16.47 15.99 15.93 16.18 15.84
6.45 6.43 6.41 6.40 6.42 6.51 6.81
8.14 8.18 8.21 8.25 8.32 8.49 8.92
0.758 0.695 0.645 0.601 0.587 0.589 0.528
9.57 8.84 8.26 7.74 7.61 7.69 6.92
The arrival of low fertility in China 75 China has not always been a country with rapid population growth. Over its long history, the population grew quite slowly due to the fact that the high birth rate was compensated for by a high death rate. Rapid population growth was not seen until the seventeeth century when mortality started to decline while fertility remained at a high level (Poston et al. 2005). This pattern is consistent with what is suggested by the demographic transition theory. Rapid population growth is a one-time phenomenon over the course of human history, and is now coming to an end after dominating China’s population dynamics for several hundred years. The total fertility rate (TFR) has commonly been used as an indicator of the fertility level of a given population. China’s measurement of its fertility was claimed to be “of very high quality” in the early 1980s (Coale 1984), but has become a controversial issue over the years, particularly since the mid-1990s. It was anticipated that the results of the 2000 population census would provide an answer to the debate. Surprisingly, the 2000 census initially reported a fertility level of only 1.22.1 This result has been widely considered “unacceptable,” and even the NBS sees it as “too low”. More debates have arisen on China’s fertility level in recent years, and the estimation ranges from as low as 1.35 to as high as 2.3 (Chen and Guo 2006). Several studies employing various methods have argued that China’s TFR was “between 1.5 and 1.6” in the year 2000 (Guo 2004; Cai 2005; Retherford et al. 2005; Zhang and Zhao 2006), while the official figure is insisted to be “about 1.7 to 1.8” (Chen and Guo 2006). Despite the ongoing debate, the NBS continues to publish the detailed fertility information by age and parity every year based on the results from the annual population-change sample surveys. The age-specific fertility rates and total fertility rate in China for 2000–2005 are shown in Table 4.2, which reveals a quite consistent trend with China’s fertility at the level of around 1.4 in the early twenty-first century. Whatever the method and data source used to determine China’s current fertility level, one thing is hardly refutable: China’s fertility has truly fallen below the replacement level of 2.1. We should be able to conclude that China’s fertility transition has definitely reached the stage of low fertility (see Figure 4.1). Table 4.2 Age-specific fertility rates and total fertility rate in China, 2000–2005 Age-specific Fertility Rate (‰)
Year 2000
2001
2002
2003
2004
2005
15–19 20–24 25–29 30–34 35–39 40–44 45–49
5.96 114.49 86.19 28.62 6.22 1.46 0.68
2.70 107.70 115.37 40.06 9.22 1.83 0.60
2.68 113.15 106.09 42.68 9.68 1.88 0.37
5.25 122.67 102.44 38.28 8.65 1.77 0.56
5.56 120.85 107.60 42.21 10.14 1.93 0.41
6.34 114.46 91.70 40.22 10.98 2.05 0.77
1.22
1.39
1.38
1.40
1.44
1.33
TFR
Sources: 2000: 2000 population census, 2001–2005: annual population-change sample surveys.
76 Baochang Gu 8.00 7.00 6.00 5.00
TFR
4.00 3.00 2.00 1.00 0.00 1950
1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 Year
Figure 4.1 Total fertility rate in China, 1950–2005. Source: NBS, 2005.
Low fertility and fertility policy2 To understand the appearance of low fertility in China we need to discuss the fertility policy implemented in the family planning program. It is well known that China’s fertility decline and its population control have relied heavily on a draconian fertility policy and a strong family planning program. As a result, China’s national fertility policy has often been known mostly as a “one-child policy”. Yet it is little understood that China’s fertility policy, while a national priority for over two decades, has evolved to contain highly localized features. Following modifications over two decades to the initial one-child policy, what is the desired or expected fertility level as implied by these policies for each of China’s provinces and for China as a whole? To quantify fertility policies in various localities and populations across the country, the term policy fertility is coined as a quantitative indicator summarizing the fertility level implied by the fertility policies implemented in a given region. Fertility policy in China can be grouped into four categories: 1 2 3 4
1-child policy (1 child per couple) 1.5-children policy (those whose first child is a girl may have a second child) 2-children policy (2 children per couple) 3-children policy (3 children per couple).
Map 4.1 highlights the geographic diversity of fertility policy measured by policy fertility for China’s prefectures, an administrative level between county and province. Tibet is not included in the calculation and mapping due to lack of data on fertility policy at the sub-provincial level. For the sake of simplicity, policy fertility is grouped into four categories:
Tibet
Jinchang City
no data
1.3–1.5
Panzhihua City
Lanzhou City
Baiyin City
Shenzhen City
Wuhai City
Langfang City
Beijing City Shenyang City Fushun City
Changde, Hunan
Wuhan City
Shanghai City
Taiyuan City
Tianjin City
Guangzhou City
Taiwan
Map 4.1 Geographic distribution of policy fertility, prefecture level, China, late 1990s
1.5–2.0
2.0–3.5
1.0–1.3
Chongqing City
Ganze Zang Liangshan Yi Autonomous Prefecture, Autonomous Prefecture, Sichuan Sichuan
Aba Zang-Qiang Autonomous Prefecture, Sichuan
Haixi Mongol-Zang Autonomous Prefecture, Qinghai
Urumqi City
Jiayuguan City
Da Hinggan Ling Prefecture, Heilongjiang
78 Baochang Gu 1 2 3 4
1.0–1.3, corresponding with regions with a predominantly 1-child policy 1.3–1.5, corresponding to areas with a mixture of 1-child and 1.5-children policies 1.5–2.0, corresponding to areas with 1.5- and 2-children policies 2 or higher, for areas with a 2- or more-children policy.
The four shades of the map represent different levels of policy fertility, from the lowest (1.0–<1.3) to the highest (2.0–3.5). It is evident from this map that the areas with the most stringent fertility policy are mainly the municipalities directly under the jurisdiction of the central government, the provincial capital cities, as well as the two provinces of Jiangsu and Sichuan since they have a province-wide “one child per couple” policy, even for the rural population. Most prefectures with a policy fertility of 1.3–<1.5 are located in East and Central China, whereas most prefectures with the highest policy fertility (above 2.0) are distributed in the Central and Western regions of the country.3 These are mostly areas with a high concentration of minority populations. The population distribution by policy fertility for China as whole is presented in Table 4.3. Slightly above a third of the population (35.4 percent) fall into the 1-child policy category. Over half of the national population (53.6 percent) fall into the 1.5-children policy category. Combined, these 2 categories encompass nearly 90 percent of China’s national population, who are subject to a belowreplacement fertility policy. Only about 10 percent of the population (9.7 percent) fall into the 2-child category, and the percentage of population subject to a 3-child policy is even more trivial, only 1 percent. What proportion of China’s population could have only one child, if they followed fully the policies as summarized above? Taking the 1-child couples in the 1.5-children policy areas into account, 63 percent of all couples in China could end up with only 1 child, 36 percent with 2 children, and only 1 percent with 3 or more. The estimated policy fertility levels for China’s provinces are presented in Table 4.4. Next to the policy fertility level for each province is the observed total fertility level as reported in China’s 2000 census (National Bureau of Statistics 2003). Table 4.4 shows that the average policy fertility among Chinese provinces varies widely, from as low as barely above 1 (1.06) in Shanghai and Jiangsu, to as high as well above 2 (2.37) in Xinjiang. China’s provinces can therefore be grouped into four categories by their fertility policy. There are 6 provinces in category 1 with Table 4.3 Demographic distribution of fertility policy, China, late 1990s Policy 1-child 1.5-children 2-children 3-children Total
Population (million)
Percentage
439.16 664.94 120.33 16.13
35.4 53.6 9.7 1.3
1,240.56
100.0
The arrival of low fertility in China 79 Table 4.4 Policy fertility and observed fertility of China’s provinces, c. 2000 Province Between 1.0 and 1.3 Shanghai Jiangsu Beijing Tianjin Sichuan Chongqing Between 1.3 and 1.5 Liaoning Heilongjiang Guangdong Jilin Shandong Jiangxi Hubei Zhejiang Hunan Anhui Fujian Shanxi Between 1.5 and 2.0 Henan Shaanxi Guangxi Gansu Hebei Inner-Mongolia Guizhou 2.0 and above Yunnan Qinghai Ningxia Hainan Xinjiang
Policy Fertility
Observed Fertility
Difference (P – O)
1.06 1.06 1.09 1.17 1.19 1.27
0.7 1.0 0.7 0.9 1.4 1.5
0.36 0.06 0.39 0.27 –0.21 –0.23
1.38 1.39 1.41 1.45 1.45 1.46 1.47 1.47 1.48 1.48 1.48 1.49
1.1 1.0 1.1 1.0 1.3 2.0 1.1 1.2 1.5 1.5 1.1 1.7
0.28 0.39 0.31 0.45 0.15 –0.54 0.37 0.27 –0.02 –0.02 0.38 –0.21
1.51 1.51 1.53 1.56 1.59 1.60 1.67
1.7 1.3 1.8 1.3 1.5 1.2 2.4
–0.19 0.21 –0.27 0.26 0.09 0.40 –0.73
2.01 2.10 2.12 2.14 2.37
2.0 1.7 1.8 1.8 1.7
0.01 0.40 0.32 0.34 0.67
Source: Observed fertility is from the 2000 population census (NBS 2003), in which the national total fertility rate was re-reported at 1.4.
a policy fertility level less than 1.3. There are 12 provinces in category 2, with policy fertility between 1.3 and less than 1.5. There are 7 provinces in category 3 with policy fertility above 1.5 and below 2. Finally, 5 provinces are in category 4, where policy fertility is above 2, among which all but 1 (Hainan) are in the Western region of China. Moreover, at the provincial level, policy-required and census-observed fertility levels are very close to each other. As seen in Table 4.4, whereas only 6 of China’s 30 mainland provinces (excluding Tibet) have a policy fertility level at 1.3 or lower, 12 provinces in the 2000 census reported a fertility level of 1.2 or lower. The differences between the policy-required fertility and the census-observed fertility for the 30 provinces are generally quite marginal.
80 Baochang Gu Relatively large differences (above 0.5 children) are found in only 3 provinces (Jiangxi, Guizhou, and Xingjian). Sub-nationally, distribution of fertility policy follows a gradient of economic development level, from the east to the west by these three broad economic development regions. Map 4.2 marks the average level of policy fertility for the three regions and depicts this policy gradient by the regional classification. The economically most developed Eastern region has the largest shares of prefectures and population, and also the highest share of the population falling under the requirement of the 1-child rule (69.6 percent). The overall policy fertility level for this region is only 1.39. The economically least developed Western region has the second largest share of prefectures (36.7 percent), but the smallest share of the national population (28.8 percent). Overall the policy fertility requirement for the region is also the most lenient, at the level of 1.56, allowing over 40 percent of the population to have 2 or more children. The policy targeted fertility for the Central region is in between at the level of 1.47. From the above discussion, it can be seen that China’s fertility policy encompasses much variation, both geographically and demographically. Nevertheless, the majority of the Chinese population (more than 70 percent) live in areas with a policy fertility level of 1.3 to 2.0 children per couple. Meanwhile, the one-child policy remains a core element of China’s fertility policy and inevitably exerts an enormous impact on China’s demographic processes. Should all couples in various policy regions follow the current fertility policies fully, more than 60 percent of all Chinese couples would still end up with only 1 child. Based on local fertility policies and corresponding population distributions, the overall average fertility targeted by the current fertility policies for China as a whole was estimated to be 1.47 at the end of the 1990s, and this remains valid today. This is a level far below the replacement level, but very close to the level reported by the National Bureau of Statistics over the years.
Low fertility and child-bearing desire What are the underlying forces of emerging low fertility in China? China’s transition from high to low and now to below-replacement fertility has often been seen as differing in fundamental ways from that in other societies, mainly because of China’s ongoing strong government intervention in and control of reproduction (Greenhalgh and Winckler 2005; Lee and Wang 1999). Can we thus conclude from the above discussion that the arrival of low fertility in China is purely or directly driven by the fertility policy implemented? If not, what roles do other factors play? And how does China differ from other countries? Gu and his colleagues (Gu et al. 2007b) argue that the low-fertility regime emerging in China and elsewhere cannot be understood without reference to the new global political economic context that has profound implications for human reproduction. China is different, but it is not a mythical “other” in the emerging global regime of below-replacement fertility. To examine the separate roles of economic pressure and value changes that are both parts of the new global
Qinghai
Shanxi
Hebei
Hainan
Zhejiang Fujian
Hongkong
Guangdong
Hunan
Guangxi
Guizhou
Jiangxi
Anhui
Shanghai
Tianjin
Beijing
Liaoning
Shandong Jiangsu Shaanxi Henan
Ningxia
Hubei Sichuan Chongying
Yunnan
Gansu
Jilin
Map 4.2 Policy-required fertility level by broad economic regions, China, late 1990s
Tibet
Xinjiang
Inner-Monglia
Heilongjiang
1.39
1.47
1.39
82 Baochang Gu economic and cultural environment, and the role of birth-control policy, which is unique for China, an empirical survey of fertility preference and fertility desire/ intention was recently designed and carried out in the Jiangsu province of China. Jiangsu province is located in the Lower Yangtze region of China, along China's east coast. With a population of 75.5 million (end of 2006), Jiangsu is one of China’s largest and most economically dynamic provinces. Geographically the eastern part of Jiangsu province surrounds China’s largest city, Shanghai. In 2006, the per capita GDP for Jiangsu province ($3,500) ranked the third highest among China’s 28 mainland provinces (excluding the 3 provincial-level municipalities of Beijing, Shanghai, and Tianjin). Jiangsu was chosen as the survey site not only for its economy but also for its demography. Jiangsu is one of only 2 provinces (along with Sichuan province) that have implemented a province-wide 1-child policy for the past 2.5 decades, requiring not only urban but also rural couples to have only 1 child (Gu et al. 2007a). Its recorded fertility level of 1.0, according to China’s 2000 census, is also among the very lowest. However, Jiangsu province’s strict birth-control policy also contains a benign element for couples in this province. As a measure of policy transition to phase out the 1-child policy after one generation, provincial birthcontrol regulations allow newly married couples in rural areas to have 2 children if one of the marriage partners is a single child, and in urban areas if both marriage partners are single children.4 After 2.5 decades of the 1-child policy, couples qualifying to have 2 children are now entering their marriage and child-bearing age. These couples are the main study targets. The data collection took place in late 2006 in 6 selected counties along the east coast of Jiangsu province, with 2 counties each representing different economic levels in this part of Jiangsu (south, middle, and north, with the northern part being the least developed among the 3). This survey was the first wave of a longitudinal study planned for this research. A multi-stage, random cluster process was used to draw a sample in each of the counties. The primary sampling unit was village in rural areas and neighbourhood in urban areas. A weighted sampling scheme was used to select randomly primary sampling units, and every woman aged 18 to 40 in the selected units was interviewed. A preliminary analysis with data collected from 2 of the 6 counties was carried out, with a sample of 5,821 women aged 18 to 40 from 17 communities. Among them, 4,741 (81 percent) were married, and 1,492 (26 percent) were qualified to have 2 children according to the local fertility policy. The survey contained questions about reproductive preferences and desire/intentions, including questions on general preferences for child-bearing and specific questions about reproductive desire/intention in the absence of the one-child policy. To some respondents, these questions were hypothetical. To others, they were real, as some of the young respondents were only children themselves, who were allowed under the existing policy to have a second child if they wished. Occupationally, at the turn of the twentieth century the study population was anything but rural. Among female respondents aged 18 to 40, participation in the labor force was very high, with over 80 percent working at the time of the survey.5
The arrival of low fertility in China 83 Nevertheless, only a small share of them (18.1 percent) was in the agricultural sector, whereas the largest share (about 50.3 percent) was working in the manufacturing sector. The share of these women working in the low-skill service sector (20.8 percent) also exceeded that in the agricultural sector. The other 10.7 percent were working in professional white-collar jobs. Economic transformation in this region has also involved a high level of geographic mobility. Among female respondents, nearly 40 percent either had an out-migration experience to another county or city seeking employment or education (31.6 percent) or was an in-migrant (8.0 percent). Among the husbands of married women, out-migration was even more prevalent – close to half (46.9 percent) either worked or went to school in other counties or cities – and 3.4 percent were in-migrants. For many of the young women in the survey area, marriage and child-bearing were no longer assumed life-course obligations dictated by parents or the larger society. As shown in Table 4.5, whereas the majority of the respondents still assigned a very high value to having children in their lives, with over 90 percent agreeing with the statement that “watching children growing up is the greatest happiness in life” (74.9 percent + 18.3 percent) and nearly 70 percent agreeing with the statement that “those with no children live an empty life” (44.2 percent + 24.9 percent), the majority of respondents also disagreed with statements such as “the purpose of getting married is to have children” (41.3 percent + 13.1 percent), “people have children because everyone else does so” (44.0 percent + 18.6 percent) or “children’s education is a great investment for parents” (36.3 percent + 17.8 percent). To examine desired fertility among the women of child-bearing age, it was asked in the survey: “what is the ideal number of children if not considering the factor of birth-control policy?” As shown in the left-hand column of Table 4.6, almost all survey respondents (99.4 percent) reported either 1 or 2 as the ideal number of children. Moreover, more respondents chose 1 child over 2 (56.3 percent vs. 43.1 percent). Preference for 1 child over 2 does not vary by the respondent’s educational attainment level, but it does vary by age, with more younger respondents choosing 1 child over 2 than older ones, and by their household economic conditions, measured here as “affordability,” which is the ratio between their household income and the median yearly educational cost for primary school children aggregated from respondents in a community. There is almost a linear negative relationship between economic constraints and preference for one child, with those at the lower end of the economic distribution more likely to choose one child. To make the question more realistic rather than hypothetical, the respondents were divided into two groups, those who are entitled under the current policy to have only one child (PQ1), and those who are entitled to have two (PQ2). Among the 4,741 currently married women in the survey sample, about a third (33.9 percent) were qualified to have a second child and the rest (66.1 percent) were only allowed to have one according to the local fertility policy. As shown in the middle columns of Table 4.6, more than half (53.0 percent) of all respondents who were qualified to have two children still report one child as the ideal. The age
84 Baochang Gu Table 4.5 Respondents’ answers to attitude questions on family and having children, Jiangsu survey, end 2006 Question
Totally Agree (%)
Somewhat Not Sure Somewhat Agree Disagree (%) (%) (%)
Totally Disagree (%)
Having children affects women’s career The purpose of getting married is to have children People having children because everyone else does so Only children could provide old-age care and companionship Watching children growing up is the greatest happiness in life Pregnancy and giving birth are a lot of trouble Children’s education is a great investment for parents An ideal family should have at least one boy Those with no children live an empty life Those who could not afford should not have children Having children would affect (negatively) marriage quality Only families with children are happy families
10.9
33.1
2.9
35.8
17.4
19.6
23.9
2.1
41.3
13.1
12.5
22.6
2.3
44.0
18.6
29.4
32.2
1.7
29.7
7.1
74.9
18.3
1.3
4.6
0.9
16.7
28.8
5.5
34.6
16.4
20.8
23.1
2.0
36.3
17.8
10.8
13.2
2.2
41.7
32.1
44.2
24.9
3.4
19.4
8.1
9.6
19.5
2.4
39.5
29.1
2.6
9.6
3.4
33.7
50.8
81.8
12.1
1.7
3.5
0.9
Note All percentage statistics are properly weighted by sampling probability.
pattern is similar to that for all survey respondents, namely younger respondents are more likely to choose one child over two. A stronger indicator of reproductive behaviour is fertility desire/intention, namely whether a respondent herself will consider having another child, if policy allows. Two patterns are evident, as shown in the right-hand column of Table 4.6. First, among them fertility desire is very low, regardless of their policy entitlement. Only 21.5 percent of those who were entitled to have only 1 child wish to have a second one. Even among those who were allowed to have 2 children, less than 40 percent intended to have another child. Second, policy does make a difference. The percentage of respondents desiring to have a second child is clearly higher among those who were qualified to do so under the current policy than among those who were not, by a margin of nearly two to one. Why did young women or couples in this part of China not want to have a second child, even when they were allowed to? Or in other words, what factors would they take into consideration for having another child? Answers to this question are
The arrival of low fertility in China 85 Table 4.6 Reproductive ideal and reproductive desire, Jiangsu survey, end 2006 Ideal Number of Children (%)
Want to Have Two Children (% saying yes)
All Respondents
PQ2
PQ1
1
1
2
PQ2
2
Age Group 18–20 21–25 26–30 31–35 35–40 Education Primary School or Less Secondary School High School College or Above Affordability <=5 5–10 10–15 15–20 >20 Total (%) Total (N)
63.8 65.4 55.9 55.0 51.0
35.7 34.6 43.2 44.7 48.1
71.1 63.0 53.5 49.4 41.3
28.9 37.0 45.6 49.7 56.8
25.1 25.7 23.4 23.7 19.3
— 35.3 34.3 35.9 48.2
55.3 55.5 58.9 57.8
43.8 44.1 39.8 42.2
53.1 52.6 57.2 45.1
46.1 46.7 41.0 54.9
20.1 20.5 23.5 30.0
43.8 35.2 36.4 51.0
65.3 59.2 56.1 52.1 51.3
33.5 40.5 42.9 47.1 48.6
45.7 56.4 55.3 48.6 50.0
49.4 43.0 43.8 49.9 50.0
23.9 19.1 18.9 23.0 27.8
50.7 33.1 35.0 39.1 43.4
56.3
43.1
53.0
46.1
21.5
37.4
3301
2495
787
696
3249
1492
Notes 1 The numbers of cases (N) reported here are not adjusted, but all summary statistics are properly weighted by sampling probability. 2 Percentages in each row of “Ideal Number of Children” do not always add up to 100 because of missing cases and omitted categories (few people choose no child or more than two children).
shown in Table 4.7. Economic condition emerges as the most common concern; altogether 60 percent chose it as a “main” or “determinant” concern, followed by policy (35.7 percent), and childcare (18.4 percent). The answer of “personal happiness,” a measure of self-fulfilment value orientation, was chosen by only about 10 percent of the sample as a major reason. A more rigorous empirical test of the separate roles of economic constraints, value orientation, and policy restrictions on fertility intention was also carried out by multivariate logistic regression (Gu et al. 2007b). Three categories of variables with respect to the macro structure, individual-level socioeconomic status, and culturally related factors are included in the multivariate analysis. The results, by and large, further confirm the finding discussed above. In particular, among women who are qualified to have two children under the current policy, those with the highest educational attainment, at the two ends of the income distribution, or with only a daughter are more likely to plan to have a second child, whereas those with
86 Baochang Gu Table 4.7 Factors to consider for having a second child, women aged 18–40 who are qualified to have two children, Jiangsu survey, end 2006
Economic Situation Career/Education Housing Health Marriage Personal Happiness Other People’s Opinion Childcare Old-age Support One-child Benefit Child’s Sex Child’s Companionship Birth-planning Policy
No (%)
Maybe (%)
Mainly (%)
Determinant (%)
14.2 35.8 48.8 43.1 52.4 60.1 78.2 47.6 51.5 55.8 66.9 51.9 35.4
25.8 41.1 31.9 33.9 27.3 30.2 17.8 34.0 35.5 30.4 23.5 37.8 28.9
55.0 21.6 18.1 21.3 18.6 9.0 3.5 17.7 11.9 13.6 9.1 10.0 31.1
5.0 1.5 1.2 1.8 1.8 0.7 0.5 0.7 1.2 0.2 0.5 0.3 4.6
Note All percentage statistics are properly weighted by sampling probability.
a higher status within the household are less likely to do so. Economic conditions remain the most important factors in considerations for having more children. The preliminary results from the Jiangsu survey as the first systematic study of below-replacement fertility in China may suggest that China’s continuing strict birth-control policy still plays a significant role, but it is a role that is much less crucial compared to concerns with economic constraints. Even for those who are allowed to have two children, most (53 percent) would consider one child as ideal, and only 37.4 percent of them reported to plan for having a second child. This suggests that though fertility policy matters in understanding China’s low fertility, it is not the whole story. Economic constraints such as higher expectation of children’s education and child-rearing cost, and changing value orientation, such as placing more emphasis on child development than on the number of children, appear to weaken people’s desire to have children, and may even lead to a fertility level lower than desired, as observed in some other Asian countries (Bongaarts 2001; Choe et al. 2004).
Low fertility and demographic impact The arrival of low fertility will certainly have a tremendous impact on China’s society. Along with the fertility decline and the improvement in life expectancy, population aging will naturally follow as a demographic consequence. When fertility declines more rapidly, aging will come about more rapidly as well. Aging has received increased attention in China mainly due to the release of the results of the 2000 census which reported that China’s population aged 60 and above was more than 10 percent and aged 65 and above was more than 7 percent, a definite
The arrival of low fertility in China 87 indication of China becoming an aging society. More recent data from the 2005 1 percent national-population sample survey report indicates that China’s population aged 60 and above accounts for 11 percent and aged 65 and above, 8 percent. More important than proportion is the number of elderly people, which implies a population aged 60 and above of about 144 million and age 65 and above of about 100 million. “Unlike the case in the developed countries however, population aging occurs in China far before the realization of its modernization” (Gu and Peng 1992). Support of the elderly is an acute social issue to be addressed, particularly in China’s rural areas. Traditionally, rural people tend to rely heavily on their children for elderly support, but with the sharp reduction in number of children, the practice is becoming no longer feasible. At the same time, the coverage of the social security program for the rural population will be limited in the near future. Moreover, because of accelerated population movement triggered by the marketoriented economy, more young people tend to leave the village for cities. As a result, while rural fertility is higher than urban fertility, the degree of population aging turns out to be higher in rural areas than in urban areas. Compared to 1982, in 2000 the proportion of elderly aged 65 and above in urban areas has increased from 4.5 percent to 6.4 percent, while in rural areas it increased from 5.0 percent to 7.5 percent (Gu 2006). By the mid-twenty-first century, more than a third of China’s population is likely to be aged 60 and above, and more than 100 million aged 80 and above. Half of the population will be aged above 45, and thus there will be fewer than two young adults to support one old person (United Nations 2007). China will become an aging society older than the oldest country in the world today. Figures 4.2 to Males
Females
85+ 80–84 75–79 70–74 65–69 60–64 55–59 50–54 45–49 40–44 35–39 30–34 25–29 20–24 15–19 10–14 5–9 0–4 8000
6000
4000
2000
0
2000
4000
Population in 00,000
Figure 4.2 Population structure of China, 2000 (shaded) and 2050.
6000
8000
88 Baochang Gu Males
85+ 80–84 75–79 70–74 65–69 60–64 55–59 50–54 45–49 40–44 35–39 30–34 25–29 20–24 15–19 10–14 5–9 0–4 100
Females
50
0
50
100
Population in 00,000
Figure 4.3 Population structure of Shanghai, 2000 (shaded) and 2050.
4.4 compare population pyramids between 2000 and 2050 for China as a whole as well as Shanghai and Beijing, the two largest metropolitan cities in China.6 They show that in a few decades, China’s population structure will turn from a pyramid to a pillar shape, and for Shanghai and Beijing, the population pyramids will even become upside down. What challenges a society may face with an aging population structure such as this is still a huge question to be answered. As a far from complete account it can be seen that the child-bearing and -rearing facilities will Males
Females
85+ 80–84 75–79 70–74 65–69 60–64 55–59 50–54 45–49 40–44 35–39 30–34 25–29 20–24 15–19 10–14 5–9 0–4 100
80
60
40
20
0
20
40
Population in 00,000
Figure 4.4 Population structure of Beijing, 2000 (shaded) and 2050.
60
80
The arrival of low fertility in China 89 likely become superfluous and services for the elderly will be urgently needed, people will tend to reduce their consumption for saving in order to pay their own old-age medical costs, and labor supply will be beginning to shrink. It is apparent that population aging is far more than an issue of elderly support, but rather it implies an overall transformation of the whole society, and calls for a reconstruction of social functions and social structures. Along with mortality decline and increase of life expectancy, the sexual disparity in life expectancy is remarkable. Table 4.8 shows that, while the overall life expectancy in China has increased over the last 2 decades from 67 years to 73 years, the disparity between male and female has become greater, from 3 years in the early 1980s to about 4.6 years in the early 2000s. It can be expected to widen further in the future. Because women tend to live longer than men, China will end up with more women than men surviving in the older ages. The higher the age the more this will be the case. This trend can also be seen in the population pyramids shown in figures 4.2 to 4.4. At the high ages, the bar for females tends to be longer than that for males. The 2004 annual population-change sample survey suggests that the sex ratio by age tends to fall below 100 at age 65, and even below 60 at age 80. In this sense, the issue of population aging is in fact an issue of aging females. But the impact of population aging is not only seen to affect elderly females but also females at young and mid ages. The responsibility for elderly care most frequently falls on the shoulders of female offspring, i.e. daughters or daughtersin-law. Should they spend more time and energy on their parents, they may have less time and energy available for their career and contributing to the wider society. When life expectancy moves upwards the care time for the elderly increases, and even results in the younger elderly taking care of the older elderly. What may represent a plus for family support would very likely represent a minus for careers and for the society (Gu and Peng 1992). How the aging situation may affect women’s status is an issue to be addressed. Another issue related to rapid fertility decline is arguably the abnormal sex ratio at birth (SRB), as well as infant mortality. China’s sex ratio among new-born babies has been rising since the mid-1980s. The 1990 population census reported China’s SRB at 111, much above the acceptable normal level observed internationally. The abnormality of SRB reported from the census engendered much concern Table 4.8 Life expectancy by sex in China for selected years, 1981–2002 Year
Total
Male
Female
Diff.(M–F)
1981 1990 1996 2000 2001 2002
66.8 68.6 70.8 71.4 72.3 72.9
66.3 66.8 68.7 69.4 70.2 70.7
69.3 70.5 73.0 73.3 74.6 75.3
–3.0 –3.7 –4.3 –3.9 –4.4 –4.6
Source: NBS, 2003.
90 Baochang Gu
55
120
50
115 110
IMR by sex (%)
45
105 40
100
35
95 90
30
85
25
80
2002
2001
2000
1996
1995
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
1980
1979
70 1978
15 1977
75
1976
20
1975
sex ratio of infant deaths
but also created a great debate as to whether the reported abnormality was a fictitious phenomenon due to statistical error or reflection of a serious reality (Zeng et al. 1993; Gu and Roy 1995). The debate lasted for ten years until the release of the results of the 2000 population census, which reported China’s SRB to be further up to 117, way above the normal level. More recently the 2005 1 percent nationalpopulation sample survey reported an even higher SRB of 118.9. The abnormality of SRB thus has been observed in China for two decades. While China is not the only population in the world with an abnormal SRB, it is the population with the most severe abnormal SRB over the longest period. The last two decades not only witnessed the rising sex ratio at birth in China but also an abnormality in infant mortality rate (IMR) by sex. As shown in Figure 4.5, the overall IMR in China has been declining, but the IMR for girls has been abnormally higher than that for boys. The sex ratio of infant deaths fell to below 100, and has fallen further in recent years. The excessive sex ratio at birth and the over-low sex ratio among infant deaths can be seen as two sides of the one coin of sexual selection in child-bearing, i.e. “before-birth solution” and “after-birth solution,” and reflects the sexual discrimination in childbirth and child survival (Li et al. 2006). It has been much discussed that among other factors, the occurrence of abnormal SRB has much to do with the rapidity of fertility decline, which intensifies the conflict between the number and sex of children people intend to have (Gu and Roy 1995).
year Male
Female
sex ratio of infant deaths
Figure 4.5 Infant mortality rate by sex and sex ratio of infant deaths in China for selected years, 1975–2002. Source: NBS 2003.
The arrival of low fertility in China 91 Last but not least, the impact of low fertility on labor supply requires a brief discussion. A smaller birth cohort will naturally convert into a smaller labor cohort in later decades. It is predicted that if current fertility continues, China will soon observe a rapid decline in labor supply – from 966 million in the late 2020s to 761 million by the mid-twenty-first century. While the size of labor force will remain huge, the reduction of its size will be quite dramatic, a rate of reduction of 100 million per decade, or 10 million per year. And the labor force itself will be aging as well (Guo et al. 2006). “Labor shortage” has been reported since the spring of 2004. Though there are various reasons, some Chinese economists nevertheless have started to claim that China’s labor force has moved from a period of “abundant supply” to a period of “limited surplus” (Cai and Wang 2006). It is even claimed that in the next few years (by about 2013) the growth of working age population in China will turn to zero (Cai and Du 2007).
Concluding remarks At the turn of the twenty-first century, below-replacement fertility has become a new global demographic reality. With over a fifth of the world’s population, China is a newcomer but an important one in the emerging global regime of belowreplacement fertility. This chapter is an expansion of an earlier paper (Gu 2007) on the fertility transition in China with respect to population growth and fertility trend, fertility policy as measured by policy fertility, as well as the impact of low fertility on population aging, gender equality, sex ratio of newborns and of infant deaths, and labor supply, with the addition of a brief report of the preliminary results from the Jiangsu survey on women’s child-bearing desires. While China’s population has entered the stage of below-replacement fertility, recognition of the arrival of low fertility in China is still slow, as also seen in the cases of Japan, Korea, Singapore, and Taiwan province. Why does the response to low fertility tend to be slow and much delayed? Jones’s explanation of the delay, stressing three reasons, deserves attention: 1 2 3
population momentum tends to show population continuing to grow despite below-replacement fertility anti-natalist mindset has been entrenched for decades and is difficult to change demographic theory has long focused on the transition from high to low fertility, but failed to address below-replacement dynamics (Jones 2007).
With the changes in life course in the low-fertility setting, sexual/reproductive health needs may have to be redefined, with crucial implications for the future direction of the family planning/reproductive-health program. For example, with a prolonged period before/out of marriage, people tend to be exposed longer to unprepared sex, leading to an increase in unintended pregnancies among unmarried women. As a result, induced abortion may become prevalent among the unmarried women as already seen in China (World Health Organization Western Pacific Region 2005; Yardley 2007). With people having fewer children, even only
92 Baochang Gu one, in their lives, in recent years there has been an increase in cases of Cesarean section in birth delivery, not only in urban hospitals but also in rural clinics (Guo et al., 2007). With the child-bearing period shortened, contraceptive use becomes prolonged for years. To ensure effective use of contraception, better quality of contraceptive services is needed to prevent the occurrence of unwanted pregnancy and child-bearing. Contraceptive failure has been cited in many Chinese surveys as the primary cause of abortion among the married. In a survey in Suzhou city in 2006, it was reported that contraceptive failure was the cause of two-thirds of abortions. Finally, when a population ages, people tend to live more years during the aging period, and although they may not be able to reproduce, this does not mean that sexual health is of no concern to them. Research has shown sexual health is relevant to old people’s happiness, longevity, and well-being. All these new sexual/reproductive-health issues need to be addressed, a new agenda of services developed, and a new mission for the Chinese family planning program identified. The arrival of below-replacement fertility in China calls for a new understanding of population dynamics and a review of population policies to address the issues arising from low fertility. “Low fertility” as a demographic phenomenon has now been observed not only in Europe but also in Asia, not only in developed countries, but also in developing countries, not only in traditionally “low-fertility” countries but also in traditionally “high-fertility” countries (Gubhaju and MorikiDurand 2003; Morgan and Taylor 2006). There are 65 countries accounting for 43 percent of the world population with a fertility below replacement, 31 countries among them with a TFR less than 1.5, and more countries are expected to reach the “very low fertility” level. “Below-replacement fertility” has gradually become a global trend in population dynamics and a concern of international society (United Nations 2000). It is even claimed to be “the norm in post-transitional societies” (Demeny 1997; Bongaarts 2001). Our understanding of the global trend toward below-replacement fertility as well as the issues faced in developing policy responses to it is still limited. China is once again at a demographic crossroads. If it wishes to avoid serious negative consequences associated with below-replacement fertility in the long run, the need to revisit China’s current fertility policy is compelling (Wang 2005).
Notes 1 The National Bureau of Statistics (2003) later adjusted the 2000 total fertility rate to 1.4 according to the short form of the census. 2 This part is based on Gu et al. 2006 and Gu et al. 2007a. 3 In a classification commonly used in China, the 31 mainland provincial units are divided into 3 broad economic regions: the most developed Eastern region, which includes 11 provinces of Beijing, Tianjin, Hebei, Liaoning, Shanghai, Jiangsu, Zhejiang, Fujian, Shandong, Guangdong, and Hainan; the least developed Western region, which includes 12 provinces of Chongqing, Sichuan, Guizhou, Yunnan, Tibet, Shaanxi, Gansu, Qinghai, Ningxia, Xinjiang, Inner-Mongolia, and Guangxi; and 8 provinces in between known as “Central” region, which includes Shanxi, Jilin, Heilongjiang, Anhui, Jiangxi, Henan, Hubei, and Hunan.
The arrival of low fertility in China 93 4 The determination of rural and urban is based on the wife’s Hukou (official household registration) status. 5 Among those unemployed at the time of the survey, about 37 percent were at school, 43 percent were housewives, and only about 17 percent fell under the conventional definition of unemployed – those who were in the job market looking for jobs. 6 Developed by Cai Yong based on data from 2000 population census.
References Bongaarts, John. 2001. “Fertility and reproductive preferences in post-transitional societies,” Population and Development Review 27, supplement: Global Fertility Transition: 260–281. Cai, Fang and Du Yang. 2007. The Coming Lewisian Turning Point and Its Policy Implications, Greenbook of Population and Labor (2007), Report on China’s Population and Labor No. 8. Beijing: Social Science Academic Press. Cai, Fang and Wang Meiyan. 2006. “Aging and labor shortage in China,” Chapter 5 in Fang Cai and Baochang Gu (eds), Greenbook of Population and Labor (2006), Report on China’s Population and Labor No. 7, Demographic Transition and Its Social and Economic Consequences. Beijing: Social Science Academic Press: 143–160. Cai, Yong. 2005. “Assessing fertility levels in China using variable-r method,” paper presented at the Population Association of America annual meeting, Boston, MA. Chen, Li and Zhenwei Guo. 2006. “Analysis on population situation in 2005,” Thematic Report 1 in Fang Cai and Baochang Gu (eds), Greenbook of Population and Labor (2006), Report on China’s Population and Labor No. 7, Demographic Transition and Its Social and Economic Consequences. Beijing: Social Science Academic Press: 3–19. Choe, Minja Kim, Robert D. Retherford and Seung-Kwon Kim. 2004. “Very low fertility in South Korea: patterns and prospects,” paper presented at the 2004 Annual Meeting of the Population Association of America, April 1–3, Boston, MA. Coale, A.J. 1984. Rapid Population Change in China, 1952–1982. Washington, DC: National Academy Press. Demeny, Paul. 1997. “Replacement-level fertility: the implausible endpoint of the demographic transition,” in G.W. Jones, R.M. Douglas, J.C. Caldwell, and R.M. D’Souza (eds), The Continuing Demographic Transition. Oxford: Clarendon Press: 94–110. Greenhalgh, Susan and Edwin A. Winckler. 2005. Governing China’s Population: From Leninist to Neoliberal Biopolitics. Stanford, CA: Stanford University Press. Gu, Baochang. 2003. “Population, reproductive health and poverty in China,” in Population and Poverty: Achieving Equity, Equality and Sustainability – Population and Development Strategies, No. 8. New York, NY: UNFPA: Chapter 4, 63–77. —— 2006. “China’s population dynamics in the new period,” Chapter 1 in Fang Cai and Baochang Gu (eds), Greenbook of Population and Labor (2006), Report on China’s Population and Labor No. 7, Demographic Transition and Its Social and Economic Consequences. Beijing: Social Science Academic Press: 61–83. —— 2007. “Low fertility in China: trends, policy and impact,” Asia-Pacific Population Journal 22(2): 73–90. Gu, Baochang and Xizhe Peng. 1992. “Consequences of fertility decline: cultural, social and economic implications in China,” Impact of Fertility Decline on Population Policies and Program Strategies. Seoul, South Korea: Korea Institute for Health and Social Affairs: 49–66.
94 Baochang Gu Gu, Baochang and Krishna Roy. 1995. “Sex ratio at birth in China with reference to other areas in East Asia: what we know,” Asia-Pacific Population Journal 10(3): 17–42. Gu, Baochang, Wang Feng, Guo Zhigang, and Zhang Erli. 2006. “Fertility policy and policy fertility in China: a quantitative analysis,” presented at the Annual Meeting of Population Association of America, March 30–April 1, 2006, Los Angeles, CA.I —— 2007a. “China’s local and national fertility policies at the end of the twentieth century,” Population and Development Review, 33(1): 129–147. Gu, Baochang, Zheng Zhenzhen, Wang Feng, and Yong Cai. 2007b. “Globalization, policy intervention, and reproduction: below replacement fertility in China,” paper presented at the Annual Meeting of Population Association of America, New York, NY, March 29–31, 2007. Gubhaju, Bhakta B. and Yoshie Moriki-Durand. 2003. “Below-replacement fertility in East and Southeast Asia: consequences and policy responses,” Journal of Population Research, 20(1): 1–18. Guo, Sufang, Sabu S. Pasmadas, Zhao Fengmin, James J. Brown, and William Stones. 2007. “Delivery settings and caesarean section rates in China,” Bulletin of the World Health Organization, October, 85(10): 755–762. Guo, Zhigang. 2004. “Study for fertility of China in the 1990s,” Population Research 28(2): 10–19 (in Chinese). Guo, Zhigang, Zhang Erli, Gu Baochang, Wang Feng, and Xie Zhenming. 2006. “Current fertility and future perspectives of population development in China,” in Zeng Yi, Li Ling, Gu Baochang, and Justin Yifu Lin (eds), 21st Century Population and Economic Development in China. Beijing: Social Science Academic Press: 67–94. Jones, Gavin W. 2007. “Very low fertility in East Asian countries: causes and policy responses,” paper presented at Shanghai Forum 2007, Crown Plaza Hotel, Shanghai, 24–27 May. Lee, James and Wang Feng. 1999. One Quarter of Humanity, Malthusian Mythology and Chinese Realities 1700–2000. Cambridge, MA: Harvard University Press. Li, Shuzhuo, Wei Yan, and Jiang Quanbao. 2006. “Girl child survival in China: past, present and prospect,” Market and Demographic Analysis 1: 2–16. Morgan, P. Philip and Miles G. Taylor. 2006. “Low fertility at the turn of the twenty-first century,” Annual Review of Sociology: 32: 375–399. National Bureau of Statistics. 2003. Highlights of Data from 2000 the Fifth National Population Census. Beijing: National Bureau of Statistics. —— 2005. 2004 China Population. Beijing: China Statistics Press. —— 2006. “Communiqué of Major Figures from 2005 1 Percent National Population Sample Survey,” China Population Statistics Yearbook 2006: 3. Beijing: China Statistics Press. —— 2007. Statistical Communiqué of National Economic and Social Development in 2006. Downloadable from http://www.stats.gov.cn/was40/gjtjj_en_detail. jsp?channelid=4920&record=14. Poston, Dudley, Baochang Gu, and Hua Luo. 2005. “The effects of the fertility and mortality transitions on the elderly and eldercare in China, and in Shanghai, Beijing, and Tianjin,” in Papers of the International Symposium on Population and Sustainable Development Strategy. Shanghai: Shanghai Academy of Social Science Press: 94–115. Retherford, R.D., M.K. Choe, J. Chen, X. Li, and H. Cui. 2005. “Fertility in China: how much has it really declined?,” Population and Development Review 31(1): 57–84. United Nations. 2000. Below Replacement Fertility. New York, NY: The United Nations. —— 2007. World Population Prospects: The 2006 Revision, Population Division, Depart-
The arrival of low fertility in China 95 ment of Economic and Social Affairs. New York, NY: The United Nations. Wang, Feng. 2005. “Can China afford to continue its one-child policy?” Asia Pacific Issues, No. 77. East-West Center, Honolulu, HI. World Health Organization Western Pacific Region. 2005. Sexual and Reproductive Health of Adolescents and Youths: A Review of Literature and Projects, 1995–2002. Manila, Philippines. Yardley, Jim. 2007. “Today’s face of abortion in China is a young unmarried woman,” New York Times, May 13. Zeng, Yi, Tu Ping, Gu Baochang, Xu Yi, Li Bohua, and Li Yongping. 1993. “Causes and implications of the recent increase in the reported sex ratio at birth in China,” Population and Development Review 19(2): 283–302. Zhang, Guangyu and Zhongwei Zhao. 2006. “Re-examining China’s fertility puzzle: data collection and quality over the last two decades,” Population and Development Review 32(2): 293–321. Zhuang, Yaer and Liping Zhang. 2003. Basic Data of China Population Since 1990. Beijing: China Population Press.
Socioeconomic development and fertility in the People’s Republic of China and the Republic of Korea
5
Dudley L. Poston, Jr., Heather Terrell Kincannon, and Jungwon Yoon Introduction In recent decades, the People’s Republic of China and the Republic of Korea have experienced fertility transitions that are among the most rapid of any countries in the world. In 1960, both countries had total fertility rates of around 6 children per woman. By 2005, these had declined to 1.7 in China and to 1.1 in South Korea (Figure 5.1). In this chapter, we address empirically the impact of social and economic development on the fertility reductions in China and Korea. County level fertility data, for the year 2000 for the over 2,400 counties in China and for the year 2004 for the 231 counties of South Korea, are utilized to examine the extent to which the variation in fertility is accounted for by factors of socioeconomic development. The fertility reduction in China has been due in part to social and economic development (Tien, 1984; Poston and Gu, 1987; among many others), and in part to a “direct and forceful government intervention” in family planning (Wolf, 1986: 101; see also Feeney and Wang, 1993). The rapid decline in fertility in Korea is also 7 6
TFR
5 4 3 2 1
72 74 19 76 19 78 19 80 19 82 19 84 19 86 19 88 19 90 19 92 19 94 19 96 19 98 20 00 20 02 20 04 20 05 19
70
19
19
19
60
0 Year China
S. Korea
Figure 5.1 Total fertility rates: China and South Korea, 1960–2005.
Socioeconomic development and fertility in the PRC and the ROK 97 attributable to both socioeconomic development and a “very successful” adoption of government-led family planning programs. We explore how social and economic development factors in these two countries have operated independently and together to influence fertility change. We address this issue cross-sectionally using county level data for the year 2000 in China and 2004 in Korea. One reason for taking this route is that annual time-series data are not available for each of the countries in the kind of descriptive detail necessary. Even though we undertake the cross-sectional investigations for time periods (2000 and 2004) after the transitions in China and Korea have already occurred, the analysis will be instructive if only because of their county level focus. Most prior cross-sectional investigations of development and fertility have been conducted among countries (for examples see, among many others, Mauldin and Berelson, 1978; Tsui and Bogue, 1978; Hernandez, 1984; Donaldson, 1991; Mauldin and Ross, 1991; and Livi-Bacci, 1997: chapter 5). We turn now to a description of fertility among the counties of China and South Korea. We describe first the patterns in China, and then in South Korea.
Patterns of fertility among the counties of China and South Korea County level fertility in China in 2000 The 2000 Population Census of China, “arguably the world’s most ambitious census ever” (Kennedy, 2001: 1), involved “10,000 tons of paper for questionnaires, 5 million enumerators, and a million supervisors” (Lavely, 2001: 755). This was the Fifth National Census of the People’s Republic of China and the first complete count of the country’s population in ten years. China’s 2000 census has been commended for its use of short-form and long-form questionnaires (Chan, 2003). Moreover, it was the first census in thirty years to introduce a “zanzhu renkou (temporary population) form to record this population (the length of stay was reduced to less than six months), whereas the remaining, ‘ordinary resident’ population (changzhu renkou) was recorded in the regular census form” (Chan, 2003). This incorporation will benefit the analysis in this chapter because it has been hypothesized that many out-of-quota births are children of the “floating population.” In the China analysis in this chapter, we use the total fertility rate (TFR) as the dependent variable for the analysis of fertility among China’s counties. In recent years, there has been some skepticism regarding the effectiveness of the conventional TFR among lowest-low-fertility populations. According to Sobotka (2004) and others (see Kohler et al., 2002), the postponement of fertility tends to depress the period fertility rates in lowest-low-fertility populations. This tempo effect may tend to distort the conventional TFR in that there is a supposed “recovery” of the TFR at the end of the fertility postponement. Thus, Sobotka believes lowest-low fertility is caused by increasing age at motherhood – a temporary trend that will disappear once postponement comes to a halt. We concur that this is a problem in many European populations where the mean age at first birth is very high (i.e., 28–30 years of age); however, it should not
98 Dudley L. Poston, Jr., Heather Terrell Kincannon, and Jungwon Yoon pose as serious a problem in China since childbirth generally occurs shortly after marriage (by, say, around 25 years of age). According to the 2000 Population Census of China, there are 2,873 Chinese counties or county equivalents. Three of these areas had missing values for their fertility rates and were consequently dropped, leaving us with 2,870 Chinese counties or county equivalents. County equivalents are typically cities or districts of cities which fit the administrative criterion of a county. The mean value of the TFR in 2000 among the 2,870 counties of China is 1.32, with a standard deviation of 0.47 (see Table 5.1). There is considerable variability among the counties in the TFR, varying from lows of 0.41 in the Xiangyang district of Jiamusi City (in Heilongjiang Province), 0.43 in the Heping district of Tianjin city, and 0.46 in the Mawei district of Fuzhou city (in Fujian Province) to highs of 3.96 in Geji County (in Tibet), 4.07 in Jiali County (Tibet), and 5.47 in Baqing County (Tibet). The range between the two extreme scores (i.e., 0.41 and 5.47) is more than 5 children per woman. Variation in the TFR among China’s counties is considerable. The districts with the three lowest TFRs are in Heilongjiang Province, Tianjin Municipality, and Fujian Province, respectively located in the northeast, north, and east regions of China. The per capita GDPs of these three provinces are higher than in most provinces of China factoring mostly to their relatively strong economies and being socioeconomically advanced. Jiamusi City, situated in the heart of Heilongjiang Province, contains the district with the lowest TFR in the nation, a value of 0.41 children per woman. This small city is home to the John Deere Company and is known for its healthy environment and clean air. Jiamusi City is also at the forefront of education, recently launching the Dongze Joy Foreign Language School, which teaches English to students between the ages of 4 and 15 years. The counties with the highest TFRs are all located in the Tibet (Xizang) Autonomous Region, a poor, rural region in west China. Baqing County, Tibet, has the highest TFR in the nation, a value of 5.47 children per woman. This county also has the highest illiteracy rate and the highest minority percentage among the 2,870 counties of China. Table 5.1 Descriptive statistics, fertility rates: 2,870 counties of China in 2000, and 231 counties of South Korea in 2004 Variable China Total fertility rate South Korea General fertility rate
Standard mean deviation
Minimum value
Maximum value
1.32
0.47
0.41 5.47 Xiangyang District Baqing County (Jiamusi City, (Tibet) Heilongjiang)
34.36
6.50
21.60 Gangseo-gu County (Pusan)
54.31 Hwacheon-gun County (Kangwon-do)
Socioeconomic development and fertility in the PRC and the ROK 99 County level fertility in Korea in 2004 The principal source of the Korean data we analyze in this chapter is the Korean Statistical Information System (KOSIS), a program of the Korea National Statistical Office (KNSO). KOSIS is an online statistical database that provides a wide range of domestic and international statistical data in long-term series delineated by month, quarter, and year. Most of the data used here are from 2004 Korean vital statistics and the 2000 Population and Housing Census of Korea, both available through KOSIS. Korean vital statistics collected in 2004 provide the data for the dependent variable, the general fertility rate (GFR), i.e., the number of births in 2004 per 1,000 women aged 15–49. All the Korean data used in this chapter were collected according to the “si,” “gun,” and “gu” county units of the seven municipalities and the nine provinces in South Korea. However, the Korean vital statistics data available to us when we wrote this chapter (early 2007) did not include age-specific birth-rate data for the counties. For this reason, we were restricted to using the GFR as the measure of Korean fertility. However, the GFR and TFR are highly related to one another (Bogue, 1969: 660–661). Indeed, the TFR may be estimated with GFR data using this formula: TFR = GFR × 0.3 When reporting GFR values for Korea, we will occasionally indicate what the TFR values are estimated to be, where more detailed fertility data is available. According to Korean vital statistics data for 2004, the country’s GFR is 34.72, which is equivalent to a TFR of about 1.04. The GFRs among the provinces and municipalities of Korea range from a low of 26.32 for the Pusan municipality to a high of 40.63 for Cheju-do Province. This GFR range is equivalent to a TFR range of approximately .79 to 1.22. However, among Korea’s counties, there is more variability in fertility than among the provinces. The mean GFR among the 231 counties is 34.36, with a standard deviation of 6.50. The county level GFRs range from a low of 21.6 in Gangseo-gu, in the Pusan municipality, to a high of 54.31 in Hwacheon-gun, in the Kangwon-do Province. This is equivalent to a TFR range of about .65 to 1.63. Not surprisingly, both the provinces and counties of Korea vary in their GFRs, but the highest GFR values are very low. Korea’s fertility is quite low all across the country. Most counties with the lowest GFRs are located in the Pusan municipality, and they are all roughly equal to TFRs less than 1.0. Indeed, the counties with the lowest fertility levels are all located in one of Korea’s seven municipalities. This should not be a surprise given that the municipalities are all highly urbanized and advanced in terms of socioeconomic development. On the other hand, the counties with the highest GFRs found are in four provinces: Kangwon-do, Chollanamdo, Kyongsangbuk-do, and Kyongsangnam-do provinces, all less urbanized and developed than the municipalities. In the next section, we discuss the theoretical rationale underlying the analysis.
100 Dudley L. Poston, Jr., Heather Terrell Kincannon, and Jungwon Yoon
Theoretical rationale The purpose of this investigation of fertility patterns in China and South Korea is to examine the effects of social and economic development on fertility among the counties of the two countries. The broader theoretical model focuses on factors impacting fertility, including the direct effects of social and economic development as well as the indirect effects of family planning programs and behaviors and other more proximate fertility variables. Development is typically viewed as providing an aggregate setting which influences fertility directly. Blake (1973) noted many years ago that social and economic structures and institutions tend to influence reproductive motivation and fertility by specifying the reward structures related with child-bearing (also see Hernandez, 1984: 11–13). In Tien’s (1984) analysis of Chinese fertility patterns, he observed that among those sub-areas experiencing the greatest fertility decline, “profound changes in socioeconomic structure may have occurred at the same time” (Tien, 1984: 385). Birdsall and Jamison (1983) and Poston and Gu (1987) have reported similar results, and recent analyses of China and South Korea by Terrell (2005) and Yoon (2006) have comparable findings. See also the extensive statements about the impacts of socioeconomic development on Korean fertility by Doo-Sub Kim (1987, 2005), Choe et al. (2005), and Ik-Ki Kim (1987). Factors of social and economic development may also influence fertility indirectly through family planning programs and behaviors and other variables more proximate to fertility (see Hernandez, 1984: 101–102, among others). In this context, family planning programs and services are seen as facilitating an already preeminent relationship. Without the prior effects of social and economic development on reward structures, there would likely be little or no effect of family planning on fertility. However, many of the analyses investigating the differential impacts of development and family planning on fertility have shown that the effects of development are usually stronger than those dealing with family planning (Schultz, 1971, 1980, 1994; Gertler and Molyneaux, 1994, 1998; Pritchett, 1994a, 1994b; Hirschman and Young, 1998). Our focus, thus, is on the effects on fertility of social and economic development. This general idea that fertility rates, as well as mortality rates, fall in response to increases in social and economic development is central to the classical theory of demographic transition, as originally described by Notestein (1953). As Mason has noted: … this theory attributes fertility decline to changes in social life that accompany, and are presumed to be caused by, industrialization and urbanization. These changes initially produce a decline in mortality, which sets the stage for – or by itself may bring about – fertility decline by increasing the survival of children, and hence the size of families. (Mason, 1997: 444) Demographic transition theory per se has been the subject of some debate among
Socioeconomic development and fertility in the PRC and the ROK 101 demographers in recent years. Hirschman (1994), for example, has asked about its overall utility. He has correctly observed that “over the past few decades, intensive research on demographic change in historical and contemporary societies has revealed complex patterns that do not fit neatly into earlier theoretical schema” (Hirschman, 1994: 204). The extensive publications of the European Fertility Project (see Coale and Watkins, 1986, for a summary) have indeed shown many exceptions to the general theory. For instance, in some of the European countries, fertility change was rapid, and in others gradual. Also, the initial declines in fertility were not always responses to prior changes in urbanization, literacy, agricultural production, and labor-force activity (see, for instance, Knodel and van de Walle, 1979, and Kertzer and Hogan, 1989: chapter 8). So extensive have been the critiques of demographic transition theory that Greenhalgh has observed that the “edifice of demographic transition theory has been cracking for well over a decade” (Greenhalgh, 1990: 86), Alter has written that it “has been dramatically shattered” (Alter, 1992: 13), and Hirschman has stated that the “theory is near death” (Hirschman, 1994: 213). And what may be the ultimate statement is Fricke’s observation about the demise of classic demographic transition theory: “Wielding the hammers that drove the nails into its coffin were demographers themselves …” (Fricke, 1997: 825). These critiques notwithstanding, many of the ideas of demographic transition theory “are hard to ignore and … live on” (Mason, 1997: 444). This is due in part to the fact that, according to Hirschman, “there is no consensus on an alternative theory to replace demographic transition theory” (Hirschman, 1994: 214). More importantly in our view is McNicoll’s point that some of its ideas persist because “at least in general terms, many of the [theory’s] associations are quite well substantiated – though allowing ample scope for dispute over emphasis” (McNicoll, 1992: 92).
Socioeconomic development and fertility in China We have already noted that the fertility measure in our analysis of county level fertility in China is the total fertility rate. Regarding the independent variables, we have reviewed prior literature about the influence of socioeconomic development and modernization on fertility patterns in China (Birdsall and Jamison, 1983; Freedman et al., 1988; Peng, 1989; Poston, 2000; Poston and Gu, 1987; Poston and Jia, 1990; Tien, 1984). Based on this and other literatures we have developed seven different independent variables. Some of China’s counties are missing data on at least one of these variables; we have data on all the variables and the fertility rate for 2,432 counties. Three variables measure urbanization and economic progress among the counties, namely, the percentage of the total population that is urban, the percentage of the total population employed in non-agricultural-related work, and the percentage of the total employed population employed in white-collar work (i.e., professionals, associate professionals, managers, and administrators). One variable measures the absence of rudimentary levels of education, namely, the percentage of the female
102 Dudley L. Poston, Jr., Heather Terrell Kincannon, and Jungwon Yoon population aged 15 and over that is illiterate. Two independent variables are intended to expose the presence of traditional family norms and cultural values, namely, the percentage divorced and the percentage of one-generation family households. The seventh and final variable is the percentage of the total population belonging to one of China’s 55 minority nationalities. Owing to collinearity issues with several of the above independent variables, we decided to estimate several models. In developing the models, we used independent variables with tolerances of .40 or higher. The three economic development variables, percentage urban, percentage non-agriculture, and percentage white collar, were highly associated with one another and could not all be included in the same model. The other four variables, percentage of one-generation family households, percentage divorced, percentage minority, and percentage female illiterate, had low relationships with one another and the economic variables, indicating that they could be included in the same models. In view of the three highly correlated economic development variables, we decided to divide our analysis into three distinct regression models. Each model is a regression of the TFR on five independent variables. The four cultural, minority, and illiteracy independent variables are the same in all three models; the single economic variable is what sets the models apart from one another. We expect the cultural and economic development variables to have negative effects on fertility, and the minority and illiteracy variables to have positive effects on fertility. Table 5.2 reports the results of all three multiple regression models. Standardized estimates are listed in parentheses under the coefficients Table 5.2 Multiple regression coefficients for the TFR on DTT independent variables: 2,432 counties and county equivalents of China, 2000 Independent Variables
Model 1
Sqrt % Population Urban
–0.060*** (–0.371)
Log10 % Population Non-agriculture
Model 2
–0.175*** (–0.405)
Log10 % Labor Force White-Collar % Households One-generation Log10 % Population (15+) Divorced Log10 % Population Minority Log10 % Female Population (15+) Illiterate Constant R-squared (adjusted) Source: 2000 Population Census of China. * P < 0.05, ** P < 0.01, *** P < 0.005
Model 3
–0.007*** (–0.154) –0.035*** (–0.065) 0.020*** (0.167) 0.120*** (0.238) 0.382 0.421
–0.008*** (–0.181) –0.008 (0.015) 0.021*** (0.179) 0.098*** (0.194) 0.651 0.434
–0.158*** (–0.259) –0.011*** (–0.236) –0.023* (–0.042) 0.019*** (0.161) 0.141*** (0.281) 0.380 0.381
Socioeconomic development and fertility in the PRC and the ROK 103 and significance values are specified with asterisks. We transformed all but one of the variables, owing to skewness and other issues. These transformations are shown in Table 5.2. Model 1 examines the impact of the four cultural, minority, and illiteracy independent variables, plus the percentage of the population that is urban on the TFR. We expect the higher the percentage of urban residents in a county, the lower the fertility. The two variables that measure traditional family norms and cultural values, namely, the percentage of households that are one generation and the percentage of the population (15+) that is divorced, are also both expected to be negatively associated with the TFR. Ethnic differences and more flexible policy requirements should lead to higher fertility rates among minority groups; thus, we anticipate the minority variable to have a positive relationship with the TFR. The last independent variable is the percentage of the female population (15+) that is illiterate, and it is expected to be positively associated with the TFR. The coefficients in Model 1 all show significant effects on fertility in the hypothesized direction: percentage urban (–), percentage one generation (–), percentage divorced (–), percentage minority (+), and percentage female illiterate (+). Each of the five coefficients is significant at P < 0.005. The R2 (adjusted) statistic is 0.4212, indicating that slightly over 42 percent of the variance in fertility is accounted for by these X variables. Standardized coefficients are shown in parentheses below the metric coefficients. We see that percentage urban has the strongest influence on fertility among these five variables, followed by percentage female illiterate, percentage minority, percentage one-generation, and percentage divorced. Model 2 uses four of the same independent variables, percentage one-generation, percentage divorced, percentage minority, and percentage female illiterate, plus percentage non-agriculture, to predict the TFR. The non-agriculture variable is expected to have a negative effect on fertility. Once again, all of the coefficients’ signs are in the expected direction, and all are significant at P < 0.005, with the exception of the percentage divorced coefficient, which has the hypothesized effect but is not statistically significant. In Model 2, the percentage of the population employed in non-agricultural work has the largest relative impact on the TFR, followed by percentage female illiterate, percentage one-generation, and percentage minority. The R2 (adjusted) is slightly larger in Model 2 than in Model 1 at 0.43, indicating that the X variables explain just over 43 percent of the variation in the TFR among the counties of China. Model 3 presents the results of a third multiple regression analysis of the TFR on the same four cultural, minority, and illiteracy variables, but this time the economic development variable is percentage of white-collar workers. This variable is expected to have a negative impact on the TFR. The coefficients reveal that all of the independent variables have significant effects on fertility in the hypothesized direction. Considering the standardized coefficients, percentage female illiteracy has the strongest effect on fertility. Given that there is such a small percentage of white-collar workers among the counties of China, it seems sensible that the effect of this economic development variable is less than the effects of those in
104 Dudley L. Poston, Jr., Heather Terrell Kincannon, and Jungwon Yoon the previous two models. The independent variables in Model 3 account for 38 percent of the variation in the TFR among the counties of China. An indication of robustness is that the adjusted coefficients of determination (R2) do not vary much among the three different models, ranging from 0.38 to 0.44. We turn now to a companion analysis of South Korea.
Socioeconomic development and fertility in South Korea In the analysis of socioeconomic development and fertility among the counties of South Korea, we used seven independent variables. Three variables are direct measures of socioeconomic development, namely, the percentage of the population engaged in agriculture, the percentage of the population in the labor force above 15 years of age, and the percentage of the population aged 25 and over who attain above a university degree. One variable gauges the status of women, namely, gender equity in educational attainment (measured as the ratio of the percentage of male educational attainment above a university degree to the corresponding percentage of females). Regarding traditional family norms and cultural values, three variables are used, namely, the sex ratio at birth, the marriage rate, and the divorce rate. As with the fertility analyses of China’s counties, we also estimated three Ordinary Least Squares (OLS) multiple regression models among the Korean counties. We first conducted some diagnostics of the variables and decided to transform two of them. These are noted in Table 5.3. We also examined the tolerances of the Table 5.3 Multiple regression coefficients for the GFR on socioeconomic status, female status, and cultural values variables: 231 counties of South Korea, 2004 Independent Variables
Model 1
Model 2
% Population Agriculture
0.278*** (0.877)
Log% Educational Attainment % Labor-force Participation Log Sex Ratio of % Educational Attainment Divorce Rate (25+) Marriage Rate (25+) Sex Ratio at Birth Constant R-squared (adjusted)
Model 3
–3.088*** (–0.295) 0.302*** (0.312) 13.059*** (0.421) 0.126 (0.019) 3.070*** (0.896) 0.004 (0.006) –18.474 0.524
13.422*** (0.433) –0.589 (–0.089) 3.458*** (1.010) 0.015 (0.022) 3.731 0.514
1.242*** (0.188) 3.511*** (1.025) 0.024 (0.036) –7.911 0.545
Source: Korean Statistical Information System (KOSIS), Korea National Statistical Office (KNSO). * P < 0.05, ** P < 0.01, *** P < 0.005; standardized coefficients in parentheses
Socioeconomic development and fertility in the PRC and the ROK 105 independent variable; if in any equation, a tolerance was under 0.4, the regression equation was then modified. All the tolerances in each of the three models were at statistically acceptable levels, all above 0.4. Moreover, as just noted, since many of the independent variables are collinear with each other, we divided the analysis into three OLS regression models. Models 1 and 2 each included five independent variables, representing three different characteristics: socioeconomic development, women’s status, and traditional norms and cultural values (see Table 5.3). In Model 1, three independent variables had significant effects on the general fertility rate (GFR), namely, labor-force participation, the sex ratio of the percentage of educational attainment, and the marriage rate. However, the signs of only two of them were consistent with the hypotheses: the sex ratio of educational attainment variable and the divorce rate. Labor-force participation was positively related to the GFR, in opposition to the demographic-transition hypothesis. In Model 2, all the coefficients were significantly related to the GFR in the hypothesized directions. Of all the variables in the first two models, the marriage rate consistently has the strongest effect on the GFR. The results from Models 1 and 2 generally support the basic theory of demographic transition. Model 3 introduces into the equation the agriculture variable, along with the divorce rate, the marriage rate, and the sex ratio at birth. The socioeconomic variable, namely, the percentage of the population involved in agriculture, is significantly associated with the GFR in the hypothesized direction. Among the cultural value variables, the marriage rate and the divorce rate are significantly and positively related to the GFR. The sex ratio at birth does not have a statistically significant effect. In all three models, the independent variables perform well in accounting for the variance in fertility rates among the 231 Korean counties in 2004. The R2 values range from .51 to .55. Although the levels of the fertility are extremely low throughout the country, it seems that socioeconomic and modernization conditions are nonetheless having an impact on fertility differentials in Korea. Moreover, our results indicate that traditional norms and cultural values about marriage and fertility seem to have a modest impact. However, the variable we use to measure son preference, namely, the sex ratio at birth, was found to be insignificant in all analyses.
Conclusion and discussion With respect to levels of social and economic development, China lags somewhat behind South Korea. Yet fertility levels in both China and South Korea have been below replacement since the 1990s. Korea’s fertility is somewhat lower than China’s. One of the major differences about the fertility reductions in China and South Korea is that the decline in China was due in part to a forceful and very powerful intervention via family planning by the Chinese government. Family planning programs were also influential in South Korea, but participation in them was voluntary.
106 Dudley L. Poston, Jr., Heather Terrell Kincannon, and Jungwon Yoon We believe that were it not for China’s far-reaching fertility control policies, it is unlikely that its fertility rate would have dropped as rapidly and as precipitously as it did in recent decades. Were it not for these policies, China’s fertility rate would not be almost as low today as South Korea’s. Nevertheless, despite these vastly different levels of social and economic development in China and South Korea, their effects on fertility have been similar. The results of the analyses in this chapter provide general support for the relevance of social and economic development as an explanation of fertility. Although the results are not uniform in the two countries regarding the influence of a single social or economic development variable, they do support the general hypothesis that among the counties of China and South Korea, the higher the levels of social and economic development, the lower the fertility rates. Our tests of the development-fertility association could be significantly improved if we were able to employ the same indicators of development in all cases. Since this was not possible, one must exercise caution in generalizing about the effects of certain kinds of development factors being more influential than others. Our less-than-perfect operationalizations also need to be mentioned. In some tests we used the illiteracy rate as an indicator of social development, and in another test we used the ratio of male to female educational attainment as a relative measure of female status. Regarding the former, illiteracy may no longer tap issues of social development as well as it did previously, given the dramatic increases in China since the late 1960s in educational attainment, particularly among women (Freedman et al., 1988; Lavely et al., 1990). In a similar vein, our failure to show any association between the sex ratio at birth, an indicator of son preference and fertility among the counties of South Korea does not necessarily mean that son preference in South Korea is unrelated to fertility change. The lack of an association may be due in part to an imperfect indicator. These issues and problems notwithstanding, our results indicate clearly and persuasively the strong influences of social and economic development factors on fertility. Among the counties of China and South Korea, cross-sectional examinations show consistent and pervasive negative associations between levels of social and economic development and fertility. Indeed, the similarities of the experiences in the two countries are often more apparent than the differences. The results shown in this chapter are hard to ignore. Development does indeed have an important influence on fertility, at least with regard to the measures and cross-sectional tests for China and South Korea reported here. We conclude that a social and economic development-based theory of fertility decline, as represented in the broader theory of demographic transition, continues to have considerable relevance and application as a general perspective for understanding variation in fertility among the counties of China and South Korea in the c. 2000 time period.
Socioeconomic development and fertility in the PRC and the ROK 107
References Alter, George. 1992. “Theories of fertility decline: a nonspecialist’s guide to the current debate,” in John R. Gillis, Louise A. Tilly, and David Levine (eds), The European Experience of Declining Fertility, 1850–1970: The Quiet Revolution. Cambridge, MA: Blackwell Publishers, Inc.: pp. 13–25. Birdsall, Nancy and D.T. Jamison. 1983. “Income and other factors influencing fertility in China,” Population and Development Review 9: 651–675. Blake, Judith. 1973. “Fertility control and the problem of volunteerism,” in Scientists and World Affairs, Proceedings of the Twenty-second Pugwash Conference on Science and World Affairs, London, September: 279–283. Bogue, Donald J. 1969. Principles of Demography. New York, NY: Wiley. Chan, Kam Wing. 2003. “Chinese Census 2000: new opportunities and challenges,” The China Review 3: 1–12. Choe, Minja Kim, Kyung-Ae Park, and Korea National Statistical Office. 2005. “Fertility decline in South Korea: forty years of policy-behavior dialogue,” paper presented at the XXV IUSSP International Population Conference, Tours, France, July 20. Coale, Ansley J. and Susan Watkins (eds). 1986. The Decline of Fertility in Europe. Princeton, NJ: Princeton University Press. Donaldson, Loraine. 1991. Fertility Transition: The Social Dynamics of Population Change. Cambridge, MA: Basil Blackwell, Inc. Feeney, Griffith M. and Feng Wang. 1993. “Parity progression and birth intervals in China: the influence of policy in hastening fertility decline,” Population and Development Review 19: 61–101. Freedman, Ronald, Zhenyu Xiao, Bohua Li, and William R. Lavely. 1988. “Education and fertility in two Chinese provinces: 1967–1970 to 1979–1982,” Asia-Pacific Population Journal 3: 3–30. Fricke, Tom. 1997. “The uses of culture in demographic research: a continuing place for community studies,” Population and Development Review 23: 825–832. Gertler, Paul J. and John W. Molyneaux. 1994. “How economic development and family planning programs combined to reduce Indonesian fertility,” Demography 31: 33–63. —— 1998. “The allocation and impact of family planning inputs in Indonesia,” paper presented at the Conference on Economic Aspects of Demographic Transition: the Experience of Asian-Pacific Countries, Taipei, Taiwan, June 19–20. Greenhalgh, Susan. 1990. “Toward a political economy of fertility: anthropological contributions,” Population and Development Review 16: 85–106. Hernandez, Donald J. 1984. Success or Failure? Family Planning Programs in the Third World. Westport, CT: Greenwood Press. Hirschman, Charles. 1994. “Why fertility changes,” Annual Review of Sociology 20: 203–233. Hirschman, Charles and Yih-Jin Young. 1998. “The decline of fertility in Thailand, Malaysia, Indonesia, and the Philippines: 1968–70 to 1988–90,” paper presented at the Conference on Economic Aspects of Demographic Transition: the Experience of AsianPacific Countries, Taipei, Taiwan, June 19–20. Kennedy, Bingham, Jr. 2001. “Dissecting China’s 2000 Census,” Population Today: 1–3. Kertzer, David I. and Dennis P. Hogan. 1989. Family, Political Economy, and Demographic Change: The Transformation of Life in Casalecchio, Italy, 1861–1921. Madison, WI: The University of Wisconsin Press.
108 Dudley L. Poston, Jr., Heather Terrell Kincannon, and Jungwon Yoon Kim, Doo-Sub. 1987. Socioeconomic Status, Inequality and Fertility. Seoul, South Korea: Seoul National University. —— 2005. “Theoretical explanations of rapid fertility decline in Korea,” The Japanese Journal of Population 3: 2–25. Kim, Ik Ki. 1987. Socioeconomic Development and Fertility Behavior in Korea. Seoul, South Korea: Seoul National University. Knodel, John and Etienne van de Walle. 1979. “Lessons from the past: policy implications of historical fertility studies,” Population and Development Review 5: 217–245. Kohler, Hans-Peter, Francesco C. Billari, and Jose Antonio Ortega. 2002. “The emergence of lowest-low fertility in Europe during the 1990s,” Population and Development Review 28: 641–680. Korea National Statistical Office (KNSO). 2006. Korean Statistical Information System (KOSIS). Retrieved February 2006 from the On-Line Statistics Database: http://kosis. nso.go.kr. Lavely, William. 2001. “First impressions from the 2000 Census of China,” Population and Development Review 27: 755–769. Lavely, William, Zhenyu Xiao, Bohua Li, and Ronald Freedman. 1990. “The rise in female education in China: national and regional patterns,” The China Quarterly 121: 61–93. Livi-Bacci, Massimo. 1997. A Concise History of World Population. Second Edition. Malden, MA: Blackwell Publishers, Inc. McNicoll, Geoffrey. 1992. “Changing fertility patterns and policies in the third world,” Annual Review of Sociology 18: 85–108. Mason, Karen Oppenheim. 1997. “Explaining fertility transitions,” Demography 34: 443–454. Mauldin, W.P. and B. Berelson. 1978. “Conditions of fertility decline in developing countries,” Studies in Family Planning 9: 89–147. Mauldin, W.P. and John A. Ross. 1991. “Family planning programs: efforts and results,” Studies in Family Planning 22: 350–367. Notestein, Frank. 1953. “Economic problems of population change,” in Proceedings of the Eighth International Conference of Agricultural Economics. London: Oxford University Press: 13–31. Peng, Xizhe. 1989. “Major determinants of China’s fertility transition,” The China Quarterly 117: 1–37. Poston, Dudley L., Jr. 2000. “Social and economic development and the fertility transitions in mainland China and Taiwan,” Population and Development Review 26: 40–60. Poston, Dudley L., Jr. and Baochang Gu. 1987. “Socioeconomic development, family planning, and fertility in China,” Demography 24: 531–551. Poston, Dudley L., Jr. and Zhongke Jia. 1990. “Socioeconomic structure and fertility in China: a county level investigation,” Journal of Biosocial Science 22: 507–515. Pritchett, Lant H. 1994a. “Desired fertility and the impact of population policies,” Population and Development Review 20: 1–55. —— 1994b. “The impact of population policies: reply,” Population and Development Review 20: 621–630. Schultz, T. Paul. 1971. Evaluation of Population Policies: A Framework for Analysis and its Application to Taiwan’s Family Planning Program. Santa Monica, CA: The Rand Corporation. —— 1980. “An economic interpretation of the decline in fertility in a rapidly developing country; consequences of development and family planning,” in Richard A. East-
Socioeconomic development and fertility in the PRC and the ROK 109 erlin (ed.), Population and Economic Change in Developing Countries. Chicago, IL: University of Chicago Press: pp. 209–288. —— 1994. “Human capital, family planning, and their effects on population growth,” American Economic Review 84: 255–260. Sobotka, Tomas. 2004. “Is lowest-low fertility in Europe explained by the postponement of childbearing?,” Population and Development Review 30: 195–220. Terrell, Heather K.M. 2005. “Fertility in China in 2000: a county-level analysis,” unpublished M.S. thesis. Tien, H. Yuan. 1984. “Induced fertility transition: impact of population planning and socioeconomic change in the People’s Republic of China,” Population Studies 38: 385–400. Tsui, Amy Ong and Donald J. Bogue. 1978. “Declining world fertility: trends, causes, implications,” Population Bulletin 33: 1–42. Wolf, Arthur P. 1986. “The pre-eminent role of government intervention in China’s family revolution,” Population and Development Review 12: 101–116. Yoon, Jungwon. 2006. “Variation in patterns of low fertility in South Korea in 2004: a county-level analysis,” unpublished M.S. thesis.
6
The 1997 Asian economic crisis and changes in the pattern of socioeconomic differentials in Korean fertility Doo-Sub Kim
Introduction Korea has witnessed a drastic decline in the level of fertility since the 1997 Asian economic crisis. The total number of births in 2005 was estimated as 438,000, revealing a 35.4 percent reduction from 678,000 births in 1997. The total fertility rate (TFR) was recorded at 1.08 in 2005, down from 1.54 in 1997 (KNSO, 2006a). The pace and extent of this ongoing decline in fertility are much greater than most demographers’ forecasts. According to the Population Reference Bureau (2006), the TFR of Korea is currently at the lowest level in the world except for Hong Kong (1.0) and Macau (0.8). In contrast to the enormous amount of fertility research that was done until the mid-1980s, there have been only a few empirical studies covering the transition to lowest-low fertility during the past decade. The nature and theoretical basis of the recent changes in Korean fertility have not yet been sufficiently explored (Eun, 2003; Jun, 2004; Kim, 2004, 2005; Kim and Song, 2006; Kim et al., 2006; Park, 2006; Yoo, 2006; Eun, 2007). We therefore do not have satisfactory explanations for the mechanisms of the recent fertility transition. Nor is it clear what causal mechanisms have operated in bringing about the recent changes in the pattern of socioeconomic differentials. The main purpose of this study is to review the recent changes in Korean fertility, and to explore the effects of the 1997 Asian economic crisis on the level of fertility and the pattern of socioeconomic differentials. Trends in fertility-related indicators during the past two decades are reviewed. Then, based on an analysis of data from the 2003 Korean National Fertility Survey, the chapter explores recent changes in the level of fertility according to the socioeconomic status of couples, including educational level, occupation, working status, and income. This study focuses on a comparative analysis of children ever born (CEB) and post-1998 CEB of married women aged 20–49. It is postulated that the patterns of socioeconomic differentials in CEB and the post-1998 CEB would not be similar if the economic crisis affected the causal mechanisms of fertility significantly. For the second round of comparative analysis, the study population was split into two groups – those married before 1997 or remarried regardless of the timing of marriage, and those married for the first time after 1997 – and the level of fertility
The 1997 Asian economic crisis 111 and the pattern of socioeconomic differentials of the two groups were compared. It is expected that this comparative analysis will strengthen our understanding of the level of fertility and the socioeconomic differentials before and after the economic crisis.
Trends in fertility-related indicators after the 1997 economic crisis The 1997 Asian economic crisis has had a profound economic impact on the growth, finance, earnings, consumption, and labor markets of Korean society. Although the economic growth rate has recovered to some extent, various statistics indicate that Korea is still under deep economic stagnation (Bank of Korea, 2006; IMF, 2006; KNSO, 2006a). The implications of the 1997 economic crisis are not just confined to economic spheres. It has also created social pressures to change the pattern of family composition, which has had a direct bearing on fertility. Mainly due to high unemployment, distortion of the labor market and expansion of poverty after the 1997 crisis, many young people have tended to delay or avoid marriage and child-bearing (Jun, 2004; Kim, 2005; Kim and Song, 2006; Park, 2006). Korea’s descent to lowest-low fertility can mostly be attributed to these factors. Changing trends in fertility-related indicators before and after the 1997 economic crisis are reviewed in this section. Rapid downward trends in the number of marriages and the crude marriage rate, after reaching a peak of 435,000 and 9.4 per 1,000, respectively, in 1996, are apparent in Figure 6.1. The number of marriages dropped 10.7 percent in 1997 from the previous year. In contrast, the number of divorces and the crude divorce rate show a rapid upward trend during the period from 1997–2003. The annual increase in the rate of divorces was recorded at 14.1 percent in 1997, and 28.0 percent in 1998. It is interpreted that the 1997 economic crisis has played an important reinforcing role for young people in delaying initial family formation, and that high unemployment and increased poverty have expedited family dissolution. As shown in Figure 6.1, the increasing trends in the number of divorces and the crude divorce rate were reversed after 2003, and dropped to 128,000 and 2.6 per thousand, respectively, in 2005. Currently, Korea maintains the second highest level of crude divorce rate, next to the United States, among OECD countries (KNSO, 2006a). The proportion of births out of wedlock has been very low in Korea. Thus, age at first marriage of women, along with the number of marriages and the crude marriage rate, has become the dominant determinant of fertility level. The singulate mean age at marriage (SMAM) of women was estimated at 25.7 in 1997. Nevertheless, a continued increase in the SMAM of women took place after 1997, and reached 27.7 in 2005 (KNSO, 2006a). A rise in age at first marriage tends to delay the timing of first birth. The mean age at first birth rose from 26.9 in 1997 to 29.1 in 2005. The mean age at second birth rose from 29.1 to 31.0 during the same period. Figure 6.2 shows a marked downward trend in fertility rate for women aged 25–29. The fertility rate for
112 Doo-Sub Kim No. of Marr/Divorce (in 1,000)
CMR/CDR 12.0
450 400
10.0
350
8.0
300 250
6.0 200 150
4.0
100
2.0
50 0
1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 Marriage
Divorce
Crude marriage rate
0.0
Crude divorce rate
Figure 6.1 Trends in the number of marriages and divorces, crude marriage rates, and crude divorce rates, 1981–2005. Source: KNSO (2006c).
women aged 25–29, estimated as 161.5 per 1,000 in 1997, sharply declined to 153.4 per 1,000 in 1998, 148.1 per thousand in 1999, and further declined to 92.3 per thousand in 2005. This implies a reduction of 8.3 percent for the prime reproductive group during the 2 years immediately after the economic crisis, and a 42.9 percent reduction of births during the period 1997–2005 (KNSO, 2006b). As shown in Figure 6.2, the fertility rate for women aged 20–24 has maintained a lower level since 1994, compared to women aged 30–34. By the year 2005, it even dropped to a lower level than that of women aged 35–39. It is interesting to note that the fertility rates for women in their thirties show slightly increasing patterns. As a result, women aged 30–34 rather than women aged 25–29 became the prime reproductive group in 2005. About 40.9 percent of births in 2005 were from women aged 30–34, while 40.2 percent were from women aged 25–29 (KNSO, 2006b). Change of the prime reproductive group implies a strong tendency for young people to delay or avoid marriage and child-bearing. High unemployment after the 1997 economic crisis and economic hardship due to labor-market insecurity appear to be the main factors explaining this phenomenon. Since the 1997 economic crisis, employment has become much less secure for young people at the ages of career formation and marriage, as well as for those in their thirties and forties. The unemployment rate for men aged 25–29 rose from 4.9 percent to 10.9 percent between 1997 and 1998. Similar patterns are found for women in corresponding age groups as well as for men aged 30–34 (OECD, 2004; Kim,
The 1997 Asian economic crisis 113 Births per 1,000 Women 200 178.8 150 104.6 100 72.7 50
84.2
92.3 82.3
64.2 13.8
0
20.4
18.9
18.6
17.9
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Age 20–24
Age 25–29
Age 30–34
Age 35–39
Figure 6.2 Trends in age-specific fertility rates, 1993–2005. Source: KNSO (2006b).
2005). Despite its drop since 1999, the level of unemployment has continued to be substantially higher than the pre-crisis rates. It is rather widely argued that security has not recovered from the economic crisis, and that the conditions of the labor market, such as wages, part-time work, lay-offs, etc., have even deteriorated in recent years. There is no doubt that the 1997 economic crisis marked an important turning point in the process of the socioeconomic development and fertility transition of Korean society. Labor-market deregulation and high unemployment associated with a poor economy have made many young people delay or avoid marriage and child-bearing. The negative effects of economic recession and increased insecurity in the labor market on the level of fertility can also be applied to those in their forties.
Data and the study population The main objective of this chapter is to explore whether and to what extent the pattern of socioeconomic differentials in Korean fertility has changed since the 1997 Asian economic crisis. The main data set to be used for the analysis is from the 2003 Korean National Fertility Survey. This study focuses on women aged 20–49. To avoid the effects of exogenous factors on fertility, the sample of this study was confined to 6,660 women who were currently married at the time of this survey. The number of CEB and the post-1998 CEB are estimated to analyze recent changes in the level of fertility and the pattern of differential fertility according to the socioeconomic status of the couple.
114 Doo-Sub Kim To give one perspective on recent changes in fertility after the economic crisis, this chapter focuses on analyzing socioeconomic differentials in CEB of women who married after 1997 and those who were in their first marriage at the time of the survey. To examine the generality of the findings, comparisons were made with the other women who married before 1997, or who were in their second marriage or higher regardless of the timing of the marriage. Ideally, of course, such analysis should control for differential exposure, by controlling for variables such as duration of marriage or parity (i.e., number of children ever born). However, the relatively small number of cases from the 2003 survey data does not allow such an analysis by selected independent variables. To strengthen the analysis, though, a later section of the chapter does compare fertility between two marriage cohorts. Basic demographic profiles of the study population are presented in Table 6.1. The mean CEB of women aged 20–49 and currently married is 1.9. This includes the mean post-1998 CEB of 0.4. When the sample was divided into two subgroups, the mean CEB of the first group (women married before 1997 or remarried) and the second group (women married for the first time after 1997) were 2.1 and 1.1, respectively. The gap in the mean age of the couples of the two groups is estimated as 10 years. The mean age at marriage of husbands and wives turns out to be 1.7 and 2.1 years higher, respectively, in the second group. The mean interval between marriage and first birth was calculated as 18.2 months for the second group, which was slightly longer than the 17.3 months of the first group. The number of cases and percentage distributions according to educational attainment and occupation of the couple are presented in Table 6.2. The second group of women who married for the first time after 1997 is relatively younger than the first group, and thus their level of educational attainment is substantially higher. The proportion of husbands and wives who received college education Table 6.1 Demographic profiles of the study population, 2003 Currently married (aged 20–49)
CEB Post-1998 CEB Husband’s age Wife’s age Husband’s age at marriage Wife’s age at marriage Interval b/w marr & first birth (months)
Married before 1997 or remarried
First married after 1997
Total
Mean
N
Mean N
Mean N
2.1 0.2 43.2 39.8 27.2 23.8 17.3
5,164 5,301 5,299 5,299 5,287 5,261 5,150
1.1 1.1 32.3 29.4 28.9 25.9 18.2
1.9 0.4 41.0 37.7 27.5 24.2 17.5
1,359 1,359 1,359 1,359 1,358 1,359 1,358
6,525 6,660 6,660 6,660 6,645 6,620 6,508
Table 6.2 Distribution of educational attainment and occupation of the two study groups, 2003 Currently married (aged 20–49) Married before 1997 or remarried
First married after 1997
Total
N
N
N
%
%
%
Husband’s education Primary school graduates Middle-school graduates High-school graduates Some college College graduates
461 689 2,190 454 1,500
8.7 13.0 41.4 8.6 28.3
9 36 589 255 468
0.6 2.7 43.4 18.8 34.5
449 699 2,714 703 1,953
6.9 10.7 41.6 10.8 30.0
Total
5,294
100.0
1,357
100.0
6,518
100.0
Wife’s education Primary school graduates Middle-school graduates High-school graduates Some college College graduates
579 892 2,637 323 862
10.9 16.9 49.8 6.1 16.3
9 32 688 280 349
0.6 2.4 50.7 20.6 25.7
571 874 3,270 598 1,205
8.8 13.4 50.2 9.2 18.5
Total
5,293
100.0
1,358
100.0
6,518
100.0
199 402 444 654 325 610 769 814 332 506
3.9 8.0 8.8 12.9 6.4 12.1 15.2 16.1 6.6 10.0
14 106 220 293 105 139 180 175 50 27
1.1 8.1 16.8 22.4 8.0 10.6 13.8 13.4 3.8 2.1
210 507 660 945 418 735 912 966 369 517
3.4 8.1 10.6 15.1 6.7 11.8 14.6 15.5 5.9 8.3
5,055
100.0
1,309
100.0
6,239
100.0
19 199 162 226 534 638 131 214 433 395
0.6 6.7 5.5 7.7 18.1 21.6 4.4 7.3 14.7 13.4
0 80 68 166 62 73 9 13 20 11
0.0 15.9 13.5 33.1 12.4 14.5 1.8 2.6 4.0 2.2
19 279 228 389 566 701 138 226 439 396
0.6 8.3 6.7 11.5 16.7 20.7 4.1 6.7 13.0 11.7
2,951
100.0
502
100.0
3,381
100.0
Husband’s occupation Legislators/senior managers Professionals Technicians Clerks Service workers Sales workers Craft workers Machine operators Simple laborers Agricultural workers Total Wife’s occupation Legislators/senior managers Professionals Technicians Clerks Service workers Sales workers Craft workers Machine operators Simple laborers Agricultural workers Total
116 Doo-Sub Kim is 53.3 percent and 46.3 percent, respectively. In contrast, the proportion of husbands and wives with middle-school education or lower is only 3.3 percent and 3.0 percent, respectively. These are substantially lower than the counterpart figures of 21.7 percent and 27.9 percent for the first group. Distribution of the couple’s occupations within the two groups reveals a pattern consistent with that of educational level. Table 6.2 shows that, for the second group (those married after 1997), clerical workers compose the largest occupational subgroup. In the case of husbands, clerical workers account for 22 percent, followed by technicians (17 percent) and craft workers (14 percent). In contrast, in the first group, the proportion of blue-collar workers is relatively higher: machine operators and craft workers compose 16 percent and 15 percent, respectively. The proportion of agricultural workers is also much larger than in the second group. As for wives, in the second group, clerical workers compose by far the largest occupational subgroup (33 percent), followed by professionals (16 percent) and sales workers (15 percent). The occupational pattern is very different in the first group, where the largest occupational groups are sales workers (21 percent), service workers (18 percent), simple laborers (15 percent), and agricultural workers (13 percent).
Analyses of CEB and post-1998 CEB There are difficulties in analyzing the pattern of differential fertility after 1998. As shown in Table 6.1, the mean value of CEB and post-1998 CEB is as low as 1.9 and 0.4, respectively. The lower the level of the dependent variable, the smaller its deviation and variance become, and thus it is more difficult to find regularities in causal structure. Despite these difficulties, the arguments set out here can be strengthened if we find significant relationships or even clues to these regularities. It is apparent in Figure 6.3 that CEB and post-1998 CEB according to educational attainment of the couple reveal contrasting patterns. As the level of a couple’s education goes up, CEB tends to decrease. However, college graduates show slightly higher CEB than those of the next group in the hierarchy. This pattern is found to be consistent for both husbands and wives. Though wives with no education show relatively lower CEB than primary school graduates, this was calculated based on only 36 cases, and cannot be generalized. In Figure 6.3, post-1998 CEB according to a couple’s level of education reveals an absolutely different pattern. Couples are more likely to have a greater number of children as level of education rises. For college graduates, however, CEB was found to be slightly lower than the next group in the hierarchy. This implies that changes in reproductive behavior after the economic crisis have been most drastic among those with middle-school education or lower. It is also indicated that the level of fertility of couples who graduated from college or with higher education decreased more sensitively than those in the next group in the educational hierarchy. My interpretation is that, with a bad economy and serious job insecurity in
The 1997 Asian economic crisis 117 CEB 3.0 2.5 2.0 1.5 1.0 0.5 0.0 no education
primary
middle
high
college
college graduate
CEB by husband's education
CEB by wife's education
Post-1998 CEB by husband's education
Post-1998 CEB by wife's education
Figure 6.3 CEB and post-1998 CEB by education of the couple, 2003. Source: KIHASA (2003).
the past decade, less-educated couples with low SES have not had many choices other than reducing their family size. Their relative deprivation becomes greater in view of the process of increasing socioeconomic inequality, and they tend to emulate the status-related and child-rearing expenditures of those with higher SES rather than having more children. Increasing socioeconomic inequality has also led the highly educated couples with high SES to undertake more statusdifferentiating expenditures to fulfill their aspirations. In particular, their psychological burden of child-education expenditures has become much more serious in the past decade. As socioeconomic inequality increases, couples are more likely to be exposed and receptive to the status-related and status-differentiating expenditures, and thus reduce their level of fertility drastically. This is particularly true for those living in large cities. If this situation continues, it is postulated that the pattern of socioeconomic differentials in fertility is likely to be substantially different in the future. Analysis of differential fertility according to couple’s occupation reveals a pattern similar to the one above. As shown in Figure 6.4, agricultural workers and simple laborers show the highest and the next highest CEB, respectively. As occupational prestige goes up, CEB tends to be smaller. However, legislators and senior managers with the highest prestige reveal relatively higher CEB than those in the next prestigious group. The level of fertility for wives with a clerical occupation turns out to be the lowest. Figure 6.5 indicates that the pattern of post-1998 CEB according to couple’s occupation is remarkably different from that of CEB. The most drastic changes
CEB 3.0
2.5
2.0
1.5
1.0 rs
an
m sr
ns
ls
e ag
i
ss
fe
o pr
a on
a ici
n
ch
te
ks
r cle
s
er
ice
k or
w
v er
les
w
sa
s
s rs rs or er ke to rk ra lab wor o e e w l pl op ft ra im tu cra ne l s i u h ric ac m ag CEB by wife's occupation s
er
k or
CEB by husband's occupation
Figure 6.4 CEB by occupation of the couple, 2003. Source: KIHASA (2003).
CEB 0.8
0.6
0.4
0.2
0.0 s
er
sr
g na
a
m
ls
p
si
es
f ro
a on
ch
te
s
an
ci ni
ks
r cle
e
se
ic rv
s
w
s
le sa
s
er
er
k or
k or
w
Post-1998 CEB by husband's occ
t
af
cr
rs
ke
r wo ac
m
e
n hi
to
op
a er
rs
e l pl ra sim ultu ric ag
rs
ke
r wo
Post-1998 CEB by wife's occ
Figure 6.5 Post-1998 CEB by occupation of the couple, 2003. Source: KIHASA (2003).
or
lab
The 1997 Asian economic crisis 119 in reproductive behavior after the economic crisis are found among those with less-prestigious occupations, including agricultural workers and simple laborers, followed by the most prestigious group – legislators and senior managers. It should also be noted that the post-1998 CEB for wives with service, sales, and craft-related occupations has declined markedly after the economic crisis. A similar pattern of recent changes in reproductive behavior is observed when fertility differentials by working status of the couple are analyzed. If we assume that the sequence of working status on the x axis in Figure 6.6 reflects the prestige or stratification hierarchy, the pattern observed in this figure becomes similar to the pattern of fertility differentials by occupation in Figure 6.4. The pattern of relationship between the working statuses of the couple and the post-1998 CEB in Figure 6.7 also appears to be similar to the one in Figure 6.5. Though unpaid family workers turn out to be an exception to the above-mentioned patterns, the number of husbands for calculation in this category was as small as 26 in Figure 6.6, and 27 in Figure 6.7, and thus cannot be generalized. In sum, it can be concluded from Figures 6.6 and 6.7 that those with working status of low prestige and those with the highest level of fertility have reduced their CEB most drastically after the 1997 economic crisis. In the next place, level of fertility of employers, whose prestige or hierarchical rank is recognized as the highest, decreased substantially. In other words, the extent of decrease in the level of fertility after the economic crisis is positively associated with level of fertility: the higher the level of fertility, the more drastic the decline in fertility. The pace of decline has been relatively slow for those with low levels of fertility. So far, the pattern of differential fertility according to husband’s education, occupation, and working status turns out to be consistent with the pattern CEB 2.4 2.2 2.0 1.8 1.6 1.4 employer
regular employees
temp/daily employees
CEB by husband's status of working
own account unpaid family workers workers CEB by wife's status of working
Figure 6.6 CEB by working status of the couple, 2003. Source: KIHASA (2003).
120 Doo-Sub Kim CEB 0.6 0.5 0.4 0.3 0.2 0.1 0.0 employer
regular employees
Post-1998 CEB by H's status of working
temp/daily employees
own account unpaid family workers workers Post-1998 CEB by W's status of working
Figure 6.7 Post-1998 CEB by working status of the couple, 2003. Source: KIHASA (2003).
according to the wife’s equivalent characteristics. In Figure 6.8, however, the post-1998 CEB by place of work reveals different patterns between husbands and wives. For wives, recent fertility is found to be highest among government officials and teachers, followed by those working at government-supported organizations and other public organizations. Though the difference is not substantial, as the size of the place of work becomes smaller, the level of recent fertility tends to be lower. Due to high unemployment and labor-market insecurity, government officials, teachers, and employees of the public organizations are considered to have the most secure and stable jobs, and therefore they have recently become the most popular occupations (Dong-A Daily News, 2006). It is believed that the pattern of recent fertility is supported by these changes in occupational preference. A similar explanation can be applied to the negative relationship between the size of workplace and the level of fertility. In contrast, the level of recent fertility tends to be highest among husbands working in mediumsized companies. In Figure 6.8, differential fertility according to husband’s place of work is not substantial except for those working at private companies and the category “others.” The markedly low level of fertility among husbands working in private companies implies that they are the group most seriously affected by recent economic stagnation. Income differentials in fertility have been an important area of research since the foundation of demography (Becker, 1960; Freedman, 1963; Mincer, 1963; Cain and Weininger, 1973; Willis, 1973; Leibenstein, 1974, 1975; Easterlin, 1975; Leriden, 1976; Mueller and Short, 1983). During the past decade or two, however,
The 1997 Asian economic crisis 121 CEB 0.6 0.5 0.4 0.3 0.2 0.1 govnt offic/ teachers
public org
large comp
Post-1998 CEB by H's place of work
mid-sized comp
small comp
private comp
others
Post-1998 CEB by W's place of work
Figure 6.8 Post-1998 CEB by place of work of the couple, 2003. Source: KIHASA (2003).
the theoretical basis and empirical findings of income differentials in fertility in Korea have not been sufficiently explored (Kim, 1987; Kim and Song, 2006; Kim et al., 2006). Differential fertility according to the level of household income and family assets is explored in this chapter. In Figure 6.9, household income shows a V-shaped pattern with fertility. At very low household-income levels, CEB tends to fall as household income rises. Beyond a certain level of income, however, the relationship between household income and CEB is positive. If income could be divided into more refined categories, the curve would very likely turn into a U-shaped curve.1 It is considered that the level of family assets reflects household income over a longer period of time or is a cumulative concept of household income. For those who own housing in an urban area, this variable is very much affected by the price of housing. In Figure 6.9, the level of family assets reveals a similar relationship pattern to that of household income. It appears that the pattern of the post-1998 CEB in Figure 6.10 is not consistent with the pattern according to education, occupation, and working status of the couple. Although the patterns of relationships in Figures 6.5, 6.7 and 6.10 appear to be similar to each other, it should be noted that the hierarchical order of household income and family assets on the x axis in Figure 6.10 is constructed in the opposite direction compared to those in Figures 6.5 and 6.7. The inconsistency, however, may be mainly due to a positively skewed distribution of household income and family assets. If we transfer the variables of household income and family assets on the x axis onto a natural logarithmic scale, the resultant relationship pattern becomes a bit closer to the results of analysis in Figures 6.5 and 6.7.
CEB 2.3 2.2 2.1 2.0 1.9 1.8 1.7 1.6
CEB by household income
+m
9 10
00
99
9
0– 50
0–
49
9 30
20
0–
29
9 19
9
0– 15
10
0–
14
9 –9 50
9 –4
9
30
–2 10
0–
9
1.5
CEB by family assets
Figure 6.9 CEB by household income and family assets, 2003. Source: KIHASA (2003). CEB 0.6 0.5 0.4 0.3 0.2 0.1
Post-1998 CEB by household income
00 10
Post-1998 CEB by family assets
Figure 6.10 Post-1998 CEB by household income and family assets, 2003. Source: KIHASA (2003).
+m
9 99 0–
50
30
0–
49
9
9 29
20
0–
9 19 0–
15
10
0–
14
9
9 –9 50
9 –4 30
9 –2 10
0–
9
0.0
The 1997 Asian economic crisis 123
Changes in fertility of those first married after 1997 To strengthen our findings on the recent pattern of differential fertility, another round of comparative analysis was conducted in this study. After selecting married women aged 20–49 from the 2003 Korean National Fertility Survey, the sample was divided into two groups. The first group contained 5,301 women who married before 1997 or who remarried regardless of their timing of marriage. The second group included 1,359 women who married for the first time after 1997. It is postulated that the level of fertility of the first group would primarily reflect reproductive behaviors before the 1997 economic crisis. Although CEB of the first group includes children born after 1998, the mean value of the post-1998 CEB was found to be as low as 0.2. As shown in Table 6.1, the mean CEB of the first group was estimated as 2.1. The second group, composed of those married for the first time after 1997, is homogeneous in its age composition and socioeconomic characteristics. It is found that 76.2 percent of husbands and 85.6 percent of wives belong to the 25–34 year age group. Thus, the proportions of those with low education are very small, as is the proportion working as legislators, senior managers, or agricultural workers. As the duration of their working life tends to be relatively short, the proportions of employers, the very rich, and those with upper-class identity are also found to be very small. These characteristics of the second group should be noted in the process of comparative analysis with the first group. It is not surprising to find that the second group reveals a pattern of socioeconomic differentials in fertility similar to the pattern of the post-1998 CEB discussed above: both reflect reproductive behavior during the period from 1997– 2003. Thus, some of the results from the analysis of the second group that are very similar to the pattern of the post-1998 CEB and which do not provide further interpretations, are not presented here. The pattern of CEB of the two groups by wife’s place of work is presented in Figure 6.11. The first group – those married before 1997 or remarried – does not show significant differences in the level of fertility. For the second group – those married for the first time after 1997 – however, the fertility level of government officials, teachers and those who work in private companies turns out to be significantly higher than others. This phenomenon is already confirmed in Figure 6.8, and can be related to job security. It is generally agreed that government officials and teachers have relatively stable jobs, less affected by labor-market deregulation and stagnation. These jobs are also considered to have a less-unfavorable atmosphere in terms of sex discrimination against women and are more accommodating with regard to maternity leave. Figure 6.11 differs considerably from Figure 6.8 in showing that those who work in private companies have a higher level of fertility. In general, recent fertility is generally low for those who work in private companies. However, this phenomenon does not seem to apply to those who married for the first time after 1997. Along with working status, occupation, and place of work, the duration of work after marriage for the wife is considered to be an important variable in fertility
124 Doo-Sub Kim CEB 2.5 2.0 1.5 1.0 0.5 0.0 govnt offic/ public teachers org
large comp
mid-sized comp
Married before 1997 or remarried
small comp
private comp
others
First married after 1997
Figure 6.11 CEB of the two groups by wife’s place of work, 2003. Source: KIHASA (2003).
analysis. Duration of work after marriage is a primary explanatory variable of fertility for women, and at the same time, is a resultant variable affected by the number of children she already has. In Figure 6.12, the duration of work shows a shallow U-shaped pattern with CEB. Without a doubt, a working wife tends to reduce the couple’s level of fertility by delaying or avoiding child-bearing. CEB 2.5
2.0
1.5
1.0
0.5 0
1–5 m 6–11 m 1–2 yr
3–4 yr
Married before 1997 or remarried
5–9 yr 10–14 yr 15–19 yr 20+ yr First married after 1997
Figure 6.12 CEB by the duration of work after marriage for the wife, 2003. Source: KIHASA (2003).
The 1997 Asian economic crisis 125 However, many empirical studies in Korea have found a positive relationship between the work experience of the wife and fertility (Kim, 1986). An increasing tendency in fertility beyond a certain level of duration of work can be easily understood if we assume an opposite direction of causality. If a woman has enough children already and needs to support them, she is likely to work for a longer duration out of economic necessity (Kim, 1986, 1987; Kwon and Kim, 2002). Recently in Korea, many young women tend to work to supplement costs for child-rearing and private education as well as to satisfy their own self-realization and selfaccomplishment. It can also be noted that the positive relationship between the work duration of the wife and fertility is partly due to an age effect. As the duration of work increases, the woman’s age is likely to be higher, and thus the number of children tends to be larger. Due to the very low level of fertility, it is expected to be difficult to perceive the effect of family assets on CEB for those who married for the first time after 1997. As shown in Figure 6.13, however, the results of the analysis are found to be contrary to our expectations. The first group does not reveal a distinctive pattern of relationships between the two variables. For the second group of those married for the first time after 1997, CEB tends to increase as the level of family assets rises. It is surprising to find in this figure that the second group shows a similar pattern to that of CEB in Figure 6.9 rather than that of the post-1998 CEB in Figure 6.10. CEB 2.5
2.0
1.5
1.0
Married before 1997 or remarried
Figure 6.13 CEB of the two groups by family assets, 2003. Source: KIHASA (2003).
+m 00
10
99
9
9
0– 50
30
0–
0–
49
29
9
9 20
19 0– 15
10
0–
14
9
9 –9
–4 30
50
9
9 –2 10
0–
9
0.5
First married after 1997
126 Doo-Sub Kim In an effort to find plausible explanations for these results, various additional analyses were conducted, though not successfully. The observed pattern of relationships from the second group of those married after 1997 can be alternatively explained as follows. CEB does not reveal substantial changes up to an asset level of 200 million won, and tends to rise only after this threshold level. It is found that most of those who own assets beyond this threshold level are house owners. Therefore, the pattern of relationship in Figure 6.13 implies that house owners are more likely to have more children than non-owners. Housing is the most important of family assets. It is very likely that level of family assets and status of house ownership show a similar pattern of differential fertility. Although house owners in the first group show the highest CEB in Figure 6.14, the differences in the level of fertility are not substantial. In contrast, for the second group – those married after 1997 – CEB tends to be distinctively higher among the categories of house owners and “others,” and lowest among those with a short-term (monthly) lease. A large portion of the “others” category includes couples living with their parents. Although they do not own housing, their fertility-related norms, values, attitudes, and behaviors are not likely to be markedly different from those of house owners. Due to pressure from their parents’ generation to have a child or children as well as the possibility of getting a great deal of help in childcare, these couples are less likely to delay or avoid child-bearing. It can therefore be argued that status of house ownership and level of fertility are negatively related for couples in the second group. CEB 2.5
2.0
1.5
1.0
0.5 own
lease-long term
lease-semi long term
Married before 1997 or remarried
lease-monthly
others
First married after 1997
Figure 6.14 CEB of the two groups by status of housing ownership, 2003. Source: KIHASA (2003).
The 1997 Asian economic crisis 127
Summary and concluding remarks In response to the very low level of fertility and the socioeconomic consequences of the resultant aging population, the Korean government is now in the process of establishing various pro-natal policies aimed at boosting TFR up to the target of 1.6, the average value of OECD countries. Major policy options focus on reducing the economic costs of child-rearing by providing child allowances, tax exemptions based on the number of children, childcare support systems, etc. Another set of pro-natal policy options is to consolidate women’s roles as mothers and workers by improving the flexibility of the labor market. A new policy direction is to improve the social status of women and gender equity, and to guarantee the involvement of fathers in childcare and rearing responsibilities. A high priority is to provide extended maternity leave, childcare facilities for female workers, more flexible working hours and short-term leave for childcare purposes (MOHW, 2006). Without a doubt, the policy directions mentioned above and related programs would be helpful in improving the childcare environment as a welfare policy. A question could still be raised whether the above approach and related programs would be effective in boosting fertility in Korea. Evidence indicates that in several European countries, including France and Luxembourg, policies increasing the compatibility of women’s work and childcare as well as economic support for childcare may have been effective in increasing the level of fertility (Gauthier and Hatzius, 1997). For example, the French government spent 56.4 billion euros on social-protection benefits in “family” and “maternity” functions in 2003. The resultant increase in TFR attributed to the fiscal measures is estimated as being between 0.1 and 0.2 (Héran, 2005). The effectiveness of a policy depends largely on its relevance to the country’s socioeconomic context. For example, Korea does not have the same level of financial resources as France, and it will be almost impossible to apply the costly French program to Korea. The rapid pace of population aging in Korea over the coming decades is expected to cause a serious drain on the public budget as more funds are required to support the elderly. It is thus not reasonable to assume that public transfers of resources will be enough to increase the maternity support budget to a sufficient level in the years to come. Another question is whether these policy directions and options are established based on results from refined analyses of changes in Korean fertility. Rather than spend limited public funds to support a variety of ineffective programs, the government must concentrate its resources on options and programs that attack the core of the low-fertility crisis. Priority should also be given to select target groups for each program, and to motivate these people to get married at earlier ages as well as to have more children. This process should be based on results from analyses of recent changes in fertility. This chapter has attempted to explore the effects of the 1997 Asian economic crisis on the level of Korean fertility and the pattern of socioeconomic differentials. Based on an analysis of data from the 2003 Korean National Fertility Survey,
128 Doo-Sub Kim the level of fertility according to the socioeconomic status of the couple, including educational level, occupation, working status, income, etc., before and after the economic crisis has been investigated. The major findings of this analysis can be summarized as follows. First, the socioeconomic status of the couple, including educational level, occupation, and working status, showed a reversed J pattern with CEB. As educational attainment goes up, the level of fertility tends to be lower. However, the level of fertility of those with the highest education is relatively higher than that of the next group in the educational hierarchy. Occupational differentials in fertility reveal a similar pattern. Agricultural workers show the highest CEB. As occupational prestige goes up, CEB tends to be smaller. However, the relatively small group of legislators and senior managers, with the highest prestige, reveals relatively higher CEB than those in the next group in the prestige hierarchy. As far as working status is concerned, with minor exceptions, the highest CEB is found among unpaid family workers, which is followed by own account workers, employers, and regular employees. Household income is found to show a V-shaped pattern with CEB. As household income rises at the low level, CEB tends to decline. After a certain level of income, however, the two variables reveal a positive relationship. Second, recent differentials in fertility after the economic crisis were found to contrast with the pattern above. Decrease in fertility has been most drastic among those with a high level of fertility, and relatively slow for those with a low level of fertility. This pattern turns out to be consistent when the educational level, occupation, and working status of the couple are used as indicators of socioeconomic status. Decrease in fertility has been most drastic among those with the lowest socioeconomic status such as middle-school graduates or lower, agricultural workers, simple laborers, unpaid family workers, and own account workers. Those with the highest socioeconomic status such as college graduates, legislators and senior managers, and employers have also shown a marked decrease in fertility after the economic crisis. As a result, level of recent fertility turns out to be highest among those with upper-middle socioeconomic status, followed by those with the highest socioeconomic status, and finally those with the lowest status. Interestingly, as far as wife’s place of work is concerned, post-1998 CEB is found to be highest among government officials and teachers, which is followed by those working at government-supported organizations and other public organizations. As the size of the workplace becomes smaller, the level of recent fertility becomes lower. This pattern can be explained in terms of job security and stability. Findings of this study on socioeconomic differentials in fertility as well as recent changes in the level of fertility can be useful in projecting demographic changes in the years to come. They also have implications for long-term and short-term pro-natalist policies and their primary target groups. Further research applying the framework of this analysis to Thailand, Malaysia, or Taiwan, countries that experienced a similar economic crisis in the late 1990s, may contribute to enhancing our understanding of the causal mechanisms of fertility.
The 1997 Asian economic crisis 129 What it has not been possible to establish in this analysis is the extent to which the dramatic fertility decline among young women represents a postponement of births and the extent to which they foreshadow a reduction in completed family size. This key issue requires further study. Finally, it should be noted that pro-natalist policies will probably not be effective in the short term, and much less effective compared to the anti-natalist policies of the past. No advanced countries have been markedly successful in boosting the level of fertility in recent times. The current level of lowest-low fertility in Korea is mainly due to a strong sense of job insecurity among young couples combined with an unpredictable economic recovery. The small amount of economic support for childcare and some family-friendly workplace arrangements for mothers are not likely to have a significant influence on young couples who have strong desire for self-realization and relief from the traditional childcare load. To encourage young people to marry earlier and have children, overall improvements in the security of the labor market and in the educational system are necessary in the long run.
Acknowledgments This work was supported by the Korea Research Foundation Grant funded by the Korean Government (MOEHRD) (KRF-2005-078-BS0004).
Notes 1 Using the 1974 Korean National Fertility Survey, Kim (1987) found a cubic relationship between income indicators (expected family income and family assets) and CEB. Among those in the lowest socioeconomic status (SES) group, the relationship between income and CEB is positive. Those in the middle SES group are likely to have fewer CEB as income increases. For the highest SES group, CEB is likely to increase slightly as income increases. He argued, however, that the range of the fertility curve varies with respect to the socioeconomic level of a specific country or target population.
References Bank of Korea. 2006. Economic Statistics (ECOS). Seoul: Bank of Korea. Downloadable from the On-Line Statistics Database: http://ecos.bok.or.kr/. Becker, Gary S. 1960. “An economic analysis of fertility,” in Demographic and Economic Change in Developed Countries, edited by National Bureau of Economic Research. Princeton, NJ: Princeton University Press: pp. 209–231. Blake, Judith. 1967. “Income and reproductive motivation,” Population Studies 21(3): 185–206. Cain, Glen G. and Adriana Weininger. 1973. “Economic determinants of fertility: results from cross-sectional aggregate data,” Demography 10(2): 205–223. Dong-A Daily News. 2006. “‘Occupation rather than personality’; criteria for spouse selection has changed in great deal during the past 10 years”. Retrieved August 5, 2006 from: http://www.donga.com. Easterlin, Richard A. 1975. “An economic framework for fertility analysis,” Studies in
130 Doo-Sub Kim Family Planning 6(3): 54–63. Eun, Ki-Soo. 2003. “Understanding recent fertility decline in Korea,” Journal of Population and Social Security, Supplement to Volume 1. Tokyo: National Institute of Population and Social Security Research, The Ministry of Health and Welfare. Downloadable from: http://www.ipss.go.jp/index-e.html. —— 2007. “Lowest-low fertility in the Republic of Korea: causes, consequences and policy responses,” Asia-Pacific Population Journal 22(2): 51–72. Freedman, Deborah. 1963. “The relation of economic status to fertility,” American Economic Review 53(3): 414–426. Gauthier, Anne and Jan Hatzius. 1997. “Family benefits and fertility: an econometric analysis,” Population Studies 51(3): 295–306. Héran, François. 2005. “Demography and population policy: is it possible to draw any lesson from the French case?,” presented at the KIPP (Korea Institute of Population Problems) Demographic Forum, Seoul, Korea, April 8, 2005. IMF. 2006. World Economic Outlook Database. Washington, DC: International Monetary Fund. Downloadable from: http://www.imf.org/. Jun, Kwang-Hee. 2004. “Chapter 3: Fertility,” in Doo-Sub Kim and Cheong-Seok Kim (eds), The Population of Korea. Daejeon: Korea National Statistical Office: pp. 65–89. KIHASA. 2003. The 2003 National Fertility and Family Health Survey. Seoul: Korea Institute for Health and Social Affairs. Kim, Doo-Sub. 1986. “Working experience of married woman and fertility in Korea,” Bulletin of the Population and Development Studies Center 15: 19–30. Seoul: Seoul National University. —— 1987. Socioeconomic Status, Inequality and Fertility. Seoul: Population and Development Studies Center, Seoul National University. —— 2004. “Chapter 1: population growth and transition,” in Doo-Sub Kim and CheongSeok Kim (eds), The Population of Korea. Daejeon: Korea National Statistical Office: pp. 1–32. —— 2005. “Theoretical explanations of rapid fertility decline in Korea,” The Japanese Journal of Population 3(1): 2–25. Tokyo: National Institute of Population and Social Security Research. Kim, Doo-Sub and Yoo-Jean Song. 2006. Effects of the “IMF Economic Crisis” on the Level and Causal Mechanism of Korean Fertility, Report to the Asan Foundation. Seoul: Institute of Population and Aging Research, Hanyang University. Kim, Tai-Hun, Sam-Sik Lee, and Dong-Hoy Kim. 2006. “Fertility differentials by demographic and socioeconomic characteristics: analysis of Korean population census data,” Korea Journal of Population Studies 29(1): 1–23 (in Korean). KNSO. 2006a. Korean Statistical Information System (KOSIS). On-Line Statistics Database: http://kosis.nso.go.kr/. Daejeon: Korea National Statistical Office. —— 2006b. Birth and Death Statistics, 2005. Daejeon: Korea National Statistical Office. —— 2006c. Marriage and Divorce Statistics, 2005. Daejeon: Korea National Statistical Office. Kwon, Tai-Hwan and Doo-Sub Kim. 2002. Understanding Population. Second Edition. Seoul: Seoul National University Press (in Korean). Leibenstein, Harvey. 1974. “An interpretation of the economic theory of fertility: promising path or blind alley?,” Journal of Economic Literature 12(2): 457–479. —— 1975. “The economic theory of fertility decline,” Quarterly Journal of Economics 89(1): 1–31.
The 1997 Asian economic crisis 131 Leriden, H. 1976. “The role of economic factors in birth-rate trends and fluctuations,” in A.J. Coale (ed.), Economic Factors in Population Growth. New York, NY: John Wiley and Sons: pp. 179–197. Mincer, Jacob. 1963. “Market prices, opportunity costs and income effects,” in C. Christ (ed.), Measurement in Economics: Studies in Mathematical Economics and Econometrics in Memory of Yehuda Grunfeld. Stanford, CA: Stanford University Press: pp. 67–82. MOHW. 2006. The First Basic Policy for Low Fertility and Aged Society, press release from the Ministry of Health and Welfare. Mueller, Eva and Kathleen Short. 1983. “Effects of income and wealth on the demand for children,” in Rodolfo A. Bulatao and Ronald D. Lee (eds), Determinants of Fertility in Developing Countries. Vol. 1: Supply and Demand for Children. New York, NY: Academic Press: pp. 590–642. OECD. 2004. OECD Statistics: Labor Force. Paris: Organization for Economic Co-operation and Development. Downloadable from: http://www.oecd.org. Park, Kyung Ae. 2006. “Fertility decline in Korea: trends, determinants, and policy implications,” Population and Society 2(2): 77–113. Population Reference Bureau. 2006. 2005 World Population Data Sheet. Washington, DC: Population Reference Bureau. Willis, Robert J. 1973. “A new approach to the economic theory of fertility behaviour,” Journal of Political Economy 81(2): S14–64. Yoo, Sam-Hyun. 2006. “On regional fertility differentials: understanding the causal mechanisms of low fertility in Korea,” Population and Society 2(2): 53–76 (in Korean).
7
Ultra-low fertility in Hong Kong A review of related demographic transitions, social issues and policies to encourage childbirth Paul Yip, C. K. Law, and Karen Cheung
Introduction Low fertility as a challenge to society At the beginning of the twenty-first century, low fertility and its impact in slowing population growth were of great concern to governments in many parts of the developed world. The total fertility rate (TFR) is one of the most commonly used measures to assess fertility of the Hong Kong population.1 A total fertility rate of 2.1 children per woman is conventionally regarded as the level of inter-generational replacement. In order to ensure exact replacement, a generation of women must each give birth to slightly more than two children.2 Hong Kong has had one of the lowest total fertility rates in the world over the last ten years, dropping to 0.927 in 2004. Indeed, when excluding babies born by non-Hong Kong residents, which constitute about 24 percent of the total births, the TFR was only about 0.76 in 2004. Low fertility is a recent phenomenon in most developed economies. After the Second World War, the “baby boom” was considered a major challenge all over the world. During this period the importance of and need to limit population growth began to receive more public attention in Hong Kong, and discussions came to a peak at about the time of the World Population Year 1974, during which several local activities were sponsored by the Family Planning Association of Hong Kong (FPAHK). The high-priority issues mainly focused on the baby boom and the need for birth control, family planning, and limitations on family size. The large influx of immigrants from mainland China in the 1960s and 1970s was also a great concern. However, over the three decades since 1976, Hong Kong has experienced a rapid fall in fertility. The total fertility rate dropped steadily from the replacement level down to the lowest level in the world. An important challenge in Hong Kong today and in the near future is to curb falling fertility rates, which in the absence of migration will lead to a downward spiral in population size. Declining fertility in Hong Kong since the 1970s What are the factors that account for the declining fertility rate in Hong Kong? Some early studies (Choi and Chan, 1973; Coughlin and Coughlin, 1963; Fan
Ultra-low fertility in Hong Kong 133 and Lee, 1986; Freedman and Adlakha, 1968; Freedman et al., 1969; Freedman et al., 1970; Hong Kong Census & Statistics Department, 1982; Ng, 1978; United Nations, 1974) associated fertility decline with changing social and economic conditions in Hong Kong. The marked reduction in family sizes can be explained partly by industrialization during the 1960s and 1970s, which had various consequences: 1 2 3
a rise in the labor-force participation of females due to rapid economic development and improved education urbanization favouring the formation of smaller families a change in people’s aspirations concerning personal financial rewards and standards of living, which together with the cost of raising children made large families too expensive.
These factors created new pressures on reproductive behavior. Couples took a more pro-active attitude towards limiting family size (Hong Kong Census & Statistics Department, 1979). An organized and persistent effort by the Government and private agencies to promote family planning also contributed to the decline in fertility. According to previous official reports, the decline in birth rates in the 1970s was due to a combination of factors. These included a decrease in the number of women of child-bearing age; a decrease in the proportion of women who marry in the productive child-bearing years, the postponement of child-bearing, and a marked fall in the fertility rate of married women (Hong Kong Census & Statistics Department, 1972 and 1979). The fertility report compiled by Hong Kong Census and Statistics Department (Hong Kong Census & Statistics Department, 1982 and 1983) also highlighted that declines in fertility over time were associated with the advanced age of women who got married, which was a gradual continuation of the trend toward a later mean age of mothers at first birth, as well as the more widespread availability of contraception and its greater effectiveness. The 1982 Special Review on Hong Kong Fertility Trends stated that those women born in 1956 (who were 25 years of age in 1981 and would be 49 years of age in 2005) would act positively to limit the size of their families and would be the first generation to experience “below-replacement level” fertility. However, once those women who preferred to have large numbers of children had passed their reproductive age and were replaced by younger women who now generally accept the practice of family planning, the fertility rate was expected to drop substantially (Hong Kong Census & Statistics Department, 1982). More recently, changing marital status distribution of child-bearing-age population and new family combinations have required special attention. Some countries have reported a considerable rise in childlessness, and in some cases (such as Germany) this has been identified as the main cause of fertility decline (in Germany, the incidence of childlessness rose from 11 percent among the cohort of women born in 1940 to 32 percent among the cohort born in 1965 (Economic and
134 Paul Yip, C. K. Law, and Karen Cheung Social Research Institute, 2004). The 1982 special review (Hong Kong Census & Statistics Department, 1982) had already suggested that the probable increase in the proportion of women remaining childless was due to the increasing popularity of no-child families and spinsterhood, and these factors would outweigh the effects arising from advances in medical technology which have given reproductive capabilities to some infertile women and some sterile husbands. It was stated, however, that the proportion could not be expected to increase continuously into the future or rise beyond a certain level. A proportion of 15 percent, based on the experience of some developed countries of women remaining childless throughout their lifetime, was regarded as the ultimate limit. In Hong Kong, the proportion of childlessness had reached a level of 15.2 percent in 2002, a rise of 13.1 percentage points from its level of 2.1 percent in 1982 (Family Planning Association of Hong Kong, 2002). The organization of the present study The review above shows that fertility transition is influenced by various factors; both individual preference and the wider socio-cultural context as well as social policies have to be taken into account. In order to adopt timely and appropriate policies that aim at opening the window on child-bearing and fostering increases in fertility, it is very important to understand the factors and mechanisms that underlie people’s fertility decisions. The section below titled “Fertility in Hong Kong” will provide an overview of fertility trends in Hong Kong. Like many other Asian regions, child-bearing within wedlock remains the mainstream. Therefore, trends in marriage are highly related to fertility as well, and this will be thoroughly explored in “Marriage and marital fertility in Hong Kong”. “Other demographic factors related to Hong Kong’s fertility” spells out other demographic factors affecting fertility in Hong Kong. All the above seem to suggest obstacles to boosting fertility. In “Concerns of Hong Kong people on bearing and rearing children,” by looking into Hong Kong people’s attitudes towards child-bearing, their aspirations and concerns, we will understand better what prevents couples in Hong Kong from having children. In “Policy measures to encourage childbirth in Hong Kong,” we then summarize different policies and practices that the Hong Kong Government has implemented for encouraging childbirth. The implications and effectiveness of those policies are also explored. The chapter is completed with a discussion of principles and methods for meeting the challenge of low fertility.
Fertility in Hong Kong There are various indicators of fertility. In this section, the number of births and TFR in Hong Kong across time will be reviewed. A comparison of age-specific fertility rates allows us to see the changes in different age groups more clearly, while figures regarding the number of births, mean age of parity, births from
Ultra-low fertility in Hong Kong 135 Mainland women, and completed and cohort fertility rates help to look at the issue from different perspectives. Number of births and total fertility rate Figure 7.1 shows the number of live births and the TFR in Hong Kong over the period from 1961 to 2004. The annual number of live births in Hong Kong has dropped by about 55 percent from 110,884 in 1961 to 49,796 in 2004. In 1961 and 1965, the TFR stayed at its highest level of 5.17 and 4.46 births per woman respectively (Hong Kong Census & Statistics Department, 1975), but by 1976, it had already fallen to 2.48 births per woman in less than fifteen years. Since then, Hong Kong has experienced a more rapid fall in fertility. The TFR fell steadily from the replacement level in early 1980 (Kwong, 1993; Tu, 2002; Yip et al., 2001; Yip and Lee, 2002) and then dipped further to 0.927 in 2004 (Hong Kong Census & Statistics Department, 2005). The total fertility rate decreased by over 73 percent from 3.46 live births per 1,000 women in 1971 to 0.93 in 2004. Age-specific fertility rate The age-specific fertility rate (ASFR) is the number of live births occurring to women (excluding foreign domestic helpers) in a given age group during a calendar year, often expressed as the number of births per 1,000 women at mid-year in 120000 Number of Births
80000
4.0
60000
3.0
TFR
5.0
Replacement level (2.1) 40000
2.0
20000
1.0
0
TFR 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
Live births
100000
6.0
0.0
Figure 7.1 Hong Kong: number of live births and TFR, 1961–2004. Source: Hong Kong Census & Statistics Department (1961, 1966, 1975, 1983, 2005). Note: TFR in 1961 and 1965 cover only each quinquennial age group between 15–44. Since 1971, TFR is computed by single-age group from age 15–49 for women.
136 Paul Yip, C. K. Law, and Karen Cheung
Live births per 1,000 women
300 250 200 150 100 50 0 15–19
20–24
25–29
30–34
35–39
40–44
1971
1976
1981
1986
1991
1996
2001
2004
45–49
Figure 7.2 Age-specific fertility rates of Hong Kong, 1971–2004. Source: Hong Kong Census & Statistics Department (1983, 2005).
these age groups. The ASFRs declined across all age groups between 1971 and 2004. For Hong Kong women in the younger age groups of 15–19 and 20–24, the fertility rates decreased by 78 percent and 79 percent respectively over the past 35 years (Figure 7.2). For women in the age groups of 25–29 and 30–34, the fertility rates also fell by 75 percent and 64 percent over the same period. This is a genuine reduction in fertility rate rather than a tempo effect. Number of births and mean age of parity As shown in Figure 7.3, there was an upward trend in the proportion of women having a first birth. It increased by about 8.4 percentage points, from 43.7 percent in 1981 to 52.1 percent in 2003.3 The proportion of women having a second birth has stayed at about 37 percent over the past two decades, with no sign of increase. However, there was a drastic reduction in women having a third birth over this period, the proportion decreasing dramatically by almost 12.7 percentage points, from 22.8 percent in 1981 to 10.1 percent in 2003. Figure 7.4 shows a gradual upward trend in the mean age of mothers by live birth order from 1981 to 2003. The mean age for women having their first live birth increased by 20 percent (5 years), from 24.8 years in 1981 to 29.8 years in 2003. The increase in the mean age was also similar for women having their second live birth (5 years), from 27.2 years in 1981 to 32.2 years in 2003. The mean age at the third and fourth live birth increased by 15.3 percent (4.5 years) and 10.4 percent (3.2 years) respectively.
Ultra-low fertility in Hong Kong 137 Percentage of live births by order of live birth
60% 50% 40% 30% 20% 10% 0% 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 1st birth
3rd birth +
2nd birth
Figure 7.3 Hong Kong: percentage distribution of live births by live-birth order, 1981– 2003. Source: Hong Kong Census & Statistics Department (2002) and Micro Known Birth Dataset.
36
Mean age at birth
34 32 30 28 26 24 1981
1986
1st parity
1991
2nd parity
2001
1996
3rd parity
4th parity
Figure 7.4 Hong Kong: mean age of mothers by live-birth order, 1981–2003. Source: Hong Kong Census & Statistics Department (2002).
138 Paul Yip, C. K. Law, and Karen Cheung Births from Mainland women Recently, there has been a significant change in women giving birth in Hong Kong. About 24 percent of births in 2004 were from non-Hong Kong residents and this has increased to 40 percent in 2006, thus affecting the appropriateness of the previous fertility statistics in describing the fertility level of Hong Kong local residents. With this in mind, the Hong Kong Census and Statistics Department has recently presented a trend booklet of Hong Kong fertility from 1975 to 2004 in order to update and enhance the approach for compiling fertility statistics, taking into account the rapid increase in births by non-local women. Figure 7.5 shows the number of live births in Hong Kong to Mainland women whose spouses are Hong Kong residents (BSHKR), as well as live births in which both parents are Chinese nationals but not Hong Kong residents (BnotHKR). The number of BSHKR has increased by 87 percent from 4,964 in 1991 to 9,285 in 2004. This latest phenomenon of childbirth in Hong Kong could be attributed to the closer ties between Hong Kong and the Mainland of China and is likely associated with a higher number of successful applicants for certificates of no-marriage registration (CAMR)4 during the late 1980s and early 1990s. This will be further explored in the “Marriage and marital fertility” section. 10000
5000
BSHKR
Number of live births (BSHKR)
8000
4000
7000 6000
3000
5000 4000
2000
3000
BnotHKR
2000
1000
Number of live births (BnotHKR)
9000
1000 0 2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
0
Figure 7.5 Number of live births in Hong Kong to Mainland women whose spouses are Hong Kong residents (BSHKR), and live births with both parents being Chinese nationals but not Hong Kong residents (BnotHKR), 2001–2004. Source: (2005).
Ultra-low fertility in Hong Kong 139 Moreover, after the ruling at the Court of Final Appeal in July 2001, babies born in Hong Kong to Chinese nationals have the right of abode in Hong Kong under Article 24 (Para. 1) of the Basic Law of Hong Kong (1990). Since then, the increase in babies born in Hong Kong with both parents being Chinese nationals but not Hong Kong residents (BnotHKR) is unprecedented and requires special attention. As shown in Figure 7.6, the proportion of such births to all births increased by 6.9 percentage points from 0.4 percent in 2001 to 7.3 percent in 2004. There are also no signs indicating that future trends will decrease. In those cases in which couples with live births born in Hong Kong to Mainland women whose spouses are Hong Kong residents (BSHKR), the proportion increased by about 11.4 percentage points from 7.3 percent in 1991 to 18.6 percent in 2004. By contrast, the proportion of live births to Hong Kong residents decreased by 18.7 percentage points from 92.7 percent in 1991 to 74.1 percent in 2004. Completed and cohort fertility rates The TFR measures the average number of children a woman would have, assuming that she will live through her reproductive years (ages 15 to 49) and bear children at rates observed in a particular period at each age. This assumption is not valid because real groups of women will never give birth exactly at these particular rates. To gain a more realistic picture, it is necessary to look at the completed fertility rate. 100%
80%
60%
40%
20%
0% 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 BHKR
BnotHKR
BSHKR
Figure 7.6 Proportion of live births in Hong Kong by parentage, 2001–2004. Source: Hong Kong Census & Statistics Department (2005).
According to a previous study (Tu, 2002), completed cohort fertility for cohorts born between 1930 and 1937 have been relatively stable and remained within an interval ranging from 4.4–5.0 children, except for the 1935 cohort (4.1 children) and 1936 cohort (3.7 children) (Hong Kong Census & Statistics Department), with no marked downward or upward trend in Hong Kong. Therefore, Hong Kong women of these cohorts did not really reduce their fertility levels. They just changed the timing of their fertility. However, after the 1937 birth cohort, the cohort fertility steadily declined from 3.6 children (1938 cohort) to 2.9 children (1947 cohort), and further down to 2.1 children (1952 cohort). These cohorts did, in fact, contribute to Hong Kong reaching low levels of fertility, especially for the TFR after the mid-1980s. Figure 7.7 shows that the average number of births a 50-year-old woman from the 1936 birth cohort had was 3.73 children, but for the 1951 birth cohort it was 2.24 children. As noted, women born in 1941 had on average 2.53 children when they reached 30 years of age, and 3.41 children by the age of 40, implying that the fertility level for the 1941 birth cohort was still much higher than the replacement level. However, the corresponding figures for women born 10 years later in 1951 were 1.64 and 2.22 children respectively. Similarly, the average number of children ever born for the 1961 birth cohort, who reached 40 years old, was 1.53 children in 2001. The 1971 birth cohort, reaching 30 in 2001, had only 0.60 children by that age, which is about one-third of the number for the cohort born 20 years earlier in 1951. Hong Kong women in recent birth cohorts gave birth to fewer children than women in any of the earlier cohorts. 4.0 20 years old 25 years old 30 years old 35 years old 40 years old 45 years old 50 years old
3.5
Cumulative fertility
3.0 2.5 2.0 1.5 1.0 0.5 0.0 1936
1941
1946
1951
1956
1961
1966
1971
1976
1981
Cohort
Figure 7.7 Average number of children ever born with the mother’s year of birth and age, Hong Kong. Source: Hong Kong Census & Statistics Department (2002).
Ultra-low fertility in Hong Kong 141
Marriage and marital fertility in Hong Kong Fertility is closely related to marriage since child-bearing within wedlock remains the mainstream in Hong Kong. An exploration of trends in marriage can therefore give us more insight into the low-fertility situation. A comparison of marital fertility rates shows a decrease in fertility among married couples. An overall trend of an increasing proportion of the never-married population, as well as a falling general marriage rate and increasing median age of first marriage have had a significant impact on the fertility decline. Further, a significant increase in the time interval between marriage and giving birth is also observed which further lowers the propensity for child-bearing. The changing socio-cultural context suggests that cross-border marriage between Hong Kong residents and Mainland Chinese citizens and births outside wedlock are on the rise. Their impact on fertility will also be explored below. Age-specific marital fertility rate and total marital fertility rate A decline in fertility among married couples can be shown by the age-specific marital fertility rate (ASMFR), which is the number of live births occurring to married women in a given age group during a calendar year compared to the total married female population (excluding foreign domestic helpers) at mid-year in that age group. Figure 7.8 shows that the ASMFRs declined sharply across all age groups during 1971 to 2001, except for the youngest age group of 15–19. The high rates at ages 15–19 were associated with low levels of marriage coupled with high rates of pre-maritally conceived births (FPAHK, 2002). For women in the age
Live births per 1,000 married women
700 1971 600
1976 1981
500
1986 1991
400
1996 300
2001
200 100 0 15–19
20–24
25–29
30–34
35–39
40–44
Figure 7.8 Age-specific marital fertility rates of Hong Kong, 1971–2001. Source: Hong Kong Census & Statistics Department (1983, 1977–2003).
45–49
142 Paul Yip, C. K. Law, and Karen Cheung groups of 20–24 and 25–29, the marital fertility rates decreased by 54.4 percent and 60.5 percent respectively (Figure 7.8). A similar reduction was also found in the age groups of 30–34 and 35–39 (52.9 percent and 61.6 percent). The steepest decline of all occurred in the age groups of 40–44 and 45–49 (by 83.8 percent and 94.7 percent respectively over the same period). The total marital fertility rate (TMFR) is the sum of the ASMFRs in a given year and represents the average number of children that would be born alive to 1,000 married women during their lifetime if they were to pass through their child-bearing ages, 20–49, experiencing the age-specific married fertility rates prevailing in that year. In order to avoid giving undue weight to the tiny number of married women in the age group 15–19, many of whom were married as a result of teenage pregnancy, the TMFR here excludes the marital fertility rates for the age group 15–19. The TMFRs decreased by 35.5 percent from 5.210 live births per 1,000 married women between the ages 20 and 49 in 1971 to 3.359 in 1981. A further decline in the TMFR has been occurring since 1981. The TMFRs decreased by 34.0 percent from 3.359 live births per 1,000 married women between 20 and 49 in 1981 to 2.216 in 2001. The overall trend points to the conclusion that the reduction in fertility rates in Hong Kong between 1981 and 2001 was primarily due to the reduction in the TMFR, which is due to postponement of low-order live births and also curtailment of high-order live births, as indicated in Figures 7.3 and 7.4, respectively (Hong Kong Census & Statistics Department, 2005). Never-married population Apart from the decrease in TMFR, the large decline in fertility is also the result of a shift of marital status distribution, which has exerted a downward driving force on the TFR and has magnified the effects of the fertility decline (Yip et al., 2001). The proportion of never-married men and women increased continuously between 1981 and 2001. In the age group 25–29, it had increased by 29.4 percentage points from 30.3 percent in 1981 to 59.7 percent in 2001 for women and by 18 percentage points from 57.5 percent in 1981 to 75.5 percent in 2001 for men (Figure 7.9) (Hong Kong Census & Statistics Department, 2001a). More strikingly, for the age groups 30–34 and 35–39 among Hong Kong women it increased significantly from 11.0 percent to 31.2 percent and from 4.5 percent to 18.8 percent from 1981 to 2001 respectively. In other words, the decline in the TFR is attributable to the fact that there were fewer married women in the most productive child-bearing years (25–29), but also in the more advanced productive years (30–39). General marriage rate and median age at first marriage The trends over the past 30 years can also be studied in relation to the general marriage rate (GMR), which followed a downward trend and reached its lowest point of 5.5 per 1,000 population aged 15 and over in 2000 (Figure 7.10). The GMR dropped by 42.6 percent from 10.1 to 5.8 between 1975 and 2001. This
100 Men 80
%
60
40
20
0 1981
1986
1991
1996
2001
15–19
20–24
25–29
30–34
35–39
40–44
100 Women 80
%
60
40
20
0 1981
1986
1991
1996
2001
15–19
20–24
25–29
30–34
35–39
40–44
Figure 7.9 Proportions of never-married population by age and sex, Hong Kong, 1981– 2001. Source: Hong Kong Census & Statistics Department (2002).
144 Paul Yip, C. K. Law, and Karen Cheung 14
50 000
10
40 000
8
6
30 000
Number of marriage
General marriage rate, per 1,000 populations aged 15 and over
12
4
20 000 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
2 GMR
Marriage
Figure 7.10 General marriage rate and number of marriages, Hong Kong, 1971–2004. Source: directly provided by Hong Kong Census & Statistics Department. Note: the number of marriages has excluded the re-registration of couples who had either customarily married in Hong Kong before the Marriage Reform Ordinance was enacted in 1971 or had married outside Hong Kong.
drastic decrease resulted from fewer people getting married within the pool of fertile women in the past 26 years. The lowest levels of GMR lasted about five years from 1998 to 2002. A small rebound in the GMR occurred in 2003 and it reached 7.1 per 1,000 aged 15 and over in 2004, which is similar to the 1997 level (Figure 7.10). In Hong Kong, postponement of marriage has a direct effect on the delay in childbearing, since out-of-wedlock births are still relatively small, albeit increasing. Women’s ages at marriage are an important determinant of the number of children a woman will be able to conceive, because they influence the proportion of reproductive years in which women are exposed to the risk of child-bearing. The data indicate that the median age of first marriages rose rapidly for both men and women in Hong Kong (Figure 7.11). The median age of first marriage of women increased by 4.2 years (17.6 percent) from 23.9 in 1981 to 28.1 in 2004; for men it rose 4.1 years (15.2 percent) from 27 to 31.1 between 1981 and 2004. This linear increase in the median age of first marriage is in line with the rise in the proportion of never-married men and women, especially those in their 20s and 30s (Figure 7.9). Undoubtedly, the increase in the proportion of the never-married population, the decline in marital fertility, and the postponement of age of first marriage have all
Ultra-low fertility in Hong Kong 145 32
Median age at first marriage
30
28
26
24
1971 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
22
Men
Women
Figure 7.11 Median age of first marriage by sex, Hong Kong, 1981–2004. Source: Hong Kong Census & Statistics Department (2002, 2005).
had a direct effect on reducing the level of fertility. Also, Yip et al. (2001) suggest that the substantial decrease in the proportion of married women in the age group 20–29 is due in part to a peculiar marriage phenomenon appearing as early as the 1990s whereby Hong Kong male residents preferred to marry in Mainland China. This will be explored later in this section. Time interval between age of marriage and prime child-bearing age Figure 7.12 shows the median age of a first marriage among women and the interval between marriage and first live birth. As the median age of first marriage among women increased, the median interval from marriage to the first live birth also rose from 14.7 months in 1981 to 27.5 months in 2001. Both increases in the median age of first marriage and the interval between marriage and first live birth lower the propensity for child-bearing. Associated with the delay in giving birth to higher ages, there was a significant increase in the time interval between marriage and birth. While 78.8 percent of the live births in 1981 occurred within the first five years of marriage, the corresponding percentage in 2001 dropped to 65.9 percent (Hong Kong Census & Statistics Department, 2002). Consequently, postponement of the first-order birth after marriage has further shortened the average child-bearing exposure period of a married woman for the second and the higher-order birth.
146 Paul Yip, C. K. Law, and Karen Cheung 28
30
25
20
Months
Median age
26
24 15
22 2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
10
Median age at first marriage among women Median interval from marriage to first order live birth
Figure 7.12 Median age for first marriage among women and interval between marriage and first-order live birth, Hong Kong, 1981–2001. Source: Hong Kong Census & Statistics Department (2002).
Cross-border marriages The on-going tendency among Hong Kong men to marry in China also explains why there has been a considerable increase in the proportion of never-married women in older age groups as opposed to that of men. This has resulted in the hypothesis of a “marriage squeeze” on Hong Kong women. Consequently, this might have led to a situation whereby women of marriageable age are unable to find a suitable man to marry (Yip and Lee, 2002). However, as shown in Figure 7.13, there was a downward trend in the number of successful applications for certificates of absence of marriage registration (CAMR)5 for men in recent years, despite the fact that a total of 266,364 Hong Kong men had successfully received a CAMR. The number of CAMRs increased by 25.9 percent from 21,655 in 1990 and reached its highest level of 27,259 in 1997. It then decreased by 70.9 percent over almost eight years, reaching its lowest level of 7,943 in 2003. The early increase in the CAMR can be associated with an outflow of Hong Kong residents to China due to greater job opportunities starting in the early 1990s. As many Hong Kong manufacturers were moving their assembly lines to the Pearl River Delta (PRD), these companies usually employed Hong Kong managers or Hong Kong-trained managers to run their operations there (Banister
Ultra-low fertility in Hong Kong 147 30000
Number of CAMR
25000 20000
15000 10000
5000 0 1988
1993
1998
Men
2003
Women
Figure 7.13 Number of successful applicants for certificate of no-marriage registration (CAMR), Hong Kong, 1990–2003. Source: directly provided by Hong Kong Census & Statistics Department.
and Kwong, 1999; Kwong, 1993; Lim and Abella, 1994; Luk, 1998). Some of these Hong Kong men entered the marriage market in China. One possible explanation of the later declining trend of the CAMR may be the recent Mainland policy of relaxing restriction for issuing a two-way permit for Mainlanders to visit Hong Kong as tourists. A more open and prosperous economy in the Mainland means that more Mainland women can easily come to Hong Kong and get married here. This argument is in line with a small rebound in the GMR in the most recent periods as shown in Figure 7.10. With respect to the increase in the proportion of never-married Hong Kong women, the large number of Hong Kong men marrying in the Mainland could have a major effect on Hong Kong women of marriageable age in that they might be unable to find suitable husbands. Taking into account that in more than 40 percent of registered marriages in Hong Kong, the age of the groom is at least five years older than that of the bride, it is estimated that the “pool” of never-married women exceeds that of never-married men by about 100,000. This imbalance in number of men and women forces some women to remain single. However, the substantial increase in the proportion of never-married Hong Kong women may be more indicative of a genuine propensity for them to remain single for reasons other than a mere shortage of potential husbands (Hong Kong Census & Statistics Department, 2002).
148 Paul Yip, C. K. Law, and Karen Cheung Births outside wedlock and childlessness In Western countries, the decline in the marital fertility rate has commonly been offset by an increase in non-marital fertility. In France and Sweden, for example, it has been observed that a large proportion of births take place outside of marriage, accounting for about 43 percent and 50 percent respectively. In Australia, the proportion of non-marital fertility is about 30 percent. In the United Kingdom, it has accounted for 40 percent of all births. According to Lesthaeghe and Moors (2000), young adults enter a cohabitating relationship first, have their first child, and then proceed to marriage. However, in Asian countries, the proportion of births outside of wedlock is low. In Japan, this proportion is extremely low at about 1 to 2 percent. The proportion of single-parent births in Singapore increased from about 0.8 percent in 1994 and 1995 to 1.2 percent in 2001 and 2002 (Singaporean Registry of Births and Deaths, 2002). In Hong Kong, the proportion of births by cohabitating parents had increased by 1.4 percentage points from 1991 to 1999, reaching its highest level of 7.5 percent in 2000. In 2003, it remained at about 5.8 percent (Figure 7.14). All these figures were much lower than in Western developed countries. Despite the fact that the increase in non-marital fertility in Hong Kong is quite small, the rising trend should be given closer attention. As Hong Kong has experienced a rapid demographic transition that has more or less mirrored the path taken in the Western countries, one of the driving forces for increases in the fertility level in future will perhaps be associated with an increase in non-marital fertility, which may become more acceptable for younger cohorts. However, other factors 8%
Proportion of live births
7% 6% 5% 4% 3% 2% 1% 0% 1981
1986
1991
1996
2001
Figure 7.14 Proportion of live births in Hong Kong by cohabitating parents, 1981–2003. Source: Micro Known Birth Dataset.
Ultra-low fertility in Hong Kong 149 such as cultural or religious barriers and a bias against single-parent families may still restrain any increase in non-marital births in Hong Kong. It has also been shown that the well-being of children from single-parent households is a major concern (Yip et al., 2004).
Other demographic factors related to Hong Kong’s fertility Due to changes in marriage patterns, other demographic factors such as divorce rates, age of re-marriage, and abortion rates are also worth noting to build a more comprehensive and complete picture of fertility in Hong Kong. Increase in divorce rates
14000
2.5
12000
2.0
10000
1.5
8000
1.0
6000
0.5
4000
0.0
2000
Divorce decrees granted (DDG)
3.0
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001
General divorce rate, per 1,000 population aged 15 and over
Figure 7.15 shows the number of divorce decrees granted (DDG) and the general divorce rate (GDR). Both have been on the rise from 1981 to 2001. The number of DDGs increased significantly from 2,060 in 1981 to 13,425 in 2001. The general divorce rate (GDR), which relates to the number of divorce decrees granted to the mid-year population aged 15 and over, increased rapidly over the same period as well. In 2001, the GDR was 2.39 per 1,000 aged 15 and over, compared with 0.53 in 1981. As more men and women divorce, a new phenomenon worth noting is re-marriage. The median age for re-marriage was considerably higher than that of first
GDR
DDG
Figure 7.15 Number of divorce decrees granted and the general divorce rate, Hong Kong, 1981–2001. Source: Hong Kong Census & Statistics Department (2002).
150 Paul Yip, C. K. Law, and Karen Cheung marriages for both genders. During 1981–2001, the median age increased from 36.3 years to 41.4 years for divorced men, and from 30.8 years to 35.0 years for divorced women. The data indicate that the median age for re-marriage increased by 3 to 8 years over the past 20 years. The difference between the median age at the first marriage and median age at re-marriage for divorced women was about 11.2 years for men and 7.5 years for women in 2001. The increase in the duration between the divorce and re-marriage for divorced women would also affect the reproductive exposure period. Trend in abortions According to the “Family Planning Knowledge, Attitude and Practice” (KAP) study6 (Family Planning Association of Hong Kong, 2002), the overall induced abortion rate was 26.3 percent (total 1,603 fertile respondents), which was slightly higher than 1997 totals. In the age groups 15–24 and 45–49, the abortion prevalence rates in 2002 (21.4 percent and 30.4 percent) were higher than those in 1997 (12.0 percent and 28.5 percent) respectively (Family Planning Association of Hong Kong, 1999; 2002), highlighting two prominent facts: 1 2
a strong increase in teen abortions for the youngest age group (15–24) a weak increase in unwanted pregnancies for the advanced age group (45–49).
Figure 7.16 illustrates the trend in the abortion ratio over time, which refers to the total number of legal induced abortions in a particular year to the total number of live births for residents of Hong Kong. The abortion ratio increased from about 38 per 100 live births in 1992 to a peak of 53 per 100 live births in 2001. It remained at the average level of about 50 per 100 live births throughout the period 1998–2003. Subject to the rapidly changing attitude towards sexual behavior, the increasing abortion ratio indicates that more effort should be put into educating the public on the importance of contraception to reduce the number of unwanted births in the future.
Concerns of Hong Kong people on bearing and rearing children Fertility is closely related to the socio-cultural and economic context, as these factors affect people’s ideas regarding family planning, personal freedom, and child-rearing. Some researchers have focused on social processes and marital structure as they relate to fertility changes (Banister and Kwong, 1999; Yip et al., 2001; Yip and Lee, 2002) and the pattern of lowest-low fertility in Hong Kong (Tu, 2002). According to the 2002 KAP study, there has been a significant increase in proportions favoring having no children, from 2.1 percent in 1982 to 15.2 percent in 2002 (Family Planning Association of Hong Kong, 2002). Those who did not want to have more children gave two major reasons:
Ultra-low fertility in Hong Kong 151
Abortion ratio, per 100 live births
55
50
45
40
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
35
Figure 7.16 Abortion to live-birth ratio, Hong Kong, 1992–2003. Source: abortion data directly provided by Department of Health (2005).
1 2
they felt satisfied with their current number of children (46.8 percent) they felt a heavy financial burden when raising children (24.6 percent).
More strikingly, about 7.9 percent of fertile women thought that there was too much responsibility involved in bringing up a child. External factors such as concerns about downturns in Hong Kong’s economy also affected their decision about having more children (6.6 percent).7 On the other hand, the vast majority of respondents in the KAP study still wanted to have two children, while a significantly high proportion of female respondents (41.8 percent) mentioned “like children” as one of the main reasons for having children. Another popular reason is to provide a companion for their present child, 21.2 percent. Passing on the family name and family heritage was ranked third (11.8 percent). None of the 1,294 fertile female respondents said they wanted to have more children to help nurture the next generation for society. The emphasis on bringing up children in order to have someone to look after them in old age was also found no longer appealing, as it was mentioned by only about 8 percent. The lack of confidence in the local education system is another reason for the low-fertility phenomenon in Hong Kong (Yip and Cheung, 2005). Therefore, despite the fact that since the 1970s, Hong Kong has had in place a policy of nine-year free compulsory education, the financial costs for pre-primary and nursery education are expensive. Instead of enjoying the nine-year free education, more parents prefer to send their children to “international” pre-primary schools8 as early as possible to secure a place. Thus, despite the high tuition fees and
152 Paul Yip, C. K. Law, and Karen Cheung associated expenses, waiting lists at top international kindergartens are extremely long. Conversely, the number of locally run kindergartens has been diminishing with a reduction in class size. This suggests that parents have decided to trade-off quality for quantity instead of having more children (Yip and Cheung, 2005). The lack of an equal and family-friendly working environment is also a prominent factor in low fertility in Hong Kong. Hong Kong people work long hours by international standards. As in other parts of Asia, Saturdays are a working day for many. A recent survey by Robert Walters, a UK recruitment consultant firm, showed that Hong Kong has the longest working hours in the world, with 43 percent of those polled saying they worked 50 to 55 hours or more a week (Harney, 2005). Hong Kong has never imposed legal requirements for overtime pay, working hours, or a minimum wage for most workers. The only people who qualify for a minimum wage are some security guards and cleaners working on outsourced contracts for the Government, and the city’s foreign domestic helpers, whose monthly wage was cut in 2004 to HK$3,270. Labor advocates are pushing for a 44-hour working week and overtime pay of one-and-a-half times the normal hourly wage (Harney, 2005). Although the total working hours may often be stipulated as 44 hours per week in an employment contract, the long working hours may still be self-created – Hong Kong’s work culture.
Policy measures to encourage childbirth in Hong Kong We have shown that people’s concerns may make them hesitant to have children. The SAR Government of Hong Kong, however, remains a stronghold for the concept of laissez-faire, believing that any dramatic policy action is likely to be expensive and ineffective in raising fertility. Thus, policy measures to encourage childbirth mainly focus on the economic context through maternal leave, tax allowances, provision of limited childcare services, and financial assistance to alleviate the parents’ economic burden of child-bearing, childcare, and education costs. However these policies also have strong limitations and these are discussed below. Maternal leave Under the Employment Ordinance, Cap. 57 (The Hong Kong Labour Department, 2007), a female employee is eligible for 10 weeks’ paid maternity leave if she has worked under a continuous contract for not less than 40 weeks immediately before the commencement of the maternity leave. If the length of employment service under a continuous contract is less than 40 weeks immediately before the commencement of maternity leave, the employee is eligible for 10 weeks’ maternity leave without pay. If an employee wants to take care of a child or adult relative when he or she is sick, an employee can, of course, take annual leave,9 but the purpose of the leave is different. This ordinance, however, does not apply to parttime and self-employed workers (The Hong Kong Labour Department, 2007). Maternity leave pay is equal to four-fifths of normal wages. It is paid on the
Ultra-low fertility in Hong Kong 153 normal pay day of the employee. With the agreement of her employer, a pregnant employee may decide to commence her maternity leave anytime from two to four weeks before the expected date of confinement. If the employee is absent from work for medical examinations related to her pregnancy, post-confinement medical treatment, or a miscarriage, any day she is absent shall be counted as sick leave. She is entitled to a sickness allowance of four-fifths of her normal wages. A key point to note, though, is that with an increasing trend towards part-time employment, self-employed status, and renewable-contract terms, more women plan not to get pregnant for fear of being fired. Child allowance In Hong Kong, child allowance for salaries tax is governed by Section 31 of the Inland Revenue Ordinance (IRO) (Hong Kong Inland Revenue Department, 2007). Under the rule, a taxpayer is eligible to claim child allowance if, at any time during a year of assessment, they maintain an unmarried child,10 who was: 1 2 3
under the age of 18, or of or over the age of 18 but under 25, and was receiving full-time education at a university, college, school, or other similar educational establishment, or of or over the age of 18, and was, by reason of physical or mental disability, incapacitated for work.
If a couple, not a husband and a wife living apart, have more than one child for whom child allowance is claimed, all child allowances must be claimed by either the husband or the wife. They must decide who shall claim the allowance. From financial year 1998/99 to 2002/03, salaries tax allowances for dependents under the age of 18 were HK$30,000 each for the first and second children and HK$15,000 each for the third through ninth child (Hong Kong Inland Revenue Department, 2007). The Ordinance was revised in 2003/04 in order to increase the child allowance for the third and subsequent child by HK$15,000 to HK$30,000. Subsequently, the Financial Secretary of the SAR Government of Hong Kong has further increased the child allowance for each child to HK$40,000 in 2004/05 and to HK$50,000 in 2007/08 (Hong Kong Inland Revenue Department, 2007). From April 2007 onwards, an additional tax allowance of HK$50,000 is given to the parents during the year of childbirth. (Hong Kong Inland Revenue Department, 2007). Childcare/pre-primary service provision and subsidies Creating a sound environment for bearing and rearing children in the workplace is not new in Hong Kong, yet existing support facilities for childcare are scarce. The need for childcare services and facilities at the workplace reflect the importance of supportive measures for reconciling work and family responsibilities, especially for the increasing number of working mothers in Hong Kong over the last decade (Yip and Cheung, 2005).
154 Paul Yip, C. K. Law, and Karen Cheung In Hong Kong, childcare/pre-primary services are operated by private-sector and local charitable organizations. The Government takes the role in subvention, regulation, and monitoring of childcare-service providers in ensuring their service quality reaches an acceptable level. In 2005/06, there were 17 childcare centres (13 daycare centres and 4 residential centres) for children aged 0 to 3 under the registration of Social Welfare Department (The Hong Kong Social Welfare Department, 2007) and 1,062 kindergartens (serving children aged 3 to 6)/ kindergarten-cum-childcare centres (for children aged 0 to 6) under the monitoring of the Education Bureau (2007). To expand the coverage of financial support for pre-primary services in Hong Kong, the Government introduced the Pre-primary Education Voucher Scheme (Voucher Scheme). Under the Voucher Scheme, the Government provides a fee subsidy for parents, legal guardians, or registered custodians to help meet school fees for pre-primary education of eligible children in the form of pre-primary education vouchers. Eligible kindergartens or kindergarten-cum-childcare centres (recognized by the Education Bureau of the Government) may redeem the vouchers of children who are enrolled in the kindergartens concerned (The Hong Kong Student Financial Assistance Agency, 2007a). In 2007/08, the amount of voucher dedicated towards fee subsidy was HK$10,000 and the amount will progressively increase to HK$16,000 by 2011/12 (The Hong Kong Student Financial Assistance Agency, 2007a). Furthermore, for lower-income families without support from the Comprehensive Social Security Assistance (CSSA), parents are eligible for the Kindergarten and Child Care Centre Fee Remission Scheme (KCFRS), which provides assistance with the costs of pre-primary services (The Hong Kong Student Financial Assistance Agency, 2007a). Subject to a means test,11 an applicant’s eligibility for assistance will be assessed, as well as the corresponding assistance level (full remission, 75 percent remission or 50 percent remission) if he or she is eligible for fee remission under the KCFRS. From September 2005 onwards, the ambit of KCFRS has further expanded to cover children attending full-time programs in childcare centres in Hong Kong (The Hong Kong Student Financial Assistance Agency, 2007a).
Discussion Is having children an individual choice or does it have collective dimensions? In our view, children should be seen, at least in part, as collective assets, ones that society can rely on to provide a sufficient and reliable workforce to maintain economic growth. We need children to generate the resources to pay for health care and welfare for the existing generation (including the human resources to provide the labor to take care of the elderly, etc). When discussing fertility levels or policies that affect the number of children couples have, the critical question is, “Who will take care of the elderly?” The answer, of course, is “today’s children”. In the absence of children, elderly support will be in danger. Another key question is, “Who will pay for elderly health care?” Again, today’s children. Therefore, the
Ultra-low fertility in Hong Kong 155 Government has a legitimate right and responsibility to provide any “preferential” treatments in order to remove the barriers to child-bearing for the sustainable development and betterment of our community. The reduction in fertility is sometimes seen as a manifestation of other social or economic problems within the community. These include the lack of job security or long-term commitment in a marital relationship, dissatisfaction with the quality of governance and public services in education, or insufficient personal resources to raise children according to accepted community standards. All of these factors are important socio-economic problems that need to be addressed in order to improve the well-being of the community. For many women, being in paid employment has become a pre-condition of having children (Krieger, 2004). Due to increasing workloads and stress levels at work, many working women in Hong Kong are forced to make a choice between having a family and having a career. Therefore, it is important to provide a family-friendly working platform for Hong Kong women. Due to financial needs, it is now the norm rather than the exception for both husband and wife to work. The challenge is removing barriers to having children among working women. We propose that a more conducive environment for child-bearing, which comprises the following four major components, should be cultivated in the community, namely: 1
2
3
Encouraging child-friendly cultural and psychological changes. Children are public assets and the community should be made aware of this fact. Interventions and policies that encourage marriage, family formation and childbirth should be supported by active and deep-rooted mass media and education information campaigns. The Hong Kong people need to adopt a child-bearing-friendly mindset and realize that this is no longer just a private issue but one that affects the interests of the entire community. Providing economic and financial support. Rearing children is costly in Hong Kong. If we see children as public assets, then a large share of the costs of rearing children should be borne by the community at large. Some of the major concerns of married couples about child-bearing are economically related, for example, housing and education costs. Child allowances will ease the burdens of couples, so that quality lifestyles for couples and children can be maintained. We advocate making use of the income-tax system, which can provide financial advantages to families with dependent children. The idea is to implement a tax rebate rather than a cash bonus to ease the burdens of working couples with children. However, a small cash bonus should also be provided as a gesture or a token of appreciation from the Government to those couples that are helping to raise the next generation of our community. Organizational changes. The ultimate ideal is to ensure that work and career do not become an activity competing with child-bearing. To this end, private employers should be induced to create a family-friendly work environment. The Government should take the lead in this effort to set a good example. France and Sweden demonstrate that work and bearing children are not necessarily competing forces. Female workforce participation is about 70 percent
156 Paul Yip, C. K. Law, and Karen Cheung
4
in France and their TFR is at 1.7. Furthermore, it is important to release the power of women in the workforce and their participation in the community’s economic activities. The notion of providing parental leave for men is not only to help their spouses with the new infant chores, but also serves as a signal to the community that men indeed have a responsibility to take care of their families. That is why employers should be reminded that leave is not only exclusively for women, but also for men. Long working hours not only have a negative effect on fertility but also adversely affect the overall well-being of the community. But shorter working hours will require not only changes in regulations, but also changes in the mindsets of Hong Kong workers, which are very much attuned to long working hours. Insufficient and expensive childcare facilities are also major concerns of working mothers. For this reason, provision of quality and affordable childcare facilities near the workplace can be a win-win situation for employees and employers. Demographic changes. Demographic changes in Hong Kong are happening rapidly and they should be closely monitored and analyzed. The potential impact on size, distribution and quality of population due to increasing crossboundary marriages and births from non-Hong Kong residents cannot be underestimated, so we must pay careful attention to these trends. Because more than 80 percent of our population growth originates from net migration (Hong Kong Census & Statistics Department, 1983 and 2002), Hong Kong has to remain attractive for potential migrants. It is of great importance that the SAR Government of Hong Kong is able to choose “desirable” people to settle here.
As implementing a population policy is a complex process, with implications that affect everyone in the community, it is vitally important to raise awareness in the community. Certainly, it is not easy to obtain consensus from different parties that have their own interests. However, the Government must realize the urgency of the issue and have the vision to put in place the suggested changes. These policies should aim to bring out the best in people, to remove gender inequality, and to work toward the betterment of the community. The emphasis of the population policy should not only be on quantity, but on quality as well.
Notes 1 Total Fertility Rate (TFR) in a given year represents the average number of children that 1,000 women would bear (ignoring mortality) if they were subject throughout their reproductive span to the prevailing fertility rates during that year. This measure represents a hypothetical average completed family size if fertility were to remain constant at a given level (Hong Kong Census & Statistics Department, 2002). 2 The sex ratio at birth is generally about 105 males to 100 females, and if all children born grew up to reach adulthood, then 2.05 would be the replacement level. However, even in the most advantageous populations, there still exists some infant and child mortality; the specified level is 2.1, taking mortality into account (Wilson, 2004; Wilson and Pison, 2004).
Ultra-low fertility in Hong Kong 157 3 To calculate the proportion of women by parity, the births from non-Hong Kong residents (the residency of mothers is not from Hong Kong) has been discounted since 1996. Due to vagueness in resident coding before 1996 in the Micro Known Birth Dataset, the authors are unable to exclude the births from non-Hong Kong residents before that time. In 2003, there were 54 percent and 9 percent for the proportion of women having a first birth and three children respectively if the births from non-Hong Kong residents are included. 4 Under the current practice, residents of Hong Kong who intend to get married with another Chinese national in Mainland China are required to obtain a Certificate of Absence of Marriage Record (CAMR) from the Immigration Department of Hong Kong. 5 For an explanation of this certificate, see Note 4. 6 The KAP study is carried out by the Family Planning Association of Hong Kong every five years to gain understanding of the perceptions and attitudes of an individual towards marriage, family formation, child-bearing, children, and the overall environment. 7 Multiple responses were allowed, if the respondent had more than one reason to explain their attitude in the survey. 8 School fees of international schools vary from a high of about HK$112,800 (such as Yew Chung Hong Kong International School) to the lowest level of HK$15,301–$20,020 (like Jimmy’s International Nursery and Kindergarten). In some international schools (such as the Chinese International School), parents have to pay a debenture (i.e., individual HK$75,000 (refundable), or annual levy of HK$7,500 (non-refundable); corporate, HK$300,000 per child (transferable when child leaves)). Note: 7.8 HK$ = 1 US$ in July 2008. 9 Every employee in Hong Kong is entitled to annual leave with pay after having been employed under a continuous contract for every 12 months. His or her entitlement to paid annual leave increases progressively from 7 days to a maximum of 14 days according to his length of service. Under the rule, an employer that without reasonable excuse fails to grant annual leave to an employee is liable to prosecution and, upon conviction, to a fine of $50,000. 10 According to S31 of IRO, a child refers to (1) the child of you or your spouse or former spouse; or (2) the adopted child of you or your spouse or former spouse; or (3) the stepchild of you or your spouse or former spouse. 11 The level of assistance is based on a sliding scale, which takes into consideration the applicant’s household size and income after deduction of rent. The KGFRS is assessed on an Adjusted Family Income (AFI) mechanism, which takes into account the applicant’s household income and the number of dependents in the family (The Hong Kong Student Financial Assistance Agency, 2007b).
References Banister, H. and Paul C. K. Kwong. 1999. Social Processes Underlying Hong Kong’s Population Development, presented at the First Annual Conference of Hong Kong Sociological Association, Hong Kong. Basic Law. 1990. Paragraph.1 of Article 24. Accessed October 10, 2007 from http://www. info.gov.hk/basic_law/fulltext/content0203.htm. Choi, C. Y. and K. C. Chan. 1973. The Impact of Industrialization on Fertility in Hong Kong: A Demographic, Social and Economic Analysis. Hong Kong: The Chinese University of Hong Kong Social Research Center. Coughlin, Richard J. and Margaret M. Coughlin. 1963. “Fertility and birth control among low income Chinese families in Hong Kong,” Marriage and Family Living, 25 (May): 171–177. Economic and Social Research Institute. 2004. Fertility and Family Issues in an Enlarged Europe, European Foundation for the Improvement of Living and Working Conditions. Budapest: Demographic Research Institute. Accessed on September 17, 2007 from: http://www.eurofound.eu.int/qual_life.
158 Paul Yip, C. K. Law, and Karen Cheung Education Bureau. 2007. Kindergarten Education in Hong Kong. Accessed on September 12, 2007 from http://www.emb.gov.hk/index.aspx?langno=1&nodeID=916. Family Planning Association of Hong Kong. 1999. Report on the survey of family planning knowledge, attitude and practice in Hong Kong 1997. Hong Kong: Family Planning Association of Hong Kong. —— 2002. Report on the survey of family planning knowledge, attitude and practice in Hong Kong 2002. Hong Kong: Family Planning Association of Hong Kong. Fan, S. C. and S. L. Lee. 1986. “The population growth in Hong Kong,” Journal of Population Studies, 9: 105–130. Freedman, R and A. L. Adlakha. 1968. “Recent fertility declines in Hong Kong: the role of the changing age structure,” Population Studies, 22(2): 181–198. Freedman, R., D. N. Namboothiri, A. L. Adlakha, and K. C. Chan. 1969. “Hong Kong: the continuing fertility decline,” Studies in Family Planning, No. 44(August): 8–15. —— 1970. “Hong Kong’s fertility decline, 1961–1968,” Population Index, 36(1): 3–18. Harney, Alexandra. 2005. “Hong Kong labours over minimum wage debate,” Financial Times, February 15, 2005. Hong Kong Census & Statistics Department. 1972. “Recent trends in marriage and fertility in Hong Kong,” Hong Kong Monthly Digest of Statistics, July: 64–66. Hong Kong: Hong Kong Government Printer. —— 1975. Births, Deaths and Marriages 1970–1973. Hong Kong: Hong Kong Government Printer. —— 1977–2003. Hong Kong Annual Digest of Statistics. Hong Kong: Hong Kong Government Printer. —— 1979. “Recent trends in fertility in Hong Kong,” Hong Kong Monthly Digest of Statistics, Dec: 59–65. Hong Kong: Hong Kong Government Printer. —— 1982. “Fertility in Hong Kong: past and future trends,” Hong Kong Monthly Digest of Statistics, April: 91–100. Hong Kong: Hong Kong Government Printer. —— 1983. Demographic Trends in Hong Kong 1971–82: Analysis based on vital registration statistics of births, marriages, and deaths on census results. Hong Kong: Hong Kong Government Printer. —— 2001a. 2001 Population Census Summary Results. Hong Kong: Hong Kong Government Printer. —— 2001b. 1999/2000 Household Expenditure Survey and the Rebasing of the Consumer Price Indices. Hong Kong: Hong Kong Government Printer,. —— 2001c. 2001 Population Census Main Report. Hong Kong: Hong Kong Government Printer,. —— 2002. Demographic Trends in Hong Kong, 1981–2001. Hong Kong: Hong Kong Government Printer. —— 2005. The Fertility Trend in Hong Kong, 1975–2004. Hong Kong: Hong Kong Government Printer. Kreiger, H. 2004. “No future without children: demographic developments in Europe,” European Ministerial Conference paper on “Family means future – joint family policy thrust in Europe,” Berlin, Germany, December 2, 2004. Kwong, Paul C. K. 1993. “Internationalization of population and globalization of families,” in Choi Po-king and Ho Lok-sang (eds), The Other Hong Kong Report. Hong Kong: Chinese University Press. Lesthaeghe, R. and G. Moors. 2000. “Recent trends in fertility and household formation in the industrialized world,” Review of Population Social Policy, 9: 121–170. Lim, L.L. and M. Abella. 1994. “The movement of people in Asia: internal, intra-regional,
Ultra-low fertility in Hong Kong 159 and international migration,” Asian and Pacific Migration Journal, 3(2–3): 209–250. Luk, Bernard H. K. 1998. “Hong Kong society toward the twenty-first century?,” American Asian Review, 16 (2, Summer): 41–65. Ng, P. 1978. Family Planning, Fertility Decline, and Family Size Preference in Hong Kong: Some General Observations. Hong Kong: The Chinese University of Hong Kong Social Research Centre. The Hong Kong Labour Department. 2007. “Maternal leave”. Accessed on September 12, 2007 from http://www.labour.gov.hk/eng/faq/cap57h_whole.htm. The Hong Kong Social Welfare Department. 2007. “Statistics of social welfare services in Hong Kong 2005/06”. Accessed on September 12, 2007 from http://www.swd.gov.hk/ doc/res_stat/2006fig.pdf. The Hong Kong Student Financial Assistance Agency. 2007a. “Pre-primary education voucher scheme”. Accessed on September 12, 2007 from http://www.sfaa.gov.hk/eng/ schemes/pevs.htm. —— 2007b. “Kindergarten and child care centre fee remission scheme”. Accessed on September 12, 2007 from http://www.sfaa.gov.hk/eng/schemes/pevs.htm. The Inland Revenue Department. 2007. “Salaries tax/personal assessment: allowances, deductions and tax rate table”. Accessed on September 12, 2007 from http://www.ird. gov.hk/eng/tax/ind_all.htm#04. Singaporean Registry of Births and Deaths. n.d. “Report on the registration of births and deaths 2002”. Singapore. Tu, Jow-Ching Edward. 2002. “Patterns of lowest-low fertility in Hong Kong,” Journal of Population and Social Security (Population), Supplement to Vol. 1: 629–642. United Nations. 1974. The Demographic Situation in Hong Kong, ESCAP Country Monograph Series No. 1, Economic and Social Commission for Asia and the Pacific. Bangkok, Thailand: United Nations. Wilson, Chris. 2004. “Fertility below replacement level,” Science, 304(5668): 207–209. Wilson, Chris and G. Pison. 2004. “More than half of the global population lives where fertility is below replacement level,” Population and Societies, INED, No. 405. October: 1–4. Yip, P. S. F., S. L. K. Cheung, Stephen Law, Iris Chi and J-M Robine. 2004. “Demographic window and economic burden in Hong Kong,” International Seminar on the Demographic Window and Healthy Aging: Socioeconomic Challenges and Opportunities, May 10–11, 2004. Yip, P. S. F. and S. L. K. Cheung. 2005. “A study of policies and practices in selected countries that encourage childbirth,” Consultancy Study on Population-Related Matters. Hong Kong: Central Policy Unit. Yip, P. S. F., J. Lee, B. Chan, and J. Au. 2001. “A study of demographic changes under sustained below-replacement fertility in Hong Kong SAR,” Social Science & Medicine, 53: 1003–1009. Yip, P. S. F., J. Lee, and T. H. Lam. 2001. A Study of Barriers on Fertility. Hong Kong: Central Policy Unit. Yip, P. S. F. and J. Lee. 2002. “The impact of the changing marital structure on fertility of Hong Kong SAR,” Social Science & Medicine 55: 2159–2169.
8
Ultra-low fertility in Singapore Some observations Yap Mui Teng
Introduction Singapore’s Total Fertility Rate (TFR) fell to 1.26 children per woman in 2003 and has since hovered at about this level. This decline to below 1.3 births per woman has been an unprecedented development and it places the country squarely among those with the “lowest-low” fertility, as defined by Kohler and his colleagues (Kohler et al. 2002). This development has taken place in spite of Government effort to raise the fertility rate through pro-natalist, pro-family policy measures. Following Kohler et al. (2002), this chapter examines the demographic and socioeconomic conditions that underlie “lowest-low” fertility in Singapore (referred to as “ultra-low fertility” in this book to avoid the impression that fertility could go no lower).
Factors contributing to the emergence of lowest-low fertility in Europe Kohler et al. (2002: 641–2) identified five factors that together led to the emergence of lowest-low fertility in Europe. These are: 1
2 3
4
demographic distortions of period total fertility measures, caused by the postponement of fertility and changes in the parity composition of the population, that have reduced the level of period fertility below the associated level of cohort fertility; economic and social changes that have made the postponement of fertility a rational response for individuals; social interaction processes affecting the timing of fertility that have rendered the population response to these new socioeconomic conditions substantially larger than the direct individual responses, and as a consequence, modest socioeconomic changes can explain the rapid and persistent postponement transitions from early to late age patterns that have been associated with recent trends towards low and lowest-low fertility; institutional settings in Southern, Central and Eastern European countries that have favored an overall low quantum of fertility;
Ultra-low fertility in Singapore 161 5
postponement-quantum interactions that have amplified the consequences of these settings and have caused particularly large reductions in completed fertility in lowest-low fertility countries in which childbearing is delayed.
This chapter attempts to examine some of these issues in the Singapore context.
Demographic trends: postponement and reduction of fertility Singapore experienced an extended post-World War II baby boom, with the TFR exceeding six children per woman at its peak in the late 1950s. From this peak, the TFR fell rapidly to reach replacement level in 1975, a period of less than 20 years. Fertility fell below replacement in 1977 and has remained below replacement level since (see Figure 8.1). As mentioned earlier, fertility fell to the lowest-low range in 2003. Marriage trends
70,000
7.0
60,000
6.0
50,000
5.0
40,000
4.0
30,000
3.0
20,000
2.0
10,000
1.0
0 1965
1975
1985 Births
1995
TFR (per woman)
Births
In Singapore, marriage remains a precondition for childbearing – the proportion of babies born out of wedlock (indicated by absence of father’s particulars in birth registration) was only 1.3 percent in 2005 (Singapore Registry of Births and Deaths, “Report on Registration of Births and Deaths” 2005). Figure 8.2 (based on Singapore Department of Statistics 2006b) below shows that marriage rates, defined as the number of marriages per 1,000 unmarried resident population, declined for males and females of all age groups between 1980 and 2000. However, while the shift in age patterns over the period 2000–2005 suggests that there has been some postponement and catching up among males, this
0.0 2005
TFR
Figure 8.1 Live births and total fertility rate, 1957–2005. Source: Singapore Registry of Births and Deaths various years, Singapore Department of Statistics 2006b. Note: Figures from 1980 onwards are residents only.
162 Yap Mui Teng Males Per 1000 unmarried resident males
160 140 120 100 80 60 40 20 0 15–19
20–24 1980
25–29 1990
30–34 2000
35–39
40–44
2005
Females Per 1000 unmarried resident females
160 140 120 100 80 60 40 20 0 15–19
20–24
1980
25–29
1990
30–34
2000
35–39
40–44
2005
Figure 8.2 Age-specific marriage rates, male and female, 1980–2005. Source: Singapore Department of Statistics 2006b.
was not the case for females. For the latter, there were fewer marriages contracted throughout the period observed. The Total Marriage Rates (TMRs) corresponding to the above age-specific marriage rates are presented in Table 8.1, which shows that while the male TMR seems to have stabilized in the early 2000s, the female TMR has continued to decline. The postponement of marriages may also be indicated by the rising median ages of grooms and brides at registration of first marriage. The male age at first marriage rose from 26.9 years in 1970 to 28.7 in 2000 and then to 29.8 in 2005
Ultra-low fertility in Singapore 163 Table 8.1 Total Marriage Rate (per 1,000 unmarried singles)
1980 1990 2000 2005
Males
Females
2494 1752 1680 1675
2027 1746 1570 1341
Source: Computed by author from Singapore Department of Statistics 2006b.
(Singapore Department of Statistics 2006b: v). The corresponding figures for females were 23.1, 26.2 and 26.9 years respectively. In 2005, 71 percent of the male and 46 percent of the female resident population were single at ages 25–29 (Table 8.2). These were higher than the 67 percent and 41 percent in 1995, and 64 percent and 40 percent in 2000. About one-third (34 percent) of males and 22 percent of females aged 30–34 were also single in 2005 – higher than in 2000. On the other hand, the proportions of singles at ages 35–39 years appear to have stabilized since 1995, at 20 percent and 15 percent for males and females respectively. The proportions of singles at ages 40–44 (which, for females, is essentially at the end of their reproductive period) also appear to have stabilized in the early 2000s, at around 15 percent for males and 14 percent for females. Table 8.3 shows the proportions single by age, sex, and educational attainment for 2000 and 2005. The proportions single rose across nearly all educational groups among all the age groups of males. These changes probably reflect the responses of Singaporean men to the economic and labor-market uncertainties and economic restructuring during this period (more of which will be discussed below). The singlehood rates among the least-qualified older males (aged 35–44) have leveled off, but they remain far higher than those for other educational groups. Among females, increases in the proportions single in the 25–29 age group were mainly among secondary and post-secondary educated women, and to a smaller extent, among the university educated. Among women aged 30 and older, however, the singlehood rates leveled off, and even declined among the Table 8.2 Proportions single by age and sex (percent) 1995
2000
2005
Age Group
Males
Females
Males
Females
Males
Females
25–29 30–34 35–39 40–44
67 34 20 12
41 20 15 13
64 31 20 15
40 19 15 14
71 34 20 15
46 22 15 14
Sources: For 1995 and 2005: Singapore Department of Statistics 2006b: Chart 4, p. 5; For 2000: Singapore Department of Statistics 2006c: Chart 1, p. 6.
164 Yap Mui Teng Table 8.3 Proportions single by age, sex, and educational attainment (percent)
Males 20–24 25–29 30–34 35–39 40–44 Females 20–24 25–29 30–34 35–39 40–44
Below Secondary
Secondary
Post-secondary
University
2000
2005
2000
2005
2000
2005
2000 2005
88 62 40 28 21
89 67 42 29 23
92 58 29 18 14
91 63 35 22 14
96 70 28 15 9
97 69 33 18 12
96 69 28 13 7
98 73 32 14 8
53 21 13 10 9
50 21 13 10 10
71 30 17 14 15
71 34 17 14 14
86 43 23 20 20
87 47 24 17 18
91 54 27 23 22
90 56 28 20 21
Source: Computed by author from Singapore Department of Statistics 2006c: Table 3, pp. 67–71. Leow Bee Geon 2001, “Census of Population 2000, demographic characteristics,” Statistical Release 1. Singapore Department of Statistics: Table 16, pp. 74–78.
better educated. There was therefore some narrowing of the gap in singlehood rates between less educated and more educated women, but the gap remained quite wide. The trend data have probably been complicated by the influx of migrants who became citizens and permanent residents (averaging, respectively, 8,300 and 38,700 annually) over the period 2001–2005. The majority of these were highly qualified “independent” migrants while the rest included spouses and minor children of Singapore citizens and permanent residents (Singapore Parliament Reports, various volumes, 2007). Indirect data based on the 2005 mid-decade mini-census suggest that migrants are more likely to be married as the proportions single among citizens are generally higher than that for the resident population (Singapore Department of Statistics 2006c). Table 8.4 shows the singlehood rates among Singapore’s three main ethnic groups: the Chinese, who constitute about 75 percent of the resident population; the Malays, 14 percent; and the Indians, about 9 percent. As the table shows, singlehood rates are highest among Chinese Singaporeans, regardless of age and gender. In 2000, these were generally followed by the Indians and then the Malays. By 2005, however, Malay male singlehood rates at ages above 30 were well above those among the Indians due to increases in singlehood rates among the Malays and declines among the Indians. The postponement of marriage among Malay males is likely attributable to the impact of economic recession and the restructuring of the Singapore economy during this period, which affected the lower socioeconomic strata the most. Malay Singaporeans tend to have the lowest socioeconomic attainment among the three main ethnic groups, and even as this community has progressed along with the others, the gap between them – for example, in median monthly household income − is also widening (see Ministry
Ultra-low fertility in Singapore 165 Table 8.4 Proportions single by age, sex, and ethnic group (percent) Chinese
Males 15–19 20–24 25–29 30–34 35–39 40–44 Females 15–19 20–24 25–29 30–34 35–39 40–44
Malays
Indians
2000
2005
2000
2005
2000
2005
100 96 67 33 22 17
100 97 73 37 22 17
100 91 53 25 13 8
100 92 62 30 17 11
100 94 58 23 13 8
100 95 64 19 11 8
100 87 45 22 17 15
100 91 51 26 17 16
97 72 26 12 9 8
98 76 33 12 9 8
98 71 23 13 9 9
99 73 23 9 9 10
Source: Computed by author from Singapore Department of Statistics 2006c: Table 2, pp. 62–66. Leow Bee Geon 2001, “Census of Population 2000, demographic characteristics,” Statistical Release 1. Singapore Department of Statistics: Table 14, pp. 66–70.
of Community Development, Youth and Sports n.d.). In terms of their occupations, the majority of Malays were in the lower occupational strata, including production and related workers (26 percent), cleaners, and laborers (11 percent), sectors that face downward wage pressures due to competition from low-waged countries in the region. Interestingly, the singlehood rates among Indians have either plateaued or even declined. The only exception was for males in the 25–29 age group. These trends in the proportions remaining single among Indian Singaporeans could be due to greater Indian immigration (i.e., foreigners who became Singapore permanent residents and citizens) during this period. There are no direct data on immigration by ethnicity. However, indirect data show that the Indian share of the resident population increased over the period 2000–2005, while the shares of Chinese and Malays declined (Department of Statistics 2006c: viii) despite the Indian community having experienced more than two decades of below-replacement fertility at levels quite close to that of the Chinese (see below). As mentioned above, new immigrants were more likely to be married, although again there are no data on ethnicity. Fertility trends A comparison of the age-pattern of fertility over time shows that the decline in fertility has been accompanied by a shift towards fertility at older ages (Figure 8.3). Fertility rates at ages 30 and older increased between 1980 and 2005, though not to the same extent as fertility in the younger age groups has fallen. Notably,
166 Yap Mui Teng
Per 1000 resident women
160 140 120 100 80 60 40 20 0 15–19
20–24 1980
25–29 1990
30–34 2000
35–39
40–44
2005
Figure 8.3 Age-specific fertility rates, 1980–2005. Source: Singapore Department of Statistics 2006b.
the peak age at childbearing has shifted to ages 30–34 years in 2005, from 25–29 previously. In fact, the cross-over point (not shown in the chart) was in 2002. The delay in entry into parenthood may also be seen from the ages of mothers at first birth obtained from birth-registration data. According to the “Report on Registration of Births and Deaths” 2002 (Singapore Registry of Births and Deaths c. 2003), the median age of first-time mothers was 29 years in 2002. The corresponding figure for 2005, the most recent year for which figures are available, was 29.2 years, down from 29.3 years in 2004 (Singapore Registry of Births and Deaths c. 2006). Decomposition of the change in TFR over the 2000–2005 period shows that the decline was due to both declining proportions married and declines in marital fertility. The proportion married declined for most age groups within the reproductive age range (Figure 8.4), but especially in the prime childbearing age, while marital fertility declined for all except the 15–19 age group (Figure 8.5). However, as the former figure shows, only minuscule proportions (1 percent in 2000 and 0.6 percent in 2005) of females aged 15–19 were married. Births to mothers in this age group accounted for only about 2 percent of resident births that occurred in the two years, some of which were to unmarried mothers. Declining marital fertility contributed more to the decline in fertility over the 2000–2005 period than did the decline in proportions married (Table 8.5). About one-third (38 percent) of the total decline in TFR during this period was due to changes in marital patterns (proportions married), while about two-thirds (64 percent) was due to decline in marital fertility. This is despite the fact that the country has adopted a pro-natalist policy since 1987 − albeit a selective one in the beginning, which was subsequently liberalized and enhanced, including two major revisions in the early 2000s (see Appendix and below). The decline
Ultra-low fertility in Singapore 167 90 80 70 Per cent
60 50 40 30 20 10 0 15–19
20–24
25–29 2000
30–34
35–39
40–44
2005
Figure 8.4 Proportion married among resident females, 2002–2005. Source: Singapore Department of Statistics 2006b.
in marital fertility rates suggest that married Singaporeans have not responded to the Government’s efforts but, as mentioned earlier with regard to singlehood rates, the early years of the new millennium were an especially difficult period, marked by uncertainty and restructuring of the economy. This could have also affected fertility decisions. As will be argued below, the search for an explanation for the unprecedented low fertility should probably go beyond the business-cycle hypothesis and look into the impact of globalization à la McDonald (2002). As with the differential singlehood patterns discussed above, the three main ethnic groups also differed in their fertility behavior (see Table 8.6). Fertility among
Per 1000 resident married women
1400 1200 1000 800 600 400 200 0 15–19
20–24
25–29 2000
30–34 2005
Figure 8.5 Resident marital fertility rates, 2002–2005. Source: Singapore Department of Statistics 2006b.
35–39
40–44
168 Yap Mui Teng Table 8.5 Decomposition of TFR change, 2000–2005 1 2 3 4 5 6 7 8 9
TFR 2000 (Actual, per 1,000 women) TFR 2005 (Actual, per 1,000 women) TFR 2005 (assuming constant marital fertility) TFR 2005 (assuming constant proportions married) Absolute change 2000–2005 [2–1] Absolute change due to changes in proportions married [3–1] Absolute change due to changes in marital fertility [4–1] Proportion due to changes in proportions married [(6/5)*100] Proportion due to changes in marital fertility [(7/5)*100]
1599.5 1249.5 1465.8 1374.9 –350.0 –133.7 –224.6 38.2% 64.2%
Source: Computed by author.
Malay Singaporeans, although on a declining trend, remained at the replacement level in 2005. The decline in the TFR to the “lowest-low” level in recent years is mainly attributable to Chinese and, to a lesser extent, Indian fertility declines. In 2005, the TFR among Chinese Singaporeans was only 1.08 children per woman, which was the same as for Korea, but marginally higher than that for Hong Kong (just under 1 child per woman) and lower than Japan’s 1.25. Data on the mean number of children ever born obtained from censuses in 1990 and 2000, and the mid-decade mini-census in 2005, show declines in family sizes in all age groups over time (Table 8.7). If ages 40–49 may be taken as the end of the reproductive span, it may be noted that the completed family size was 2.1 children – well above the period TFR. As Table 8.8 shows, Malay Singaporeans have the largest family size in 2005, at 2.9 children per woman, followed by the Indians (2.2) and finally the Chinese (2.0). The slight increase in Malay family size since 2000 probably reflects the rise in Malay TFR following the relaxation of the antinatalist policy in the late 1980s (see Table 8.6). However, it is doubtful whether the overall family size can be sustained in the future. This is particularly the case as more Singaporean women acquire higher education, and the better educated are likely to have fewer children (Table 8.9). It is worth noting, though, that the Table 8.6 Total Fertility Rates by ethnic group Year
1980 1985 1990 1995 2000 2005
All Ethnic Groups*
1.82 1.61 1.83 1.67 1.60 1.25
Ethnic Group Chinese
Malay
Indian
1.73 1.46 1.65 1.51 1.43 1.08
2.19 2.28 2.69 2.50 2.54 2.07
2.03 1.78 1.89 1.71 1.58 1.25
Sources: Singapore Ministry of Community Development, Youth and Sports n.d.: Chart 9.3, p. 13, (based on Department of Statistics). * Includes “Others”.
Ultra-low fertility in Singapore 169 Table 8.7 Mean number of children born among ever-married resident females Age Group (years)
Mean Number of Children Born 1990
2000
2005
Total 15–29 30–39 40–49 50 and over
2.8 1.0 1.9 2.8 4.7
2.5 0.8 1.8 2.2 3.9
2.4 0.8 1.6 2.1 3.3
Source: Singapore Department of Statistics 2006b: Table 8, p. 12.
differential in children born by educational attainment has also narrowed over the years, so the downward pressure on fertility rates exerted by the increasing proportion with higher education may not be as strong as in the past. The effect of the changing tempo of fertility on the observed period TFR has been analyzed by Navaneetham and Lutz (2002). By computing Adjusted Total Fertility Rates (à la Bongaarts and Feeney), Navaneetham and Lutz found that the adjusted fertility for second- and third-order births increased from the year 1987 to 1989 and started declining again from 1989. The adjusted fertility rate for first-order births, on the other hand, increased from 1988 to 1993 before starting to decline. The decline in first-order births was the main contributor to the decline in the adjusted TFR from 1993 onwards, contributing 55 percent to the overall decline between 1993 and 1998, while second- and third-order births contributed 29 percent and 16 percent respectively. This was in contrast to the 1970s and early 1980s when the decline was mainly caused by declines in second- and thirdorder births. The narrowing of the gap between the Adjusted and observed TFRs in 1998 led Navaneetham and Lutz to conclude that fertility decline in the late 1990s was due to the quantum effect rather than changes in the timing of childbearing. Comparing cohort TFRs for cohorts born in the 1950s and 1960s, they also suggested that the younger cohorts that have been postponing childbearing are unlikely to be able to make up for lost time and therefore unlikely to attain the Table 8.8 Mean number of children born by resident ever-married females aged 40–49 by ethnic group
All Ethnic Groups Chinese Malays Indians
1990
2000
2005
2.8 2.6 3.5 2.9
2.2 2.1 2.8 2.2
2.2 2.0 2.9 2.2
Sources: Leow Bee Geok 2001, “Census of Population 2000,” Advance Data Release. Singapore Department of Statistics, Tables 6 and 7, pp. 66–67 for 1990 and 2000; figures for 2005 are computed by author based on “General Household Survey 2005, Statistical Release 1, Socio-Demographic and Economic Characteristics,” Singapore Department of Statistics 2006: Table 4, pp. 72–76.
170 Yap Mui Teng Table 8.9 Mean number of children born among ever-married resident females aged 40–49 by education Mean Number of Children Born Year
Below Secondary
Secondary
Post-secondary
University
1990 2000 2005
3.0 2.4 2.3
2.1 2.1 2.1
2.1 2.0 2.0
2.0 1.9 1.9
Source: Singapore Department of Statistics 2006b: Table 9, p. 12.
family sizes attained by the older cohorts (at 1.8 for the former as compared to over 2 attained by the latter).
Economic and social changes and institutional settings The Singapore economic success story has been documented elsewhere and need not be recounted here. It was one of the four newly industrialized economies (NIEs) that experienced what came to be known as the Asian economic miracle. Some indicators of socioeconomic development within the country that are likely to have had an impact on fertility are shown in Table 8.10. Recognizing that its people are its only resource, one of the early emphases of the Singapore Government was investment in education − although the expansion of tertiary education was a more recent phenomenon. The mean years of schooling increased from 4.7 years in 1980 to 8.6 years in 2000, rising further to 9.3 years in 2005. Over the same period, the proportion of resident population aged 25 and over with university qualifications rose from 2.7 percent to 12.1 percent and 17.5 percent respectively. According to the Singapore Department of Statistics (2006a: p. 61), “Educational progress (over the last decade) was particularly significant Table 8.10 Some trends in Singapore’s human resources
1980 1985 1990 1995 2000 2005
Mean Years of Schooling among resident non-students aged 25 and over
Labor-force Participation Rate (% of resident population aged 15 and over)
Professional, Managerial, Technical Workers (% of employee residents aged 15 and over)
Male
Female
Male
Female
Male
Female
M–F
5.6 6.5 7.3 8.4 9.2 9.9
3.7 4.9 5.9 7.2 8.1 8.8
81.5 79.9 79.0 78.4 81.1 78.2
44.3 44.9 53.0 50.1 55.5 56.6
20.1 24.8 27.4 41.6 37.8 39.0
13.9 16.3 19.5 27.0 32.6 34.9
6.2 8.5 7.9 14.6 5.2 4.1
Source: Singapore 2006a, pp. 53–54.
Ultra-low fertility in Singapore 171 among younger adults. The proportion of university graduates among residents aged 25−39 years increased from 12 percent in 1995 to 33 percent in 2005”. In comparison, the increase was only from 4.1 percent to 9.1 percent among those aged 40 and over. Higher education has, however, also resulted in a decline in the labor-force participation rate of those aged 20−24 years, from 77 percent in 2000 to 72 percent in 2005 (Singapore Department of Statistics 2006a: p. 55). According to the report as well, gender disparity in educational attainment has narrowed although females continue to have fewer years of schooling than males (Singapore Department of Statistics 2006a: p. 52). More significantly, it also noted that “(f)emales are on par with their male counterparts in the pursuit of higher education. In 2005, females constituted about half of the 10,300 graduates from university”. As Table 8.10 shows, over one-third of women in the workforce in 2005 were professional, managerial, and technical workers, as compared to only 14 percent in 1980. The gender gap among such workers is also narrowing. Even as progress has been made in the roles and statuses of women outside the home, however, working women in Singapore are said to be faced with “three paradoxes” − they are called upon to be economically active and to contribute to the economy and their family (not surprising, given the small population size and the high educational attainment of women), yet at the same time, they are also expected to be traditional wives and mothers (Lee et al. 1999, see also Quah 1994). Additional income-tax relief for children is given to working mothers to encourage them to remain in the workforce, a policy that was first adopted even as the Government sought to limit family sizes to no more than two children per family in an anti-natalist phase that lasted from the mid-1960s until the mid1980s. As Lee et al. have noted, the adoption of the pro-natalist policy from the mid-1980s has probably increased the burden on women as the measures adopted to promote larger family sizes typically assigned the care-giver role to mothers. Thus, a childcare subsidy is given to children of working mothers enrolled in childcare centres, while this was not available to non-working mothers until 2002 – the unspoken assumption being that childcare was the mother’s role and centerbased services are subsidized only where it is used as a substitute for maternal care due to her employment. When part-time work, unpaid childcare leave, and sick childcare leave were introduced in the civil service in 1987 as an example to other employers, these were available only to females and not to males. When paternity leave was introduced in the civil service in 2000, it was only for three days – something which mothers, and some fathers, felt was hardly adequate. However, the policy setting in which Singaporeans have to make their marriage and fertility decisions is hardly an unchanging one. In response to Singaporeans’ complaints about time and financial constraints as reasons for not having children or not having more of them, the Singapore Government has made several adjustments to the package of pro-natalist measures, sometimes in a major way (see Appendix). For example, a cash incentive in the form of the Baby Bonus was introduced for the first time in 2000, something which the Government had resisted doing because of concerns over potential abuse. Savings in the Child Development Co-Savings account, to which the Government also contributes
172 Yap Mui Teng as part of the Baby Bonus scheme, may also be used to pay for childcare and kindergarten costs, and subsequently healthcare costs. In 2004, the subsidy for center-based infant care (which is generally costlier than childcare) was raised substantially to S$400 per month. In 2004, as well, a grandparent care-giver relief was introduced for working mothers whose children are cared for by their parents or parents-in-law. Working mothers receive additional tax relief and additional tax rebates that are not available to other taxpayers. Working mothers who employ foreign domestic helpers to care for their children are entitled to claim incometax relief equal to twice the amount of foreign-worker levy paid – a privilege that men and singles are not entitled to. The quantum of the levy, which the Government had used as a price mechanism to ameliorate demand for foreign domestic helpers, was itself reduced. The Government is also promoting work–life balance by adopting a five-day work week in the civil service (something that has been resisted previously) and introducing a Work–Life Works fund that employers can tap to ease the cost of introducing work–life measures. The quantitative family size target of at least three children per family, as embodied in the slogan “Have three, or more if you can afford it” adopted in 1987, was quietly dropped in 2004 and the current target is to create a conducive environment for families at each stage of family formation.
Developments since 2000
16 14 12 10 8 6 4 2 0 –2 –4
3.5
TFR (per woman)
3.0 2.5 2.0 1.5 1.0 0.5 0.0 1980
1985 TFR
1990
1995
GDP Gr
Unemployment Rate
2000
Figure 8.6 TFR and GDP growth/unemployment rates, 1980–2005. Source: Singapore Department of Statistics 2006a.
2005
GDP Growth/Unemployment Rate (%)
Perhaps most significant among recent developments that could have affected fertility trends in Singapore are the economic fluctuations and rise in unemployment due to globalization and economic restructuring in the early 2000s. As Figure 8.6 shows, the years since the Asian Financial Crisis in 1997 have seen economic growth in Singapore fluctuate widely between negative and 10 percent growth rates. Singapore has been affected by the global economic downturn as well as the impact of SARS and the September 11 attack on the US. Singapore
Ultra-low fertility in Singapore 173 also experienced its highest unemployment rates in recent times, and structural unemployment, which was previously a situation for the older, less educated, also affected white-collar, managerial workers. Whereas in the past, the TFR tracked economic performance, this was not the case in the recent years – while the economy has improved, and unemployment was on the way down, the TFR has, at best, leveled off. Economic uncertainty is likely to lead the young to postpone the onset of childbearing, as Kohler et al. (2002) explain: “Postponement can reduce the uncertainty about the costs and benefits of children, and also the uncertainty associated with the economic situation and the stability of unions in early adulthood”. Moreover, in low-fertility countries, where only one or two children (if any) are desired, “the timing of fertility is relatively flexible. The desired onset of childbearing can range over two decades in the life span from the later teenage years to the mid and late 30s (potentially also later). The timing of fertility in lowest-low-fertility countries is therefore likely to be sensitive to changes in socioeconomic conditions, especially at low parities” (Kohler et al. 2002: pp. 652–53). In Singapore, while youth unemployment worsened over the period 1993–2003, rising from 5.5 percent to 7.8 percent, the ratio of the rates of youth unemployment to overall unemployment actually fell (Ministry of Manpower c. 2005). Moreover, “unlike in some developed countries, youths in Singapore generally do not experience long unemployment spells” (ibid.). Be that as it may, social interaction among youths and between youths and older Singaporeans may influence their perception of uncertainty (as has been posited by Kohler et al. in the case of European youths). Globalization affects the labor market in another way, as McDonald (2002) explained. The new labor market under globalization rewards the successful but is also riskier. In such an environment, there is greater incentive for the young to invest in education to enhance their human capital and competitive edge; for couples to reduce the risks of job loss and dependency by adopting a dual-income strategy; and for women to remain in the workforce. According to McDonald, “The demographic outcomes of the emergence of the new economy are longer dependence of children upon their parents, increased participation and longer duration of education, difficulty in forming relationships, delayed marriage or non-marriage, delay of childbearing and childlessness” (McDonald 2002). In his view, countries with very low fertility, including Singapore, experience these outcomes to varying degrees.
Speculations about the future Singapore is a small, open economy and it cannot afford to be closed to global forces. As such, it is likely to be exposed to the global up- and downturns. However, the Government is also taking steps to bolster the economy and position the country for future growth by investing heavily in the training of its people to take advantage of new opportunities. This is shown by the resource allocation in the most recent (2007) Singapore Government budget (www.mof.gov.sg/
174 Yap Mui Teng budget_2007/budget_speech/index.html). It sticks to its philosophy that job creation is the best way to help its middle class. With better economic conditions, perhaps there will be less incentive to postpone marriage and childbearing. If this is the case, it could combine with the favorable demographic structure created by the surge in births in the years following the 1987 reversal of the anti-natalist policy to at least increase the number of babies born over the next decade or so. As shown in the chart (Figure 8.1) at the beginning of this chapter, the number of babies born annually rose sharply in 1988 and for the next decade or so remained above the pre-1988 level. The average annual number of babies born jumped to about 47,900 annually over the period 1988–1997, from about 40,300 annually over the period 1980–1987. These larger cohorts have begun entering their reproductive ages – the babies born in 1988 turned 15 in 2003 while those born in 1997 will reach the same age by 2012; more importantly, they will be in the prime ages for childbearing (25–34 years) from about 2012–2031. If this momentum generated by age-structure changes is not dissipated by extended changes in birth timing, mainly in the form of delay in childbearing, an echo of the baby boom may be expected to take place. Whether further delays in birth timing take place will depend on whether the forces delaying childbearing have run their course. These include the structural economic changes, educational trends in association with labor-market trends, further incentives offered by Government, and ideational changes making for a further decline in family size desires, i.e., if the younger Singaporeans want and have fewer children or if more of them remain unmarried or childless.
References Bongaarts, J. and G. Feeney (1998). “On the quantum and tempo effect of fertility,” Population and Development Review 24(2): 271–291. Geok, Leow Bee (2001). “Census of population 2000,” Advance Data Release. Singapore: Singapore Department of Statistics. Kohler, Hans-Peter, Francesco C. Billari, and Jose Antonio Ortega (2002). “The emergence of lowest-low fertility in Europe during the 1990s,” Population and Development Review 28(4): 641–680. Lee, Jean, Kathleen Campbell, and Audrey Chia (1994). The Three Paradoxes; Working Women in Singapore. Singapore: Association of Women for Action and Research. McDonald, Peter (2002). “Low fertility: unifying the theory and the demography,” paper prepared for Session 73, Future of Fertility in Low Fertility Countries, 2002 Meeting of the Population Association of America, Atlanta, May 9–11, 2002. Navaneetham, Kannan and Wolfgang Lutz (2002). “Fertility trends in Singapore: period and cohort analysis,” manuscript. Quah, Stella R. (1994). “Gender roles: striving for coherence,” in Stella R. Quah, Family in Singapore; Sociological Perspectives. Singapore: Times Academic Press: pp. 177–217. Saw, Swee-Hock (2005). Population Policies and Programmes in Singapore. Singapore: Institute of Southeast Asian Studies. —— (1990). Changes in the Fertility Policy of Singapore, IPS Occasional Paper No. 2. Singapore: Institute of Policy Studies and Times Academic Press. Singapore Department of Statistics (2006a). “Singapore 2006, Statistical Highlights”.
Ultra-low fertility in Singapore 175 Downloaded February 17, 2007 from: http://www.singstat.gov.sg/pdtsvc/pubn/softcopy/ sh2006.pdf. —— (2006b). “Population Trends 2006”. Downloaded February 15, 2007 from: http:// www.singstat.gov.sg/pdtsvc/pubn/softcopy/population2006.pdf. —— (2006c). “General Household Survey 2005, Statistical Release 1: Socio-Demographic and Economic Characteristics”. Singapore: Singapore Department of Statistics. Singapore Ministry of Community Development, Youth and Sports (n.d.). “Progress of the Malay Community in Singapore since 1980”. Downloaded January 22, 2008 from: http://www.mcys.gov.sg/MCDSFiles/download/ProgressofMalayCommunity.pdf. Singapore Ministry of Finance (2007). “Singapore Budget 2007”. Downloaded on February 20, 2007 from: http://www.mof.gov.sg/budget_2007/budget_speech/index.html. Singapore Ministry of Manpower (c. 2005). “Singapore Report for Symposium on Globalization and the Future of Youth in Asia”. Downloaded February 21, 2007 from: http:// www.mhlw.go.jp/topics/2005/05/dl/tp0512-1b12.pdf. Singapore Parliament Reports (2007). Downloaded on January 27, 2008 from: http://www. mhlw.go.jp/topics/2005/05/dl/tp0512-1b12.pdf. Singapore Registry of Births and Deaths (various years). “Report on Registration of Births and Deaths”.
Enhanced Child Relief (ECR) Additional income-tax relief (on top of normal child relief) given to highly qualified working mothers to incentivize them to remain in the workforce raised from flat 5 percent of earned income for first three children to 10 percent for second child and 15 percent for third child, subject to $10,000 cap for each child; at the same time, criterion for eligibility lowered to 5 GCE O Level passes.
Effective Year of Assessment 1988
Income-tax Reliefs, Rebates Normal income-tax relief for third child raised to $750 (up from $500), equal to that for first two children. Enhanced Child Relief amounting to 15 percent of earned income extended to fourth child born on or after January 1, 1988; educational qualifications lowered to 3 GCE O Level passes. $20,000 tax rebate for third child born on or after January 1, 1987 introduced. Rebate may be offset against either or both spouses’ income-tax liabilities. Additional tax rebate equal to 15 percent of earned income in lieu of maternity leave (which remained limited to the first two births) for working wives who opt for separate income-tax assessment. Delivery and hospitalization expenses for fourth child may be offset against parents’ earned income, subject to maximum of $3,000.
Effective 1984 Primary School Registration (Graduate-mother Priority Scheme) registration exercise Children of women with university degrees or professional qualifications given priority in Primary 1 registration exercise (measure discontinued in 1985). Social Development Unit (SDU) Social Development Unit (SDU) established to provide opportunities for social interaction and match-making services to single officers in Government departments, statutory boards, and Government-linked companies that employ university graduates. 1987 March “Have three, or more if you can afford it” “Stop at two” policy replaced by “Have three [children], or more if you can afford it”.
Year: 1984 Effective Year of Assessment 1985
Note: Monetary units are Singapore dollars. In 2005, S$1 = US$0.6; in July 2008, S$1 = US$0.735
Chronological development of pro-natalist measures in Singapore
Appendix
Appendix continues
Childcare Subsidy $100 monthly subsidy for children of working mothers attending full-day program at childcare center (amount halved for part-time program). Subsidy paid directly to childcare center rather than to parents. Medisave for Delivery and Hospitalization Charges Use of Medisave (savings in the medical savings account of provident fund savings) for delivery and hospitalization charges for third child. Accouchement Fees for Post-partum Sterilization In Lower Class Wards in Government Hospitals Waiver of accouchement fees if post-partum sterilization is performed after third or higher order birth instead of second birth. Public Housing Upgrading for Three-child Families Rules eased for families living in three-room or larger public flats who have a third child born on or after January 1, 1987 to upgrade to larger flat. Part-time Work for Female Civil Servants Part-time work (21 hours per week) for up to 3 years at half-monthly salary. Childcare Leave for Female Civil Servants Up to four years unpaid leave to care for young children, subject to exigencies of service. Sick Childcare Leave for Female Civil Servants Full-pay unrecorded leave of up to 5 days per child and maximum of 3 children (total 15 days annually) to look after sick children under age 6. Unrecorded Sterilization Leave for Civil Servants Eligibility for one week unrecorded post-sterilization leave restricted to those with no O Level passes. Abortion Counselling Compulsory pre- and post-abortion counselling for secondary educated women with fewer than three children introduced. Pre-sterilization Counselling Compulsory counselling for men and women with only one or two children introduced.
Primary School Registration Disincentives against third and higher order births removed.
Foreign-maid Levy Relief for Working Mothers Income-tax relief equal to two times annual foreign-maid levy paid, for working mothers only. Second-child Tax Rebate Tax rebate, on a sliding scale, for second child born on or after January1, 1990 to mothers aged below 31, from $20,000 if child was born before age 28 to $5,000 for birth before age 31. Period of claim for all rebates extended to nine years.
Income-tax Reliefs, Rebates Normal child relief for first, second, and third child doubled to $1,500. New relief of $1,500 for fourth child born on or after January 1, 1988. Enhanced child relief for second, third, and fourth child raised to 15 percent, 20 percent, and 25 percent respectively, subject to $15,000 cap. $20,000 tax rebate extended to fourth child born on or after January 1, 1988. Additional tax rebate of 15 percent in lieu of maternity leave extended to fourth child, claimable only by mothers.
Effective April 2001 Third-child Paid Maternity Leave Eight-week paid maternity leave extended to third child. Paid by Government (rather than employer), subject to maximum of $20,000. Flexible Work for Government Departments Government departments permitted to adopt flexible work hours provided total time worked per week equals 42 hours. Paternity Leave for Civil Servants Three days’ paternity leave for first three children.
2000 Effective April 2001 Baby Bonus Two-tier payment, involving an outright cash component and a co-saving component matched dollar for dollar by the Government, subject to a cap, given annually for second and third children born on or after April 1, 2000. Co-saving component, payable over six years, to be used for childcare or child-development purposes. Savings may be used for all children.
1990 Effective Year of Assessment 1991 Effective Year of Assessment 2001
1989 Effective Year of Assessment 1990
Chronological development of pro-natalist measures in Singapore
Appendix continued
Effective Year of Assessment 2005
Effective Year of Assessment 2005
Effective Year of Assessment 2005
2004
2002
Appendix continues
Working-mother Child Relief Replaces Enhanced Child Relief and Additional Tax Rebate. Reliefs as follows: first child – 5 percent of earned income second child – 15 percent of earned income third child – 20 percent of earned income fourth child – 25 percent of earned income. Grandparent Care-giver Relief Working mothers eligible for $3,000 grandparent care-giver relief if their parents or in-laws look after their children below age 12.
Parenthood Tax Rebates Rebate for second, third or fourth child born or adopted on or after January 1, 2004. Age requirement for second birth removed. Time limit on claim also removed.
“Have three, or more if you can afford it” replaced by “Singapore. A great place for families”. Numerical target for family size removed; new objective: to create a more conducive environment for starting and raising families. Baby Bonus Baby bonus extended to first and fourth children born on or after August 1, 2004 − cash gift of $3,000 and $6,000 for first and fourth children respectively; co-savings component extended to fourth children subject to $12,000 cap.
Childcare Subsidy for Non-working Mothers $75 monthly subsidy for children of non-working mothers enrolled in childcare centers.
Public Housing for Newly-weds For young couples purchasing four-room public flats for the first time, 20 percent down payment to be paid in two stages: 10 percent on signing of agreement, remaining 10 percent on taking possession of flat. Paid Marriage Leave Three days’ paid marriage leave for Government employees marrying for the first time.
Maid Levy Concession Monthly foreign domestic-maid levy reduced from $345 to $250 per foreign maid for families with children aged below 12 or elderly aged 65 years and above. Capped at two foreign domestic maids at any one time. Use of Medisave Extended to delivery and hospitalization charges for fourth children born on or after August 1, 2004. Use for pre-delivery and delivery medical expenses for fifth and subsequent children provided parents have combined Medisave balance of at least $15,000 at the time of delivery. Withdrawal limit for assisted conception procedures raised from $4,000 to $6,000, $5,000 and $4,000 for the first three cycles. Statutory Childcare Leave Parents with children under age seven entitled to two days per year of statutory employer-paid leave. Five-day Work Week for Civil Servants Hours previously observed on Saturday redistributed and collapsed into a five-day week with longer hours from Monday to Friday. CPF Housing Top-up Grant for Singles Getting Married Introduced Singles who previously received Government-housing grant may apply for top up to quantum given to couples if they decide to marry. Work–Life Works!(WOW) Fund $10 million Work–Life Works! Fund established for funding training of HR personnel, infrastructural adjustment, engaging work–life consultants, or for employing or deploying staff to drive work–life implementation in an organization.
Infant-care Subsidy Monthly subsidy for infant care raised to $400 (from $150).
Paid Maternity Leave Extended to working women giving birth to their fourth child. Duration extended from 8 to 12 weeks for the first 4 births. For first two births, government pays for the extra four weeks, subject to $10,000 cap. For third and fourth births, Government funds full 12 weeks, subject to $30,000 cap. The additional four weeks may be utilized flexibly within six months of birth. Four weeks’ maternity leave (to be paid by Government and capped at $10,000) for adoptions on or after August 1, 2004 provided employer agrees to provide such leave on a voluntary basis.
Chronological development of pro-natalist measures in Singapore
Appendix continued
9
From population control to fertility promotion A case study of family policies and fertility trends in Singapore Paulin Tay Straughan, Angelique Chan, and Gavin Jones
Introduction Singapore is a city-state with a total population of about 4.5 million and a resident population of about 3.6 million (Singapore Department of Statistics 2006). The current imagery of Singapore as a cosmopolitan multicultural society at the forefront of technological advancements and international commerce is a stark contrast to the much smaller Third World city it was pre-1960s. Yet, in the 40 years or so since this small city-state gained independence, the transformation from developing to developed status has had major implications, particularly for the family. In this short period, the total fertility rate fell from 4.7 in 1965 (the year of independence) to 1.26 in 2006 (Singapore Department of Statistics 2002). Singapore is now one of the most rapidly aging societies in the world, and as a nation that is totally dependent on manpower as its main resource, population concerns are one of the most serious issues it faces. It is significant to study Singapore’s demographic trends from a sociological perspective. First, as it is a small nation, the impact of state policies on fertility can be monitored with relative ease. Second, as a multicultural society with three main ethnic groups among its permanent residents (75.2 percent Chinese, 13.6 percent Malay, 8.8 percent Indians, and 2.4 percent of other ethnic identities), it is interesting to see how cultural factors intertwine with other developments to affect family life. Of particular concern is the effect of rapid modernization on work– family dynamics and gender role identity. The study of effects on fertility outcomes is necessarily complex as the processes involved are dynamic, particularly when medical advances have now rendered childbirth a choice rather than an act of nature. Because couples can now choose – to a certain extent – when they want children, and how large a family they want to grow, it is important that we understand what factors inhibit and facilitate such decisions. Macro factors like state policies, growth rate, and global outlook inevitably play a significant role in shaping fertility patterns. This is particularly so for a small country like Singapore where international trade is of paramount significance for the health of the national economy. To stay competitive, the state has to manage the fine balance between nurturing a culture that favors competitive work ethics (to attract investors), and one that values investment in family (to achieve a sustainable labor
182 Paulin Tay Straughan, Angelique Chan, and Gavin Jones force). Yet, these ideals are inevitably perceived as being contradictory to each other. The contradictions are more evident when traditional family structures have to make way for new family arrangements that favor the needs of the new economy. One critical development is the foray of women into paid work. As Singapore is highly dependent on human capital as a resource, women are encouraged to enter the labor force, and remain there through retirement. This push towards paid work has important implications for work–family dynamics and gender ideologies. In pre-industrial Singapore, the family functioned through a strong reliance on a clear division of labor along gender lines which allowed men to focus on work commitments outside of the home while women managed domestic affairs and raised children. With the advent of industrialization and modernization, the family had to adapt to meet the challenges that arose. In turn, these developments affected how family is perceived, and the role of children in the family. Cultural attitudes towards family distinctive to the major ethnic groups in Singapore also affect how group members perceive family and their fertility decisions. For example, Chinese couples favor certain zodiac signs as conducive for procreation. So in the year of the dragon, the birth rate for the Chinese subgroup goes up. Conversely, in the year of the white tiger – a zodiac sign that does not favor children – the birth rate goes down. However, while this emphasis on desirable and less desirable years for giving birth affects the specific timing of births, it probably has little influence on total lifetime births. Furthermore, while cultural factors are still influential among middle-aged married couples, we see less differentiation among the younger adults. For example, in a recent study on young adults and their attitude towards marriage and children, we observed a convergence of attitudes among the respondents from different ethnic groups (see Straughan et al. 2006). Past trends in marriage and procreation consistently showed higher fertility rates for Malays, while the Chinese favored smaller families. Sociologists often attribute this ethnic difference to, among other reasons, the more traditional Malay culture that continued to value family formation and strong community ties. However, Malay fertility rates have fallen to replacement in the early 2000s, and the focusgroup discussions with young adults showed career aspirations as top priority across the different ethnic groups. This is very likely due to equal access to formal education, which tends to homogenize attitudes and beliefs regarding the family. This chapter will focus on family policies and their effectiveness in promoting procreation in Singapore. To complete the macro analysis, we will share the findings from a recently concluded large-scale survey on attitudes and barriers leading to low fertility among married couples. In the discussion, we will address how family policies can be invoked to overcome the barriers to growing larger families.
Population and family policies in Singapore The post-independence years, 1965–1975 Like population trends elsewhere, fertility patterns in Singapore correlate strongly with economic development. Figure 8.1 in the previous chapter details the total
From population control to fertility promotion 183 fertility rate from 1960 to 2006. In the 1960s, trade and commerce were the primary economic activities. Birth rates were high, and the leading concern of the Government was population control. Riding the tide of global rapid industrial growth in the late 1970s, the focus of policy was on training manpower for the needs of manufacturing industries that dominated the economic sphere, and grew to provide more than a third of all employment in the mid-1970s. As more jobs were created and the demand for skilled labor increased, formal education became an important channel for social mobility. Given the limited manpower base for the nation-state, all able citizens – regardless of gender – were encouraged to pursue the requisite skills through formal education and vocational training and to join the labor force. Many had no option but to heed the call, for the Government adhered steadfastly to an anti-welfare philosophy. Concurrently, the family policies were focused on population control. The Government believed that as an emerging economy, it was important to curb population expansion so that the state’s resources could be efficiently focused on economic advancements. The concern with arresting population growth was fuelled by the high growth period between 1947 and 1970, which saw the population of Singapore double (Wong and Yeoh 2003). The Family Planning and Population Board (FPPB) was formed in 1966 when the Government took over control of all responsibilities for family planning, and the tag line was “Stop at Two” (Wong 1979). Family policies were focused on disincentives to those who had more than two children, and incentives for those who conformed (for a detailed description of the population control policies, see Saw 2005). Period of rapid economic growth and population decline, 1975–1985 By 1975/1976, the total fertility rate (TFR) had dropped to replacement level, and has remained at below replacement level since. By 1983, the TFR had fallen to an all-time low of 1.61 – well below the replacement rate. Why were the population-control policies so effective? One reason is the comprehensive approach that governed the policies. Incentives covered a range of services and benefits, including public housing (which provides for more than 80 percent of the housing needs of the population), personal income tax, accouchement fees, maternity leave, and even education for children (see Saw 2005). But perhaps a more significant factor that facilitated the downward fertility trend was the underlying economic climate. In her reflections on Singapore’s family planning program in this period of the nation’s development, Wong (1979) argued that the social and cultural factors governing the nation-state at that time acted as catalysts for population-control initiatives. Certainly, the availability of birth-control options (including cheap and safe abortions) and the social acceptance of contraceptive use helped women gain control over their fertility decisions. This relieved them of unwanted pregnancies, and gave them freedom to pursue careers outside of the family. The timing of opportunities for Singapore women to achieve self-actualization could not
184 Paulin Tay Straughan, Angelique Chan, and Gavin Jones have been better. Post-independent Singapore saw tremendous opportunities for economic development and industrialization. Job opportunities were abundant as factories sprouted throughout the island, and agricultural space was transformed into industrial parks. To meet the need for skilled labor, formal education was accessible to all regardless of race or gender. Women were trained, and soon became an indispensable part of Singapore’s labor force. The female laborforce participation rate rose steadily, from 28.2 percent in 1970 to 44.8 percent in 1984 (see http://www.singstat.gov.sg/stats/themes/economy/hist/labour.html for details). While the opportunities for educated women were tremendous, the effects were devastating for fertility trends. Marriage rates and the TFR for this group of women fell sharply, and it was inevitable that the static populationcontrol policies had to be revamped. The tone of family policies did not change for almost a decade after the replacement level of fertility was breached. Finally, in 1984, the Government reversed its population-control polices for a select segment of the population. Popularly known as the eugenics phase of population policies in Singapore, new pro-natalist family policies were introduced to encourage tertiary educated women to marry and have more children. The Social Development Unit (SDU) was set up in 1984 to facilitate this. Concurrently, monetary incentives were offered to less-educated women who abided by the small family ideal encouraged by the state. In short, a two-pronged population policy was adopted: those who were deemed to be capable and able parents were encouraged to have more children, and those who were not were encouraged to stop at two – the replacement level. However, this did not go down well with the population in general (see Wong and Yeoh 2003: 10). And the TFR continued to fall. By 1986, population control was belatedly recognized to be no longer needed, and the Government closed down the Family Planning and Population Board as it was deemed redundant. Balancing work–family interweave, 1986–1999 Though the TFR continued to drop, new population polices adopted during this period were still only selectively pro-natalistic. To set the tone for a change in population policies, a new slogan was adopted. “Have three or more children, if you can afford it” became the tag line for this era, and the leadership shifted focus from eugenics to economics. There was growing awareness of the change in ideology towards having children. Meantime, married couples were moving away from larger families to a concentration on providing quality nurturing for children. Sustained economic growth had resulted in accumulation of excess, and popular culture began to portray an ideology of the child that valued the intrinsic. In pre-industrial societies, children were often viewed as pre-adults who were economically worthless. With industrialization and modernization, family size decreased and mortality rates fell. With a lowered dependence on child labor, children were now valued for their emotional and social worth: “pricing the priceless child” (Zelizer 1994).
From population control to fertility promotion 185 Much of this ideological work on re-defining the child was reflected in Singapore’s population-policy tag lines. Campaign slogans began to impress on the society the importance and significance of children in families. These included “Children – life would be empty without them,” “Life’s fun when you’re a dad and mum,” and “The most precious gift you can give your child is a brother or sister” (Wong and Yeoh 2003: 12). This period saw the dominance of paid work over family in young couples’ scale of values, and the extrinsic rewards from career appealed to the educated young adults seeking returns from investment in formal education. Marriage and procreation took a back seat, and age at first marriage for both brides and grooms rose, as did the proportion of singles. To address this alarming trend, part of the new population policies focused on the never-married segment of society. Media messages like “Why build your career alone? Family life helps” and “Make room for love in your life” for “Life would be lonely without a family” engaged the singles and highlighted the Government’s concern that the push towards economic productivity may have made Singaporeans too work focused (Wong and Yeoh 2003). Several Government-led initiatives attempted to reinstate the family to a centrally valued position. In 1985, the National Family Week – an annual event – was established. This was followed by the inception of the National Advisory Council on the Family and Aged (NACFA) in 1989, which was supported by two other committees – the Committee on the Family (COF) and the Committee on the Aged (COA). Together, these people-sector-led civic groups served to provide a continual focus on family issues and recommend policy directions for the Government. The new population policies focused on four main dimensions: tax incentives, housing, delivery costs, and childcare (see Lien 2002 for details). Tax incentives and housing incentives that were in place to encourage small families a decade earlier were now restructured and refined to encourage those who could to have more children. To facilitate living arrangements for larger families, housing policies gave priority to large families for upgrade to bigger HDB flats. The tax incentives were designed with a dual intent: to persuade middle and higher income couples to have more children, and to encourage women to stay engaged in the workplace while they raised their children. For example, under the Enhanced Child Relief, a progressive increase in percentage of tax relief was given for each child based on the mother’s earned income. Prior to the new population policies, couples could only use Medisave, a compulsory saving scheme for medical expenses, to cover delivery expenses for their first two children. The intent then was to encourage couples to stop at two children, as delivery costs for the third child would have to be paid in cash. The new population policies extended use of Medisave to cover delivery costs of the third child. To help working mothers manage the demands of work and family, a centerbased childcare subsidy was initiated and working mothers were given a set amount for each child they placed in a licensed childcare center (for the first four children). Various leave schemes were also introduced in the civil service for married female officers only. These included no-pay leave for childcare up to a maximum of four
186 Paulin Tay Straughan, Angelique Chan, and Gavin Jones years for each child below four years old, part-time employment for up to three years (regardless of the child’s age), and full-pay unrecorded leave of five days per year to look after a sick child below six years old. These policies relating to maternal employment had a significant impact on the ideology of the family in Singapore. While the state’s intent was to lure mothers back into the workforce to ensure that its manpower needs were met, the message sent to both men and women was very clear: childcare is the sole responsibility of the mother. This continues to have serious implications for marriage and procreation trends. When women weigh the pros and cons of family formation, many often end up overwhelmed by the perceived costs of retreating from their careers and the direct costs of child-rearing. It is not surprising that the foray of women into paid work is not reciprocated by men’s greater involvement in family matters (Silberstein 1992). Costs of child-rearing are permutated as individual costs, and borne disproportionately by women. And given that women play the primary role in fertility decisions, it is not surprising the TFR continues to fall. The second significant consequence of the population policies arose from the philosophy of promoting self-sufficiency and encouraging those who could afford it to have more children. Underlying this is the concern that all children have a right to a decent family life where their basic needs are met and they have a good chance of succeeding as adults. This philosophy would later reinforce the ideology of the sacred child where couples who are well positioned economically believe that they cannot afford to have children because they perceive that they do not have sufficient resources (both tangible and intangible) to provide for the needs of a child. Taken together, these ideologies placed tremendous stress on women. The state upholds an ideology of childhood that requires that the child should be well looked after, and the mother has been assigned this important role. Up to this point in the population policies, fathers continued to be economic providers only, with serious implications for fertility trends in Singapore, as women continued to struggle between the economic benefits of paid work and the demands of maternal duty. Shifting the focus on the family, 2000 and beyond In 2000, the Government introduced what is popularly termed “the baby bonus” scheme, which was further enhanced in 2004. Under this scheme, a Children Development Account is opened for a family upon the birth of the couple’s first four children. A Baby Bonus of S$3,000 cash is given upon the birth of the first child, and for subsequent children, the Government will contribute S$500 each year into the account and up to another $1,000 each year to match contributions dollar-for-dollar from the parents. This contribution will add up to S$9,000 for the second child and S$18,000 for the third and fourth child (Ministry of Community Development, Youth and Sports, 2004a).1 In addition to monetary incentives, the Government also announced that working mothers will get twelve weeks of maternity leave for the first four children (Ministry of Community Development, Youth and Sports, 2004a; Straits Times, August 21, 2000).
From population control to fertility promotion 187 To fill the gap at home when the wife enters paid work, the state initiated the foreign domestic workers scheme which allowed families to employ full-time, live-in maids from around the region (mainly from Indonesia and the Philippines). About 20 percent of households in Singapore hire foreign domestic maids to help manage their households. When a household hires a foreign domestic maid, a tax of S$250 per month is levied by the state. To help dual-earner families cope with childcare and family responsibilities, tax relief for the Foreign Maid Levy was introduced to encourage working married women to have children (Ministry of Community Development, Youth and Sports, 2004a). Public-housing-allocation policies were also introduced to facilitate larger families. The Third Child Priority Housing Scheme was set up to give priority to families who wish to upgrade to bigger public housing (HDB) apartments upon the birth of the third child (Ministry of Community Development, Youth and Sports, 2004a). While foreign maids are an option for dual-earner families, the state is mindful that this remains a temporary solution, and that families should seek a long-term solution that is more stable and not dependent on the availability of foreign labor (Huang and Yeoh 1996). The logical solution to childcare needs is seen to lie in support from grandparents, which will ensure that the children are socialized by adults from the same cultural backgrounds. The Grandparent Caregiver Tax Relief was initiated in 2004 to encourage grandparents as primary caregivers when the mother is engaged in paid work. To facilitate the three-generation extended family, several public-housing policies were also initiated (see Ministry of Community Development, Youth and Sports, 2004a). Not all grandparents, however, are enthusiastic about taking on the role of raising their grandchildren (Teo et al. 2006: Chapter 9). Overall, the pro-family policies announced in 2004 under the new Prime Minister Lee Hsien Loong addressed three main areas: family time, childcare options, and work–family balance (details of the new family policies can be found at http://fcd.ecitizen.gov.sg.htm). One significant variation from previous policies is the removal of mothers’ education qualification from the tax incentives. The fertility rate has slipped so low that it is no longer feasible to maintain selective pro-natalist policies. Perhaps the most significant change in the post-2000 population policies is the ideological change regarding men’s role in the family. Prior to this, childcare leave was only provided for mothers as it was expected that the mother would be responsible for looking after young children. Fathers were expected to play the traditional role of economic provider. In 2000, paternity leave was introduced as part of the new population policies. Although it grants only 3 days of leave for fathers (compared to 12 weeks of maternity leave for mothers), the significance was symbolic because, for the first time, men were considered to be directly involved in childcare. In addition, men are now eligible to take unrecorded childcare leave which was previously only available to their female counterparts in the civil service. This sends a clear message that childcare is a shared responsibility between husbands and wives. Table 9.1 summarizes the inception of the various family population policies,
188 Paulin Tay Straughan, Angelique Chan, and Gavin Jones and notes the corresponding TFR in that year. For more details on the policies, see Appendix 1 in Chapter 8. Despite the introduction of the various initiatives, TFR continues to drop. A slight relief was brought by the years of the dragon in 1988 and 2000, where birth rate rose slightly for the Chinese who believed that dragonyear sons were auspicious. The family policies post-2000 have signaled that it is ideologically, socially, and politically correct to support investment in family. This was further enhanced when state leaders stood up for the social significance of the family in public speeches. As a husband, a father and now a grandfather, I can only tell you that a family adds warmth and meaning to our lives. Friends are important, but a family is indispensable. We would be so much lonelier if we did not have a partner with whom to share our achievements and anxieties, our joys and sorrows. Table 9.1 Summary of Singapore family policies and TFR (1980–present) Policy
Year
TFR
Zodiac Sign
•• Continuation of “Stop at two policy”
1980 1981 1982 1983 1984
1.82 1.78 1.74 1.61 1.62
Monkey Rooster Dog Pig Rat
1985
1.61
Ox
1986
1.43
Tiger
1987
1.62
Rabbit
1988 •• Setting up of National Advisory Council on the 1989 Family and Aged (NACFA for short) •• Tax rebates for couples 1990 •• Aged Parent Relief 1991 •• Relief for Handicapped •• Tax rebates for foreign-maid levy •• HDB housing schemes implemented to encourage family 1992 formation •• First Timers •• Fiancé/Fiancée •• Married Child Priority •• Third Child Priority
1.96 1.75
Dragon Snake
1.83 1.73
Horse Sheep
1.72
Monkey
•• •• •• •• •• •• •• •• •• •• •• ••
Graduate Mothers Scheme introduced Sterilization Cash Incentive Scheme Establishment of SDU for university graduates Enhanced Child Relief Scheme Multi-tier Family Housing Scheme Graduate Mothers Program scrapped Setting up of Social Development Service (SDS) for non-graduates Establishment of National Family Week Closing of Singapore Family Planning and Population Board First major introduction of pro-natalist policies Phased in terms of affordability “Have 3 or more if you can afford it”
Policy
Year
TFR
Zodiac Sign
•• Small Families Improvement Scheme (SFIS) set up 1993 •• Maintenance of Parents Bill 1994 •• New inter-ministry committee set up to study problems of broken families •• Family Court set up 1995 •• Government moves to help families •• HDB introduced housing schemes aimed at young couples 1996 1997 •• CPF top-up grant of S$40,000 given to singles when 1998 they marry •• Committee of Family (COF) reconstituted to a 1999 people-sector-led civic group •• Launching of Family & Community Development @ eCitizen portal 2000 •• Introduction of “Baby Bonus” •• Increase in maternity leave and tax rebate for working mothers •• Public Education Committee on Family was formed •• Setting up of Work–Life Unit by MCDS •• MCDS also formed the Tripartite Committee on Work–Life Strategy with the Ministry of Manpower, the National Trades Union Congress, and the Singapore National Employers Federation •• $50 million boost to educate public on the value of 2001 family life •• Public Education Committee on Family renamed to 2002 Family Matters! 2003 •• Current COF was reconstituted again in August 2003 •• Members from different backgrounds to provide feedback and advice on policy reviews and new programs impacting on the family •• Major policies for the family announced during National 2004 Day Rally •• Baby Bonus enhanced and Medisave maternity package introduced •• Tax rebates and family-friendly policies implemented •• Family Conference was held in Nov 2004 •• State of the Family Report released by COF •• Committee of Aging Issues set up in Dec 2004 •• National Population Committee set up •• HOPE scheme set up to replace SFIS •• More family-friendly benefits doled out after the Budget 2005 Speech by PM Lee •• Baby Bonus adjusted for flexibility •• Edusave top-up 2006 •• National Family Council (NFC) established •• Taking over of Committee on the Family and Family Matters! by NFC •• Government announced three-pronged approach to tackle population problem •• Adjustment made to Work–Life Works! Fund 2007
1.74 1.71
Rooster Dog
1.67
Pig
1.66 1.61 1.48
Rat Ox Tiger
1.47
Rabbit
1.60
Dragon
1.41
Snake
1.37
Horse
1.26
Sheep
1.25
Monkey
1.25
Rooster
1.26
Dog
NA
Pig
190 Paulin Tay Straughan, Angelique Chan, and Gavin Jones The house would be so much emptier without the laughter of children. How miserable we would be if we have no children to look after us when we grow old and weak. Goh Chok Tong, 2000 National Day Rally Speech It’s fulfilling to bring up a child. You can have the most successful career, you can be the richest man on Earth or the most powerful man or woman on Earth, but if you don’t have a family and don’t have children, I think you’re missing something. It starts off with missing changing nappies when you have a child, but then you miss watching a child learning to walk, learning to talk, going to school, getting sick, depending on you, walking with you, playing with you. You are teaching him, doing homework with him or her. Then you’ll find that he’s got his own temperament, character, personality, he’s different from you. He’s got his own ideas and after a while, one day, you are helping him or her do homework and he says, “No, no, you do it like this” and I look puzzled and I don’t quite know what he is talking about, but he does and you know he’s taken a step forward and he’s on his own. And then they grow up a bit more and you have to, come a time when they ring up and say, “Can you pick me up at the cineplex?” Then you will worry whether they have found girlfriends or boyfriends. Then you will worry if they have not found girlfriends or boyfriends … Lee Hsien Loong, 2004 National Day Rally Speech The key question remains – is it enough to encourage young adults to marry early and start their families early?
Social and cultural barriers to low fertility – a sociological perspective Macro analysis – environment factors and TFR Family policies can only be effective if they address the root cause of the problem. In the introduction to this chapter, we highlighted both macro and micro issues that might affect fertility decisions. From a macro perspective, in addition to family policies, concerns about health of the economy that affect employability and outlook on the future are assumed to be important. Taking annual economic growth (in percentage) as an indicator of economic health, and male and female labor-force participation rates as indicators of employability, we ran a regression model to predict the fertility rate. The bivariate correlations show that TFR is positively correlated to male laborforce participation rate and percentage growth in the economy, and negatively correlated to female labor-force participation rate (see Tables 9.2 and 9.3). In the multiple regression model, only two factors were statistically significant in explaining variation in TFR – percentage economic growth and female labor-force
From population control to fertility promotion 191 Table 9.2 Macro data on economic health and employment Year
Fertility Rate
% growth
Female LFPR*
Male LFPR*
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1.82 1.78 1.74 1.61 1.62 1.61 1.43 1.62 1.96 1.75 1.83 1.73 1.72 1.74 1.71 1.67 1.66 1.61 1.48 1.47 1.60 1.41 1.37 1.26 1.25 1.25 1.25
9.7 9.7 7.1 8.5 8.3 –1.4 2.1 9.8 11.5 10.0 9.2 6.6 6.3 11.7 11.6 8.2 7.8 8.3 –1.4 7.2 10.1 –2.4 4.2 3.1 8.8 6.6 7.9
44.3 44.8 45.2 45.7 45.8 44.9 45.6 47.0 47.8 48.4 48.8 48.0 48.6 48.0 48.6 46.8 49.9 49.5 49.4 50.7 50.2 51.6 50.6 50.9 51.3 52.0 54.3
81.5 81.1 81.5 81.6 81.2 79.9 79.4 78.6 78.5 78.6 77.5 79.4 79.4 78.8 79.2 77.2 78.7 78.2 77.4 77.8 76.6 77.7 77.2 76.1 75.7 74.4 76.2
Source: Singapore Department of Statistics 2006. * Labor-force participation rate.
Table 9.3 Bivariate correlations between TFR and economic-health indicators Variables Overall GDP (S$m) Female labor-force participation rate Male labor-force participation rate
Total Fertility Rate (TFR) N Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N
* Correlation is significant at the 0.05 level (2-tailed). ** Correlation is significant at the 0.01 level (2-tailed).
27.0 –.760(**) .000 27.0 –.652(**) .000 27.0 .646(**) .000 27.0
192 Paulin Tay Straughan, Angelique Chan, and Gavin Jones participation rate. Overall, the model with the three predictor variables explained 56.3 percent of the variation in TFR (see Table 9.4). That economic growth is directly correlated with the TFR is not surprising. In contemporary society when fertility decisions are – to a certain extent – dependent on individual choice, many would delay starting a family or adding to a family when the economic outlook is bleak. Growing a family is an affirmation of one’s optimistic outlook on the future of one’s society. This relates also to contemporary ideology of child, which prescribes that the sacred child deserves a stable environment to grow up in, and that his/her needs are adequately met. As expected, when female labor-participation rate increases, TFR decreases. This shows the difficulties faced by women as they struggle to straddle the divide between work and home. The absence of a dedicated full-time mother in the context of an ideology of child which prescribes delicate care poses a contradiction that is difficult to resolve. Hays (1997; 1996) described this expectation on mothering as the ideology of intensive motherhood. The ideology, supported by various influential social agents (like the state, childcare experts, popular culture and upper-class women) argues that the mother is the best caregiver for the child, and a good mother must engage intensive methods of parenting. These methods are time consuming and frequently very expensive, so few can really afford them. While the macro analysis is useful and affirms our hypotheses, when it comes to informing interventions, it is of limited value. It is therefore important to augment these findings with an individual perspective. Given constant environmental conditions, what explains the variation in family size within the same community? Surveying individuals – perceived impact of pro-family policies in fertility decisions In 2005, the authors carried out a large-scale survey of a probability sample of 1,512 married Singapore residents (including citizens and permanent residents) who were married in 1980 or later. We chose 1980 as the window for sampling for an important reason. Given that median age at first marriage in the 1980s was early–mid-twenties, the defined target group of couples who married in 1980 to Table 9.4 Regression model predicting TFR – macro factor Coefficients(a) Model
B
Beta
1
.521 .020** –.028* .029 0.563**
.401** –.366* .281
(Constant) percentage growth in economy* female labor-force participation rate** male labor-force participation rate Adjusted R-Square
(a) Dependent variable: fertility rate. * Statistically significant at 0.10 level. ** Statistically significant at 0.01 level.
From population control to fertility promotion 193 1985 would be between 40–50 years old. They would have lived through the three phases of family policies, and would have just completed their procreation phase. Reflections from this group would be helpful when we compare them with those who married after 1985, when population policies favored larger families. The fieldwork was completed in the first quarter of 2007, and data analysis is in progress. The response rate for the survey was 65 percent. For this chapter, we extracted information on perceptions of the new profamily policies and investigated how these perceptions influence decisions on having children in the near future. As procreation decisions are determined by women more than men, for this chapter, we selected the responses from female respondents. We employed three indicators relating to the new pro-family policies announced by the Government at the end of 2004. The first measured awareness. The second registered respondents’ opinions on whether these policies would help increase fertility among Singaporeans. The final indicator measured a more direct impact of these announcements on the respondents’ fertility decisions. We asked if the new pro-family policies will encourage respondents to have more children. The frequency distributions are detailed in Tables 9.5 and 9.6. While almost 40 percent were well informed about the new family policies, it is interesting to note that almost a quarter had not heard about the new initiatives at all. Given the limited geographical expanse of this nation-state and the communication channels in place, this is indeed surprising. Interestingly, while 68 percent felt that the fertility rate for the nation would increase because of these new initiatives, only 24 percent indicated that the pro-family policies will have an effect on their personal fertility decisions. Who are the women who are likely to have a child in the near future? We investigated this through a logistic regression model. The responses were coded in dichotomous form, with 0 indicating ‘No” and 1 indicating “Yes”. About a quarter of the respondents indicated affirmative in this dependent variable – i.e., they are likely to have a child in the near future. The logistic regression model was built with 11 predictor variables. The first set measured ideology of motherhood, which looked at the perceived role of “mother”. Two indicators measuring expectations of motherhood were used here – whether the mother is expected to be on-call 24 hours a day for her children, and whether the mother is the best caregiver for her children. Both were Table 9.5 Awareness of pro-family policies announced by the Government at the end of 2004 Percent Yes, I have read up on them Yes, I heard about them but do not know the details No, I have not seen the announcements Total N
39.8 37.6 22.6 100.0 811
194 Paulin Tay Straughan, Angelique Chan, and Gavin Jones Table 9.6 Perception of and attitudes towards pro-family policies Will pro-family policies help increase the number of children Singaporeans will have?
Would respondent consider having more children with pro-family policies in place?
Definitely yes Yes No Definitely no
Percent 16.0 52.4 26.3 5.3
Percent 4.4 19.8 55.6 20.2
Total N
100.0 525
100.0 520
dichotomous variables, with 1 indicating affirmation of the statements. The second set measured the expectations of child – which ties directly to the articulated reasons for wanting children. This index is made up of 3 indicators (see Table 9.7). The index ranged from 4 to 12 points, with high scores correlating with a belief that children should bring extrinsic returns (which runs contradictory to the sacred-child ideology). The Cronbach Alpha for this index was 0.682. The third set of predictors measured marital satisfaction. This 14-item scale was validated in several surveys in Singapore (see Straughan 2005; Straughan et al. 2000). The scale has a Cronbach Alpha of 0.867, and is detailed in Table 9.8. The fourth attitudinal predictor looked at satisfaction with division of domestic labor – also a dichotomous variable with 1 indicating satisfaction. The rationale for the inclusion of these attitudinal variables is to observe their collective effect on fertility decisions. The slate of attitudinal predictive variables are all intricately interlinked. Contemporary expectations of child require the commitment of a full-time caregiver in the form of the biological mother. The child should also be nurtured in a social environment that is blissful and free from conflict (as indicated by the level of marital health). And as demands on the caregiver are intense, we must take note of how well the mother is coping at home (thus, satisfaction with division of domestic labor). In addition to attitudinal measurements, we controlled for demographic and other factors in the model. These include the respondent’s age, number of biological children, work status (working full-time or not), importance of religion, household income, and ethnicity. Table 9.7 Items in scale measuring extrinsic expectations of child* 1 2 3
We have children so that we will not grow old alone. I expect my children to take care of me when I get old. I expect my children to provide for me financially when I get old.
*Respondents were asked to indicate their level of agreement (on a 4-point Likert Scale) to the statements listed in this table.
From population control to fertility promotion 195 Overall, the model performed well (see Table 9.9A). It was able to predict the outcome accurately 90 percent of the time, and the Nagelkerke R-Square (a pseudo R-square indicator in logistic regression) was 0.631. This suggests that about 63 percent of the variation in the dependent variable was explained by the model. Six predictor variables were statistically significant (see Table 9.9B). The results showed that women who were likely to have a child in the future had the following profile: 1 2 3 4 5 6 7
did not embrace intensive mothering, i.e., that a good mother must be on-call 24 hours a day for her children; had fewer children; worked full-time; had higher marital satisfaction; believed that having children would result in extrinsic returns; not Chinese; younger.
These results are not surprising, and at the most basic level, show that fertility decisions are affected both by attitudes on family ideals, and also by social background characteristics. As hypothesized, the dominant ideologies of child that highlight the intrinsic worth of having children and intensive mothering practices run contradictory to increased fertility. Practically, those in happy marriages are more receptive to growing families. Women working full-time indicated stronger inclination towards having children in the future, very likely because they have Table 9.8 Marital satisfaction scale* Statements Measuring Marital Satisfaction a b c d e f g h i j k
I know what my spouse expects of me in our marriage I worry a lot about my marriage If I could start all over again, I would marry someone other than my present spouse I can always trust my spouse My life would seem empty without my marriage Ever since I got married, I lost my personal freedom My marriage has a bad effect on my health I often become upset, angry, or irritable because of things that occur in my marriage I feel competent and fully able to handle my marriage I expect my marriage to give me increasing satisfaction the longer it continues I must look outside my marriage for those things that make life worthwhile and interesting l I have definite difficulty confiding in my spouse m Most of the time, my spouse understands the way I feel n I am definitely satisfied with my marriage *Respondents were asked to indicate their level of agreement (using a 4-point Likert Scale) to the statements listed in this table.
Table 9.9A Logistic regression model predicting future procreation decision, performance of model Observed
Predicted Likely to have child in future Dummy No
Step 1 Likely to have child in future Dummy
Percentage Correct
Yes
No
458
25
94.8
Yes
48
162
77.1 89.5
Overall Percentage
Table 9.9B Logistic regression model predicting likelihood of having child in future Predictor Variables Mother as 24 hour job* (Dummy – 1=Yes) Mother as Best Caregiver (Dummy – 1=Yes) Number of Biological Children* (Scale) Working Full Time* (Dummy – 1=Yes) Importance of Religion (Dummy – 1=Yes) Satisfaction with Division of Domestic Labour (Dummy – 1=Yes) Marital Satisfaction* (Scale) Perception of Child as Extrinsic Value* (Scale) Household Income Ethnicity (Chinese as reference group) Indian* Malay* Others* Age of Respondent* (Scale) Constant* Cox & Snell R Square Nagelkerke R Square * Statistically significant at 95 percent confidence level.
B
Sig.
Exp(B)
–.378
.150
.685
–.090
.737
.913
–1.547
.000
.213
.762
.019
2.142
.079
.785
1.083
.246
.637
1.279
–.057
.019
.944
–.226
.006
.798
–.019
.594 .001
.981
.888 1.593 1.086 –.145
.018 .000 .048 .000
2.429 4.920 2.964 .865
9.537
.000
13860.806
0.446 0.631
From population control to fertility promotion 197 delayed child-bearing in lieu of career advancement. And Chinese women continue to be most resistant to larger families. Taken as a whole, these findings tell us we can count on younger, non-Chinese working women with few children and in happy marriages, and who still have traditional expectations about children and do not subscribe to intensive mothering to help raise the fertility rate in Singapore. Realistically, though, the proportion of such women in Singapore is not very high.
Discussion This chapter has taken a critical look at the evolution of family policies in Singapore. Family policies play two key roles. First, they entice compliance through attractive incentives. Second, they set the ideological tone on the value of family for the society. In these regards, family policies in Singapore have been very successful. Family policies during the population-control period were focused on fiscal incentives, and these were much valued under the economic conditions at that time. The nation-state was developing and household income was relatively lower. Family policies that were tied to housing policies were particularly potent as many did not yet own their homes. Ideologically, the family policies succeeded in establishing small, self-sufficient families as the preferred family types, and responsible parenting as the governing norm. The family policies from 1965 to 1985 were indeed text-book examples of what government policies can achieve under ideal environments. As the nation-state progressed, and after extended periods of sustained growth, fiscal policies lost some of their attraction. In the period since 1985, Singaporeans have made remarkable progress in terms of formal education, acquired skill sets, home ownership, and many other social indicators. As a result, family policies that focus primarily on fiscal returns as carrots find fewer converts. Just as the economic and social conditions of the 1960s and 1970s facilitated the success of the population control measures, these socio-economic factors are also responsible for the limited success of the new pro-family policies. The sustained economic growth since the 1980s (interrupted by brief periods of economic slow down) has resulted in a fairly affluent population – especially among the educated young adults. Armed with tertiary education and the requisite professional skills, they find themselves entrenched in careers that reward total commitment. The rewards to paid work are tremendous – comprising both extrinsic and intrinsic returns. So for many, the enticement of the new profamily policies which promise cash incentives for compliance is limited. After all, they will stand to lose much more financially if they retreat from career investment. While the fiscal dimension of the new pro-family policies has had limited success thus far (as reflected in the continued decline of the TFR), the ideological dimension has had more success. First, with the introduction of paternity leave and childcare leave for both men and women, the Government has set the normative expectations for men’s involvement in the family. Second, the emphasis and
198 Paulin Tay Straughan, Angelique Chan, and Gavin Jones commitment to work–life balance has set the tone for work–family engagement. While it would have been disastrous for an employee to take time off for family commitments in the early years, employers now handle such requests delicately in the more family-friendly environment. Several initiatives announced in this period consistently point to the importance of family time. Specifically, these initiatives are intended to attract women to the workforce while growing their families. Work–life strategies (defined as strategies employers implement to facilitate flexible work arrangements, employeesupport schemes and family care leave) are promoted by the Government as good business practices (for details, see http://www.businesstoolkit.com.sg/). For example, two dedicated grants were set up to entice businesses to these best practices. The Pro-family Business Grant (PFB) gives up to S$20,000 to help businesses achieve “pro-family” qualities in their organization. The Government also committed S$10 million to the new WoW! (Work–life Works!) Fund. This aims to provide financial support to companies, and facilitate the development and implementation of family-friendly work practices. Under WoW!, companies can be reimbursed up to 70 percent of the cost of family-friendly initiatives at the workplace, up to a cap of S$30,000. The initial take-up was slow, with less than 10 percent of the initial S$10 million committed through 32 projects in the first year (Straits Times, August 23, 2005). However, interest picked up and the entire inaugural fund was committed by April 2007. This encouraged the Government to top up the fund with another S$10 million (Ministry of Manpower online news portal at http://www.mom.gov.sg/publish/momportal/en/communities/ workplace_standards/work-life_harmony/WoW_Fund.html). While larger organizations with larger manpower bases may find it attractive to embrace such family-friendly workplace practices, small and medium-sized enterprises (SMEs), which employ more than half of Singapore’s workforce, tend to see family-friendly practices as “luxuries, distracting from their main aim – survival” (Straits Times, August 23, 2005: p. H18). Similarly, other existing schemes for part-time and flexi-work opportunities have not been encouraging (Straits Times, January 18, 2005). This suggests that either women themselves are reluctant to reduce their economic participation rate, or that their employers are not easily able to adopt flexi-work schemes because of the rigid organizational structure. The reality is very likely a combination of both. For example, adoption of flexi-work arrangements varied by sectors, and remains unpopular with employers in the manufacturing, construction, and food and beverage industries (Ministry of Community Development, Youth and Sports 2004b). Finally, the Government – after much deliberation – instituted the five-day work week in 2004, one of the proposals tabled by the Committee on the Family chaired by Mr Noel Hon, Managing Director of NEC Singapore. The Government continued to engage prominent business leaders in people-sector driven committees that looked after many family related social issues. In May 2006, the National Family Council was formed, taking over the roles of the Committee on the Family and the Family Matters Singapore Committee. It serves as an advisory and consultative body for family related matters, and the inaugural committee
From population control to fertility promotion 199 included several prominent business leaders. Through such participation, the Government hopes to solicit buy-in from the private sector in its quest to grow a conducive business culture that favors family commitments. While the TFR did not rise – indeed, it fell sharply – in the years after the 2000 pro-family policies were put in place, it is important to note that the fall has been arrested since 2003. It is difficult to know the part played by the affirmation of the family by the state. While there have been modest ideological gains in repositioning the family, there is still a lot of work to be done to promote a pro-family environment where investment in family does not lead directly to tremendous personal costs. Currently, the perceived cost of raising a child has been inflated so much that no fiscal incentive can be enough to attract the pragmatic-minded. One estimate from a bank pegs the direct financial cost of raising one child at over S$300,000 – and, of course, direct financial costs are only one element in the overall perceived costs of raising a family. Part of the reason for this high perceived cost is the inherent contradictions in various aspects of the social reality relating to the family. Some of these contradictions will now be highlighted. Ideology of paid work versus ideology of the family The recent restructuring in remuneration packages across all sectors replaced seniority with performance indicators as key factors for calculation of rewards and incentives. While it is true that flexi-work policies are more widely practiced now, and these are considered family-friendly as they free the worker from a rigid schedule that ties her to time and spatial considerations, the performancebased remuneration system runs counter to these gains. As with any output-driven reward system, it is not when you do it, but how much you can produce that matters. Though the message sent clearly favors the efficient and effective worker, many on the ground see this as an open-ended work call. The result is a work culture that is characterized by a 24/7 commitment and a fostering of overwork through the gradual eradication of clear work–family boundary (see Lopata 1993). This is, to a large extent, fueled by the increasingly borderless economy we serve – one that crosses several time zones which results in work being relevant at all times. With advances in technology, work becomes portable, thus further blurring the spatial separation between home and office. Hochschild and Machung (1989) wrote about the second shift for women, detailing how working women serve the first shift in the office, and return home for a second shift in home work. With the advances in telecommunications and computing gadgets, it is now the norm to put in a third shift – turning on the laptop with built-in modem after the kids are in bed. E-mail is one example of the long arms of the office – you are only a click away from your bosses. And given that contract work is now the norm, most employees are not secure enough to ignore the “you’ve got mail” icon. In the attempt to make labor cost more efficient, the family wage – where one income (usually the man’s) would support the worker and his dependents – has now been replaced by less costly individual pay packages. This, combined with rising aspirations, means that more and more families are dependent on two incomes.
200 Paulin Tay Straughan, Angelique Chan, and Gavin Jones Just as ideology of work demands a 24/7 commitment from the employee, the ideology of the family – as exemplified by the ideology of child and intensive mothering – also demands a 24/7 commitment. Realistically, one cannot give 100 percent to both work and family concurrently. Gender role expectations: homemaker versus career woman Singapore, like other Asian societies, is still very much governed by patriarchy. Even as women continue to make gains in terms of career achievements and formal education, men continue to be viewed as head of household. The gender expectations for men have remained fairly constant – they are expected to take care of their family financially, through active engagement in paid work. While they are encouraged to play more active roles in the home and especially in childcare, many continue to hide in their work responsibilities, leaving the management of home and children to their wives. The 2000 Singapore census showed that 40.9 percent of all married couples were dual-earners compared to 40.2 percent where only the husbands worked (Leow 2004). A significant proportion of married women are engaged in paid work. How do they manage these contradictory demands? Hochschild and Machung (1989) detailed the emergence of the “Super-mom syndrome”. The Super-mom, an image of a woman who has successfully embraced the multiple roles of career woman/ ideal wife/model mom, exemplifies how a single individual can absorb such a mosaic of responsibilities without help from others. Hochschild wrote, “There is no trace of stress, no suggestion that the mother needs help from others. She isn’t harassed. She’s busy, and it’s glamorous to be busy” (Hochschild and Machung 1989: p. 23). However, this image of the Super-mom serves only to suppress the problems inherent in the multiple responsibilities imposed on the working mother. It succeeds in highlighting the competence of women, and reinforces the message that women are special beings who are able to take on all the domestic responsibilities over and above that of their job. There is no mention of the presence of a sound social arrangement that enables the woman to embrace a career as well as to nurture a family (Straughan et al.2005). This imagery is constantly reinforced by popular culture, and, as a result, the myth is perpetuated, so much so that society assumes that working mothers need very little help to manage. Thus, the conflict between work and family is assumed to be non-existent because Super-moms are expected to take care of everything. And if they cannot manage, perhaps they should reconsider their full-time status in the workforce. So women end up having to choose between motherhood and career. While the returns to motherhood are tremendous, it is a long-term investment, and the rewards are largely intrinsic. Career investments, on the other hand, yield tangible results in a much shorter period of time. Increasingly, women are opting to invest in work, than in the family. This is demonstrated through the trend data on the proportion that remain single, and the proportion of childless married couples or DINKS (double-income, no kids).
From population control to fertility promotion 201 Ideology of child: freedom to grow? The final contradiction we shall highlight is that relating to the social expectations of “child”. Obviously, one reason why many find fertility decisions difficult is the perceived demands of child-rearing on the parents. The ideology of child requires that parents devote themselves to giving the best to the child. While one would expect that this would include ensuring a blissful period of innocent childhood where the child is allowed the luxury to dream, play freely, and just enjoy not having serious responsibilities, this is sorely missing from the Singapore adaptation of the ideology. Given our strong emphases on self-sufficiency and achievement orientation, we have conceptualized parental responsibilities to include the demand that parents give the child a head start in life, so that (s)he can make remarkable gains in the education system. In our focus-group interviews, many respondents talked about their fear of parenthood because, through the child, the parents will be drawn directly into the very competitive education system. Approaches that promise head starts involve serious commitment of time and money from the parents. Parents in Singapore routinely take leave to be present for their children when undergoing the stress of the Primary School Leaving Examination (PSLE), a crucial hurdle influencing the future career of these pre-teenagers. When those contemplating parenthood see the stress and anxieties parents go through when their children are in the system, many decide that they would not want to participate in the mayhem. How then do we move forward from here? The next phase of family policies will have to address these contradictions in our society. These ideological contradictions make it difficult for many to choose between having children and remaining childless. When we force people to choose, many will go the route of least resistance. It is not easy to raise a family in contemporary society. Yet, resisting family formation is not an option any society can accept as the norm. The only option is to continue to work at making sense of these contradictions and freeing the path towards family formation.
Notes 1 Note that in 2007, one Singapore dollar was equivalent to approximately 68 cents US.
References Goh, C.T. (2000). National Day Rally Speech. Downloaded on October 1, 2007 from: http://stars.nhb.gov.sg/public/index.html. Government of Singapore (2008). Family and Community Development @ eCitizen. Downloaded on February 1, 2008 from http://fcd.ecitizen.gov.sg/. Hays, S. (1996). Cultural Contradiction of Motherhood. New Haven, CT: Yale University Press. —— (1997). “The ideology of intensive mothering: a cultural analysis of the bestselling ‘Gurus’ of appropriate childrearing,” in Elizabeth Long (ed.), From Sociology to Cultural Studies: New Perspectives. Malden, MA: Blackwell Publishers Ltd.
202 Paulin Tay Straughan, Angelique Chan, and Gavin Jones Hochschild, A.R. and A. Machung (1989). The Second Shift: Working Parents and the Revolution at Home. New York, NY: Viking. Huang, S. and B.S.A. Yeoh (1996). “Ties that bind: state policy and migrant female domestic helpers in Singapore,” Geoforum, 27(4): 479–493. International Finance Corporation (2008). Singapore Business Toolkit. Downloaded on February 1, 2008 from http://www.businesstoolkit.com.sg/. Lee, H.L. (2004). National Day Rally Speech. Downloaded on October 1, 2007 from: http://stars.nhb.gov.sg/public/index.html. Leow, B.G. (2004). Census of Population 2000: Households and Housing. Singapore: Singapore Department of Statistics. Lien, L. (2002). “Marriage and procreation: to intervene or not – a policymaking perspective,” paper presented at Workshop on Fertility Decline, Below Replacement Fertility and the Family in Asia: Prospects, Consequences and Policies, National University of Singapore, April 10–12, 2002. Lopata, H.Z. (1993). “The interweave of public and private: women’s challenge to American Society,” Journal of Marriage and the Family, 55(1): 176–190. Ministry of Community Development, Youth and Sports (2004a). “Singapore – A Great Place for Families”. Singapore: Ministry of Community Development, Youth and Sport. —— (2004b). “Seminar on work-life practices and the business bottom line,” press release. Downloaded on October 1, 2007 from: http://app.mcys.gov.sg/web/corp_press.asp. Ministry of Manpower (2005). Report on Labour Force in Singapore 2004. Singapore: Ministry of Manpower. —— (2008). WoW! Fund. Downloaded on February 1, 2008 from http://www.mom.gov.sg/ publish/momportal/en/communities/workplace_standards/work-life_harmony/WoW_ Fund.html. Saw, S.H. (2005). Population Policies and Programmes in Singapore. Singapore: Institute of Southeast Asian Studies. Silberstein, L.R. (1992). Dual-career Marriage: a System in Transition. Hillsdaye, NJ: L. Erlbaum. Singapore Department of Statistics (2002). Twenty-Five Years of Below Replacement Fertility: Implications for Singapore. Singapore: Department of Statistics. —— (2004). Singapore 2004: Statistics Data Sheet. Downloaded on October 1, 2007 from: http://www.singstat.gov.sg/keystats/annual/poster.pdf. —— (2005). Population Indicators. Downloaded on October 1, 2007 from: http://www. singstat.gov.sg/keystats/annual/indicators.html#Population%20Indicators. —— (2006). Population Trends. Singapore: Department of Statistics. —— (2007). Yearbook of Statistics. Singapore: Department of Statistics. —— (2008). Labour Force Participation Rate. Downloaded on February 1, 2008 from http://www.singstat.gov.sg/stats/themes/economy/hist/labour.html Straits Times. Various issues. Straughan, P. (2005). Report on Marriage and Divorce in Singapore. Singapore: Ministry of Community Development, Youth and Sports. Straughan, P., A. Chan, and G.W. Jones (2006). “Where is the stork? Sociological insight on barriers to fertility,” Social Transformations in Chinese Societies, 2 (Sociology for Change) (2006), Netherlands: 143–162. Straughan, P., S. Huang, and B.S.A. Yeoh (2005). “Family ideology and practice: implications for marital satisfaction,” unpublished manuscript. —— (2000). “Work, family and marital satisfaction: Singapore women’s perspectives,”
From population control to fertility promotion 203 paper presented at Conference on Families in the Global Age: New Challenges Facing Japan and Southeast Asia, Singapore October 4–6, 2000. Teo, Peggy, Kalyani Mehta, Leng Leng Thang, and Angelique Chan (2006). Ageing in Singapore: Service Needs and the State. London: Routledge. The Sunday Times. July 10, 2005. Wong, A.K. (1979). “The national family planning programme and changing family life,” in E.C.Y. Kuo and A.K. Wong (eds), The Contemporary Family in Singapore. Singapore: Singapore University Press Wong, T. and B.S.A. Yeoh (2003). “Fertility and the family: an overview of pro-natalist population policies in Singapore,” Asian MetaCentre Research Paper Series No. 12. Singapore: Asia Research Institute. Zelizer, V.A. (1994). Pricing the Priceless Child: the Changing Social Value of Children. Princeton, NJ: Princeton University Press.
10 Fertility in Pacific Asia Looking to the future Gavin Jones, Paulin Tay Straughan, and Angelique Chan
Why should we worry about low fertility? This book has examined the demographic situation in a very important region of the world where population decline is in prospect. It may seem strange to be so concerned about low fertility and the possibility of population declines. Over recent decades, the main worry among international agencies and national planners has been about the consequences of rapid population growth, and although this growth is slowing worldwide, there is still great concern about the perils to the very survival of humanity resulting from further population growth and the demands this is placing on the world’s resources and ecosystems. World population has grown by almost 37 percent over the past two decades, from 5 billion to 6.7 billion. It is expected to increase to at least 9 billion before it levels off. In launching the 570-page Global Environmental Outlook report in 2007, the head of the UN Environment Program, Achim Steiner, said, “the human population is now so large that the amount of resources needed to sustain it exceeds what is available at current consumption patterns” (UNEP, 2007). In such a context, why worry about the prospect for a shrinking population in the countries of East Asia? After all, Japan, China, South Korea, Taiwan, and Vietnam have long been concerned about their high population density, and one group of scholars in China in 1981 produced a study claiming that the appropriate population for China was only 650–700 million, a figure only just over half the population reached today (Song Jian, 1981). There are several issues to deliberate but first, what do the latest United Nations Population Division figures say about likely future growth of the world’s population, and specifically that of East and Southeast Asia? It depends on which variant is used. There are three variants – high, medium, and low. The low projection assumes that for countries with very low TFRs, the TFR will not alter drastically over the projection period, ending up with a TFR of 1.35 children per woman, whereas for countries with TFRs at replacement level or higher, the TFR will gradually decline towards 1.35, though not necessarily reaching this level by 2050. In contrast, in the medium projection (the one considered by most commentators and, at least by implication, by the UN demographers themselves, the most likely to be followed), countries with very low fertility are projected within about a decade to experience a
Fertility in Pacific Asia 205 gradual rise in fertility towards a level of 1.85 by the year 2050, though not necessarily all reaching that level. Based on the latest evidence of fertility trends in East Asian countries, we would argue that the low projection seems to be the best aligned with what is happening in this region, at least up to this point. For most of Europe, we would argue that the low projection may be too low, and the medium projection too high. Fertility rates have edged up since 1995 in some European countries, including France, the Netherlands, and even some of the lowest fertility countries – Spain, Italy, and Germany. But they remain very low in countries such as the Czech Republic and Poland (Prioux and Mandelbaum, 2005). As for the rest of the world, it is hard to say which projection is most likely to be followed. In any event, Table 10.1 shows the current and projected population for the world and for East and Southeast Asia in 2050 according to the UN low and medium projections. Importantly, even in the low projection, the world’s population would still increase by 20 percent between 2005 and 2050. Such is the tyranny of population momentum, which carries growth based on past high fertility well into the future. But the major regions of Europe and East Asia would experience population declines of some 23 percent and 11 percent respectively. And while Southeast Asia would still experience a growth of 15 percent, different countries Table 10.1 United Nations population projections: world and selected regions and countries, 2005–2050 1 Low projection Region
World Europe E Asia SE Asia Japan S Korea China Thailand
Population (‘000) 2005
2050
6,514,751 731,013 1,522,472 557,669 128,085 47,872 1,312,979 63,003
7,791,945 566,032 1,357,942 642,210 88,645 35,891 1,201,666 56,623
9,191,288 664,181 1,591,241 766,608 102,512 42,329 1,408,847 67,378
% increase
% increase in age groups 0–14
15–29 30–64 65+
19.6 –22.6 –10.8 15.2 –30.8 –25.0 –8.5 –10.1
–36 –52 –56 –50 –65 –73 –55 –56
–16 –54 –47 –33 –64 –70 –45 –48
48 –26 –11 54 –43 –33 –8 –7
213 58 197 347 53 229 232 219
41.1 –9.1 4.5 37.5 –20.0 –11.6 7.3 6.9
–1 –17 –25 –16 –35 –51 –24 –22
11 –33 –26 –6 –44 –50 –24 –25
60 –18 –3 67 –36 –25 1 2
213 58 197 347 53 229 232 219
2 Medium projection World Europe E Asia SE Asia Japan S Korea China Thailand
6,514,751 731,013 1,522,472 557,669 128,085 47,872 1,312,979 63,003
Source: Calculated from data in United Nations, 2007.
206 Gavin Jones, Paulin Tay Straughan, and Angelique Chan would experience very different growth trajectories. Thailand’s population would decrease by 10 percent, whereas the Philippines’ would increase by 39 percent. But it is not just the overall population trends to 2050 that are of concern in the ultra-low-fertility countries. The real issues lie in two key factors: 1 2
the shifts in the age structure and in the growth or decline in particular age groups which these projections imply the downward spiral in population beyond 2050 built into the population structure.
For Japan to reach a 2050 population 31 percent below the 2005 population, as projected in the United Nations low projection, implies the remarkably large decrease of two-thirds in numbers in the 0–14 age group, an almost equally large decline in the 15–29 age group and a decline of one-third in the 30–64 age group. But a rise of 53 percent is projected in the 65+ age group (see Table 10.1). Japan’s percentage changes by age groups are a little more extreme than those of Europe. Interestingly, although South Korea is projected to experience a slightly smaller percentage decline in population than Japan, the changes in age groups will be even more spectacular: a decline of over 70 percent in the age groups below 30, and an increase of 229 percent in the 65+ age group. The reason for these sharper changes is that by 2005, Japan had already progressed further towards an aging population, whereas the trend towards aging will be much more dramatic in Korea, because of the compression of the sharp fertility decline into a shorter period. If the increases in fertility underlying the UN medium projection for the very low-fertility countries are realized, East Asia’s population (and that of China) will increase slightly between 2005 and 2050, though even in this projection Japan and South Korea would face substantial population declines, especially in the age groups below 30, which in the Korean case would halve in number. In China and Thailand, too, these age groups would decline by one-quarter over the projection period. As noted earlier, we expect population growth in Pacific Asia to be lower than in the medium projection. To summarize briefly the age structure changes evolving over the next 45 years, Figure 10.1 shows the share of the 0–14 and 65+ age groups in the populations of East Asia and Europe over the 2005–2050 period, according to the UN low and medium projections. East Asia, of course, is dominated by China, but also includes, among others, Japan and Korea. Europe and East Asia will follow roughly similar trends, but the timing and pace will differ. The 0–14 and 65+ age groups are notionally the “dependent” age groups, though of course in reality this is only a broad approximation of reality. In any case, in the low projection the increase in the share of the elderly population in East Asia – especially after about 2015 – is sharper than the decline in the share of the child population, thus implying a decline as well in the share of the working-age population. The crossover point – when the share of the elderly will first exceed the share of children – is occurring in Europe more or less as we write, but in East Asia it will be delayed until the early 2020s. Remarkably, by 2050 a yawning gap
Medium variant 30
25
Percentage
20
15
10
5
20 50
20 40
20 35
20 30
20 25
20 20
20 15
20 10
20 05
0
Year 65+ (Europe)
0–14 (Europe)
0–14 (E. Asia)
65+ (E. Asia)
Low variant 35 30
Percentage
25 20 15 10 5
50 20
20
45
40 20
20
35
30 20
25 20
20 20
15 20
10 20
20
05
0
Year 65+ (Europe)
65+ (E. Asia)
0–14 (E. Asia)
0–14 (Europe)
Figure 10.1 Trends in percentage of population in two broad age groups. Source: Calculated from data in United Nations, 2007.
208 Gavin Jones, Paulin Tay Straughan, and Angelique Chan will have developed, with the share of the elderly in East Asia nearing 30 percent and the share of children not much above 10 percent. In the medium projection, the general trends are in the same direction as those in the low projection, but in muted form, because of the assumption that fertility in very low-fertility countries will edge up. Thus, the share of the elderly in both Europe and East Asia will not go as high as in the low projection, and the share of the child population will level off at about 15 percent instead of 10 percent. This projection also shows a decline in the share of the working-age population, but not as great or as rapid as in the case of the low projection. East Asia reaches the “crossover point” in the early 2030s, not as rapidly as in the case of the low projection. This projection has less negative implications for the dependency situation in the East Asian region than does the medium projection. The projections to 2050 do not capture the further downward spiral in population size implicit in the assumption of continued well-below-replacement levels of fertility. Should such rates continue at 30 percent below replacement (or a TFR of just below 1.5) the population would fall in half every 60 years, and would thus virtually disappear within 200 years in the absence of in-migration. These projections are, of course, based purely on extrapolations of fertility levels far into the future. They will not be realized, but they are the raw material on which newspaper and magazine articles will be increasingly based, and which politicians and planners will be reacting to when formulating policy. The prospect of maintaining a Japanese or Korean population only through substantial migration and increasing “dilution” of the ethnic identity of the original population is unlikely to be warmly received in either country. The alternative of providing pro-natalist incentives sufficiently strong (and costly) to have the chance of a real effect on fertility is likely to be increasingly entertained.
The factors making for continuing low fertility In thinking about the likelihood of a resurgence in fertility in East Asia, with or without strong government policy, we cannot ignore the increasingly late initiation of child-bearing due to delayed marriage, and indeed the removal of a not inconsequential proportion of women from the child-bearing population altogether as a result of non-marriage. Marriage change has played a considerable role in the recent fertility declines in a number of East Asian countries. Since there is essentially no childbearing outside marriage in these countries, marital status directly affects fertility, whereas its effect on fertility is more blurred in northern and western Europe. While delayed marriage can be a direct response to a desire to avoid or delay child-bearing, there are many factors affecting marriage in Pacific Asia aside from the desire (or lack of desire) for children. Hypergamy remains strong, and in general the notion of who should marry whom is perhaps less flexible than in Western countries. As a result, involuntary non-marriage is likely to be more common. In turn: involuntary childlessness resulting from involuntary non-marriage may be playing a substantial role in the low fertility currently recorded in Pacific
Fertility in Pacific Asia 209 Asia. It therefore behooves fertility analysts to avoid exclusive concentration on fertility goals, or even worse, on the fertility goals of married people. (Jones, 2007: 472) The tendency in the countries referred to in this book for young people to delay marriage because of the perceived need to become settled in a career and build up some human and financial capital “pushes potential childbearing into an age range where the obstacles and opportunity costs are likely to loom even larger, and where decreasing fecundity of women in their late 30s and 40s becomes an (underrecognized) factor” (Jones, 2007: 472). China is the exception here, as the proportion of women remaining single on their thirtieth birthday remains very low. Once couples have married, there is a wide range of factors inhibiting their desire for child-bearing. These include uncertainty of continued employment, the financial costs of child-rearing, the conflict of work and family responsibilities, exacerbated by family-unfriendly workplace practices and attitudes, the ideology of the “quality” child and the pressure this places on mothers in particular, lack of appropriate government policies to support child-rearing, gender ideologies on housework and child-rearing, and other pressures of child-raising in big city environments. Possibly rising individualism plays a part, though the evidence for this is not compelling.
Can we expect an upsurge in fertility? Newspaper reports in the region in the 2006–2008 period show that commentators (and government officials responsible for population policies) are searching for signs that the sharp declines in fertility in the five-year period 2000–2005 (shown in Table 1.1) are being reversed. Upturns in fertility, or in the number of births recorded, are duly noted, but what is rarely remarked on is how slight these upturns have been. In South Korea, TFR rose from 1.08 in 2005 to 1.26 in 2007. In Japan, TFR rose from 1.26 in 2005 to 1.34 in 2007, and in Singapore, from 1.25 in 2005 to 1.29 in 2007. The South Korean rise is equivalent to 18 per cent of the gap to be filled if fertility is to return to replacement level; the Japanese rise to 10 per cent of the gap; and the Singapore rise to 5 per cent of the gap. Certainly, signs of a bottoming out in fertility rates are welcome, and the increases in South Korea and Japan are not trivial, being equivalent to a filling in of more than one-third of the gap between the fertility lows and a TFR of 1.5, often used as the figure below which fertility levels are seen to be in crisis. However, before concluding that slight increases in births such as these are harbingers of a major rebound in fertility, it would be necessary to show significant changes in the key underlying determinants of fertility. As we have seen in the various chapters in this book, the factors influencing fertility trends in the different low-fertility Asian countries have varied, but the fertility depressing factors noted in the previous section appear to have rather general relevance. One or more of these factors would need to be modified greatly if there is to be much likelihood of a resurgence of fertility in Pacific Asia. Greater acceptance of cohabitating relationships and of child-bearing outside of marriage could in theory
210 Gavin Jones, Paulin Tay Straughan, and Angelique Chan facilitate higher levels of fertility, but it is doubtful whether the factors making for “involuntary non-marriage” would be greatly weakened if cohabitation were more acceptable – rather, they would be likely to translate into “involuntary non-marriage and non-cohabitation”. There may be more scope for change in some of the complex set of factors summarized in the final paragraph of the previous section. However, we believe that East and Southeast Asia is likely to continue for some time in the vanguard of ultra-low fertility in the international setting.1 The key reasons have been made clear in this book. The first has to do with the economic context in these countries. Their very economic success is built on a model whereby women need and want to be in the workforce, but pressures the workplace exerts on them require great sacrifices of time and potential income if they are to raise a family in the way that East Asian societies expect. These are highly competitive economies and governments are determined to increase productivity and keep wages down. Employers remain relatively unforgiving of the divided loyalties inherent in the effort to combine child-raising with working, and society remains unsupportive of those who want to pursue non-material goals or who are not totally devoted to pressing their children to maximum performance.2 Moreover, in recent years greater uncertainty has entered the labor market, making many young people more apprehensive about taking on the commitments that starting a family entails (McDonald, 2007 and this book, Chapter 2). The second reason why fertility has fallen so low is that governments of East Asian countries have yet to commit seriously to the costly policies required to make child-bearing more appealing to potential parents – certainly they show much less commitment to such policies than do governments in Europe. The hard fact to be absorbed is that whereas there has been basic stability in Western European fertility over the last two decades (Caldwell, 2006), this is not true of East Asia, where a further dip occurred in the first five years of the twentyfirst century, bringing East Asian fertility levels below those of Europe as a whole. In East Asia, too, the most recent evidence is that fertility is leveling off. The only difference is that the stability at this level is lower than in most of Europe, and therefore more problematic. Could there be factors making for a substantial rise in fertility in any of these Asian low-fertility countries? Certainly, the possibility of such a rise in fertility should not be dismissed. Three possible factors that could lead to a significant increase in fertility may be considered. Cohort versus tempo effects Not much research has been done on these effects in Asian countries; much more has been done on Europe (see, for example, Bongaarts and Feeney, 1998; Kohler et al., 2002; Sobotka, 2004; Billari and Kohler, 2004). The key point is that trends in TFR are not necessarily moving in tandem with trends in cohort fertility. Period fertility is lower than cohort fertility when the mean age of child-bearing rises and the reverse is true when the mean age of child-bearing declines (Ryder, 1983). In effect, when successive cohorts delay child-bearing their births are spread out over
Fertility in Pacific Asia 211 a longer period than would be the case if the timing were constant; the result is a reduction in period fertility. Thus, it has been argued that in European countries, if part of the recent fertility decline has been due to a postponement of births, then the underlying cohort fertility will turn out to be higher than the period fertility. In other words, each cohort of women will end up having more children than the TFRs indicate, because the current TFRs are artificially suppressed by postponement of births that women will eventually have. Once the rise in mean age at childbearing ends – as it inevitably must – “the corresponding fertility-depressing effect stops, thus putting upward pressure on period fertility” (Bongaarts, 2002: 439). This factor must also be kept in mind in understanding fertility trends in Asian countries, in almost all of which ages at marriage and mean age at child-bearing have been steadily rising. The key uncertainty in forecasting increases in period fertility as a result of the cessation of increases in mean age at child-bearing in Asian countries is that it is not clear when the increases in the underlying factor – mean age at marriage – are likely to end. However, it is also important to recognize that when period fertility is as low as it has been now for some time in the Pacific Asian countries, the cohort effect is unlikely to raise it to anywhere near replacement level – rather, two or three percentage points increase in TFR is all that could realistically be expected.3 Impact of policy Most analyses of the possible impact of policy argue either that policy is likely to have little effect in raising fertility, or that it is likely to have a substantial impact only if much greater effort and resources are put into such policies than has been the case to date (see, e.g., Caldwell et al., 2002; McDonald, this book, Chapter 2). It is clear from the policy initiatives in Pacific Asian countries reported on in the chapters in this book that the efforts to raise fertility in Asian countries have so far been less serious than in a number of countries of Europe where, on the whole, the measures adopted have had only limited impact. How policy will change in future in Pacific Asia, of course, remains to be seen. Attitudinal changes towards work and family It could be argued that a sea change in attitudes will take place, as people become dissatisfied with what they are getting from neo-liberal economic policy and its attendant work patterns, from consumerism, and from a perceived decline in family values. A desire to return to “family values” could be bolstered by propaganda programs in favor of family formation that governments may engage in when facing the prospect of a downward spiral in population size. The extent to which individual child-bearing decisions are likely to be influenced by dire warnings of the immanent contraction and eventual disappearance – or dilution through immigration – of the Japanese or Korean populations remains a great unknown. Another – perhaps equally plausible – hypothesis is that desire for children will decrease below the normative two children because of the increasing unfamiliarity young people have with interacting with children in societies with very low fertility,
212 Gavin Jones, Paulin Tay Straughan, and Angelique Chan and the possibility of a “child-unfriendly culture” developing in situations where children are rare. As Lutz (2008) has noted, the German-speaking countries, which in the 1970s were the first to see fertility fall to very low levels, are now the first to show well-below-replacement family size ideals among younger cohorts. The beginning of a decline in ideals is also apparent in Italy and Spain, where fertility declined to really low levels only in the 1980s. Importantly, there is evidence of low-fertility desires in some East Asian settings as well. In China, when a couple marries, both of whom were only children, they are now permitted to have two children. But a study in Beijing shows that only 18–24 percent of such couples want to have the second child (Hou Yafei, 2007); and the Jiangsu survey reported on in Chapter 4 of this book shows that less than 40 percent of such couples there want to have a second child.
Research needs This book has made a modest attempt to focus greater attention on the trend to ultra-low fertility among very substantial populations in Asia. Compared with the low-fertility countries of Europe, there is a dearth of quality studies on: 1 2 3 4
the demography of what has been happening understanding of these trends from a behavioral point of view understanding their sociological, economic, and institutional underpinnings assessing the impact of government policy to date, or the potential impact of policy changes in the future.
It is not difficult to list a number of research areas that require much more development:
•• Studies of trends in family size ideals and their determinants in the lowfertility Asian countries.
•• Better understanding of the constraints on child-bearing, through survey
••
••
research, more qualitative approaches, and studies of the institutional setting within which child-bearing decisions are made. One element that requires much more research is the actual financial and opportunity costs of raising children among different socio-economic classes in East Asian settings. Without better knowledge on this, it is difficult to assess the likely impact of different levels of subsidization of child-bearing through child-support payments or tax breaks. Studies on the links – direct and indirect – between trends in marriage and trends in fertility. To what extent is delayed marriage and non-marriage voluntary or involuntary? Is there a trend towards greater acceptance and practice of cohabitation, and if so, is child-bearing outside of formal marriage likely to increase? On a more technical demographic level, more research is needed into the cohort and tempo effects on the levels of the Total Fertility Rate in Pacific Asian countries over recent decades when marriage and the onset of childbearing have been progressively delayed, and the possible effects on the TFR of a cessation in this rise in onset of child-bearing.
Fertility in Pacific Asia 213
•• More systematic study of policy responses in countries of the region. What
••
are the common elements? How much have Asian governments studied family policy in Europe, Australia, etc. in thinking through their own policy responses? What are the key differences between Asian and European countries in the extent of overt pro-natalism in family policy? What are the key issues related to the high proportion of women and men remaining never married throughout much of the period when their parents and grandparents generation would have been married? What are the levels of sexual activity among the never married? What barriers, if any, are placed on access to contraception and reproductive-health services for people in this category?
Conclusions The range of demographic situations in the world has never been wider, with TFR ranging from as high as 6 or 7 in some sub-Saharan African countries to below 1 in Hong Kong and some Chinese cities. Most of the world falls in the more demographically “moderate” middle, though more countries are being added to the very low-fertility group. The general parameters of world population distribution half a century from now are fairly clear, the main elements being a massive rise in the proportion living in sub-Saharan Africa and a sharp decline in the proportion living in Europe and East Asia. It is not these changes in proportions that matter (though they will certainly result in shifts in the international political order), but the issue of whether very high fertility in the African case and very low fertility in the Europe and East Asia cases is likely to have adverse consequences for the welfare of their populations. There are strong grounds for arguing that in both cases it will have adverse consequences. The aim of this book has been to examine the issues in the case of the very lowfertility Pacific Asian countries. The factors underlying the very low fertility have been shown to relate to the broad socio-economic and institutional settings in these countries, as well as to cultural factors. If they are to have much chance of success, policies to redress the low fertility need to be based on a thorough understanding of this setting. As these issues have been researched far more comprehensively for European countries than they have for Pacific Asian countries, sustained research is needed to guide future family and fertility policy in Pacific Asia.
Notes 1 South Asia is a different story altogether. It is one of the regions of the world where strenuous efforts are needed to lower fertility. However, South Asia has not been totally left behind – in Sri Lanka, and some states of India, fertility is already below replacement level. 2 Caldwell (2006: 239) succinctly explains the demographic balance reached in Western Europe and the “offshoots” of North America, Australia, and New Zealand thus: the strains of balancing labor market and domestic work “are offset by higher joint incomes and the consequent ability to play a fuller role in the consumer society, to keep up with the Joneses, and to ensure that one’s children keep up with the Joneses’ children”. The failure to reach such a balance in Pacific Asia, we would argue, has much to do with the
214 Gavin Jones, Paulin Tay Straughan, and Angelique Chan equally strong consumerist pressure to maximize joint incomes, even stronger pressure to keep up with the Tans or the Kims and to ensure that one’s children keep up with the Tans’ or the Kims’ children, but in the context of relatively family-unfriendly workplaces and limited family support from government. 3 In Europe, the tempo-adjusted TFR is typically about 0.2 or 0.3 higher than the TFR (Sobotka, 2004: Tables 2 and 3). In Taiwan, the tempo effect has been estimated to have been responsible for lowering the observed TFR by about 0.25 in the late 1970s and early 1990s, but by more – about 0.4 – in the mid and late 1980s when change in the mean ages of all birth orders was most rapid (Bongaarts and Feeney, 1998: 282).
References Billari, Francesco C. and Hans-Peter Kohler, 2004, “Patterns of low and very low fertility in Europe,” Population Studies, 58(2): 161–176. Bongaarts, John, 2001, “Fertility and reproductive preferences in post-transitional societies,” in Rodolfo A. Bulatao and John B. Casterline (eds), Global Fertility Transition, Supplement to Population and Development Review, Vol. 27, 2001. —— 2002, “The end of the fertility transition in the developed world,” Population and Development Review, 28(3): 419–443. Bongaarts, John and Griffith Feeney, 1998, “On the quantum and tempo of fertility,” Population and Development Review, 24(2): 271–292. Caldwell, John C., 2006, “Perspective: the Western fertility decline: reflections from a chronological perspective,” Journal of Population Research, 23(2): 225–242. Caldwell, John C., Pat Caldwell and Peter McDonald, 2002, “Policy responses to low fertility and its consequences: a global survey,” Journal of Population Research, 19(1): 1–24. Hou Yafei, 2007, “A study on the willingness of childbirth among the single child group in Beijing,” paper presented in Panel on Diversity of Population Development and Health Security, Beijing Forum, Beijing University, November 2–4. Jones, Gavin W., 2007, “Delayed marriage and very low fertility in Pacific Asia,” Population and Development Review 33(3): 453–478. Kohler, Hans-Peter, Francesco C. Billari and Jose Antonio Ortega, 2002, “The emergence of lowest-low fertility in Europe during the 1990s,” Population and Development Review, 28(4): 641–680. Lutz, Wolfgang, 2008, “Has Korea’s fertility reached the bottom? The hypothesis of a ‘low fertility trap’ in parts of Europe and East Asia,” Asian Population Studies, 4(1): 1–4. McDonald, Peter, 2007, “The emergence of low fertility as a policy issue,” Asia-Pacific Population Journal, 22(2): 5–9. Prioux, France and Jonathan Mandelbaum, 2005, “Recent demographic developments in France,” Population (English edition), 60(4): 371–414. Ryder, Norman, 1983, “Cohort and period measures of changing fertility,” in Rodolfo A. Bulatao and Ronald D. Lee (eds), Determinants of Fertility in Developing Countries, Vol. 2. New York, NY: Academic Press: pp. 737–756. Sobotka, Tomas, 2004, “Is lowest-low fertility in Europe explained by the postponement of childbearing?,” Population and Development Review, 30(2): 195–220. Song Jian, 1981, “Population development – goals and plans,” in Liu Zheng, Song Jian and others, China’s Population: Problems and Prospects. Beijing: New World Press. UNEP, 2007, Global Environmental Outlook. Nairobi: UNEP. United Nations, 2007. World Population Prospects: the 2006 Revision. New York, NY: United Nations Department of Economic and Social Affairs, Population Division.
Index
1991 Childcare Leave Act 64, 67 1995 Childcare and Family Leave Act 64 1997 Asian financial crisis 24, 35–6, 110–4, 119, 123, 127–8, 172 age-specific: fertility rates (ASFRs) 29–30, 37n1, 41–2, 75, 113, 134–6, 166; first-marriage 50; marital fertility rate (ASMFR), 141–2; marriage rates 162 “Angel Plan” 8, 64 ASFRs, see age-specific fertility rates ASMFR, see age-specific marital fertility rate Asian economies 28, 32 baby boom 132 birth rates 2, 4, 8, 16–7, 42, 75, 182–3, 188 birth-control 132, 183; policy, 82–3, 86 CEB, see children ever born certificates of absence of marriage registration (CAMR), 146–7, 157n4 child-bearing 7–9, 11, 13, 15, 17, 24, 30, 32–3, 43, 59, 63, 80, 82–3, 88, 91–2, 100, 111–3, 124, 126, 133–4, 141–2, 144–5, 152, 155, 157, 161, 166, 169, 173–4, 197, 208–12; raising 7–9, 17, 210; rearing 9, 11, 13, 15, 17, 32, 41, 43, 49, 63, 65, 86, 88, 117, 125, 127, 150, 186, 201, 209 childbirth 10–1, 69, 98, 134, 153, 155, 181 childcare 8–11, 15, 34, 37n1, 67, 85, 126–7, 129, 152–4, 156, 171–2, 185, 187, 192; leave 8–11, 14, 60, 64–6, 68, 69, 187 children ever born (CEB), 110, 113–4, 116–7, 118, 119–21, 122, 123–6, 128 China 5, 7, 35–6, 73–5, 77, 78–80, 81, 82, 84, 86–91, 96–8, 100–6, 145–7, 157n4, 204, 205, 206, 209, 212
contraception 2, 43, 92, 133, 213 decomposition 28–9, 168 demographic: change 25, 101, 128, 156; theory 91, 100–1; transformation, 73; transition 6, 75, 100, 105–6, 148; trends 3, 25, 161, 181 economic: changes 23–5, 27, 174; development 2–3, 80, 96–7, 100, 102–3, 106, 133, 184; growth 35–6, 63, 154, 172, 183–4, 190–1, 197; recession 113, 164; restructuring 23–4 education: secondary 12; tertiary 12 family: planning 2, 96, 100, 105, 132–3, 150, 183: programs 2–3, 6, 11–2, 18n2, 76, 91–7, 100, 105, 183; policies 7, 15, 17, 182–4, 187–8, 190, 193, 197, 201, 213: pro- 187, 192–3, 194, 197–9 fertility: below replacement 5–7, 17, 73, 75, 80, 86, 91–2, 105, 133, 165, 183, 208; decisions 15, 134, 167, 171, 182–3, 186, 190–4, 197, 201; decline 12, 28–9, 31, 40–1, 43, 46, 76, 86, 89–90, 96, 100, 119, 133, 141–2, 206; lowest-low fertility 36, 40, 97, 110–1, 129, 150, 160–1, 173; marital 57–8, 65–6, 69, 144, 148, 166; patterns, 100–1, 181–2; policy, 73, 76, 78, 80, 83, 86, 91–2, 213; reduction 6; replacement level 4, 6–7, 11, 14, 28, 40, 66, 132, 135, 140, 156, 161, 168, 209; transitions, 96, 110, 113, 134; trends, 73, 111, 165, 172, 183–4, 186, 205, 211; ultra-low 1, 3–4, 11–2, 14, 17, 160, 206, 210, 212 fertility rates 1–3, 10, 14, 24–5, 27, 32–3, 37, 96–8, 100, 103, 106, 111–2, 133, 135–6, 139, 141–2, 160, 165, 169,
216 Index 182, 205, 209: general (GFR), 99, 104, 105: total (TFR) 4, 6–7, 9, 16–7, 25, 29, 37, 40–3, 46, 63, 64, 66–7, 69, 73, 75, 76, 92, 96, 97–9, 101–4, 110, 127, 132, 134–5, 139–40, 142, 156, 160–1, 166, 168–9, 172, 173, 181, 183–4, 186, 188, 190–1, 192, 198–9, 204, 209–13, 214n3 gender: equality 23, 73, 91; equity 15, 23, 27, 33, 104, 127; relations 9 general marriage rate (GMR), 142, 144, 147 globalization 24, 59, 167, 172–3 GMR, see general marriage rate “Great Leap Forward” 6 Hong Kong 1, 3–4, 25, 28, 30, 32, 110, 132–9, 140, 141, 143–56, 157n4, 168, 213 immigration, 3, 165, 212 industrialization, 100, 133, 182, 184 infant mortality rate (IMR) 90 Japan 1, 4, 7, 8–9, 11–2, 17, 25–6, 28, 30, 31, 33, 36, 40–1, 43, 44, 47–8, 52–4, 56, 58–9, 63, 69, 91, 204, 205, 206, 209; bubble economy burst/collapse (1989) 35, 53, 59 Korea, see South Korea Labor: force 13, 16, 25, 30, 40, 53, 58, 67, 82, 91, 102, 104, 182, 184; market 23, 28, 33–6, 53, 111–3, 127, 129, 163, 173–4, 210: deregulation 24, 35–6, 113, 123 labor-force participation 104, 105, 133; rates (LFPRs) 12–3, 27, 171, 184, 190, 192 Lee Kuan Yew 18 LFPRs, see labor-force participation rates “low-fertility trap” 25, 29, 66 MAM, see mean age at marriage maternity leave 8–10, 123, 127, 152–3, 183, 186 mean age at marriage (MAM) 45, 46, 60, 114, 211; by sex (SMAM) 43, 44, 45–6, 111 migration 17, 132, 208; internal/in- 73, 208; net 156; out- 83 modernization 87, 101, 105, 181–2, 184
natalist policies: anti- 1–2, 4–6, 7, 10, 17, 129, 168, 174; pro- 1, 3–12, 14–7, 18n3, 63, 65, 67, 69, 127–8, 160, 166, 171, 184, 187, 208, 213 New Angel Plan 8, 64–5 “new single concept” 47–8 newly industrialized economies (NIEs) 170 OECD countries, 111, 127 oil shock of 1973 43, 53–4 “one-child policy” 5, 76, 78, 82 Organization of the Petroleum Exporting Countries (OPEC) 43 Pacific Asia 1, 206, 208–9, 211–3 “parasite singles” 28, 47–8, 52 parity progression ratio (PPR) 41–3, 45 population: aging 5, 40–1, 73, 86–9, 91, 127; control 5–6, 76, 183: policies 16–7, 18n2, 183–4; decline 1, 3, 43, 183, 204–6; density 204; dynamics, 75, 92; equilibrium 16; explosion 1; growth 1–3, 5, 16–7, 73–5, 132, 156, 183. 204, 206; policies, 1, 4–5, 9, 11, 18n3, 73, 92, 156, 184–7, 193, 209; rural, 73, 78, 87; trends 16, 182 post-World War 2 baby boomers 41, 161 PPR, see parity progression ratio SDU, see Social Development Unit Second World War, see World War 2 sex ratios 9; at birth (SRB), 89–90, 104–6, 156n2 Singapore 1, 3–4, 7, 10–2, 14, 18n3, 25, 28, 30, 32, 35, 91, 160–1, 170–1, 173, 183, 186–8, 209 singlehood rates 163–5 SMAM, see Mean Age at Marriage by Sex Social Development Unit (SDU) 184, 188 Socioeconomic: changes 160; development 73, 96–7, 99–101, 104–5; structure 100 South Korea 1, 3–4, 7, 9, 11–2, 14, 25, 28, 30–1, 33–6, 91, 96–7, 98, 99–100, 104–6, 110–1, 125, 127, 129, 168, 204, 205, 206, 209 SRB, see sex ratios at birth Sub-Saharan Africa 1, 213 Taiwan 1, 3–4, 7, 11, 25, 91, 128, 204, 214n3 TFR, see fertility rate total TMFR, see total marital fertility rates TMRs, see total marital rates
Index 217 Total: marital fertility rate (TMFR) 42–3, 69; marriage rates (TMRs) 162, 163 urbanization, 100–1, 133
work-: family balance 33; life balance 17 working mothers 11, 14, 153, 156, 172 world population 3, 40, 92, 204 World War 2 43, 53, 132