73438 Toward Gender Equality in East Asia and the Pacific A Companion to the World Development Report WO RLD BA NK E A S T A SIA A N D PACI FIC REGIO N A L REP O RT S Well known for their economic success and dynamism, countries in the East Asia and Pacific region must tackle an increasingly complex set of challenges to continue on a path of sustain- able development. Learning from others within the region and beyond can help identify what works, what doesn’t, and why, in the search for practical solutions to these challenges. This regional report series presents analyses of issues relevant to the region, drawing on the global knowledge and experience of the World Bank and its partners. The series aims to inform public discussion, policy formulation, and development practitioners’ actions to turn challenges into opportunities. World Bank East Asia and Pacific Regional Report Toward Gender Equality in East Asia and the Pacific A Companion to the World Development Report © 2012 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved 1 2 3 4 15 14 13 12 This work is a product of the staff of The World Bank with external contributions. Note that The World Bank does not necessarily own each component of the content included in the work. The World Bank there- fore does not warrant that the use of the content contained in the work will not infringe on the rights of third parties. The risk of claims resulting from such infringement rests solely with you. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions This work is available under the Creative Commons Attribution 3.0 Unported license (CC BY 3.0) http:// creativecommons.org/licenses/by/3.0. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution—Please cite the work as follows: World Bank. 2012. Toward Gender Equality in East Asia and the Pacific: A Companion to the World Development Report. Washington, DC: World Bank. doi: 10.1596/978-0-8213-9623-0 License: Creative Commons Attribution CC BY 3.0 Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. All queries on rights and licenses should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. ISBN (paper): 978-0-8213-9623-0 ISBN (electronic): 978-0-8213-9626-1 DOI: 10.1596/978-0-8213-9623-0 Cover photo: © www.istockphoto.com / Bartosz Hadyniak Library of Congress Cataloging-in-Publication Data Toward gender equality in East Asia and the Pacific.   p. cm. — (World Bank East Asia and Pacific regional report)   “A companion to the World development report.�   Includes bibliographical references.   ISBN 978-0-8213-9623-0 — ISBN 978-0-8213-9626-1 (electronic)   1. Women’s rights—Asia.  2. Women’s rights—Pacific Area.  3. Women in development—Asia.  4. Women in development—Pacific Area.  5. Sex discrimination against women—Asia.  6. Sex discrimination against women—Pacific Area.  I. World Bank.  II. World development report. III. Series: World Bank East Asia and Pacific regional report. HQ1236.5.A78T69 2012 323.3’4095—dc23 2012030808 Contents Foreword. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv Acknowledgments Abbreviations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvii Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Why does gender equality matter for development?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Recent progress, pending challenges. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Why do many gender inequalities persist? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Emerging opportunities and risks in an increasingly integrated world. . . . . . . . . . . . . . . . . 16 Toward gender equality in East Asia and the Pacific: Directions for policy . . . . . . . . . . . . . 18 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 1. The State of Gender Equality in East Asia and the Pacific. . . . . . . . . . . . . . . . . . . . . . . . . . 33 Why does gender equality matter for development?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Recent progress, pending challenges. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Fostering new opportunities, managing emerging risks . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Gender equality in East Asia and the Pacific: A roadmap to the report . . . . . . . . . . . . . . . . 53 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 2. Gender and Endowments: Access to Human Capital and Productive Assets . . . . . . . . . . . . 61 Substantial progress toward gender equality in education . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Improvements and remaining gender issues in health. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 Gender equality in productive assets: An unfinished agenda . . . . . . . . . . . . . . . . . . . . . . . . 85 Policies to promote gender equality in endowments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 v v i    C o n t e n t s 3. Gender and Economic Opportunity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Limited effects of growth on gender gaps in economic opportunity. . . . . . . . . . . . . . . . . . 105 Gender differences in economic activity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 What determines gender inequality in economic opportunities?. . . . . . . . . . . . . . . . . . . . . 114 Policies to promote gender equality in economic opportunity . . . . . . . . . . . . . . . . . . . . . . 130 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 4. Agency: Voice and Influence within the Home and in Society . . . . . . . . . . . . . . . . . . . . . . 145 Agency—important for gender equality and development . . . . . . . . . . . . . . . . . . . . . . . . . 145 The state of women’s agency in East Asia and the Pacific. . . . . . . . . . . . . . . . . . . . . . . . . . 147 Explaining progress and pending challenges. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158 Policy approaches to promote gender equality in agency . . . . . . . . . . . . . . . . . . . . . . . . . . 169 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173 5. Gender and the Region’s Emerging Development Challenges. . . . . . . . . . . . . . . . . . . . . . . 179 Globalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180 Migration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185 Urbanization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 Aging populations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195 Information and communication technologies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 6. Promoting Gender Equality in East Asia and the Pacific: Directions for Policy. . . . . . . . . 213 Promoting gender equality in human development. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215 Taking active measures to promote gender equality in economic opportunity . . . . . . . . . . 221 Taking measures to strengthen women’s voice and influence . . . . . . . . . . . . . . . . . . . . . . . 234 Fostering new opportunities, managing emerging risks . . . . . . . . . . . . . . . . . . . . . . . . . . . 238 Filling knowledge gaps. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240 Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241 References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 Boxes 1.1 Defining and measuring gender equality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 1.2 Defining and measuring agency: Women’s voice, influence, and participation. . . . . . . . 48 2.1 Various parts of the world experience reverse gender gaps in education . . . . . . . . . . . . 64 2.2 Recent improvement in the sex ratio at birth in the Republic of Korea . . . . . . . . . . . . . 82 3.1 Is higher female labor force participation always a good thing?.. . . . . . . . . . . . . . . . . 107 3.2 Gender-related beliefs on appropriate employment in Fiji, Papua New Guinea, Indonesia, and Vietnam. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130 4.1 Gender and land tenure in a plural legal environment: The case of the Solomon Islands. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166 4.2 Strengthening state mechanisms for gender equality. . . . . . . . . . . . . . . . . . . . . . . . . . 168 6.1 Reducing maternal mortality rates through improved staffing and quality of service delivery. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 6.2 The design of infrastructure projects has important gender dimensions . . . . . . . . . . . 223 6.3 Leveling the playing field in access to resources: Lessons from Papua New Guinea and Mongolia����������������������������������������������������������������������������������������227 C o n t e n t s   v ii 6.4 The impact of active labor market policies on female employment: Evidence from across the world��������������������������������������������������������������������������������������������������231 Figures O.1 The East Asia and Pacific region has experienced rapid economic growth. . . . . . . . . . . 2 O.2 Poverty reduction in the East Asia and Pacific region has been impressive. . . . . . . . . . . 2 O.3 Girls’ secondary school enrollments have converged to those of boys. . . . . . . . . . . . . . 6 O.4 Maternal mortality rates have declined in most countries in the region. . . . . . . . . . . . . 7 O.5  Female labor force participation is high by global standards but also varies substantially across the region. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 O.6  Across the region, female-headed households own less land than male-headed households. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 O.7 Women in East Asia and the Pacific still earn less than men . . . . . . . . . . . . . . . . . . . . . 8 O.8  In urban China and Indonesia, gender wage gaps are largest among low wage earners . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 O.9 Enterprises with female managers tend to be smaller. . . . . . . . . . . . . . . . . . . . . . . . . . . 9 O.10 Who decides how wives’ cash earnings are used varies widely across the region. . . . . 10 O.11 Women’s representation in parliament is low, especially in the Pacific. . . . . . . . . . . . . . 11 O.12 Violence against women is high in the region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 O.13 Women are concentrated in certain fields of study, such as education and medicine, but are underrepresented in law and engineering. . . . . . . . . . . . . . . . . . . . . 12 O.14 In Lao PDR, women—particularly those with young children—must balance household and market work. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 O.15 In Indonesia, female-led enterprises are clustered in lower-productivity and less capital-intensive industries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 O.16 Male- and female-led firms report similar constraints in Indonesia. . . . . . . . . . . . . . . 15 O.17 Men and, in some cases, women believe that men make better political leaders than women. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 O.18 The share of female workers in export-oriented firms is relatively high. . . . . . . . . . . . . 17 O.19 China has seen remarkable growth in Internet use since 2000, but women’s use trails men’s. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 O.20 The old-age dependency ratio is increasing in most East Asian countries . . . . . . . . . . 19 1.1 The East Asia and Pacific region has experienced rapid economic growth. . . . . . . . . . 34 1.2 Poverty reduction in the East Asia and Pacific region has been impressive. . . . . . . . . . 34 1.3 Girls’ secondary school enrollments have converged to those of boys. . . . . . . . . . . . . 38 1.4  Tertiary school enrollments of females have converged to those of males in East Asia and the Pacific . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 1.5 Gender gaps in secondary school enrollment vary substantially across countries . . . . 39 1.6  Gender gaps in education have reversed in several countries, particularly at the tertiary level. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 1.7 Minority populations in Vietnam often experience lower educational enrollments. . . 40 1.8  In Indonesia, gender gaps in enrollment do not vary substantially by household wealth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 1.9 Fertility rates have declined across the world . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 1.10 Under-five mortality rates have declined sharply for both boys and girls. . . . . . . . . . . . 41 1.11 Maternal mortality rates have declined across the world. . . . . . . . . . . . . . . . . . . . . . . 42 1.12 Maternal mortality rate has declined in most countries in the region . . . . . . . . . . . . . 42 1.13 East Asia has a highly skewed male-to-female ratio at birth . . . . . . . . . . . . . . . . . . . . 44 1.14 The East Asia and Pacific region has high female labor force participation rates. . . . . . . 44 v i i i    C o n t e n t s 1.15 Female labor force participation varies substantially across countries . . . . . . . . . . . . 45 1.16  The fraction of the workforce employed in agriculture has declined in the East Asia and Pacific region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 1.17 The evolution of sectoral composition by gender varies across countries. . . . . . . . . . . 46 1.18 Women are more likely than men to be temporary workers. . . . . . . . . . . . . . . . . . . . . 46 1.19 Enterprises with female management tend to be smaller . . . . . . . . . . . . . . . . . . . . . . . 47 1.20  In Lao PDR, women—particularly those with young children—must balance household work commitments with market work. . . . . . . . . . . . . . . . . . . . . . 47 1.21 Women in East Asia still earn less than men . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 1.22  Women in East Asia and the Pacific have more control over earnings and household decisions across all wealth quintiles than women from other developing regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 1.23  Female representation in parliament in East Asian and Pacific countries has hardly changed since the 1990s. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 1.24 Women’s representation in parliament is low, especially in the Pacific . . . . . . . . . . . . 50 1.25  Men, and in some cases women, believe that men make better political leaders than women. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50 1.26  More women hold top management positions in East Asia and the Pacific than in other developing regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 1.27 Violence against women is high in East Asia and the Pacific. . . . . . . . . . . . . . . . . . . . 52 1.28 The dependency ratio is increasing in most East Asian countries . . . . . . . . . . . . . . . . 53 2.1  Enrollment for both genders has been converging even among the poorest populations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63 2.2 Gender gaps in youth literacy are smaller than gender gaps in adult literacy. . . . . . . . . 63 B2.1.1 The biggest enrollment disadvantage for boys in the Phillippines is among the poor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 2.3 Girls in some ethnic minority groups in Lao PDR lag even further in enrollment. . . . . 65 2.4 Girls in some ethnic minority groups in Vietnam lag even further in enrollment . . . . . 65 2.5 Even girls in wealthier households in Cambodia and Lao PDR lag behind boys. . . . . . 67 2.6  Women are concentrated in certain fields of study, such as education and health, but are underrepresented in law and engineering. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 2.7 Girls tend to outperform boys in several subjects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 2.8 Maternal mortality is lower in higher-income countries . . . . . . . . . . . . . . . . . . . . . . . . 73 2.9 Female infant mortality is lower in higher-income countries. . . . . . . . . . . . . . . . . . . . . 73 2.10  Most East Asia and Pacific region countries do not have female-skewed under-five mortality and infant mortality, except China. . . . . . . . . . . . . . . . . . . . . . . . 74 2.11  Women in wealthier households are more likely to have births assisted by trained medical staff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 2.12 Contraceptive prevalence varies across East Asian and Pacific countries. . . . . . . . . . . . 77 2.13  The percentage of births attended by skilled professionals varies across East Asian and Pacific countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 2.14  Women in rural areas are less likely to have births assisted by trained medical staff. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 2.15  Sex ratios at higher order births are still of concern, even though the overall sex ratio at birth has approached the normal range in the Republic of Korea. . . . . . . . 79 2.16 Men are more likely to smoke than women. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 2.17 Men are more likely to drink than women. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 2.18 Tobacco use is positively correlated with mortality due to lung cancer. . . . . . . . . . . . . 84 2.19  Alcohol consumption is positively correlated with mortality due to alcohol use disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 C o n t e n t s   i x 2.20 The poor are slightly more likely to engage in smoking. . . . . . . . . . . . . . . . . . . . . . . . 85 2.21 The probability of owning land is not substantially lower for female-headed households than for male-headed households . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 2.22 Female-headed households own less land in terms of land size . . . . . . . . . . . . . . . . . . 87 2.23 Vietnamese men owned more agricultural land plots than did women in 2008. . . . . . 88 2.24 Female-headed households are less likely to own livestock . . . . . . . . . . . . . . . . . . . . . 88 2.25 Female-headed households are slightly less likely to borrow from a financial institution. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 2.26 Women are slightly less likely than men to report having an account at a formal financial institution.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 3.1  Female labor force participation in the Republic of Korea rose for women of all ages between 1960 and 2005 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 3.2  Labor force participation is greater in rural areas than in urban areas for both males and females . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 3.3  Female labor force participation rates in Malaysia have risen over time among 20- to 55-year-olds, but continue to decline during child-bearing years. . . . . 109 3.4  Women in rural Vietnam with children under age 2 are substantially less likely to participate in the labor market than those without young children. . . . . . . . 110 3.5  Women earn less than men in the majority of East Asian and Pacific countries and in all sectors of the economy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 3.6  Male- and female-owned firms in the formal sector do not display substantial differences in productivity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 3.7  In the informal sector, gender-based differences in productivity are more pronounced than in the formal sector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 3.8 Women are more likely than men to work as unpaid family workers. . . . . . . . . . . . . . 112 3.9  Women are slightly more likely to be employed in the informal sector than men. Rural-urban differences in informality are greater than gender differences within rural or urban areas. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 3.10 Men and women work in different sectors throughout the East Asia and Pacific region. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 3.11 The pattern of female ownership by firm size varies across countries . . . . . . . . . . . . . 115 3.12 In Indonesia, female-led enterprises are clustered in lower-productivity and capital-intensive industries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 3.13 Female labor force participation is high by global standards but also varies substantially across the region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 3.14 The fraction of the gender wage gap explained by differences in characteristics is smaller than the fraction of the wage gap attributable to differences in returns in Indonesia, 2009. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 3.15 Differences in productivity across informal and formal firms in Vietnam are larger than differences across male- and female-led enterprises. . . . . . . . . . . . . . . . . 123 3.16 Self-reporting shows little difference in access to finance between male- and female-run firms, with the exception of Timor-Leste and Tonga. . . . . . . . . . . . . . . . 123 3.17 Male- and female-led informal firms report similar constraints in Indonesia . . . . . . 124 3.18 In Cambodia, women—particularly those with young children—balance household work commitment with market work . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 3.19 Women in Timor-Leste spend more time on domestic activities than men, and these differences are found among richer as well as poorer households. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128 4.1  There is a positive relationship between economic development and civic activism. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 x    C o n t e n t s 4.2  There is no clear relationship between economic development and women’s representation in parliament . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 4.3  Who decides how wives’ cash earnings are used varies across the region. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 4.4  A majority of wives control decisions regarding their own health care and household purchases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 4.5  Women in East Asia have greater control over decisions than in other regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150 4.6 Fertility rates in the region have declined over the past three decades. . . . . . . . . . . . . 151 4.7 Adolescent fertility is especially high in the Pacific . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 4.8 The rate of unwanted and mistimed pregnancies is especially high in the Pacific . . . . 152 4.9  Many firms have female participation in ownership, but fewer have a female top manager. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 4.10 Civic activism has grown in the low- and middle-income countries in the region . . . . 153 4.11 Women’s representation in parliament has grown in much of East Asia but has experienced little change in the Pacific Islands . . . . . . . . . . . . . . . . . . . . . . . . 155 4.12 Women’s representation in parliament is low, especially in the Pacific. . . . . . . . . . . . . 155 4.13 Violence against women is high in the region.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156 4.14 The incidence of sexual intimate-partner violence is significantly higher in the Pacific. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 4.15 Gender-based violence can have inter-generational consequences. . . . . . . . . . . . . . . . 157 4.16 Many men and women in the region believe that men make better political leaders than women. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161 5.1  Women are more likely to work in export-oriented firms than in non-export-oriented firms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181 5.2  Employment in textile and apparel in Vietnam grew substantially between 2000 and 2008 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182 5.3 Women predominate in the garment sector in four East Asian countries. . . . . . . . . . . 182 5.4  The share of international female migrants has increased over time in Indonesia and Vietnam . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186 5.5  The urban population now accounts for more than 50 percent of the world’s population. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 5.6 Urbanization is expected to be rapid in East Asia . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192 5.7  The rates of urban growth are predicted to vary substantially across countries in the Pacific. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192 5.8  In most East Asian and Pacific countries, urban areas have better access to improved sanitation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 5.9  Rural areas have lower access to improved water sources than urban areas in the majority of countries in the region. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193 5.10 The old age dependency ratio is expected to increase for both men and women in the next two decades; the female ratio will exceed the male ratio in the future. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 195 5.11 The gender gap in life expectancy at birth is lower in the East Asia and Pacific region than in many other regions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196 5.12 Life expectancy at birth in the region has improved for both men and women since 1990, although gender gaps have widened in some countries. . . . . . . . . . . . . . . 196 5.13 Women ages 65 and above are far more likely to be widowed than men. . . . . . . . . . . 197 5.14 Women’s time devoted to housework and caregiving activities is not significantly greater in households with elderly members. . . . . . . . . . . . . . . . . . . . . . . 199 C o n t e n t s   x i 5.15 Internet use has increased substantially in the East Asia and Pacific region and around the developing world since 2000. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 5.16 The number of cell phone subscribers per 100 people in the East Asia and Pacific region has grown substantially since 2000. . . . . . . . . . . . . . . . . . . . . . . . . . . . 200 5.17 Internet use has grown quickly in many countries in the region but has grown slowly in others. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 5.18 The number of cell phone subscribers in the population has grown across most of East Asia but has remained limited in some Pacific countries. . . . . . . . . . . . . 201 5.19 Women in the region are less likely to be mobile phone subscribers than men. . . . . . . 202 5.20 Internet use has grown for both men and women in China, although a persistent gender gap in access remains. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202 Tables 2.1 Labor market returns to studying engineering are high relative to studying education. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69 2.2 School curricula in a number of East Asian countries have gender stereotyping. . . . . . 71 2.3 The East Asia and Pacific region, mainly driven by China, is characterized by its large number of missing girls at birth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 2.4 Most East Asian and Pacific countries do not differentiate by gender in inheritance and property laws. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 3.1 Legal retirement ages in the East Asia and Pacific region. . . . . . . . . . . . . . . . . . . . . . . 119 4.1 Legislation against gender-based violence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164 4.2 Political affirmative action in East Asia and the Pacific. . . . . . . . . . . . . . . . . . . . . . . . 167 5.1 Evidence shows no systematic gender differences in consumption poverty among the elderly, regardless of family status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197 6.1 Selected policy approaches to tackle excessive tobacco and alcohol use. . . . . . . . . . . . 221 Foreword O ver the past few decades, the East access than men to a range of productive Asia and Pacific region has been assets and services, including land, finan- the most economically dynamic cial capital, agricultural extension services, region in the world. In most countries and new information technologies. Sub- in the region, incomes have grown dra- stantial employment segregation by gender matically, and with that growth, absolute remains. Women are less likely than men to poverty has declined rapidly. Most of the work in formal sector jobs and more likely region’s economies have also shifted away to work in poorly remunerated occupations from agriculture and toward manufactur- and enterprises. Despite the closing of educa- ing and services. Rapid growth, structural tion gaps, women still earn less than men for transformation, and poverty reduction have similar work all across the region. Women been accompanied by progress toward gen- in East Asian and Pacific countries still have der equality in several key areas. Economic a weaker voice and less influence than men, development has led to the closing of gender whether within the household, in the private gaps in school enrollments and a decline in sector, in civil society, or in politics. And maternal mortality rates: girls in the region women across the region remain vulnerable as a whole now enroll in secondary schools to gender-based violence, often at the hand of at a higher rate than boys, and maternal an intimate partner. mortality has fallen by half over the past The main message of this book, Toward 20 years. Access to economic opportuni- Gender Equality in East Asia and the ties has also increased, particularly among Pacific— a regional companion to the World younger, better educated women. In many Development Report 2012: Gender Equal- ways, women in East Asian and Pacific ity and Development—is that policy makers countries are better positioned than ever in the region need to understand why prog- before to participate in, contribute to, and ress in closing gender gaps has been mixed benefit from development. and to implement corrective policies where Yet, the experience of the region illustrates gaps remain persistent. The reason is that also how growth and economic development gender equality is both an important devel- are not enough to attain gender equality opment objective in its own right as well as in all its dimensions. Women still have less good development policy. A growing body of xiii x i v    F o r e w o r d evidence shows that promoting gender equal- public action in the countries of East Asia ity in access to productive resources, eco- and the Pacific. First, promoting gender nomic opportunity, and voice can contribute equality in human development remains to higher economic productivity, improve the important where gender gaps in education economic prospects and wellbeing of the next and health outcomes remain large. Second, generation, and lead to more effective devel- policies to close gender gaps in economic opment policy making. Yet, gender equality opportunity have a critical role. Such mea- in many areas does not happen automatically. sures are often warranted on both equity Thus, gender-aware public policy is required and efficiency grounds. Third, initiatives to if countries are to achieve both gender equal- strengthen women’s voice and influence— ity and more rapid development. and to protect them from violence—are As a regional report, Toward Gender also called for across the region. Strength- Equality in East Asia and the Pacific focuses ening women’s agency will enhance the on issues that are particularly pertinent to quality of development decision making the region. Among other things, the report and, thus, development broadly. A nd, examines the gender dimensions of several finally, public policy can foster the oppor- emerging trends in the region—increased tunities and manage the risks associated global economic integration, the rising use with emerging trends in the region; taking of information and communication tech- a gender-aware approach to policy making nologies, migration, urbanization, and rapid in this area will lead to better gender— population aging—all of which are generat- and development—outcomes. ing new opportunities, but also new risks, for This report shows that in East Asia and promoting gender equality. The report also the Pacific, as in other parts of the world, contributes to the development of new data gender equality is both the right development and evidence on gender and development, objective as well as good development policy. significantly strengthening the ability of countries to formulate evidence-based policy Pamela Cox in this area. Vice President Drawing on this evidence base, the East Asia and Pacific Region report identifies four priority areas for The World Bank Acknowledgments T his study has been prepared by a mul- The team is grateful for the ongoing sup- tisectoral, multidisciplinary team led port and guidance provided by members of by Andrew Mason under the guid- the East Asia and Pacific Region’s manage- ance of Bert Hofman and Sudhir Shetty. ment, including James Adams, Pamela Cox, The report was written by a core team com- Vikram Nehru, John Roome, Emmanuel prising Reena Badiani, Trang Van Nguyen, Jimenez, Linda Van Gelder, Magda Lovei, Katherine Patrick, Ximena Del Carpio, and Xiaoqing Yu, Annette Dixon, Victoria Andrew Mason, with significant contribu- Kwakwa, and Coralie Gevers. The team tions from Jennifer Golan. Patricia Fer- also appreciates the support of the members nandes, Anne Kuriakose, Rea Chiongson, of the Region’s Gender Practice Group, past and Daniel Mont also provided substantive and present, including Nina Bhatt, Helene inputs. New data analysis for the report was Carlsson Rex, Markus Kostner, Eduardo carried out by Juan Feng, Reno Dewina, and Velez, and Lester Dally. Furthermore, the Flora Nankhuni, using the East Asia and team benefitted from the ongoing interac- Pacific Region’s Poverty Monitoring Data- ­ oordination, and support of the World tion, c base. Background papers for the report were Development Report 2012 team, namely written by Sarah Iqbal, Nayda Almodovar Ana Revenga, Sudhir Shetty, Ana Maria Reteguis, Yasmin Klaudia bin Humam, Munoz, Carolina Sanchez-Paramo, Luis Ben- Josefina Posadas, Reena Badiani, John veniste, Markus Goldstein, Jishnu Das, and Rang, Benedikte Bjerge, Chris Sakellariou, Aline Coudouel. and Dongxiao Liu. Qualitative studies on During the report’s preparation, the team women’s economic decision making in Fiji, received helpful comments from four peer Indonesia, Papua New Guinea, and Vietnam reviewers: Gillian Brown, Jeni ­ K lugman, were coordinated by Carolyn Turk and Patti Pierella Paci, and Martin Rama. Several L. Petesch as part of a 22-country research country teams, represented by their Coun- effort carried out under the auspices of the try Gender Coordinators (including Laura World Development Report 2012 on gender Bailey, Edith Bowles, Yulia Immajati, Vanna ­ equality and development. Production sup- Nil, Solvita Klapare, Erdene Ochir Badarch, port was provided by Lynn Yeargin, Cath- and Pamornrat Tansanguanwong), also pro- ryn Summers, and Mildred Gonsalvez. vided valuable feedback at various stages. xv x v i    A c k n o w l e d g m e n t s In addition, the team received construc- on preliminary findings and messages tive comments and inputs from Keith Bell, from the participants of the International Shubham Chaudhury, Qimiao Fan, Mary Association for Feminist Economics, held H allward Dreimeier, Mathew Verghis, ­ in June 2011 in Hangzhou, China; from R obert Jauncey, Gladys Lopez-Acevedo, ­ country team ­colleagues in Cambodia, the David Newhouse, ­ C armen N­ iethammer, Lao People’s Democratic Republic, and Bob R ­ ijkers, Sevi Simavi, Monica das Thailand who participated in a video semi- Gupta, I ­ sabella Micali Drossos, Thuy Thi nar in September 2011; from participants Thu Nguyen, Tehmina Khan, and Car- in the Asia-Pacific Economic Cooperation los Sobrado. Mark Ingebretsen, Patricia (APEC) Women and the Economy Sum- Katayama, and Andrés Meneses, from the mit held in San Francisco in September Office of the Publisher, provided excellent 2011; and from colleagues, counterparts, support in the design and publication of the and other stakeholders who participated report. Mohamad Al-Arief and Carl Hanlon in events in Australia and New Zealand provided invaluable guidance on a dissemi- in November 2011; Japan, Indonesia, and nation and communication strategy. Vietnam in December 2011; and in Papua The team benefited from early consulta- New Guinea, the Solomon Islands, and Fiji tions with policy makers, civil society rep- in March 2012. resentatives, academics, and development Generous financial support for the prepa- partners in Indonesia, Mongolia, T ­ hailand, ration and dissemination of this report was and Vietnam. It also gained from feedback provided by AusAID. Abbreviations ADAPT An Giang Dong Thap Alliance for the Prevention against Trafficking ADB Asian Development Bank AED Academy for Education Development AGI Adolescent Girls Initiative ALMPs active labor market policies ASEAN Association of South East Asian Nationals AusAID Australian Agency for International Development BFC Better Factories Cambodia BMI body mass index BPS Badan Pusat Statistik BREAD Bureau for Research and Economic Analysis of Development CAPWIP Center for Asia-Pacific Women in Politics CAREM-Asia Coordination of Action Research on AIDS, Mobility-Asia CCT conditional cash transfers CDD community-driven development CEDAW  Convention on the Elimination of All Forms of Discrimination against Women CGA Country Gender Assessment CIDA Canadian International Development Agency CWCC Cambodian Women’s Crisis Centre CWDI Corporate Women Directors International DFID Department for International Development (United Kingdom) DHS Demographic and Health Surveys EAP East Asia and Pacific EC European Commission ECA Europe and Central Asia ECPAT End Child Prostitution in Asian Tourism EFA education for all EPZ Export Processing Zone FAO Food and Agriculture Organization of the United Nations FCND Food Consumption and Nutrition Division xvii x v i i i    A b b r e v i a t i o n s FODE Flexible Open and Distance Education GBV gender-based violence GDC Gender and Development for Cambodia GDP gross domestic product GEM gender equity model GNP Grand National Party G-PSF Government-Private Sector Forum GRID Global Resource Information Data Base HEF Health Equity Fund HNP Health Nutrition and Population HSI Hang Seng Index ICT information and communication technology ICRW International Center for Research on Women IDEA International Development Evaluation Association IFC International Finance Corporation ILO International Labour Office INSTRAW United Nations International Research and Training Institute for the Advancement of Women IOM International Organization for Migration ISCO International Standard Classification of Occupations ISD Indices of Social Development KDP Kecamantan Development Program KILM Key Indicators of the Labour Market LAC Latin America and the Caribbean Lao PDR Lao People’s Democratic Republic MDGs Millennium Development Goals MENA Middle East and North Africa MIC middle income countries MMR maternal mortality rate MPDF Mekong Project Development Facility MSMEs micro, small, and medium enterprises NCRFW National Commission on the Role of Filipino Women NGOs nongovernmental organizations NIPH National Institute of Public Health NIS National Institute of Statistics NSD National Statistics Directorate NSO National Statistics Office OECD Organisation for Economic Co-operation and Development ORC Opinion Research Corporation PEKKA Indonesia Women Headed Household Program PISA Programme for International Student Assessment PPP purchasing power parity SAR special administrative region SciCon Science Connections SEZ Special Economic Zone SHG self-help group SIGI Social Institutional and Gender Index SMEs small and medium enterprises SPC Secretariat of the Pacific Community SRB sex ratios at birth STI sexually-transmitted infections SUSI Survey of Cottage and Small-Scale Firms (Survei Usaha Terintegrasi) A b b r e v i a t i o n s   x ix ThaiHealth Thai Health Promotion Foundation TIMSS Trends in International Mathematics and Science Study UN United Nations UNDP United Nation Development Programme UNESCO United Nations Educational, Scientific and Cultural Organization UNFPA United Nations Fund for Population Activities UNICEF United Nations Children’s Fund UNIFEM United Nations Development Fund for Women VCCI Vietnam Chamber of Commerce and Industry WBL Women, Business, and the Law WDR World Development Report WHO World Health Organization ISO 3166 country name abbreviations AUS Australia CHN China FJI Fiji FSM Federated States of Micronesia HKG Hong Kong SAR, China IDN Indonesia JPN Japan KHM Cambodia KIR Kiribati KOR Republic of Korea LAO Lao People’s Democratic Republic MHL Marshall Islands MMR Myanmar MNG Mongolia MYS Malaysia NZL New Zealand PHL Philippines PLW Palau PNG Papua New Guinea PRK Democratic Republic of Korea PYF French Polynesia SGP Singapore SLB Solomon Islands THA Thailand TMP Timor-Leste TON Tonga TUV Tuvalu TWN Taiwan, China VNM Vietnam VUT Vanuatu WSM Samoa Overview I n recent decades, women across the decades have been associated with reduced globe have made positive strides toward gender inequalities in several dimensions. gender equality. Literacy rates for young The region grew at 7 percent on average women and girls are higher than ever before, between 2000 and 2008 (figure O.1), the while gender gaps in primary education structure of the region’s economies has shifted have closed in almost all countries. In the away from agriculture toward manufacturing last three decades, over half a billion women and services, and extreme poverty has fallen have joined the world’s labor force (World dramatically. Indeed, the share of the region’s Bank 2011c). Progress toward gender equal- population living on less than US$1.25 a ity in East Asia and the Pacific has been day has declined by more than 50 percent similarly noteworthy. Most countries in the since 1990—from among the highest rates region have either reached or surpassed gen- of poverty in the world to among the lowest der parity in education enrollments. Health (figure O.2). Growth, structural transforma- outcomes for both women and men have tion, and poverty reduction have been accom- improved significantly. Female labor force panied by considerable progress toward participation rates in the region are rela- gender equality in several key areas, particu- tively high. Yet, despite considerable progress larly education and health. Many countries in this economically dynamic region, gender in the region have experienced closing gen- disparities persist in a number of important der gaps in school enrollments and declining areas—particularly in access to economic maternal and child mortality rates. opportunity and in voice and influence in society. For policy makers in East Asian But growth and development have not been and Pacific countries, closing these gender enough to attain gender equality in all its gaps represents an important challenge dimensions. Women still have less access to achieving more inclusive and effective than men to a range of productive assets and development. services, including land, financial capi- tal, agricultural extension services, and The East Asia and Pacific Region’s signifi- new information technologies. Substantial cant economic growth, structural transfor- employment segregation by gender remains. mation, and poverty reduction in the last few Women are less likely than men to work in 1 2    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE O.1  The East Asia and Pacific region has experienced rapid Pacific countries still have a weaker voice and economic growth less influence than men, whether in house- hold decision making, in the private sector, in average annual GDP growth rate civil society, or in politics. Moreover, women 10 across the region remain vulnerable to 8 gender-based violence, often at the hand of an intimate partner. 6 4 This report clarifies empirically the rela- tionship between gender and develop- percent 2 ment and outlines an agenda for public 0 action to promote gender equalit y in East Asian and Pacific countries. The –2 report was written as a companion to the –4 World Development Report 2012: Gen- der Equality and Development (World –6 1980–84 1985–89 1990–94 1995–99 2000–04 2005–09 Bank 2011c) and is intended as a tool to help policy makers in the region promote East Asia and Paci c Europe and Central Asia both gender equality and more effective Latin America and the Caribbean Middle East and North Africa South Asia Sub-Saharan Africa development. Following the World Devel- opment Report 2012 , this report focuses on gender outcomes in three domains: Source: World Development Indicators (WDI) database. (a) endowments—human and productive capital; (b) economic opportunity—par- FIGURE O.2  Poverty reduction in the East Asia and Pacific region ticipation and returns in the economy; and has been impressive (c) agency—women’s voice and influence in all facets of society. poverty trends by region 90 The report makes several distinct contri- percentage of the population living below 80 butions to policy makers’ understanding of 70 gender, development, and public policy in 60 East Asian and Pacific countries. $1.25 a day 50 First, the analysis focuses on issues that •   40 are particularly relevant to the region. 30 Compared with other developing regions, 20 for example, female access to basic edu- 10 cation is no longer a first-order concern 0 1981 1984 1987 1990 1993 1996 1999 2002 2005 2008 in most East Asian and Pacific countries. Gender stereotyping and gender “stream- East Asia and Paci c Europe and Central Asia ing� in education still represent critical Latin America and the Caribbean Middle East and North Africa challenges, however, and thus receive par- South Asia Sub-Saharan Africa ticular emphasis in the report. S econd, the report examines the gender •   Source: PovcalNet. dimensions of several emerging trends that are important to the region’s development: formal sector jobs and more likely to work in increased global economic integration, poorly remunerated occupations and enter- rising use of information and commu- prises. And despite closing of education gaps, nication technologies (ICTs), migration, women continue to be paid less than men urbanization, and rapid population aging. for similar work. Women in East Asian and These trends have gender dimensions that O V E R V I E W   3 are not commonly accounted for by policy Development Report 2012 (World Bank makers but that will generate a distinct set 2011c, 3), “Gender equality is smart econom- of challenges for promoting gender equal- ics.� Indeed, the literature shows that greater ity going forward. gender equality in endowments, access to T hird, the East Asia and Pacific region is •   economic opportunities, and agency can (a) vast and diverse, with important differ- contribute to higher productivity, income ences in economic and social characteristics growth, and poverty reduction; (b) improve that affect progress toward gender equality. the opportunities and outcomes of the next The report accounts for intraregional diver- generation; and (c) enhance development sity in a way that is not possible in a global decision making. This section explores the report. Particular emphasis is placed, where evidence on these three pathways, in turn. possible, on the challenges faced by coun- tries in the Pacific as distinct from those in Gender equality can contribute to East Asia. higher productivity and income growth Finally, the report has undertaken exten- •   sive empirical analysis of gender equality For households and economies to function using a newly created database of house- at their full potential, resources, skills, and hold surveys for the region. In doing so, talent should be put to their most produc- the report has contributed significantly tive use. If societies allocate resources on to the development of indicators and evi- the basis of one’s gender, as opposed to dence on gender, development, and public one’s skills and abilities, this allocation policy that were not available previously. comes at a cost. Indeed, the economic costs of gender inequalities—whether caused by the persistence of traditional norms or Why does gender equality by overt discrimination—can be consider- matter for development? able. A recent study commissioned for the Gender equality matters intrinsically. Nobel World Development Report 2012 found prize–winning economist Amartya Sen trans- that in the East Asia and Pacific region, formed the discourse on development when output per worker could be 7 to 18 percent he argued that development not only is about higher across a range of countries if female raising people’s incomes or reducing poverty entrepreneurs and workers were to work in but rather involves a process of expanding the same sectors, types of jobs, and activi- freedoms equally for all people (Sen 1999).1 ties as men and to have the same access to Viewed from this perspective, gender equal- productive resources (Cuberes and Teignier- ity is intrinsically valued. The near-universal Baqué 2011). ratification and adoption of the Convention Evidence suggests that misallocation of on the Elimination of All Forms of Discrimi- female skills and talent commonly begins nation against Women (CEDAW)2 —and the before women enter the labor force, when subsequent commitment of the international families and societies underinvest in girls’ community to Millennium Development schooling. A number of cross-country stud- Goals 3 and 5—underscores a near-global ies have found a robust inverse relationship consensus that gender equality and women’s between the size of the gender gap in edu- empowerment are development objectives in cation and gross domestic product (GDP) their own right. growth, controlling for average education levels and other factors associated with eco- Gender equality also matters for develop- nomic growth (see, for example, Klasen 2002; ment. A growing body of empirical litera- Knowles, Lorgelly, and Owen 2002). 3 ture from around the world demonstrates Moreover, to the extent that young women that promoting gender equality is also good (or men) choose fields of study on the basis development policy, or as stated in the World of their gender rather than their abilities, 4    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C this too will exact costs not only on indi- an increasingly competitive world, coun- viduals’ employment and earnings, but also tries will need to harness their resources on a country’s economic productivity more efficiently by improving opportunities for broadly. all and allocating labor on the basis of Gender inequalities in access to pro- skill instead of gender. Gender inequal- ductive assets also have costs in terms of ity, whether in endowments, economic productivity and income. Microeconomic opportunities, or agency, reduces a coun- studies from a number of countries across try’s ability to compete in this increasingly developing regions show that female farm- globalized economic environment (World ers and entrepreneurs are inherently no less Bank 2011c). productive than male farmers and entrepre- neurs; rather, they tend to have less access Promoting gender equality is also an to productive inputs.4 A recent study by investment in the next generation the Food and Agriculture Organization of the United Nations estimates that equaliz- A large body of cross-country and country- ing access to productive resources between specific literature shows that healthier, bet- female and male farmers could increase ter educated mothers have healthier, better agricultural output in developing countries educated children, which can be expected by 2.5 to 4.0 percent (FAO/Sida Partnership to positively affect children’s future pro- Cooperation 2010). ductivity and economic prospects. The A number of studies show that gender- effects begin even before childbirth. In based violence also imposes significant Timor-Leste, highly educated mothers costs on the economies of developing coun- and those in the wealthiest households are tries through lower worker productivity more likely to have their babies delivered and incomes, lower human capital invest- by skilled attendants than less educated ments, and weaker accumulation of social mothers and those from poorer house- capital (Morrison, Ellsberg, and Bott 2007). holds (NSD, Ministry of Finance, and ICF In addition to indirect costs, gender-based Macro 2010). Similarly, Demographic and violence has large direct economic costs on Health Survey data show that Cambodian society. A study in the United States found women with little education are relatively that the direct health care costs of intimate less likely to receive prenatal care and assis- partner violence against adult women were tance from trained health personnel during more than $4 billion USD in 1995 (USCDC birth deliveries than women with more 2003). Reducing gender-based violence education (Johnson, Sao, and Hor 2000). A would thus have significant positive effects mother’s health and nutrition status is also on the region’s economies by reducing found to strongly affect children’s physical health care costs and increasing investments health as well as cognitive and noncogni- in women’s human capital, female worker tive abilities, which can have long-lasting productivity, and women’s accumulation of developmental and societal consequences social capital. (Naudeau et al. 2011). As the global economy becomes more Higher labor force participation as well integrated, the productivity effects associ- as income earned and assets held by women ated with greater gender equality are likely have also been shown to have positive to be increasingly important to East Asian effects on the next generation. In Indonesia, and Pacific countries. Recent studies on the for example, women with a higher share of relationship between gender and trade sug- household assets before marriage tend to gest that gender inequalities have become use more prenatal care and are more likely financially detrimental for countries in a to have their births attended by skilled world of open trade (Do, Levchenko, and health care providers (Beegle, Frankenberg, Raddatz 2011). To participate effectively in and Thomas 2001). Similarly, in China, O V E R V I E W   5 increasing adult female income by 10 per- as good leaders as men, studies suggest that cent of the average household income raised capturing these gender-based differences the fraction of surviving girls by 1 percent- in perspective can lead to not only more age point and increased years of schooling representative but also better decision mak- for both boys and girls. In contrast, a simi- ing. Evidence from South Asia suggests that lar increase in male income reduced survival development policy making can benefit from rates and educational attainment for girls greater gender equality in voice. As an exam- with no impact on boys (Qian 2008). Stud- ple, a study of women elected to local govern- ies from across developing and developed ment in India found that female leadership regions (for example, from places as diverse positively affected the provision of public as Brazil, Ghana, South Africa, and the goods at the local level in ways that better United States) show that income in the hands reflected both women’s and men’s preferences of women positively affects their female chil- (Chattopadhyay and Duflo 2004). Similarly, dren’s health (Duflo 2003; Thomas 1995); studies from rural India and Nepal found that commonly, the marginal effects of income when women who were previously excluded and assets in the hands of mothers are larger from decisions about local natural resource than effects of similar income and assets in management gained greater voice and influ- the hands of fathers. ence, local conservation outcomes improved Reductions in gender-based violence significantly (Agarwal 2010a, 2010b). through greater female agency can also have Women’s collective agency can also be important intergenerational benefits. Sev- transformative, both for individuals and eral studies show that experiencing domes- for society as a whole. For example, for a tic violence between parents as a child group of ethnic minority women in rural contributes to a higher risk of both women China, information sharing among them experiencing domestic violence as adults has helped empower them and raise their and of men perpetrating violence against social standing in the Han-majority com- their spouses (Fehringer and Hindin 2009). munities into which they married (Judd In Timor-Leste, 56.4 percent of women 2010). In a more formal setting, over the last who were victims of spousal violence had 15 years, migrant domestic workers in Hong a father who beat their mother (NSD, Min- Kong SAR, China, have been engaged in istry of Finance, and ICF Macro 2010). In civic action focused on local migrant work- Cambodia, women who reported that their ers’ rights as well as international human mothers experienced domestic violence were rights (Constable 2009). These efforts have more likely to experience physical and psy- contributed to the enactment of laws that chological domestic violence as well (NIPH, now provide migrant domestic workers in NIS, and ORC Macro 2006). Efforts that Hong Kong SAR, China, with some of the increase women’s safety and security and most comprehensive legal protections in the that reduce domestic violence can thus lead world. to lower intergenerational transmission of violence within families. Recent progress, pending challenges Strengthening women’s voice can Over the last few decades, most East Asian enhance the quality of development and Pacific countries made considerable decision making progress toward gender equality in several Several studies show that women and men dimensions. In other dimensions, gender have different policy preferences (Edlund disparities have been more persistent. This and Pande 2001; Lott and Kenny 1999). section reviews recent progress and pend- Despite perceptions in some East Asian and ing challenges in the region, noting where Pacific countries that women do not make economic growth and development have 6    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C contributed to advances and where they have sharply declined, and under-five mortality been insufficient. rates have halved since 1990 for both boys and girls. Noteworthy gains have been made in birth attendance by health professionals. Growth and development have been In addition, the East Asia and Pacific region accompanied by reduced gender has seen substantial declines in the mater- inequalities in several dimensions nal mortality rate, from approximately 200 Many gender gaps in education have deaths per 100,000 live births in 1990 to closed. Over the last few decades, boys’ and 100 in 2008 (figure O.4). girls’ schooling outcomes have c ­ onverged at levels that are high by international Gender gaps in labor force par ticipa- standards. East Asia and the Pacific has tion have narrowed. Female labor force performed better than other developing ­ p articipation in East Asian and Pacific regions, in terms of both increasing female ­ countries is high by international standards and male educational enrollments and rais- (figure O.5), and among younger cohorts, ing the female-to-male enrollment ratio. In female labor force participation has tended 2010, the region had the highest primary to rise over time. Moreover, as countries school ratio of female-to-male enrollments grow and develop, women are increasingly among all developing regions; at the sec- moving into jobs in the nonagricultural ondary level, only Latin America and the sector and are migrating to urban areas in Caribbean had a higher female-to-male search of better employment opportunities. enrollment ratio (figure O.3). Trends and patterns of labor force partici- pation look similar to those observed in the Key health outcomes have improved. Dur- United States and other countries of the ing the past half century, the region has Organisation for Economic Co-operation experienced significant advances in sev- and Development (OECD) during their eral health indicators. Fertility rates have economic transformations. In many ways, women in East Asia and the Pacific are better positioned today than ever before to participate in and contribute FIGURE O.3  Girls’ secondary school enrollments have converged to their countries’ development. to those of boys female-to-male ratio of secondary school gross enrollment Despite progress, important 1.1 challenges to promoting gender 1.0 equality remain 0.9 Progress has been uneven across the 0.8 region. Substantial variation remains across ratio countries, both in overall enrollment rates 0.7 and in female-to-male enrollment ratios. 0.6 Countries such as Cambodia, the Lao 0.5 People’s Democratic Republic (Lao PDR), 0 and Papua New Guinea still have relatively 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 low enrollment levels and low female-to- male enrollment ratios, particularly at East Asia and Paci c Europe and Central Asia the secondary school level. Furthermore, Latin America and the Caribbean Middle East and North Africa alt houg h cou nt ries have ex perienced South Asia Sub-Saharan Africa convergence in enrollment among the young, substantial gaps still remain in the educa- Source: WDI database, 2011 data. tional endowments of adult populations. O V E R V I E W   7 Maternal mortality remains high in FIGURE O.4  Maternal mortality rates have declined in most lower-income countries and in several Pacific countries in the region countries (figure O.4). In Lao PDR, for example, maternal mortality rates were still 1,400 modeled estimates of maternal mortality more than 500 deaths per 100,000 births in 1,200 deaths per 100,000 live births 2008, among the highest rates in the world. Indonesia’s maternal mortality rate remains 1,000 high compared to other countries in the 800 region at similar levels of development. 600 Substantial differences in labor force 400 participation occur across countries in the region, even among countries with similar 200 income levels. Relative to their income levels, 0 1990 1995 2000 2005 2008 countries such as China and Vietnam have substantially higher rates of female labor Lao PDR Cambodia Timor-Leste Indonesia force participation than the world average, Papua New Guinea Philippines Vietnam Solomon whereas participation is near the world aver- Islands Mongolia China Malaysia age in countries such as Indonesia and the Philippines, and below average in countries Thailand Fiji such as Fiji and Malaysia (­figure O.5). Source: WDI database, Gender Statistics, 2010 data. Within countries, interactions between gen- der and other socioeconomic characteristics FIGURE O.5  Female labor force participation is high by global can often exacerbate disparities. Economi- standards but also varies substantially across the region cally disadvantaged and minority populations often experience lower educational enroll- female labor force participation rate (ages 15–64) 90 ments, for example. In Vietnam, school par- Lao PDR ticipation among 15- to 17-year-olds is sub- 80 Cambodia stantially higher among the Kinh and Hoa Vietnam Papua New Guinea China Australia (Chinese) majorities than among many of 70 the 52 ethnic minority populations. Among Mongolia Thailand Japan 60 Timor Leste Tonga the more economically disadvantaged and Singapore percent Indonesia less well integrated Hmong, Dao, and Khmer Philippines Korea, Rep. 50 minorities, far fewer girls attend school than boys (Baulch et al. 2002). 5 Geographic dis- 40 Malaysia tance, or remoteness, can also serve to com- Fiji pound gender disadvantage. Women in remote 30 rural areas commonly have limited access to 0 health care, significantly raising the risks asso- 0 5 6 7 8 9 10 11 12 13 ciated with pregnancy and childbirth. While log GDP per capita (2005 PPP dollars) Vietnam has experienced noteworthy declines global relationship, 2009 in maternal mortality, on average, over the last decade, progress has been much slower Sources: World Bank staff estimates using Key Indicators of Labour Market (KILM) labor force data in remote and ethnic minority regions (World (International Labour Organization) and purchasing power parity adjusted GDP per capita (in logs and at 2005 prices) from the Penn World Tables. Bank 2011b). Note: GDP = gross domestic product, PPP = purchasing power parity. The data shown for each country are from 2009 and the estimated U-shaped relationship uses data from across the world. Some gender disparities fail to close— the most concerning issues is that despite or close very slowly—with development growth and development, the problem More than a million girls and women per of missing girls remains significant. The year are “missing� in East Asia. Among term “missing women� was first coined 8    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE O.6  Across the region, female-headed households own to what is observed in developed coun- less land than male-headed households tries. Sen argued that this imbalance in sex ratios reflected severe forms of gender bias hectares of land owned, by sex of household head in affected societies. Biological differences 2.5 between males and females imply that approximately 105 boys are born for every 2.0 100 girls. Nonetheless, China, Vietnam, and until recently, the Republic of Korea 1.5 have experienced substantial deviations hectares from the biological norm, and the trend 1.0 over time, particularly in China, has been alarming. In China, the number of girls 0.5 who are missing per year at birth increased from 890,000 in 1990 to 1,092,000 in 0.0 2008. Missing girls as a fraction of the total number of female births increased from 20 lia, 20 ste, 20 m, 20 dia, 20 DR, 19 sia, na go 08 07 06 08 08 99 ne oP e bo -L 8.6 percent in 1990 to 13.3 percent in 2008 et on do m La or Vi M Ca In Tim (World Bank 2011c). female-headed households male-female gap Source: World Bank staff estimates using household income and expenditure surveys. Gender disparities still exist in access to and control of productive resources. Gender disparities in access to and control of land FIGURE O.7  Women in East Asia and the Pacific still earn less and farm inputs are pervasive in the region than men despite growth and development. Women remain less likely to own land than men. ratio of female-to-male hourly wages, by sector And when women, or specifically, female- Cambodia, headed households, do own land, they 2008 typically have smaller holdings (figure O.6).6 ­ Indonesia, Female-headed households also tend to have 2009 poorer access to other productive inputs and Lao PDR, 2008 support services, including livestock hold- Philippines, ings and access to agricultural extension 2006 services. Thailand, 2009 Despite high labor force participation, Timor-Leste, important gender inequalities in economic 2007 opportunity remain. Women still earn less Vietnam, 2006 than men in nearly all sectors in all coun- 0.0 0.2 0.4 0.6 0.8 1.0 1.2 tries in the region (figure O.7). Gender ratio wage gaps increase with age, reflecting in part lower levels of experience among government industry women caused by workforce interrup- services agriculture tions and reduced working hours during childbearing years, as well as gender dis- Source: World Bank staff estimates using household income and expenditure surveys. parities in education among older cohorts. Gender wage gaps in the region are also by Amartya Sen (1999) to refer to the phe- st rong ly i n f luenc ed by o c c upat ion a l nomenon that many low-income countries and sectoral segregation, mirroring pat- have far fewer women than men relative terns seen in the United States and other O V E R V I E W   9 OECD countries. Together, differences in FIGURE O.8  In urban China and Indonesia, gender wage gaps are education endowments, experience, and largest among low wage earners industrial and occupational segregation explain up to 30 percent of observed gen- percentage difference between male and female hourly wages 50 der wage disparities in East Asian countries (Sakellariou 2011). 45 Gender wage gaps in the region are often 40 greatest among men and women with rela- 35 tively low education and skill levels. Several 30 percent studies from East Asian and Pacific coun- 25 tries point toward “sticky floors,� that is, 20 wider wage gaps at the bottom than at the 15 top of the earnings distribution (figure O.8). 10 The finding of sticky floors contrasts with studies from OECD countries, which more 5 commonly find “glass ceilings,� that is, 0 larger wage gaps among higher-earning men 1st 2nd 3rd 4th 5th 6th 7th 8th 9th and women. income decile Women are more likely to work in small China, 2004 Indonesia, 2009 firms, to work in the informal sector, and to be concentrated in lower-paid occupa- Sources: Chi and Li 2007; Sakellariou 2011. tions and sectors. Within firms, women are more likely than men to be temporary workers. Such employment segregation FIGURE O.9  Enterprises with female managers tend to be smaller affects a number of economic outcomes by gender, including earnings, returns to percentage of small, medium, and large rms education and experience, social security 40 with a female manager coverage, and exposure to shocks. Sub- 35 stantial gender-based occupational and sectoral segregation is seen in all countries 30 and does not decline with development. 25 In fact, employment segregation tends to percent increase as economies become more diverse 20 with development. Economic growth and, 15 in particular, urbanization appear to make occupational and sectoral segregation by 10 gender more pronounced, particularly dur- 5 ing the early stages of economic structural 0 change. m s ia ne es na Similarly, female-led enterprises tend pi n et do ilip Vi In Ph to be smaller and more precarious than small rm medium rm large rm m ale-led enterprises (figure O.9) . The ­ micro, small, and medium firm sectors are Source: World Bank staff estimates using Enterprise Surveys database for 2006–11. important segments of most East Asian and Pacific economies and contribute a substan- tial fraction of GDP. Female-led enterprises across the region, particularly in the infor- have fewer employees and assets, and are mal sector, have lower profits, are less likely more likely to be home based or to operate than m ­ ale-led enterprises to be registered, out of nonpermanent premises. Although 1 0    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C female-owned and -managed enterprises Solomon Islands and 69 percent of part- are not inherently less productive, they tend nered women in Vanuatu report that they to be smaller, less capitalized, and located have experienced some sort of controlling in less remunerative sectors. behavior by their partners. This includes preventing them from seeing family, wanting Women in the region still have less voice and to know where they are at all times, forbid- influence than men. Women’s household ding contact with other men, and control- decision-making power in East Asia and the ling their access to health care (SPC and Pacific is relatively high, but levels of auton- NSO 2009; VWC 2011). omy vary across the region. Women’s auton- Women’s voice and inf luence in the omy in the household can be measured in public domain—as measured by repre- several ways, including control of assets, sentation in national and local political freedom of physical mobility, and voice in assemblies—remains low. The share of decision making. By several of these mea- female parliamentarians in East Asian and sures, including control over large house- Pacific countries is just below the global hold purchases and visiting family and average, at approximately 18 percent in relatives, women in East Asian countries 2011. Despite economic growth and devel- appear to have relatively high autonomy opment in the region, this figure has barely compared with women in other develop- changed since 1990. Although the share of ing regions (World Bank 2011c). Women women in national assemblies varies con- in the Pacific have relatively less control siderably across the region, it is systemati- over their own earnings, however. Over 15 cally lower in the Pacific than in East Asia percent of women in the ­ M arshall Islands, (figure O.11). Indeed, in no country in 15 percent in Samoa, and 13 percent in the Pacific does the share of parliamen- Tuvalu report that their husbands control tarians who are female exceed 10 per- their cash earnings ­ (figure O.10). Moreover, cent, and four countries—the Federated 58 percent of partnered women in the States of Micronesia, Nauru, Palau, and the Solomon Islands — have no female parliamentarians. FIGURE O.10  Who decides how wives’ cash earnings are used The prevalence of gender-based violence varies widely across the region is high in the region, and particularly so in the Pacific where the prevalence of domestic percentage of wives, husbands, and couples deciding how wives’ violence is among the highest in the world. cash earnings will be used As can be seen in figure O.12, 68 percent 100 of ever-married women 15– 49 years of 90 age in Kiribati, 64 percent in the Solomon 80 Islands, and 60 percent in Vanuatu have 70 experienced physical or sexual violence 60 at the hands of an intimate partner (SPC, percent 50 Ministry of Internal and Social Affairs, 40 and Statistics Division 2010; SPC and NSO 30 2009; VWC 2011). Although no nationally 20 10 representative data exist for Papua New 0 Guinea, studies conducted at the subna- tional level suggest that domestic violence oa ia lu a s te s nd ne di es va es m bo is just as prevalent (Ganster-Breidler 2010; on pi Isa Tu -L Sa ilip m or d all In Ca Tim Lewis, Maruia, and Walker 2008). This Ph sh ar M mainly wife jointly mainly husband violence is a linchpin to a bigger story; vio- lence against women represents the extreme Sources: Demographic and Health Surveys, various years. deprivation of voice and freedom among O V E R V I E W   1 1 FIGURE O.11  Women’s representation in parliament is low, especially in the Pacific percentage of seats held by women in the lower or single house, 2011 45 40 35 30 25 percent 20 15 10 5 0 . P rea an G s  rsh an ea T ar m b ia Sin Chi a ilip re La nam P a o mo P r u a N Isla u ng a Sa olia Tu a Kir alu lay i em Ko Japsia ’s . on d ga n a Vie ines rd r-L R co este wo es rld N a ts. an s Is u Th Rep. ple ep Ma bat M y l and mo i Mo ong pu n ala No Timo o PD all uat Ind ilan e w nd od Ca es m P h po u tri S Ma V uin v eo , R i d. p un t a Fe, sia ic ne l cro So Mi ,D rea Ko Source: PARLINE database (Inter-Parliamentary Union). women and, as such, is often associated FIGURE O.12  Violence against women is high in the region with a lack of agency in other dimensions. percentage of women who have experienced Awareness is increasing that men and boys— sexual or physical intimate-partner violence 80 not just women and girls—face gender- 70 specific risks. Some countries in the region 60 have started to experience a reverse gender 50 percent gap in education; girls’ secondary school 40 enrollment now exceeds that of boys in 30 China, Fiji, Malaysia, Mongolia, the Philip- 20 pines, Samoa, and Thailand. Reverse gender 10 gaps at the tertiary level are sometimes even 0 starker: in Thailand, 122.4 females were nd Ph ste s Ca an a on ati Va s u oa sh valu ds ne d di at an an m p rib a e bo nu pi enrolled for every 100 males in 2008. In ail Ja -L Tu Sa Isl Isl Ki ilip m or Th all Tim addition, men across the region experience m ar lo So M higher levels of morbidity and premature mortality related to substance abuse. The East Asia Paci c prevalence of smoking and drinking among males in East Asian and Pacific countries Sources: Demographic and Health Surveys, various years, and government surveys. Note: Data for Thailand are for Bangkok and Nakhonsawan only, and data for Japan are for is much higher than the prevalence among Yokohama only. females. 1 2    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C Why do many gender whereas females are expected to submit inequalities persist? to their husbands and be caregivers and homemakers. Low household incomes, weak service Weak systems of service delivery also delivery, and traditional norms can constrain progress in education—overall as impede gender equality in education well as for girls. In Cambodia, Lao PDR, and health and Papua New Guinea, for example, school Where gender gaps in education are still enrollments are low overall, and gender gaps observed, low income coupled with high persist. Low enrollment rates in Papua New costs of education can limit household Guinea also reflect limited physical access demand for schooling. Traditional gender to schools and high dropout rates. Poor and norms and practices also strongly influ- sparse school infrastructure, poor teacher ence household schooling decisions. Par- attitudes and attendance, lack of teachers ticipants of focus group discussions in a in remote areas, and negative pupil behav- qualitative research exercise in Papua New ior all contribute to low overall enrollments. Guinea report, for example, that parents Long distances to schools have been observed value boys’ education over girls’ education. to make school attendance costly in both The reason is that males will carry the C ambodia and Lao PDR, particularly for ­ family name and become household heads ­ istances raise safety girls, because long travel d FIGURE O.13  Women are concentrated in certain fields of study, such as education and medicine, but are underrepresented in law and engineering 90 percentage of students who are women, by eld of study 80 70 60 50 percent 40 30 20 10 0 n lth law g n h y nc d og tio in tio at cie an a es er m ol he a ra l s ts, uc ne hn d ist cia ar an ed gi in ec so es, en ce dm lt iti ien ra sa an tu sc es ul m sin ric hu ag bu Indonesia, 2009 Thailand, 2006 Vietnam, 2006 Source: Sakellariou 2011. O V E R V I E W   1 3 concerns among parents. The lack of toilets at deliveries in the forest, including beliefs that many schools makes attendance more difficult women do not need prenatal care or delivery for girls than boys. supported by skilled attendants. These factors Gender streaming in education largely take a heavy toll on women during pregnancy reflects societal norms and expectations and pose higher risks of mortality related to and has implications for gender inequali- birth complications. ties in job placement and earnings. Substan- Strong son preference, intensified by tial differences remain in the composition declining fertility and the availability of pre- of education between men and women in natal sex-identification technology, underlies the region (figure O.13). Economic returns the observed skewed sex ratios at birth in a or comparative advantage of females and few East Asian countries (China, Vietnam, males in different fields of study do not and to a lesser extent now, Korea). Parents’ appear to explain education streaming. choices to keep and care for boys over girls Social norms about appropriate work for can depend on social norms and values, dif- women and men, role models in the labor ferent economic opportunities by gender, market, and gender stereotyping in school and the benefits parents expect from a son curricula play important roles. In East Asia compared to a daughter, including material and the Pacific, teaching materials more support in old age. Although many societies frequently portray males than females in have some mild degree of preference for sons, active and leadership roles. Women are the interplay of culture, state, and political often depicted as secretaries, assistants, processes can generate extreme patrilineality nurses, and teachers whereas men are por- and highly skewed child sex ratios. In addi- trayed as doctors, politicians, or police tion, the manifestation of son preference is officers. Gender streaming in education ulti- influenced by public policies—for example, mately affects the type of work that women China’s one-child policy and Vietnam’s two- and men do and, importantly, affects their child policy—and the spread of prenatal respective abilities to take advantage of exist- sex-determination technology since the early ing and emerging economic opportunities. 1980s. For these reasons, economic growth Poor service delivery and cultural norms and development alone do not necessarily about birthing practices contribute in large reduce son preference and sex ratios at birth part to high maternal mortality in sev- in East Asia. eral East Asian and Pacific countries. Poor Gender norms about masculinity play access to quality obstetric health services, a strong role in influencing the excessive particularly in remote rural areas, places tobacco and alcohol consumption observed women at higher risk for maternal death. among men in many parts of the region. Rural areas tend to be less well served Smoking and drinking are commonly viewed by the health system, and rural residents as masculine behaviors. Men and boys feel have much lower access to birth deliver- substantial pressure to accept gender stereo- ies attended by trained staff than do urban types that they should be strong and tough. residents. Evidence shows that poor health In contrast, social disapproval of women infrastructure and long distances to the who smoke keeps the smoking prevalence nearest health center are both important among women very low in Vietnam, for barriers to reducing maternal mortality in example. Survey data indicate that the pri- Cambodia and Lao PDR. In addition, culture mary reason why most Vietnamese women and tradition play an important role in the do not use tobacco is the expectation that choice of health practices, such as the location “women shouldn’t smoke.� Survey respon- of childbirth, the use of skilled birth atten- dents consider this factor more influential dants, and sterilization practices. Preferences in affecting smoking behaviors than health can vary from birth deliveries at home to concerns. 1 4    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C Norms regarding women’s household women’s greater need for workplace flex- roles and disparities in productive ibility to facilitate management of their dual resources constrain economic household and market roles. opportunity Female-headed households in the region tend to have less access to land because of Gender norms related to the allocation of the interaction of complex legal, social, and time to household work affect women’s economic factors. In the majority of coun- opportunities in the labor market because tries in East Asia, statutory law does not they are expected to take primary responsi- differentiate property inheritance by gender. bility for home and family in addition to any However, parallel statutory and customary market role. Responsibility in the household legal ­systems in a number of East Asian and fundamentally affects all outcomes in the Pacific Island countries mean that women market sphere—from where women work are not treated equally to men in the imple- and what they do to how much they earn. mentation of the law. Gender inequalities Women work longer total hours than men persist also in access to other productive and devote significantly more time to domestic inputs and support services—from livestock and caregiving activities (figure O.14), par- holdings to agricultural extension services. ticularly in households with small children. Although evidence on access to credit is Many women temporarily leave the labor mixed across the region, female entrepre- force when they must care for small children neurs in several countries, including Timor- or the elderly. Trade-offs between household Leste and Tonga, report greater difficulty and market work can be particularly stark than their male counterparts in accessing in rural areas, where women can spend credit. Such disparities in access to produc- long hours on domestic chores because of tive resources continue to impede gender poor infrastructure and a lack of alterna- equality in access to economic opportunity. tive childcare options. Indeed, differences in A substantial share of the gaps in pro- the types of work that women and men do, ductivity and profits between female- and along with higher rates of female presence male-led firms can be accounted for by in the informal sector, are in part caused by ­ gender-based segregation of enterprises by sector, firm size, and firm characteristics. FIGURE O.14  In Lao PDR, women—particularly those with young This ­“sorting� of firms is found among both children—must balance household and market work formal and informal enterprises and reflects both gender norms regarding time alloca- tion to household and market work and 10 hours spent per day on household and market work differential access to productive inputs. In 9 Indonesia, for example, the food, retail, and 8 garment manufacturing sectors—where 7 female entrepreneurs are most likely to 6 locate—are among the least capital-intensive hours 5 and productive sectors (figure O.15). By con- 4 trast, the transport sector—where male entre- 3 preneurs are most likely to locate—has higher 2 productivity and capital intensity. 1 0 Broader constraints to business devel- without a child with a child without a child with a child opment, such as cumbersome registration men women procedures, affect both female- and male-led enterprises (figure O.16). The most impor- tant issues vary by country, but, within any market work housework and family care given country, both male and female entre- preneurs often identify similar challenges— Source: World Bank staff estimates using Lao Socio-economic Survey, 2008. competition, difficulty in accessing finance, O V E R V I E W   1 5 and lack of electricity—and in comparable FIGURE O.15  In Indonesia, female-led enterprises are clustered in magnitudes. Evidence suggests that such con- lower-productivity and less capital-intensive industries straints may be more onerous among small and informal firms than among larger firms, productivity and assets per worker, by sector however, so to the extent that female-led firms 50 are smaller and more likely to be informal, 45 value added (rupiah, thousands) they are likely to be more adversely affected. 40 35 Gender inequalities in endowments and 30 economic opportunity limit women’s 25 agency… 20 15 A woman’s agency is affected in fundamen- 10 tal ways by her endowments and access to 5 economic opportunities. Gender inequalities 0 in educational attainment, economic assets, g es rt od il and own earnings can hinder women’s abili- rin po ta vic fo re tu ns er ac ties to influence their circumstances in the tra rs uf he an ot home, to enter and participate effectively in m politics, or to leave bad or dangerous house- productivity per worker assets per worker hold situations. In Indonesia, for example, women with little or no education are less Source: World Bank staff estimates using Indonesia Family Life Survey 2007/2008. likely to participate in decisions involving Note: The graph shows productivity and assets per worker in five industries for firms with fewer than their own health care, to make household five workers. Productivity is measured by value added. purchases, or to engage regularly in social FIGURE O.16  Male- and female-led firms report similar constraints activities than women with at least a second- in Indonesia ary education (BPS and ORC Macro 2003). In China and Cambodia, women with less education are less likely to enter politics—for 35 percentage of rms reporting constraints reasons of norms or statute—than women 30 with higher levels of education (Maffii and Hong 2010; Wang and Dai 2010). Moreover, 25 worldwide evidence suggests that a woman’s 20 percent ownership and control of her own assets and income is associated with a decreased risk of 15 intimate-partner violence (Agarwal and 10 Panda 2007; ICRW 2006; Pronyk et al. 2006; Swaminathan, Walker, and Rugadya 5 2008). Women’s income can also positively 0 affect their accumulation of assets, which in capital marketing raw transport energy paying turn positively affects their ability to leave materials wages an abusive partner, to cope with shocks, male female and to invest and expand their earnings and economic opportunities (World Bank Source: World Bank staff estimates using Survey of Cottage and Small-Scale Firms (SUSI) 2002. 2011c). society. As previously noted, traditional … as do traditional norms regarding norms about women’s roles within the women’s roles … home constrain their economic opportuni- Social norms and practices can limit wom- ties and thus their decision-making power en’s voice and influence in the home or in within the home. Traditional gender norms 1 6    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE O.17  Men and, in some cases, women believe that men region, the legal environment is affected by make better political leaders than women not only statutory but also customary law. Plural legal environments, where both percentage of men and women who think men make better statutory and customary laws are prac- 80 political leaders ticed, can create important challenges 70 to promoting gender equality in voice and i ­nfluence. Statutory laws, customary percentage of men and 60 women over age 30 50 (and sometimes religious) laws and prac- tices can affect women’s voice and influence 40 in different ways when they bestow differ- 30 ent rights by gender. Moreover, in prac- 20 tice, the interaction between statute and 10 custom can mean that women’s legal status 0 varies substantially across ethnic (and reli- gious) groups, even within a single country. a sia ia d at d ay m in an St ite ys na rw ne Ch This interaction can affect women’s rights in es ala ail Un et No do Th Vi M In male female ­ marriage and divorce, reproductive health, education, asset ownership, inheritance, and Source: World Values Survey database, 2005–2009 data. Note: Data for Indonesia, Malaysia, the United States, and Vietnam are for 2006; data for China and freedom of mobility, among other things, Thailand are for 2007; data for Norway are for 2008. which in turn can fundamentally affect the extent of women’s agency. and social expectations also shape people’s In some countries, inadequate legal protec- views about women’s roles in the public tion, weak implementation and enforcement, sphere. Surveys conducted in several East and social tolerance enable gender-based vio- Asian countries indicate, for example, that lence. Although more than three-quarters a majority of men—and sometimes a major- of countries in East Asia have strengthened ity of women—think that men make better legislation on domestic abuse in recent years political leaders than women (figure O.17). (including, for example, Indonesia, Korea, Lao Similarly, in parts of rural China, many PDR, Thailand, and Vietnam), in the Pacific, people still think of women as less capable more than 60 percent of countries still lack suf- (disuzhi), and local norms dictate that they ficient legislation on domestic violence (UNDP should confine their activities to the domes- 2010). Even when countries have appropriate tic settings (Wang and Dai 2010). legislation in place, women remain unpro- tected by the legal system because the laws remain largely unenforced. A recent study … and complex legal environments found, for example, that officers in the Fiji and, often, weak access to justice Police Force Sexual Offences Unit often have The legal setting, along with people’s access unwelcoming attitudes when dealing with to justice, establishes the underlying environ- female victims (UNFPA 2008). The same is ment in which women (and men) can exercise true in some areas in Cambodia, where many agency in the home and in society. Whether local officials still believe that a husband can women and men are equally supported under threaten his wife despite the laws in place the law, and whether their rights are pro- (UNDP Cambodia and VBNK 2010). tected in practice thus critically affect their voice and influence in society. Laws and access to justice also create the environment Emerging opportunities in which women and men can (or cannot) and risks in an increasingly access resources and economic opportunity integrated world and accumulate assets, which also affects Several emerging trends in the region will their agency. In several countries in the present both new opportunities and new O V E R V I E W   1 7 risks to achieving gender equality. East FIGURE O.18  The share of female workers in export-oriented firms Asia and the Pacific are at the forefront is relatively high of several global trends: increasing global economic integration, rising availability percentage of workers who are female in a rm, and use of ICTs, increased domestic and 70 by export status of rm cross-border migration, rapid urbanization, 60 and population aging. Because these trends have gender dimensions, they will affect the 50 evolution of gender equality in the region. 40 percent In many ways, these trends will bring with them new opportunities for gender equality. 30 For example, increased economic integra- 20 tion, greater access to ICTs, and increased migration will likely all contribute to 10 increased income earning opportunities 0 for women. Along with new opportunities, Cambodia China Indonesia Thailand however, these emerging trends will bring exporter non-exporter new risks. Increasing global integration will likely Source: World Bank staff estimates using Enterprise Surveys database for 2002–06. continue to be an important source of Note: Share of female workers = female workers/total workers. ­ n onagricultural employment growth for women, who are already highly represented in export-oriented sectors (figure O.18). And network and share information about start- greater employment and earnings in export- ing and running a business. Limited evidence orientated industries can contribute to greater suggests that women in the region may still female independence and autonomy in deci- have lower access to information technolo- sion making. At the same time, increasing gies than men, however. Although data from global integration can increase economic risk China show rapid growth in access to ICT and uncertainty, as shocks are quickly trans- services, Chinese women are still less likely mitted across integrated markets. A number to use the Internet or to subscribe to mobile of studies find that while shocks do not neces- phone services (figure O.19). Widening gen- sarily have more adverse impacts on women der gaps in access to ICTs raise the risk of than men, they do have gender-differentiated rising disparities in economic opportunity effects on outcomes as diverse as employment, and voice going forward. earnings, labor force participation, educa- High economic growth and increased tion, health, and nutrition (see, for exam- economic integration over the past three ple, Bruni et al. 2011; Rodgers and Menon decades have spurred significant migra- forthcoming). tion across the region. Women consti- Advances in ICTs are opening up oppor- tute nearly half of all migrants in East tunities for both men and women throughout Asia and the Pacific and are increasingly the region. New and emerging technologies, if migrating in search of better economic accessible, can help increase women’s w­ elfare opportunity. Female migrants dominate a through a number of channels by opening number of occupations and sectors, includ- new economic opportunities, empowering ing labor-intensive manufacturing and women by breaking down information barri- export-oriented industries, and domestic ers, facilitating engagement of women in iso- work. Migration can provide women with lated communities in distance learning, and increased economic opportunity, give them enabling them to take collective action. In the chance to improve their knowledge and Malaysia, for example, female entrepreneurs skills, and increase their agency through have created self-help cyber communities to raising their contributions to family income. 1 8    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE O.19  China has seen remarkable growth in Internet use compensate for the loss of extended family since 2000, but women’s use trails men’s networks. Finally, the high-income economies in 90 internet use in China, by gender East Asia are experiencing rapid population 80 aging. Most emerging countries in the region number of  Internet users, millions have also begun this process; dependency 70 ratios are already increasing in many middle- 60 income countries in East Asia and the Pacific. 50 Old-age dependency is expected to increase 40 even more quickly in the coming decades 30 (figure O.20). Population aging is likely to have gender-differentiated effects at all age 20 levels. Gender differences in time devoted 10 to caring for the elderly imply that in the 0 absence of institutionalized care services, 2000 2001 2002 2003 2004 2005 2006 women are likely to bear the brunt of the male female increased demand for elder care (Dwyer and Coward 1992; Ofstedal, Knodel, and Chayovan Source: CNNIC (China Internet Network Information Center) Internet Statistics. 1999). In addition, while women tend to live longer than men, gender differences in edu- cation and labor force participation imply At the same time, migration brings with it that women are less likely to be vested in important gender-specific risks. For exam- formal pension systems and may have fewer ple, many female migrants work as domes- assets to ensure a basic level of well-being in tic workers, an occupation with particularly old age. weak worker protections in most countries. At present, these emerging trends have Female migrants are also disproportionately gender dimensions that are not commonly susceptible to human trafficking. accounted for by policy makers. Nonethe- Many East Asian and Pacific countries are less, these trends will generate a distinct set experiencing unprecedented levels of urban- of challenges for promoting gender equality ization as migrants move to urban areas in going forward. An important role for public search of economic opportunity. Between policy, therefore, will be to support women 2000 and 2015, Indonesia, China, and (and men) in taking advantage of emerging Cambodia are predicted to see an increase ­ opportunities while protecting them against of the population residing in urban areas by the emerging risks. 17, 13, and 9 percentage points, respectively (UN 2010). Urbanization affects all aspects of life, from the nature of employment to Toward gender equality in East the availability of services to one’s ability Asia and the Pacific: Directions to rely on extended family and community for policy networks for support. These changes almost The collection of evidence points to four certainly have gender-specific impacts. While priority areas where public policy can con- urban areas can open up a wider range of tribute to greater gender equality and more economic opportunities for both men and effective development in East Asian and women, women’s ability to take advantage of Pacific countries: new opportunities is likely to depend more fundamentally on the nature and availabil- First, promoting gender equality in human •   ity of urban services—for example, whether development remains a priority where gen- transportation systems facilitate their safe der gaps in education are large or health travel to job sites or affordable child care can outcomes are poor; closing gaps in human O V E R V I E W   1 9 development, where they persist, is likely FIGURE O.20  The old-age dependency ratio is increasing in most to yield high returns. East Asian countries S econd, taking active measures to close •   gender gaps in economic opportunity 0.45 old-age dependency ratio is often warranted on both equity and 0.40 efficiency grounds. Which policy levers 0.35 will yield the highest returns depends on 0.30 the structure of the country’s economy and which specific constraints are most 0.25 ratio binding. 0.20 •   T hird, taking measures to strengthen 0.15 women’s agency—and to protect them 0.10 from violence—is also called for across 0.05 the region; strengthening women’s voice 0.00 1990 2000 2010 2020 2030 2040 2050 and influence will contribute to the qual- ity of development decision making and China Fiji Indonesia thus to development more broadly. Malaysia Mongolia Papua New Guinea Fourth, public policy has a critical role in •   Philippines Thailand Vietnam fostering new opportunities and managing emerging risks associated with increas- ing global economic integration, the rising Sources: Data for 1990–2010: World Bank 2010; data for 2020–50: HNPStats Population Projections database. role of ICTs, increasing migration, rapid Note: The old age dependency ratio is defined as the ratio of the elderly (ages 65 and above) to the urbanization, and population aging. working age population (ages 15–64). The following sections examine policy vary by country context, but the evidence approaches to promoting gender equality in makes clear that both demand-side and East Asia and the Pacific in these four prior- supply-side factors are responsible for these ity areas, drawing on recent experience from poor human capital outcomes. the region and beyond. Policies can thus have an impact by improv- ing service delivery (for example, through infrastructure, staffing, incentives, and use of Promoting gender equality in human ICTs) and implementing demand-side inter- development ventions (for instance, through cash transfers to poor households, ­ information campaigns, Closing persistent gender gaps in human and improved accountability). For example, development Indonesia’s school construction program in In countries with unequal gender out- the 1970s significantly increased education comes in education and health, the priority attainment and future earnings (Duflo 2000). remains improving these outcomes. In East In Cambodia, a scholarship program targeted Asia and the Pacific, gender gaps in human at girls and a related program targeted at boys development at the national level tend to and girls from low-income households led to persist where overall outcomes are low. an increase in school enrollment of at least In such cases, public action to strengthen 20 percentage points (Filmer and Schady 2008, countries’ education and health systems will 2009). Evidence indicates that to reduce be called for to improve gender (as well as maternal mortality, interventions that ensure overall) outcomes. For countries with more basic infrastructure and improve account- localized gender disparities, for example, ability for service delivery are important. among specific ethnic groups or in remote, Approaches to providing services that take rural regions, more targeted interventions into account traditional norms and practices may be warranted. The exact constraints also show promise. For example, Malaysia 2 0    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C has adopted programs that provide guidance expectant parents to bypass the law, and that traditional birth attendants on hygiene enforcement of such laws is difficult. Existing practices, diagnosis of complicated cases, and evidence suggests a more promising approach information on the importance of prenatal is to adopt policies that aim to enhance fam- care. ily perceptions of the value of daughters. While general policies to promote economic Reducing gender streaming in education development may play a role, Korea’s recent For East Asian and Pacific countries where experience suggests that introducing inter- gender equality in access to human capital ventions to influence norms and facilitate is no longer the dominant concern, address- the spread of new social values may also be ing education quality—specifically, gender important, rather than relying on efforts to streaming in education—will have high raise female education and labor force partic- returns. Although concerted efforts in both ipation alone. Information campaigns, finan- education and the labor market will be cial incentives, and improved social security needed to break gender “silos� in education for the elderly can all contribute to changing and, consequently, in the economy, signifi- societal preferences and behaviors. China has cant steps can be taken within the educa- been adopting several of these types of pro- tion system. One important step involves grams. For example, the National Population reform of school curricula to address the and Family Planning Commission scaled up transmission of gender stereotypes through the Chaohu pilot through the national Care the education system. Cambodia, Lao PDR, for Girls campaign in 24 counties with severe and Vietnam have reviewed curricula and gender imbalance. This campaign went revised learning materials or encouraged beyond advocacy and media publicity alone; better practice without gender stereotyp- direct financial incentives for parents to raise ing. Other active interventions may be war- daughters have also been introduced. Prelim- ranted, including information campaigns, inary evidence suggests that these programs financial or nonfinancial incentives, and have had some impact on reducing imbal- efforts to create gender-friendly environ- ances in the sex ratio at birth. ments in higher education. Providing infor- mation on wages and career paths in these Addressing male-specific gender issues programs before individuals choose their Paying attention to male- as well as female- courses may help reduce gender streaming in specific gender issues is appropriate for training. In the United States, for example, reasons of basic welfare as well as for devel- the Science Connections program offered opment effectiveness. In this context, initial monthly science workshops for girls plus signs of reverse gender gaps in education a summer science weekend for families to in several countries should be monitored increase girls’ knowledge about and interest closely. While the long-term implications of in nontraditional careers in science. Schol- male disadvantage in education are still to arships that support women (and men) in be understood, depending on the underly- entering nontraditional fields may provide ing causes, it could have both economic and another avenue for breaking down gender social consequences. Moreover, excessive silos in education. tobacco and alcohol consumption among males in many parts of the region deserves Promoting balanced sex ratios at birth policy attention; the social costs are usu- In the few countries with “missing girls� at ally higher than private costs because of the birth, rooted in the prevalence of son prefer- negative effects of these behaviors on other ence, active measures are needed to address members of the society. Possible measures to the issue. Even where laws against sex-selective tackle this challenge include providing infor- abortion have been enacted, strong incen- mation about the health risks of excessive tives to select the preferred gender still induce tobacco and alcohol consumption, enacting O V E R V I E W   2 1 or increasing taxes on tobacco and alcohol, facilitating a more equitable division of child- imposing regulatory measures on adver- rearing responsibilities and allowing women tising, and restricting smoking in public to have the same opportunities as men for sites. The Thai Health Promotion Foun- advancing their careers in the formal sector. dation, for example, uses alcohol excise tax Within the region, only Cambodia, Indone- revenues to support the operation of an alco- sia, and the Philippines currently have provi- hol control center and a research center on sions for paternity leave. While the principles alcohol consumption, to support advertising behind paternity (and parental) leave are campaigns to reduce alcohol-related traffic appealing, evidence from the OECD on the accidents, and to promote abstinence and take-up of paternity leave is mixed, suggest- increase knowledge about the links between ing that providing paternity leave alone is not alcohol use and domestic violence. sufficient to change the current gender divi- sion of child-rearing responsibilities within households; rather, such leave policies need to Taking active measures to close gender be combined with other approaches to break- gaps in economic opportunity ing down gender norms regarding household Mitigating trade-offs between women’s caregiving. household and market roles Women often face stark time trade-offs Breaking down gender silos in the labor between household and market work, par- market ticularly in rural areas. In such contexts, A key element of breaking down gender programs targeted at reducing women’s time silos in the labor market involves support- on household work—for example, through ing young women and men to invest in skills investment in infrastructure—are likely to on the basis of their productivity rather than increase women’s ability to engage in mar- on the basis of gender norms and percep- ket-based income-earning opportunities. In tions regarding “appropriate� occupations. Lao PDR, for example, evidence indicates Beyond efforts to reduce gender streaming in that having access to electricity extends the education, programs that help both women hours available for both productive and lei- and men understand employment options sure activities, particularly for women and outside of gender silos will likely improve the girls (World Bank 2011a). allocation of talent toward jobs in ways that Policies that support women in balancing improve both equality of economic opportu- their caregiving and market roles are also nity and productivity. In Kenya, for exam- important in strengthening their access to ple, a micro and small enterprise voucher economic opportunity. Access to affordable program, called Jua Kali, provided its female and accessible child care can be critical in beneficiaries information about wages in a this regard. Community child care centers, range of occupations. Preliminary evalua- particularly those targeted at low-income tion of the program suggests that 5 percent neighborhoods, have been found to increase of women who received the information maternal employment in a number of Latin switched to more lucrative (often “male�) American countries. The importance of jobs as compared to those who did not affordable child care, particularly as urban receive the information (Hicks et al. 2011). areas expand, can be seen from recent Breaking down social norms and percep- experience in the region. In Mongolia and tions about gender roles in the workplace is China, reductions in subsidized child care in an area where the public sector can lead by the 1990s and 2000s have significantly and example, particularly with respect to enabling negatively affected female labor force par- women as leaders and managers. The pub- ticipation in urban areas. lic sector is in a unique position to establish Parental and paternity leave can pro- good practice in this regard by encouraging mote greater parity between the sexes by women’s professional advancement, either 2 2    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C through direct measures such as targets or G ender inequalities in access to informa- •   quotas or through specialized training pro- tion and training, extension services, and grams. In this context, the government of other productive inputs constrain the pro- Malaysia has put in place a system of quotas ductivity of female-led enterprises, both for female managers in the public sector. In within and outside agriculture. In Papua Mexico, the government initiated a system New Guinea, where training and exten- of grants to firms to address gender-related sion services are more likely to reach male employment issues in the workplace, includ- than female farmers, a recently launched ing fostering greater female participation in agriculture training program includes management. several components focused on clos- ing the gap between rural women’s eco- Eliminating resource constraints on female- nomic needs and the inputs and services led farms and enterprises provided. Similarly, in Cambodia, the Despite progress, women continue to have Cambodia-Australia Agricultural Exten- less access to a range of productive resources sion Project has addressed two key con- than do men as a function of their gender straints that have limited female farmers’ rather than because of their innate produc- access to agricultural support: the project tive capabilities. Public policies thus have an has increased by 80 percent the number important role to play in promoting gender of female extension workers and under- equality in the control of productive inputs— takes special initiatives that account for whether land, agricultural extension, tech- female farmers’ relatively lower levels of nology, or financial capital. Improving literacy. women’s access to productive assets can play W hile evidence on access to finance •   an important role in raising enterprise pro- in East Asian and Pacific countries is ductivity in both the farm and nonfarm sec- mixed, women do face particular chal- tors. Following are some examples. lenges in accessing credit, especially given •   S everal countries in the region have their poorer access to land, an impor- made headway in recent years in increas- tant source of collateral. Beyond finan- ing ownership and control of land. In cial constraints, training programs that response to concerns about persistent improve business skills may be imple- gender inequalities in land, several coun- mented to address gender differences in tries—including Indonesia, Lao PDR, entrepreneurial capital. In this context, and Vietnam—have recently adopted an impact assessment of a women’s entre- gender-sensitive reforms in land titling. preneurship training program in Aceh, Since the 2004 Land Law in Vietnam, all Indonesia, found that business planning new land tenure certificates must include and management training helped promote the names of both spouses. Qualita- greater confidence among women train- tive assessment of the reform’s effects in ees, create or strengthen social networks, three provinces suggests that joint titling and identify ways to improve the business improves opportunities for women to environment for female entrepreneurs access loans, empowers women in case of (ILO 2008). disputes, and leads to more mutual deci- As in the case of education and health, sion making (World Bank 2008). Because broad systemic weaknesses—whether in the the reasons for women’s lower access form of cumbersome registration procedures, to land differ across the region—from weak systems of financial intermediation, unfavorable legal frameworks to cultural or lack of electricity—affect both female- norms and practices that deem land to be a and male-led enterprises. Evidence suggests “male asset�—effective policies to increase that such constraints may be more onerous female land holdings need to account for among small and informal firms than among context-specific constraints. larger firms and, therefore, may constrain O V E R V I E W   2 3 female-led enterprises disproportionately. As of active labor market policies in East Asia a result, interventions that focus on improv- and the Pacific is limited, studies from Latin ing the overall investment climate and partic- America and the Middle East suggest that ularly on promoting small business develop- well-designed active labor market policies can ment will be important. Addressing systemic help improve women’s employment outcomes. as well as gender-specific constraints will Affirmative action policies have also been thus be critical to promoting gender equality used to overcome gender-specific barriers to in economic opportunity. employment, whether caused by implicit or overt discrimination in hiring and promotion. Creating an enabling environment for The literature reflects some debate regarding gender equality in employment the benefits and costs of affirmative action, Public policy can strengthen the enabling but the collection of evidence (largely from environment for gender equality in formal high-income countries) suggests that carefully employment. An important element of this designed policies can help break down bar- is to ensure that women and men face a riers to female employment with little or no level legal playing field with respect to jobs adverse effects on firm productivity (World and sectors. Labor regulations that result in Bank 2011c). Affirmative action in hiring asymmetries in the employability and costs and promotion in the public sector can have of hiring male and female workers can be important demonstration effects. In 2004, the found across the region. Ostensibly protec- government of Malaysia introduced a quota tive legislation, in the form of restrictions for the public sector of 30 percent female rep- on women working at night, working over- resentation across all decision-making levels, time, and working in so-called dangerous including positions such as department heads sectors, serves in practice to inhibit women’s or secretary-general (ASEAN 2008). Whereas economic participation. Priority should be in 2006, women held 24.6 percent of top posi- given to reducing labor market restrictions tions in the public sector, by 2010, the fig- that limit women’s employment options. ure had risen to 32.0 percent. More recently, Where the original concerns motivating the Malaysian government set a target for these policies are still valid—for example, 30 percent of corporate board positions to be health and safety issues—measures should held by women by 2016. be taken to ensure that these concerns are addressed more directly through work- Taking measures to strengthen place safety codes, provision of safe and reliable women’s voice and influence transport infrastructure, and so on. Active labor market policies are another Measures to increase women’s endowments means of overcoming gender differences in and economic opportunity, such as those access for formal employment. For example, described previously, contribute to strength- wage subsidies may induce employers to hire ening their voice within the household and female workers whom they may not have in society. Educated women in good health, otherwise considered, due to lack of infor- with assets and income, are better able to act mation about their workplace productivity. on their preferences and influence outcomes This intervention thus provides the opportu- that affect themselves and others in society. nity to reduce gender stereotypes by enabling In addition, several other policy approaches employers to observe women’s skills directly, can directly promote women’s agency and and it can facilitate women gaining valuable reduce gender-based violence. labor market experience. Skills training pro- grams may also enable women and men to Supporting initiatives to transform gender move into professions outside of gender silos, norms and practices particularly when paired with apprenticeship While gender norms may be persistent, they opportunities. Although evidence on the effect are far from static. Individual experiences 2 4    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C as well as large-scale political and economic Strengthening the legal and institutional processes are capable of bringing about environment dramatic, and often rapid, social change. Nearly all countries in the region have In East Asia, increasing economic integra- acceded to CEDAW, signaling commit- tion and rising access to ICTs are not only ment to adhering to internationally agreed- transforming the economic landscape but upon norms regarding gender equality.7 also increasing flows of information in ways An important pending agenda is to ensure that may serve to transform gender norms that domestic legislation and the institu- in the region. Similarly, migration and rapid tions of implementation and enforcement urbanization across the region are bringing are aligned with countries’ commitments. with them the possibility of newly defined Where comprehensive legal reform is not roles for men and women, as women and possible, governments should identify pri- men alike are exposed to new ideas and pro- orities for action. For example, in contexts duction modalities. where women’s agency within the home The education system can be a vital remains weak, a case exists for reforms to source to change social norms that perpetu- focus on rights in marriage and divorce, ate gender inequality. The integration of inheritance and maintenance laws, and gender equality principles into the school protection of women from gender-based and professional curricula can address violence, which can strengthen the enabling the value system of children early on and, environment for greater voice and influence over time, transform social norms (Utomo in the household. As part of the process of et al. 2009). Evidence of the positive effects monitoring progress toward gender equal- of changing the curriculum is available for ity, governments should undertake regular adults in Thailand where gender sensitiv- assessments to ensure that legal gaps are ity was integrated into the curriculum in being filled and that relevant laws are being the Chulalangkorn medical school. Evalu- enforced. In countries where multiple sys- ation of the program showed that respon- tems of law coexist, assessing customary dents were more aware of gender issues practices to ascertain whether they limit and tended to apply gender concepts and women’s agency and then developing strate- concerns in their work and personal lives gies to address these limiting factors will be (WHO GWH 2007). important. Provision of information through televi- Strengthening the capacities of institu- sion programming can also play a critical tions to enforce the law and increasing role in changing social norms, especially knowledge of the law are also critical. Both with respect to fertility and gender-based financial and human investments need to violence. Evidence shows that people can be made to ensure that public sector per- be prompted to rethink gender roles in soci- sonnel, such as judges and police, have the ety when they are exposed to new informa- knowledge and capacity to actively enforce tion and experiences that challenge existing laws intended to protect women and to pro- norms. In Brazil, despite strong traditional mote gender equality. Concerns have been norms in favor of having many children, raised that police forces in several countries increased exposure to the opposite behavior in the region, including Indonesia, Malay- by popular women in soap operas led to a sia, and Thailand, have been too passive in measurable decline in fertility (Chong and investigating trafficking and enforcing anti- La Ferrara 2009; La Ferrara, Chong, and trafficking laws (U.S. Department of State Duryea 2008). In India, increased expo- 2011). The Cambodian Women’s Crisis sure to television contributed to decreased Center began a community program that acceptance of wife beating, lower fertility increases the awareness of violence against rates, and noticeable shifts away from son women and of the legal rights of women, preference (Jensen and Oster 2008). including the law on domestic violence O V E R V I E W   2 5 and protection that was implemented in these measures can be effective in increas- 2005. The program promotes initiatives to ing female representation in elected bodies. strengthen enforcement of the law by pro- They can also help transform people’s views viding information and training to local about the efficacy of female political lead- authorities and developing community sup- ers (Beaman et al. 2009). At the same time, port networks. electoral quotas do impose constraints on the democratic process. It is thus important Increasing women’s access to justice to keep this—as well as the expected ben- Financial costs and travel distances often efits of increasing female participation in are de facto barriers to women accessing politics—in mind when determining when justice, even when the appropriate laws and how to enact such measures. and institutions are in place. In that context, developing and implementing innovative Pursuing a multipronged approach to methods to improve access to the judi- reducing gender-based violence cial system can help women exercise their Reducing gender-based violence requires agency in the courts when needed. The use action on a number of fronts: efforts to of mobile courts, for example, such as those increase women’s voice within the house- in rural areas of China and Indonesia, pro- hold; enactment and enforcement of appro- vides a solution to the problem of accessi- priate legislation and strengthening of women’s bility and security for women who wish to access to justice; provision of adequate sup- exercise their rights in the legal system but port services for victims of violence; and who are unable to travel to the court. Tech- use of the media to provide information on nology, such as telephone hotlines and web- women’s rights, to increase social aware- sites, can be used to undertake basic legal ness, and to shift social norms with respect transactions. For women with few economic to violence. resources, waiving or subsidizing the costs Countries that take a strong stance on of legal aid can help reduce financial barri- gender-based violence legislation and enforce- ers to accessing the judicial system. ment can make positive strides against such violence in short periods of time. Cam- Enabling women’s participation in politics bodia saw a significant decrease in the inci- and policy making dence of domestic violence between 2000 and Active measures to promote women’s par- 2005, largely attributed to strong efforts by ticipation in policy making can be effective the country’s Ministry of Women’s Affairs, in increasing female representation in local which introduced draft domestic violence and national assemblies in many contexts. A legislation in 2001. Four years later, in Octo- range of affirmative action mechanisms have ber 2005, the National Assembly adopted been used in developing countries. Quotas the legislation. The new law criminalized have been used in several countries, some- acts of domestic violence, provided for the times in the form of constitutional changes protection of victims, and enabled neighbors to reserve a specified number of posts for or local organizations to intervene if they women and sometimes through legislative witnessed domestic violence. As a comple- and political party quotas. Quotas can be ment to the law, women’s organizations and informal (and voluntary) or mandated for- other nongovernmental organizations car- mally at the subnational or national level ried out information campaigns to dissemi- (Dahlerup 2006). The idea is to provide nate information on people’s basic rights and temporary measures to break down bar- responsibilities under the law. riers to the entry of women into politics. Governments also need to provide ade- The suitability and impact of different quate support services for victims when vio- forms of quotas or targets differ depending lence does occur. This support can include a on the specific context. Data suggest that range of services, from police and judiciary to 2 6    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C health and social services. In Malaysia, the population aging. Many of the emerging government established integrated one-stop opportunities can be fostered through the crisis centers in hospitals that provide easy types of efforts to promote greater gender access to medical care and social services, and equality in endowments, economic opportu- facilitate reporting of violence to specially nity, and agency discussed in this report. For trained police officers (World Bank 2011c). example, where gender gaps in ICT use are growing, ensuring that women have access Creating space for women’s collective to these new technologies may require active agency measures similar to those discussed earlier While promoting women’s individual agency to ensure equal access to other types of pro- is important, supporting women’s collective ductive resources. Managing emerging risks, agency can be an effective way to empower however, may require additional policy women to advocate for and promote effective approaches, such as the following. public action toward gender equality. Expe- Greater economic integration will bring rience from around the region highlights with it increased exposure to employment the potential of enabling women’s collective shocks that will have gender-differentiated agency. For example, during the debate in effects. Adequately addressing the risks asso- Cambodia leading to the 2005 Law on Pre- ciated with economic integration will require vention of Domestic Violence and Protec- designing social protection programs that tion of Victims, the Cambodia Committee take into account the different risks faced of Women, a coalition of 32 nongovernmen- by female and male workers. Building on tal organizations, persistently lobbied the the lessons from recent economic crises, sev- government and the Ministry of Women’s eral developing countries, including some Affairs to secure the legislation’s passage. from East Asia and the Pacific, have begun Similarly, in Fiji, the Fiji Women’s Crisis to recognize the gender dimensions of risk Center campaigned successfully for the and shocks in the design of programs. In implementation of a nationally representa- Indonesia, for example, a conditional cash tive quantitative survey on violence against transfer program, Keluarga Harapan, targets women; the results of this survey are sched- households with members who are particu- uled to be released in 2012. Partnerships larly vulnerable during times of crisis, such as with the private sector, including women’s pregnant and lactating women. business associations, can provide an impor- As female migration across the region tant space for women to interact, learn, and increases economic oppor tunities for advocate for gender equality. A recent initia- women, it creates new concerns about tive to increase women’s participation in the female migrants’ welfare. Protecting female private sector’s dialogue with government in migrants from exploitative situations, Cambodia, through the Government Private including sex work and human trafficking, Sector Forum, has induced policy makers will also require a gender-aware approach. to undertake new initiatives addressing the Greater protection through better laws, needs of female entrepreneurs. enforcement, and monitoring; improved information flows; and safety nets will bet- ter address the vulnerabilities specific to Fostering new opportunities, managing women traveling abroad. Specific areas for emerging risks action include improving legal protections for A new and important challenge for policy female migrants, strengthening the monitor- makers will be to help foster emerging ing and credibility of recruitment agencies, opportunities and, in particular, to manage and developing and providing welfare and emerging risks associated with increasing support services to assist female migrants. economic integration, increasing access to Governments in both sending and receiving ICTs, migration, rapid urbanization, and countries will need to actively address the O V E R V I E W   2 7 issue of human trafficking through preven- different policy options. To fill knowledge tion, protection, and prosecution. Gender- gaps, additional gender-disaggregated data awareness training for people involved in need to be collected. Moreover, additional the migration process will improve their empirical analysis, both on long-standing ability to identify and assist abused female gender issues and the gender implications of migrants, including those trafficked or at emerging trends in the region, will enable risk of being trafficked. policy makers to better promote both gender Growing urbanization in the East Asia equality and more effective development. and Pacific region has presented women not only with increased economic opportunities Notes but also with particular challenges, such as  1. Sen (1999) defines freedoms and “unfree- limited access to child care and higher secu- doms� in five categories: (a) political free- rity risks in urban areas. Thus, policy mak- doms, (b) economic facilities, (c) social oppor- ers need to ensure that child care, education, tunities, (d) transparency guarantees, and infrastructure, transportation, and water and (e) protective security. sanitation policies take into account women’s  2. Adopted in 1979 by the United Nations specific social and cultural needs. Rigorous General Assembly, CEDAW is often referred laws and policies to protect women in urban to as the international bill of rights for women. areas from the risk of violence and exploita- The convention defines what constitutes dis- tion are also warranted. crimination against women and provides an Rapid population aging in the region is agenda for national action to end such dis- crimination. To date, it has been ratified by likely to have important gender-differentiated 187 countries worldwide (http://www.un.org/ effects, among other things, because older womenwatch/daw/cedaw/). women may increasingly find themselves It is important to interpret these studies with   3.  ­ l iving as widows. Along with risks from caution, given the difficulty in establishing urbanization and the breakdown of extended a causal relationship between gender equal- family support networks, these women are ity in education and growth in cross-country likely to find themselves at increased eco- studies. nomic risk, having accumulated relatively Evidence from Africa and Latin America, for   4.  few assets and mostly lacking access to for- example, suggests that ensuring equal access mal social security. In this context, designing to productive assets and technologies could old-age income security programs that can significantly raise agricultural production and household income (Goldstein and Udry 2008; protect women from destitution in old age Quisumbing 1995; Udry 1996). will have an increasingly important role. In Globally, poverty and gender often interact to   5.  addition, policy makers should consider ways compound gender inequalities (World Bank to strengthen care for the elderly to ensure 2011c). In East Asia and the Pacific, poverty that women do not bear an undue burden of does not appear to be as important a contrib- caregiving as the region’s population ages. utor to gender disadvantage in education as elsewhere, however. Survey data from several countries in the region indicate that gender Filling knowledge gaps gaps in enrollment do not vary substantially Finally, while much has been learned from or systematically across income quintiles. recent global and regional evidence on  6. For data on land holdings, by gender, in China, see de Brauw et al. (2011). Data on gender equality and development, much other countries are based on World Bank staff remains to be understood empirically to calculations, using household survey data. help inform effective public action. Con-  7. As of the end of 2011, only six countries in tinuing to close data and analytical gaps the world had not ratified CEDAW. Two of will thus be important to better understand those countries are in the Pacific: Palau and policy priorities, the effects of specific Tonga (CEDAW, http://www.un.org/women- interventions, and the costs and benefits of watch/daw/cedaw/). 2 8    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C References CNNIC (China Internet Network Information Center) Internet Statistics. Beijing, China. http:// Agarwal, Bina. 2010a. “Does Women’s Propor- www1.cnnic.net.cn/en/index/0O/index.htm. tional Strength Affect Their Participation? Constable, Nicole. 2009. “Migrant Workers and Governing Local Forests in South Asia.� World the Many States of Protest in Hong Kong.� Development 38 (1): 98–112. Critical Asian Studies 41 (1): 143–64. ———. 2010b. Gender and Green Governance: Cuberes, David, and Marc Teignier-Baqué. 2011. The Political Economy of Women’s Presence “Gender Inequality and Economic Growth.� Within and Beyond Community Forestry. Background paper for the World Development New York: Oxford University Press. Report 2012, World Bank, Washington, DC. Agarwal, Bina, and Pradeep Panda. 2007. Dahlerup, Drude. 2006. “Introduction.� In “Toward Freedom from Domestic Violence: Women, Quotas and Politics, edited by Drude The Neglected Obvious.� Journal of Human Dahlerup, 3 –31. London and New York: Development 8 (3): 359–88. Routledge. ASEAN (Association of Southeast Asian Nations). de Brauw, Alan, Jikun Huang, Linxiu Zhang, 20 08. “ASE A N Continues to Empower and Scott Rozelle. 2011. “The Feminization Women�. http://www.aseansec.org/Bulletin- of Agriculture with Chinese Characteristics.� Feb-08.htm#Article-2. Background paper for the World Development Baulch, Bob, Truong Thi Kim Chuyen, Dominique Report 2012, World Bank, Washington, DC. Haughton, and Jonathan Haughton. 2002. Do, Quy-Toan, Andrei Levchenko, and Claudio “Ethnic Minority Development in Vietnam: Raddatz. 2011. “Engendering Trade.� Policy A S o cio e conom ic Persp e c t ive.� Pol ic y Research Working Paper 5777, World Bank, Research Working Paper 2836, World Bank, Washington, DC. Washington DC. Duflo, Esther. 2000. “Child Health and Household Beaman, Lori, Raghabendra Chattopadhyay, Resources in South Africa: Evidence from Esther Duf lo, Rohini Pande, and Petia the Old Age Pension Program.� American Topalova. 2009. “Powerful Women: Does Economic Review 90 (2): 393–98. Exposure Reduce Bias?� Quarterly Journal of — — —. 2003. “Grandmothers and Granddaughters: Economics 124 (4): 1497–540. Old-Age Pensions a nd I nt ra household Beegle, Kathleen, Elizabeth Frankenberg, and Allocation in South Africa.� World Bank Duncan Thomas. 2001. “Bargaining Power Economic Review 17 (1): 1–25. within Couples and Use of Prenatal and Dwyer, Jeffrey, and Raymond Coward. 1992. Delivery Care in Indonesia.� Studies in Family “Gender and Family Care of the Elderly: Planning 32 (2): 130–46. Research Gaps and Opportunities.� In Gender, BPS (Badan Pusat Statistik–Statistics Indonesia) Families, and Elder Care, edited by Jeffrey W. a n d O RC M a c r o . 2 0 0 3 . I n d o n e s i a Dwyer and Raymond T. Coward, 151– 62. Demographic and Health Survey 2002–2003. London: Sage. Calverton, MD: BPS and ORC Macro. Edlund, Lena, and Rohini Pande. 2001. “Why Bruni, Lucilla, Andrew D. Mason, Laura Pabon, Have Women Become Left-Wing? The Political and Carrie Turk. 2011. “Gender Impacts of the Gender Gap and the Decline in Marriage.� Global Financial Crisis in Cambodia.� World Quarterly Journal of Economics 117 (3): Bank, Washington, DC. 917–61. Chattopadhyay, Raghabendra, and Esther Duflo. Enterprise Surveys database. World Bank / 2004. “Women as Policy Makers: Evidence International Finance Corporation, Washington, from a Randomized Policy Experiment in DC. http://www.enterprisesurveys.org/. India.� Econometrica 72 (5): 1409–43 FAO (Food and Agriculture Organization)/Sida Chi, Wei, and Bo Li. 2007. “Glass Ceiling or Partnership Cooperation. 2010. National Sticky Floor? Examining the Gender Pay Gap Gender Profile of Agricultural Households, across the Wage Distribution in Urban China, 2010: Report Based on the Lao Expenditure 1987–2004.� MPRA Paper 3544, University a n d C o n s u m pt i o n S u r ve y s , N a t i o n a l Library of Munich, Germany. Agricultural Census and Population Census. Chong, Alberto, and Eliana La Ferrara. 2009. Vientiane, Lao PDR: FAO. “Television and Divorce: Evidence from Fehringer, Jessica, and Michelle J. Hindin. 2009. Brazilian Novelas.� Journal of the European “Like Parent, Like Child: Intergenerational Economic Association 7 (2–3): 458–68. Transmission of Partner Violence in Cebu, the O V E R V I E W   2 9 Philippines.� Journal of Adolescent Health 44 Judd, Ellen R. 2010. “Family Strategies: Fluidities (4): 363–71. of Gender, Community and Mobility in Rural Filmer, Deon, and Norbert Schady. 2008. West China.� China Quarterly 204: 921–38. “Getting Girls into School: Evidence from a Klasen, Stephan. 2002. “Low Schooling for Scholarship Program in Cambodia.� Economic Girls, Slower Growth for All? Cross-Country Development and Cultural Change 56 (3): Evidence on the Effect of Gender Inequality in 581–617. Education on Economic Development.� World ———. 2009. “School Enrollment, Selection and Bank Economic Review 16 (3): 345–73. Test Scores.� Policy Research Working Paper Knowles, Stephen, A. K. Lorgelly, and Dorian Series 4998, World Bank, Washington, DC. Owen. 2002. “Are Educational Gender Gaps Ganster-Breidler, Margit. 2010. “Gender-Based a Brake on Economic Development? Some Violence and the Impact on Women’s Health Cross-Country Empirical Evidence.� Oxford and Well-Being in Papua New Guinea.� DWU Economic Papers 54 (1): 118–49. Research Journal 13. L a Ferra ra , E l ia na , A lber to C hong, a nd Goldstein, Markus, and Christopher Udry. Suzanne Duryea. 2008. “Soap Operas and 2008. “The Profits of Power: Land Rights and Fertility: Evidence from Brazil.� BREAD Agricultural Investment in Ghana.� Journal of Working Paper 172, Bureau for Research and Political Economy 116 (6): 981–1022. Economic Analysis of Development, Duke Hicks, Joan, Hamory Hicks, Michael Kremer, University, Durham, NC. Isaac Mbiti, and Edward Miguel. “Vocational Lewis, Ione, Bessie Maruia, and Sharon Walker. Education Voucher Delivery and Labor Market 2008. “Violence against Women in Papua Returns: A Randomized Evaluation Among New Guinea.� Journal of Family Studies 14: Kenyan Youth,� Report for Spanish Impact 183–97. Evaluation Fund (SIEF) Phase II, World Bank, Lott, John R., and Lawrence W. Kenny. 1999. Washington, DC. “Did Women’s Suffrage Change the Size and HNPStats Population Projections database. World Scope of Government?� Journal of Political Bank, Washington, DC. http://go.worldbank. Economy 107 (6): 1163–98. org/H4UN4D5KI0 Maffii, Margherita, and Sineath Hong. 2010. ICRW (International Center for Research on “Political Participation of Indigenous Women Women). 2006. Property Ownership and in Cambodia.� Asien 114–15 (April): 16–32. Inheritance Rights of Women for Social Morrison, Andrew, Mary Ellsberg, and Sarah Protection: The South Asia Experience. Bot t. 20 07. “Addressing G ender-Based Washington, DC: ICRW. Violence: A Critical Review of Interventions.� ILO (International Labour Organization). 2008. World Bank Observer 22 (1): 25–51. Women’s Entrepreneurship Development Naudeau, Sophie, Naoko Kataoka, Alexandria Aceh: Gender and Entrepreneurship Together Valerio, Michelle Neuman, and Leslie Elder. (GET Ahead) Training Implementation: 2011. Investing in Young Children: An Impact Assessment. Jakarta, Indonesia: ILO Early Childhood Development Guide for Jakarta Office. Policy Dialogue and Project Preparation. ———. Key Indicators of the Labour Market Washington, DC: World Bank. (KILM) database. International Labour Orga­ National Center for Injury Prevention and nization, Geneva, Switzerland. http://kilm.ilo Control. 2003. Costs of Intimate Partner .org/kilmnet/. Violence against Women in the United States. Indonesia Family Life Survey. 2007/2008. RAND Atlanta, GA: Centers for Disease Control and Family Life Surveys, IFLS-4. http://www.rand Prevention. .org/labor/FLS/IFLS/ifls4.html. NIPH (National Institute of Public Health Jensen, Robert, and Emily Oster. 2008. “The [Cambodia]), N IS (National Institute of Power of TV: Cable Television and Women’s Statistics [Cambodia]) and ORC Macro. Status in India.� University of Chicago, IL. 2006. Cambodia Demographic and Health http://home.uchicago.edu/eoster/tvwomen Survey 2005. Phnom Penh, Cambodia, and .pdf. Calverton, Maryland, USA: NIPH, NIS, and Johnson, Kiersten, Sovanratnak Sao, and Darith ORC Macro. Hor. 2000. Cambodia 2000 Demographic NSD (National Statistics Directorate [Timor- and Health Survey: Key Findings. Calverton, Leste]), Ministry of Finance [Timor-Leste], and MD: ORC Macro. ICF Macro. 2010. Timor-Leste Demographic 3 0    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C and Health Survey 2009–10. Dili, Timor- Islands]. 2009. Solomon Islands Family Leste: NSD and ICF Macro. Health and Safety Study: A Study on Violence Ofstedal, Mary Beth, John E. Knodel, and against Women and Children. Nouméa, New Napaporn Chayovan. 1999. “Intergenerational Caledonia: SPC. Support and Gender: A Comparison of Four SUSI (Survei Usaha Terintegrasi). 2002. Integrated Asian Countries.� Southeast Asian Journal of Survey of Cottage and Small-Scale Firms, Social Sciences 27 (2): 21–41. Indonesia. BPS. http://dds.bps.go.id. PARLINE database on national parliaments. Swaminathan, Hema, Cherryl Walker, and Inter-Parliamentary Union, Geneva. http:// Margaret A. Rugadya, eds. 2008. Women’s www.ipu.org/parline-e/parlinesearch.asp and Prope r t y R ights, H I V and A I DS , and http://www.ipu.org/wmn-e/world.htm. Domestic Violence: Research Findings from PovcalNet (online pover t y analysis tool). Two Rural Districts in South Africa and Development Research Group, World Bank, Uganda. Cape Town: HSRC Press. Washington, DC. http://iresearch.worldbank Thomas, Duncan. 1995. “Like Father, Like Son, .org/PovcalNet/index.htm. Like Mother, Like Daughter, Parental Resources Pronyk, Paul M., James R. Hargreaves, Julia and Child Height.� Papers 95-01, RAND Reprint C. Kim, Linda A. Morison, Godfrey Phetla, Series, Rand Corporation, Santa Monica, CA. Charlotte Watts, Joanna Busza, and John Udry, Christopher. 1996. “Gender, Agricultural D. H. Porter. 2006. “Effect of a Structural Produc tion, and the T heor y of the Intervention for the Prevention of Intimate- Household.� Journal of Political Economy partner Violence and HIV in Rural South 104 (5): 1010–14. Africa: A Cluster Randomized Trial.� Lancet UN (United Nations Department of Economic 2368 (9551): 1973–83. and Social Affairs, Population Division). 2010. Qian, Nancy. 2008. “Missing Women and the World Urbanization Prospects: The 2009 Price of Tea in China: The Effect of Sex-Specific Revision. New York: United Nations. http:// Earnings on Sex Imbalance.� Quarterly esa.un.org/unpd/wup/index.html. Journal of Economics 123 (3): 1251–85. UNDP (United Nations Development Programme). Quisumbing, Agnes. 1995. “Gender Differences 2010. Asia-Pacific Human Development in Agricultural Productivity: A Survey of Report. Power, Voice and Rights: A Turning Empirical Evidence.� FCND Discussion Paper Point for Gender Equality in Asia and the No. 5, Food Consumption and Nutrition Pacific . Colombo, Sri Lanka: Macmillan Division, International Food Policy Research Publishers India Ltd. for UNDP. Institute, Washington, DC. UNDP (United Nations Development Program) R o d g e r s , Ya n a , a n d N i d h i y a M e n o n . Cambodia and VBNK. 2010. Talking About Forthcoming. “Impact of the 2008–2009 Domestic Violence: A Handbook for Village Twin Economic Crises on the Philippine Facilitators. Phnom Penh: UNDP Cambodia Labor Market.� World Development. and the Agencia Española de Cooperación Sakellariou, Chris. 2011. “Determinants of the para el Desarrollo. Gender Wage Gap and Female Labor Force UNFPA (United Nations Population Fund). 2008. Participation in EAP.� Paper commissioned for An Assessment of the State of Violence against Toward Gender Equality in East Asia and the Women in Fiji. Suva, Fiji. Pacific: A Companion to the World Development USCDC (United States Centers for Disease Control Report, Washington, DC: World Bank. and Prevention). 2003. “Costs of Intimate Sen, Amartya. 1999. Development as Freedom. Partner Violence against Women in the United Oxford, U.K.: Oxford University Press. States.� USCDC, National Center for Injury SPC (Secretariat of the Pacific Community), Prevention and Control, Atlanta, Georgia. Ministry of Internal and Social Affairs U.S. Department of State 2011. Trafficking in [Republic of Kiribati], and Statistics Division Persons Report 2011. Washington, DC: U.S. [Republic of Kiribati]. 2010. Kiribati Family Department of State. Health and Support Study: A Study on Utomo, Iwu, Peter McDonald, Terence Hull, Violence against Women and Children. Ida Rosyidah, Tati Hattimah, Nurul Idrus, Nouméa, New Caledonia: SPC. Saparinah Sadli, and Jamhari Makruj. 2009. SPC (Secretariat of the Pacific Community) and “Gender Depiction in Indonesian School Text NSO (National Statistics Office) [Solomon Books: Progress or Deterioration.� Australian O V E R V I E W   3 1 Demographic and Social Research Institute, World Bank. 2008. Vietnam: Analysis of the Australian National University. Impact of Land Tenure Certificates with Both VWC (Vanuatu Women’s Centre) in partnership Names of Wife and Husband: Final Report. with the Vanuatu National Statistics Office. Washington, DC: World Bank. 2011. Vanuatu National Survey on Women’s ———. 2010. World Development Indicators Lives and Family Relationships. Port Vila, 2010. Washington, DC: World Bank. Vanuatu: VWC. ———. 2011a. “Lao PDR Country Gender Wang, Zhengxu, and Weina Dai. 2010. “Women’s Assessment.� World Bank, Washington, Participation in Rural China’s Self-Governance: DC. Institutional, Socioeconomic, and Cultural ———. 2011b. “Vietnam Cou ntr y G ender Factors in a Jiangsu County.� Discussion Assessment.� World Bank, Washington, Paper 69, China Policy Institute, University of DC. Nottingham, U.K. ———. 2011c. World Development Report W HO GW H (World Health Organization, 2012: Gender Equality and Development. Department of Gender, Women and Health). Washington, DC: World Bank. 2007. Integrating Gender into the Curricula World Values Survey database. World Values for Health Professionals, Meeting Report, 4–6 Survey Association, Stockholm, Sweden. http:// December 2006. Geneva: WHO. www.worldvaluessurvey.org/. The State of Gender Equality in East Asia and the Pacific 1 I n recent decades, women across the globe decades have been associated with reduced have made advances toward gender equal- gender inequalities in several dimensions. The ity. Literacy rates for young women and region grew at 7 percent on average between girls are higher than ever before, and gen- 2000 and 2008 (figure 1.1), the structure der gaps in primary education have closed in of the region’s economies has shifted away almost all countries across the world. In the from agriculture toward manufacturing and past three decades, over half a billion women services, and extreme poverty has fallen dra- have joined the world’s labor force (World matically. Indeed, the share of the region’s Bank 2011b). population living on less than US$1.25 a day Strides toward gender equality in East has declined by more than 50 percent since Asia and the Pacific have been similarly note- 1990—from the highest poverty head count worthy. Most countries in the region have rate in the world to among the lowest (fig- either reached or surpassed gender parity in ure 1.2). Growth, structural transformation, education enrollments. Health outcomes for and poverty reduction have been accompa- both women and men have improved signifi- nied by considerable progress toward gender cantly. Female labor force participation rates equality in several key areas, particularly in the region are relatively high. Yet despite education and health. Many countries in the considerable progress in this economically region have experienced closing gender gaps dynamic region, gender disparities persist in in school enrollments and declining maternal a number of important areas—particularly in and child mortality rates. access to economic opportunity and in voice But growth and development have not and influence in society. For policy makers been enough to attain gender equality in all in East Asian and Pacific countries, closing its dimensions. Women still have less access these gender gaps represents an important than men to a range of productive assets challenge to achieving more inclusive and and services, including land, financial capi- effective development. tal, agricultural extension services, and new The East Asia and Pacific region’s signifi- information technologies. Substantial employ- cant economic growth, structural transfor- ment segregation, or sorting, by gender still mation, and poverty reduction in the past few remains an issue. Women are less likely than 33 3 4    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE 1.1  The East Asia and Pacific region has experienced in East Asian and Pacific countries still have rapid economic growth a weaker voice and less influence than men, whether in household decision making, in the average annual GDP growth rate private sector, in civil society, or in politics. 10 Moreover, women across the region remain 8 vulnerable to gender-based violence, often at the hand of an intimate partner. 6 The East Asia and Pacific region is vast 4 and diverse, with large differences in eco- nomic and social progress, including toward percent 2 gender equality. Achievements in growth and 0 development have not been uniform across the region. Although average annual gross –2 domestic product (GDP) growth during the –4 first decade of the 2000s neared 10 per- cent in China, it was close to zero in several –6 1980–84 1985–89 1990–94 1995–99 2000–04 2005–09 small Pacific Island countries. By the end of the decade, levels of GDP per capita among East Asia and Paci c Europe and Central Asia the low- and middle-income countries of the Latin America and the Caribbean Middle East and North Africa region also varied widely, from US$623 in South Asia Sub-Saharan Africa Timor-Leste to US$8,373 in Malaysia (WDI database). Nor has progress toward gender Source: World Bank 2010b. equality been uniform. Despite widespread progress toward gender equality in schooling, FIGURE 1.2  Poverty reduction in the East Asia and Pacific a few (mostly low-income) countries continue region has been impressive to face challenges in closing gender gaps in basic education. In spite of broad improve- percentage of the population living below 90 $1.25 a day ments in health outcomes, China—and, to 80 a lesser extent, Vietnam—face significant 70 imbalances in the ratio of boys to girls at 60 birth, a function of prenatal sex selection stemming from the strong preference for sons percent 50 40 in those societies. As a result, the region has more than a million “missing� girls at birth 30 per year. 20 A number of Pacific Island countries 10 face particular challenges with respect to 0 promoting women’s voice, influence, and 1981 1984 1987 1990 1993 1996 1999 2002 2005 2008 e mpowerment—in both the private and ­ East Asia and Paci c Europe and Central Asia public spheres. For example, although gen- Latin America and the Caribbean Middle East and North Africa South Asia Sub-Saharan Africa der-based violence remains pervasive in the region, its prevalence in the Pacific is among the highest in the world. Data from Solomon Source: PovcalNet. Islands, Kiribati, and Vanuatu indicate that more than 60 percent of adult women have men to work in formal sector jobs and more experienced physical or sexual violence dur- likely to work in poorly remunerated occupa- ing their lifetime, often at the hands of an tions and enterprises. And despite the ­ closing intimate partner (SPC, Ministry of Internal of education gaps, women continue to be and Social Affairs, and Statistics Division paid less than men for similar work. Women 2010; SPC and NSO 2009; VWC 2011). T H E S T A T E O F G E N D E R E Q U A L I T Y I N E A S T A S I A A N D T H E P A C I F I C    35 Moreover, female representation in politics in in fertility and mortality are dramatically the Pacific is among the lowest in the world. changing the demographic profile of the Although female political participation is rel- region, and many countries will face rapid atively low worldwide—only 19.6 percent of population aging in the coming years. The the world’s parliamentarians were women in region also experiences rising adoption of December 2011—four of the eight countries new information and communication tech- in the world with no female parliamentarians nologies (ICTs), which are breaking down were located in the Pacific (PARLINE data- information barriers, opening up new eco- base, 2011 data). nomic opportunities, and enabling collec- A growing body of literature also suggests tive action in many contexts. that patterns of growth—not just levels—can These trends are likely to have important affect gender equality by affecting incentives gender dimensions, generating both new to invest in and to create opportunities for opportunities and new risks for gender equal- women and girls. Cross-country evidence ity. For instance, while increased economic indicates that gender gaps—in education and integration has contributed to higher demand labor force participation, for example—tend for female labor in several East Asian coun- to be smaller in countries that export more tries, it has also increased workers’ vulner- in relatively female labor–intensive sectors ability to external shocks, with different (Do, Levchenko, and Raddatz 2011). Recent effects on females and males. Similarly, while studies have found that natural resource increased migration has opened up new eco- extraction, namely oil, reduces and discour- nomic opportunities in the region, it has also ages female labor force participation, which been accompanied by new risks—among the in turn reduces their political influence (Ross most severe is the risk of human trafficking. 2008; World Bank 2012a). Such distinc- This report clarifies empirically the rela- tions in economic structure are relevant to tionship between gender equality and devel- the East Asia and Pacific context. East and opment and outlines an agenda for public Southeast Asian economies rely heavily on action to promote gender equality in East export-­ o riented manufacturing growth, Asian and Pacific countries. Written as a com- whereas economies of the Pacific Islands are panion to the World Development Report dominated by natural resource extraction, 2012: Gender Equality and Development tourism, and remittances. These differences (World Bank 2011b), the report makes sev- in economic incentives generated by distinct eral distinct contributions to policy makers’ patterns of development may help to explain, understanding of gender, development, and at least in part, differences in progress toward public policy in the region. First, the analysis gender equality in East Asian and Pacific focuses on those issues and policy challenges countries, particularly with respect to voice that are particularly relevant to East Asian and influence in society. and Pacific countries. For example, com- Several emerging trends in East Asian pared with other developing regions, female and Pacific countries will present both access to basic education is no longer a first- new opportunities and new risks to achiev- order issue in many parts of the region. At ing gender equality. Much of the region is the same time, gender stereotyping and gen- characterized by high levels of openness der “streaming� in education still represent and economic integration with the rest of critical challenges. These factors affect wom- the world. Much of the region is also char- en’s and men’s aspirations and behaviors and acterized by migration and rapid urban- contribute to persistent employment segrega- ization as the center of economic activi- tion. Second, the report examines the gender ties has moved from rural to urban areas. implications of several key emerging trends Increasing numbers of the region’s citizens in the region: increased global economic cross national boundaries in search of bet- integration, migration, urbanization, rapid ter economic opportunities. Swift declines population aging, and rising use of ICTs. 3 6    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C These trends have important gender dimen- as a dimension of agency, defining violence sions that are not commonly accounted for against women as the extreme deprivation of by policy makers but that will generate a dis- agency. tinct set of challenges for promoting gender equality going forward. Third, the report accounts for intraregional diversity in a way Why does gender equality matter that is not possible in a global report. Par- for development? ticular emphasis is placed, where possible, Gender equality matters intrinsically. Nobel on the challenges faced by countries in the prize–winning economist Amartya Sen trans- Pacific as distinct from those in East Asia. formed the discourse on development when Finally, the report has undertaken extensive he argued that development is not only about empirical analysis of gender equality using raising people’s incomes or reducing poverty, a newly created database of household sur- but rather involves a process of expanding veys for the region. In doing so, the report freedoms equally for all people (Sen 1999).1 has contributed significantly to the develop- Viewed from this perspective, gender equal- ment of indicators and evidence on gender, ity is intrinsically valued. The near-universal development, and public policy that were not ratification and adoption of the Convention available previously. on the Elimination of All Forms of Discrimi- This chapter examines the state of gender nation against Women (CEDAW)2 —and the equality in the East Asia and Pacific region, subsequent commitment of the international highlighting both recent progress and pend- community to Millennium Development ing challenges. Following the World Devel- Goals 3 and 5—underscores a near-global opment Report 2012, the discussion focuses consensus that gender equality and women’s on gender equality in three domains: (a) empowerment are development objectives in endowments, (b) economic opportunity, their own right. and (c) agency (see box 1.1). Endowments Gender equality also matters for develop- are defined here as human capital and other ment. A growing body of empirical litera- productive assets that allow individuals to ture from around the world demonstrates live healthy and productive lives. To analyze that promoting gender equality is also good gender equality in endowments, the discus- development policy, or as stated in the World sion focuses on education and health as well Development Report 2012 (2011b, 3), “Gen- as other productive assets, such as land. Eco- der equality … is smart economics.� Indeed, nomic opportunity pertains to an individu- the literature shows that greater gender al’s ability to fully and freely participate in equality in endowments, access to economic and receive returns from their work in the opportunities, and agency can (a) contribute economy. The report focuses on a range of to higher productivity, income growth, and economic indicators, including labor force poverty reduction; (b) improve the opportu- participation, earnings, and employment nities and outcomes of the next generation; segregation, whether in the labor market or and (c) enhance development decision mak- in self-employment. Agency is defined as the ing. This section explores the evidence on ability of women and men to express them- these three pathways, in turn. selves (exercise voice) in accordance with their preferences and to take actions on their Gender equality can contribute to own behalf to influence their surroundings. higher productivity and income Since people exercise agency in all aspects of growth life, the report focuses on multiple dimen- sions: agency within a household and in sev- For households and economies to function eral aspects of the public domain, including at their full potential, resources, skills, and civil society, the private sector, and politics. talent should be put to their most produc- The report also focuses on safety and security tive use. If societies allocate resources on T H E S T A T E O F G E N D E R E Q U A L I T Y I N E A S T A S I A A N D T H E P A C I F I C    37 BOX 1.1  Defining and measuring gender equality Gender refers to the social, behavioral, and cultural of outcomes measures equality of results (World attributes, expectations, and norms that distinguish Bank 2011b). Both concepts can be useful, depend- men and women. Gender equality refers to the extent ing on the domain. to which men’s and women’s opportunities and out- In some domains, such as in health and educa- comes are constrained—or enhanced—solely on the tion, where gender equality in outcomes may be basis of their gender. This book focuses on gender inherently valued, it is reasonable to focus on equal- equality in three domains: (a) endowments —human ity of outcomes. In contrast, equality of opportuni- and productive capital; (b) economic opportunity — ties may be the more relevant conception of gender participation and returns in the economy; and (c) equality in the economic sphere, where people’s pref- agency —the voice and influence of women in mul- erences may lead to different outcomes, even if their tiple dimensions in the private and public domains. opportunities are equal. Despite these distinctions, Gender equality can be conceptualized in two it is often difficult to distinguish opportunities from ways: in terms of equality of opportunities and outcomes empirically. As such, though the book will equality of outcomes. Equality of opportunities rely on both conceptions of gender equality in its measures inequalities that arise from circum- analysis, data limitations often necessitate that the stances beyond the control of individuals. Equality evidence focuses on measuring outcomes. the basis of one’s gender, as opposed to women (or men) choose fields of study on the one’s skills and abilities, this comes at a basis of their gender rather than their abili- cost. Indeed, the economic costs of gender ties, this too will exact costs not only on indi- inequalities—whether caused by the persis- viduals’ employment and earnings, but also tence of traditional norms or by overt dis- on a country’s economic productivity more crimination—can be considerable. A recent broadly. study commissioned for the World Develop- Gender inequalities in access to productive ment Report 2012 found that in the East assets also have costs in terms of productiv- Asia and Pacific region, output per worker ity and income. Microeconomic studies from could be 7 to 18 percent higher across a a number of countries across developing range of countries if female entrepreneurs regions show that female farmers and entre- and workers were to work in the same sec- preneurs are inherently no less productive tors, types of jobs, and activities as men, and than male farmers and entrepreneurs; rather, have the same access to productive resources they tend to have less access to productive (Cuberes and Teignier-Baqué 2011). inputs.4 A recent study by the Food and Agri- Evidence suggests that misallocation of culture Organization of the United Nations female skills and talent commonly begins estimates that equalizing access to productive before women enter the labor force, when resources between female and male farmers families and societies underinvest in girls’ could increase agricultural output in develop- schooling. A number of cross-country ­ studies ing countries by 2.5 to 4.0 percent (FAO/Sida have found a robust inverse relationship Partnership Cooperation 2010). between the size of the gender gap in educa- A number of studies show that gender- tion and GDP growth, controlling for ­ average based violence also imposes significant costs education levels and other factors associated on the economies of developing countries with economic growth (see, for example, through lower worker productivity and Klasen 2002; Knowles, Lorgelly, and Owen incomes, lower human capital investments, 2002). 3 Moreover, to the extent that young and weaker accumulation of social capital 3 8    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE 1.3  Girls’ secondary school enrollments have converged violence against adult women were more to those of boys than $4 billion USD in 1995 (USCDC 2003). Reducing gender-based violence would female-to-male ratio of secondary school gross enrollment thus have significant positive effects on the region’s economies by reducing health care 1.1 costs and increasing investments in women’s 1.0 human capital, female worker productivity, 0.9 and women’s accumulation of social capital. 0.8 As the global economy becomes more ratio integrated, the productivity effects associ- 0.7 ated with greater gender equality are likely to 0.6 be increasingly important to East Asian and 0.5 Pacific countries. A recent study on the rela- tionship between gender and trade suggest 0 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 that gender inequalities have become finan- cially detrimental for countries in a world East Asia and Paci c Europe and Central Asia of open trade (Do, Levchenko, and Raddatz Latin America and the Caribbean Middle East and North Africa 2011). To participate effectively in an increas- South Asia Sub-Saharan Africa ingly competitive world, countries will need to harness their resources efficiently by improv- Source: World Bank 2011c. ing opportunities for all and allocating labor based on skill rather than by gender. Gender FIGURE 1.4  Tertiary school enrollments of females have inequality, whether in endowments, economic converged to those of males in East Asia and the Pacific opportunities, or in agency (voice), reduces a country’s ability to compete in this increas- female-to-male ratio of tertiary school gross enrollment ingly globalized economic environment 1.4 (World Bank 2011b). 1.2 1.0 Promoting gender equality is also an investment in the next generation 0.8 A large body of cross-country and country- ratio 0.6 specific literature shows that healthier, bet- 0.4 ter educated mothers have healthier, better educated children, which can be expected 0.2 to positively affect children’s future pro- ductivity and economic prospects. The 0 1990 2000 2005 2007 2009 effects begin even before childbirth. In Timor-Leste, highly educated mothers and East Asia and Pacifica Europe and Central Asiaa those in the wealthiest households are more Latin America and the Caribbean Middle East and North Africa likely to have their babies delivered by South Asia Sub-Saharan Africa skilled birth attendants than less educated mothers and those from poorer households Source: UNESCO Institute for Statistics (UIS) Data Centre, 2009, 2011 data. (NSD, Ministry of Finance, and ICF Macro a. East Asia and Pacific includes developed countries. In this figure, Mongolia is included in Europe and Central Asia, not in East Asia and the Pacific. 2010). Similarly, Demographic and Health Survey data show that Cambodian women (Morrison, Ellsberg, and Bott 2007). In addi- with little education are relatively less likely tion to indirect costs, gender-based violence to receive prenatal care and assistance from has large direct economic costs on society. trained health personnel during birth deliv- A study in the United States found that the eries than women with higher education direct health care costs of intimate partner (Johnson, Sao, and Hor 2000). A mother’s T H E S T A T E O F G E N D E R E Q U A L I T Y I N E A S T A S I A A N D T H E P A C I F I C    39 health and nutrition status is also found to FIGURE 1.5  Gender gaps in secondary school enrollment vary strongly affect children’s physical health as substantially across countries well as cognitive and noncognitive abilities, which can have long-lasting developmental female-to-male ratio of secondary school and societal consequences (Nadeau et al. gross enrollment, most recent year 2011). Thailand, 2009 Higher labor force participation, greater Vanuatu, 2009 income earned, and more assets held by Philippines, 2009 women have also been shown to have positive Mongolia, 2009 effects on the next generation. In Indonesia, Malaysia, 2008 for example, women with a higher share of China, 2009 household assets before marriage tend to use more prenatal care and are more likely to Vietnam, 2008 have their births attended by skilled health Indonesia, 2009 care providers (Beegle, Frankenberg, and Timor-Leste, 2009 Thomas 2001). Similarly, in China, increas- Cambodia, 2008 ing adult female income by 10 percent of the Solomon Islands, 2007 average household income raised the fraction Lao PDR, 2008 ­ ercentage point and of surviving girls by 1 p Papua New Guinea, 1998 increased years of schooling for both boys 0 0.2 0.4 0.6 0.8 1.0 1.2 and girls. In contrast, a similar increase in ratio male income reduced survival rates and edu- cational attainment for girls with no impact Source: World Bank 2011c. on boys (Qian 2008). Studies from across developing and developed regions (for exam- ple, from places as diverse as Brazil, Ghana, South Africa, and the United States) show that income in the hands of women posi- FIGURE 1.6  Gender gaps in education have reversed in several tively affects their female children’s health countries, particularly at the tertiary level (Duflo 2003; Thomas 1995); commonly, the marginal effects of income and assets in the female-to-male ratio of tertiary school hands of mothers are larger than the effects gross enrollment, most recent year of similar income and assets in the hands of fathers. Tonga, 2004 Reductions in gender-based violence Mongolia, 2009 through greater female agency can also have Malaysia, 2008 important intergenerational benefits. Several Philippines, 2008 studies show that experiencing domestic vio- Thailand, 2009 lence between parents as a child contributes Fiji, 2005 to a higher risk of both women experienc- China, 2009 ing domestic violence as adults and of men Vietnam, 2009 perpetrating violence against their spouses Indonesia, 2009 (Fehringer and Hindin 2009). In Timor- Lao PDR, 2008 Leste, 56.4 percent of women who were vic- Timor-Leste, 2009 tims of spousal violence had a father who Vanuatu, 2004 beat their mother (NSD, Ministry of Finance, Papua New Guinea, 1999 and ICF Macro 2010). In Cambodia, women Cambodia, 2008 who reported that their mothers experienced 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 domestic violence were more likely to expe- ratio rience physical and psychological domes- tic violence as well (NIPH, NIS, and ORC Source: UIS Data Centre, 2009, 2011 data. 4 0    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE 1.7  Minority populations in Vietnam often experience Strengthening women’s voice can lower educational enrollments enhance the quality of development decision making percentage of 15- to 17-year-olds enrolled 90 in secondary school, by ethnicity, 2008 Several studies have shown that women 80 and men have different policy preferences 70 (Edlund and Pande 2001; Lott and Kenny enrollment rate (%) 60 1999). Capturing these gender-based dif- 50 ferences in perspective can lead to not only 40 more representative but also better decision 30 making. Evidence from India indicates that 20 private firms can benefit from greater gen- 10 der equality among the ranks of senior man- 0 agement. Other evidence from South Asia Kinh/ Tay Central Hmong/ Khmer ­ s uggests the same is true with respect to Chinese ethnic Dao development policy making. As an example, majority minority a study of women elected to local government in India found that female leadership posi- male female tively affected the provision of public goods at the local level in ways that better reflected Source: World Bank 2011a. both women’s and men’s preferences (Chat- topadhyay and Duflo 2004). Similarly, stud- ies from rural India and Nepal found that FIGURE 1.8  In Indonesia, gender gaps in enrollment do not when women who were previously excluded vary substantially by household wealth from decisions about local natural resource management had greater voice and influ- percentage of 13- to 15-year-olds enrolled in school, ence, local conservation outcomes improved 100 by sex and expenditure quintile, 2009 significantly (Agarwal 2010a, 2010b). 90 Women’s collective agency can also be 80 transformative, both for individuals and for society as a whole. For example, for a group enrollment rate (%) 70 60 of ethnic minority women in rural China, 50 information sharing among them has helped 40 empower them and raise their social standing 30 in the Han-majority communities into which 20 they married (Judd 2010). Migrant domestic 10 workers in Hong Kong SAR, China, have been engaged in civic action focused on local 0 I II III IV V migrant workers’ rights as well as interna- expenditure quintile tional human rights over the last 15 years male female (Constable 2009). These efforts have contrib- uted to the enactment of laws that now pro- Source: World Bank estimates using Indonesia National Socioeconomic Survey 2009. vide migrant domestic workers in Hong Kong SAR, China, with some of the most compre- hensive legal protections in the world. Macro 2006). Efforts that increase women’s safety and security and that reduce domes- Recent progress, pending tic violence can thus lead to lower intergen- challenges erational transmission of violence within Over the past few decades, many East Asian families.5 and Pacific countries have experienced T H E S T A T E O F G E N D E R E Q U A L I T Y I N E A S T A S I A A N D T H E P A C I F I C    41 considerable progress toward gender equal- FIGURE 1.9  Fertility rates have declined across the world ity, at least in some dimensions. In others, gender disparities have been more persis- fertility rate, by region tent. This section reviews recent progress 7.5 and pending challenges in achieving gender equality in endowments, economic oppor- 6.5 tunity, and agency in the region. It empha- births per woman 5.5 sizes where growth and development have contributed to advances in promoting gen- 4.5 der equality and where this has not been suf- ficient. The section also highlights the con- 3.5 siderable diversity of experience within the 2.5 region as well as within countries. The basic gender profile developed here provides the 1.5 1960 1970 1980 1990 2000 2008 foundation for the deeper analysis of gender, development, and public policy presented in East Asia and Paci c Europe and Central Asia subsequent chapters. Latin America and the Caribbean Middle East and North Africa South Asia Sub-Saharan Africa Endowments: Human and productive capital Source: World Bank 2010b. Economic growth and poverty reduction in the region have been associated with FIGURE 1.10  Under-five mortality rates have declined sharply rapid increases in female enrollment and for both boys and girls convergence in the rates of school enroll- ment, across both genders and at all levels under- ve mortality rate (probability of dying of ­e ducation. In 2010, the region had the 240 by age ve per 1,000 live births) h ighest female-to-male enrollment ratio of ­ 200 all developing regions at the primary level. under-five mortality rate At the secondary enrollment level, only Latin 160 America and the Caribbean had a higher ratio (figure 1.3). Although the female-to- 120 male enrollment ratio in the East Asia and 80 Pacific region is still below 1 at the tertiary level, it has been rising consistently over the 40 past two decades (figure 1.4). 0 However, both overall enrollment rates M F M F M F M F M F M F and female-to-male enrollment ratios vary China Lao PDR Papua Philippines Timor- Vietnam substantially across countries. Countries New Guinea Leste such as Cambodia, the Lao People’s Demo- cratic Republic, and Papua New Guinea still 1990 2008 have relatively low enrollment levels and low Source: World Health Organization (WHO) Global Health Observatory Data Repository. female-to-male enrollment ratios, particularly Note: M = male, F = female. The under-five mortality rate is defined as the probability of death at the secondary school level. Despite conver- derived from a life table and expressed as the rate per 1,000 live births. gence in enrollment among the young, sub- stantial gaps still remain in the educational countries such as China, Malaysia, Mongo- endowments of adult populations. At the lia, the Philippines, Thailand, Vanuatu, and same time, the gender gap in education has Vietnam (figure 1.5). Reverse gender gaps at reversed in several countries; girls’ secondary the tertiary level are sometimes even starker enrollment rates now exceed those of boys in (figure 1.6). 4 2    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE 1.11  Maternal mortality rates have declined across the experience lower educational enrollments. In world Vietnam, school participation among 15- to 17-year-olds is substantially higher among modeled estimate of maternal mortality the Kinh and Hoa (Chinese) majorities than 1,000 among many of the 52 ethnic minority popu- lations. Among the more economically dis- deaths per 100,000 live births 800 advantaged and less well integrated Hmong, Dao, and Khmer minorities, far fewer girls 600 attend school than boys (figure 1.7, and Baulch et al. 2002). 400 Globally, poverty and gender often interact to compound gender inequalities. Data sug- 200 gest that, in East Asia and the Pacific, poverty is not as important a contributor to gender 0 disadvantage in education as elsewhere in the 1990 1995 2000 2005 2008 world, however. Household survey data from East Asia and Paci c Europe and Central Asia several countries indicate that gender gaps in enrollment do not vary substantially or Latin America and the Caribbean Middle East and North Africa systematically across income quintiles. Indo- South Asia Sub-Saharan Africa nesia, for example, actually shows a slight female advantage in enrollments among 13- to Source: World Bank 2010b. 15-year-olds from poorer households, but a slight female disadvantage exists among the FIGURE 1.12  Maternal mortality rate has declined in most wealthiest households (figure 1.8). Overall, countries in the region data from the region suggest that gender gaps in enrollment tend to be smaller than enroll- modeled estimates of maternal mortality ment gaps across income or wealth groups. 1,400 Although gender disparities in enrollment 1,200 deaths per 100,000 live births have closed, education streaming persists. 1,000 Males and females differ in the types of edu- 800 cation obtained. Data from Indonesia, Thai- land, and Vietnam indicate that the fields of 600 engineering and law are dominated by males, 400 whereas the fields of education, health, and 200 business administration are dominated by 0 females.6 This “gender streaming� in edu- 1990 1995 2000 2005 2008 cation contributes to persistent inequalities Lao PDR Cambodia Timor-Leste Indonesia between women and men in access to eco- nomic opportunities. Papua New Guinea Philippines Vietnam Solomon Islands Growth and development in the region Mongolia China Malaysia during the past half century have also been Thailand Fiji associated with substantial improvements in key health indicators. Fertility rates have Source: World Bank 2010b. sharply declined, both in the region as well as across the world (figure 1.9). Under- Within countries, interactions between five mortality rates have more than halved gender and other socioeconomic charac- since 1990 for both boys and girls (figure teristics tend to exacerbate disparities in 1.10). Noteworthy gains have been made in education. For example, economically dis- birth attendance by health professionals. In advantaged and minority populations often 2006, 87 percent of births were attended by T H E S T A T E O F G E N D E R E Q U A L I T Y I N E A S T A S I A A N D T H E P A C I F I C    43 physicians, compared to 47 percent in 1992. number of missing girls at birth per year in Gains of this magnitude were not witnessed China, calculated by comparing the sex ratio at in any other region. In addition, the East Asia birth in China to those in high-income coun- and Pacific region has also seen declines in tries, increased from 890,000 in 1990 to the maternal mortality rate, from approxi- 1,092,000 in 2008 (World Bank 2011b). mately 200 deaths per 100,000 births in Men face gender-specific health risks as 1990 to 100 in 2008 (figure 1.11). The region well. For example, men are more likely to has experienced consistent increases in both experience higher morbidity and premature male and female life expectancy at birth since mortality related to substance abuse, war and 1960. Female life expectancy in the region conflict, and violence. Cambodia experienced has increased from 48 to 74, and male life considerable declines in the male population expectancy has increased from 45 to 70. during the Khmer Rouge regime of Pol Pot; As with education, progress in health so did Vietnam during its war era. Moreover, has been uneven across the region. Mater- differences between men and women in the nal mortality remains high in lower-income incidence of tobacco use are higher in East countries and in parts of the Pacific, espe- Asia and the Pacific than in other regions of cially in Cambodia, Lao PDR, Papua New the world; the gender differential in heavy Guinea, and Timor-Leste (figure 1.12). In Lao episodic alcohol consumption is also par- PDR, for example, maternal mortality rates ticularly stark. Male abuse of tobacco and were approximately 580 deaths per 100,000 alcohol in the region has important effects births7 in 2008, among the highest in the on men’s health and mortality rates, which world. Indonesia’s maternal mortality rate in turn can impose significant costs on eco- remains high compared to other countries in nomic productivity and growth. the region at similar levels of development. Gender disparities still exist in access to Women in remote rural areas commonly have and control of productive resources. Gender limited access to health care, which signifi- disparities in access to and control of land or cantly raises the risks associated with preg- farm inputs are pervasive around the world nancy and childbirth. Although Vietnam has, and remain issues in the region, despite sig- on average, experienced noteworthy declines nificant growth and development. Women in maternal mortality over the past decade, remain less likely to own land (or hold for- progress has been much slower in remote and mal land titles) than men. Moreover, data ethnic minority regions (World Bank 2011a). from Cambodia, China, Indonesia, Lao Moreover, more than a million girls and PDR, Mongolia, Timor-Leste, and Vietnam women per year are “missing� in East Asia. indicate that when women—or, specifically, Despite growth and development, the prob- female-headed households—do own land, lem of missing girls remains. The term missing they typically have smaller holdings. 8 A women was first coined by Sen (1992) to refer recent study of women’s land holdings in post- to the phenomenon that many low-income ­ tsunami Aceh similarly found that women’s countries have far fewer women than men, land holdings were considerably lower than relative to what is observed in developed coun- men’s (World Bank 2010).9 Female-headed tries. Sen argued that this imbalance in sex households in the region also tend to have ratios reflected severe forms of gender bias in poorer access to other productive inputs and affected societies. At birth, biological differ- support services, including livestock holdings ences between males and females imply that and access to agricultural extension services. approximately 105 boys are born for every Despite improvements in women’s access 100 girls. Nonetheless, China, Vietnam, and, to microcredit, important challenges remain until recently, the Republic of Korea have in accessing enterprise finance. Women also experienced substantial deviations from the have traditionally had less access to capital biological norm. Moreover, the trend over time than men. This disparity has been com- in China has been alarming (figure 1.13). The pounded by their poorer access to land, an 4 4    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE 1.13  East Asia has a highly skewed male-to-female important source of collateral. In response ratio at birth to gender disparities in access to credit, the microfinance movement has focused on ratio of the number of male to female live births, 1962–2008 increasing women’s access to capital across 1.25 the world. Of the 106.6 million poorest cli- China ents worldwide who have been reached by 1.20 microcredit initiatives by the end of 2007, 83.2 percent were women (Daley-Harris 1.15 2009). Evidence on access to credit among female and male entrepreneurs tells a more ratio nuanced story. Among micro- and small 1.10 firms in Indonesia, both female-and male- run enterprises cite access to finance as their 1.05 most significant business constraint, with the share of female-run firms ­ reporting this 1.0 constraint only slightly higher than the share 1962 1967 1972 1977 1982 1987 1992 1997 2002 2007 2008 of male-run firms (30 ­ versus 25 p ­ ercent, China Europe and Central Asia respectively). Among small and medium East Asia and Paci c Middle East and North Africa enterprises in nine East Asian and Pacific countries, only in Timor-Leste and Tonga Latin America and the Caribbean Sub-Saharan Africa do a greater share of ­ female-led enterprises South Asia ­ ignificantly more report access to credit as a s important constraint than reported by their Source: HNP Stats (Health Nutrition Population Statistics) database, 2010 data. male counterparts.10 Economic opportunity: Participation and returns in the economy FIGURE 1.14  The East Asia and Pacific region has high female The East Asia and Pacific region has the labor force participation rates highest average female labor force participa- tion rate and highest ratio of female-to-male labor force participation rate labor force participation in the developing of women 15–64 years of age, 2008 (%) world. In 2008, 70.1 percent of females were participating in labor market activi- East Asia and Paci c ties (figure 1.14) and the gender gap in labor Sub-Saharan Africa force participation was approximately 15 percentage points. In comparison, the aver- Europe and Central Asia age female labor force participation rate in Latin America and the Caribbean was 55 Latin America and percent in 2008, and in Europe and Central the Caribbean Asia it was 58 percent. Gender gaps in labor South Asia force participation were 27 and 16 percent- age points, respectively. Middle East and Both the levels of female participation North Africa and their trends over time vary substantially 0 20 40 60 80 100 in the region. In Cambodia, China, Lao female labor force participation rate (%) PDR, Thailand, and Vietnam, for example, female labor force participation was over Source: World Bank 2010b. 75 percent in 1980. Between 1980 and 2008, these countries witnessed declines in female T H E S T A T E O F G E N D E R E Q U A L I T Y I N E A S T A S I A A N D T H E P A C I F I C    45 participation of 5 to 10 percentage points FIGURE 1.15  Female labor force participation varies despite strong economic growth. Except for substantially across countries China, these declines were mirrored in simi- lar drops in male participation, leaving the Female labor force participation rate of ratio of female-to-male participation rates 15–64 years of age, 2008 (%) female labor force participation rate (%) unchanged. Participation rates in Korea, Indo- 90 nesia, and Malaysia were significantly lower 80 than the regional average: in 1980, only 45 70 percent of working-age females participated 60 in the labor market. Female participation 50 rates have increased over time in Korea and 40 Indonesia, while they have remained stagnant 30 in Malaysia despite strong economic growth. 20 Female labor force participation varies sub- 10 stantially in the Pacific, ranging from over 0 ilip sia Va odia Ko one s Sa Fiji re sia Tim To p. or nga on e Gu nd Vi inea m m na La atu DR M moa a N Th olia d e M Lest 75 percent in Vanuatu to 40 percent in Fiji Re na In pin Ca Chi ew aila Ph alay oP nu g b et a, - in 2008 ­ (figure 1.15). Although time series data for these countries are limited, female pu Pa participation rates appear to have risen over time in countries with lower initial rates. In Source: World Bank 2010b. Fiji, female participation rates rose from 25 percent in 1980 to 40 percent in 2008, and FIGURE 1.16  The fraction of the workforce employed in in Tonga they rose from 45 percent to 57 per- agriculture has declined in the East Asia and Pacific region cent over the same period. Structural transformation in the region’s share of total employment by sector economies has changed the type of work 100 conducted. The region has seen a signifi- 90 cant decline in the fraction of the workforce services in the agricultural sector over the past half 80 century, from approximately 60 percent in 70 1991 to just over 40 percent in 2008 (figure 60 1.16). Between 1960 and 2000, agriculture’s industry percent share of total employment declined from over 50 80 percent to under 50 percent in Thailand, 40 while in Indonesia the share declined from 70 30 percent to approximately 45 percent (Butzer, Mundlak, and Larson 2003). Nonetheless, 20 agriculture agriculture remains important: in 2008, it 10 was still the largest sector of employment in 0 the region (figure 1.16). 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 Women’s labor market responses to struc- 19 19 19 19 19 19 19 19 19 20 20 20 20 20 20 20 20 20 tural transformation have, in part, reflected Source: World Bank 2010b. country-specific patterns of development. Thailand, for example, moved from a heavy concentration of workers in agriculture in seen in the United States during the early 1980 to a rising employment share in the stages of the transition away from agricul- industrial and service sectors (figure 1.19). ture (Goldin 1995; Mammen and Paxson The early 1990s saw a substantial move- 2000). In I­ ndonesia, by contrast, female ment of females away from agriculture and labor force participation increased by out of the workforce. Similar patterns were 9 percentage points between 1980 and 2007, 4 6    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE 1.17  The evolution of sectoral composition by gender with the bulk of the increase coming from varies across countries the service sector (figure 1.17). Participation in industry and agriculture was fairly simi- lar in 1980 and 2007. percentage of total employment by sector and gender Labor market “sorting�—or employment 100 segregation—along gender lines is pervasive, by industry, occupation, formality, and flex- 80 ibility of employment. Such sorting affects a 60 number of economic outcomes, by gender, percent including earnings, social security coverage, 40 the intensity of work conducted, returns to education and experience, and exposure to 20 shocks. In many countries in the region, women are more likely than men to conduct 0 unpaid family labor in agriculture and in female male female male female male female male the informal sector (Asia Foundation et al. 1980 2007 1980 2007 2006; Asian Development Bank and World Thailand Indonesia Bank 2005; World Bank 2011a). In addi- agriculture industry tion, women are more likely to be found in services not working some occupations—such as teaching and nursing—and are less likely to be found in others, such as mining. Within the manufac- Source: World Bank 2010b. turing sector, women are more likely to be found in industries such as textiles and food processing, and are also found in large and export-oriented firms. Within firms, women are more likely than men to be temporary FIGURE 1.18  Women are more likely than men to be workers (figure 1.18). temporary workers Female- and male-led enterprises also tend to have distinct characteristics. Female-led % of temporary workers among male and female employees enterprises across the region tend to be smaller 40 than male-led enterprises (figure 1.19). They are more likely to operate in the informal sector (less likely to be registered) and to be 30 home based or operate out of nonpermanent premises. In general, female-led enterprises percent 20 across the region have fewer employees and assets and command lower profits. Although 10 female-owned and -managed enterprises are not inherently less productive, they tend to be less capitalized and located in less- 0 remunerative sectors. For instance, in Indo- Nonexp Nonexp Nonexp Nonexp Exporter Exporter Exporter Non exp Exporter Exporter nesia, female-led enterprises are more likely to locate in the food and garment manufac- Cambodia China Indonesia Thailand Vietnam turing sectors. These sectors are among the least capital intensive and least productive; female male male-led enterprises are more likely to locate in sectors such as transportation and other Source: World Bank estimates using Enterprise Surveys database, 2002–06 data. Note: Nonexp = nonexporter. The percentage of temporary workers is calculated as a ratio of female services, which are among the most capital temporary workers to female total workers. The analysis includes only manufacturing firms. intensive, most productive sectors. T H E S T A T E O F G E N D E R E Q U A L I T Y I N E A S T A S I A A N D T H E P A C I F I C    47 Women in East Asian and Pacific coun- FIGURE 1.19  Enterprises with female management tend to be tries continue to undertake the majority smaller of household work, in addition to market work, a function of longstanding gender percentage of small, medium, and large rms with a female manager norms regarding the division of labor within 40 households. In many contexts, this tradition 35 constrains women’s economic opportuni- ties, affecting their labor force participa- 30 tion, choice of sectors and occupations, time 25 worked, and, ultimately, earnings. Global percent 20 evidence indicates that women tend to work longer hours than men once both market 15 and nonmarket work are taken into account 10 (Ilahi 2000; World Bank 2001, 2011b). The composition of work also varies substantially 5 by gender. Men devote relatively more time 0 to market work, while women devote more m s sia ne na ne pi et time to domestic activities. Evidence from do ilip Vi In Ph East Asia is consistent with global patterns. small rm medium rm large rm Recent data from Lao PDR indicate that both gender differences in hours worked and in the Source: World Bank estimates using Enterprise Surveys database, 2006–11 data. division of labor between market and non- market work are exacerbated once families FIGURE 1.20  In Lao PDR, women—particularly those with have children (figure 1.20). young children—must balance household work commitments Cross-country evidence on wages indicates with market work that in the low- and middle-income countries of East Asia, women earn between 70 and 80 percent of what men earn for similar work 10 hours spent per day on household and market work (figure 1.21). These gender wage gaps are 9 partially attributable to differences in educa- 8 7 tion, experience, and industrial choice across 6 men and women. Differences in education hours 5 endowments, experience, and industrial and 4 occupational segregation explain up to 30 3 percent of the gender wage gap in East Asian 2 countries (Sakellariou 2011). 1 While gender wage gaps have evolved over 0 time, they have not always narrowed with without a child with a child without a child with a child growth and development. In Vietnam, the men women process of economic transition from a cen- trally planned to a market-based economy market work housework and family care has been associated with a sharp reduction in the gender pay gap among salaried employ- Source: World Bank estimates using the Lao Expenditure and Consumption Survey (LECS) (LSB Lao ees. The average gender pay gap halved PDR 2008). between 1993 and 2008, with the majority of the contraction evident by 1998 (Pham and contrast, in the Philippines, the wage gap Reilly 2007; Sakellariou 2011). Pay gaps still widened between 2000 and 2009, a change persist between men and women, however, that has been partly attributed to grow- with women earning on average 75 percent ing differences between men and women in of the male wage in 2009 (Pierre 2012). By terms of their returns to education and other 4 8    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C characteristics (Sakellariou 2011). In Indo- recent evidence suggests that the gender gap nesia, the average wage gap and the gap by declined between 1997 and 2009 (Sakellar- age cohorts widened between 1976 and 1999 iou 2011). This complex relationship between (Dhanani and Islam 2004), although more growth, development, and the gender wage gap reflects a number of factors at the coun- try level, including changes in the structure of FIGURE 1.21  Women in East Asia still earn less than men the economy and in labor demand; the gen- der composition of the labor force; changes ratio of female to male wages, 2007 in the relative education, skill, and experi- ence among male and female workers; and Mongolia 0.81 labor market institutions and policies, each Thailand 0.79 of which affect how remuneration is evolving by gender. Philippines 0.76 Cambodia 0.75 Agency: Women’s voice and influence East Asian and Pacific countries have expe- Vietnam 0.71 rienced progress and pending challenges Indonesia 0.70 in achieving gender equality in agency — women’s voice and influence —as with China 0.70 endowments and economic opportunity discussed above. Agency refers to the abil- Korea, Rep. 0.52 ity of women and men to take action on 0 0.2 0.4 0.6 0.8 1.0 their own behalf, in accordance with their ratio preferences, and to influence outcomes that affect them in both private and public Source: World Bank 2010b. domains (box 1.2). The ability to influence BOX 1.2  Defining and measuring agency: Women’s voice, influence, and participation All individuals in a society have ideas and prefer- within the household (the extent to which her voice ences on how to use scarce resources and live their is heard) and, potentially, her preferences. lives. The ability of women and men to express The concept of agency has multiple dimensions and act on their preferences is affected by—and that have been measured empirically. The form also affects—their ownership and control over of agency that is most frequently measured is the ­ endowments and their access to economic oppor- decision-making power of men and women (Mason tunities (Kabeer 1999). This ability to act on one’s 2005; McElroy 1990). Agency may be more explic- preferences and influence outcomes is referred to as itly measured by examining women’s mobility in the agency. Changes in individuals’ characteristics in public domain, their participation in public action, a household, community, or society will affect the and the incidence of gender-based violence (Kabeer strength of their voices relative to others, and may 1999). Finally, gender differences in bargaining also influence their preferences. This influence, in power within a household have been assessed by turn, will affect choices made at the household, com- examining the extent to which people’s choices munity, and societal levels. For example, an increase change when factors affecting their bargaining in the education of a woman will affect investments power, such as education, relative earnings, or asset in the education and health of her children through holdings, also change (Duflo 2003; Quisumbing and its effect on both her relative bargaining power Maluccio 2003; Thomas 1990, 1995). T H E S T A T E O F G E N D E R E Q U A L I T Y I N E A S T A S I A A N D T H E P A C I F I C    49 one’s life by making choices and taking FIGURE 1.22  Women in East Asia and the Pacific have more action is also a key dimension of well-being control over earnings and household decisions across all wealth in and of itself (Sen 1999). This report quintiles than women from other developing regions focuses on agency in three domains: (a) agency in the household and individual visit to decisions, examined through household large purchases family/relatives owning earnings 100 decision making, control of resources, and share of women with some control reproductive decisions; (b) agency in the 90 public sphere, examined through gender- over decisions (%) based participation and representation in 80 the private sector, civil society, politics, 70 and public institutions; and (c) safety and security as an expression of agency, exam- 60 ined through the prevalence of gender- based violence. Although the East Asia and 50 Pacific region has made progress in sev- 40 eral domains of agency, progress has been Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 Q1 Q2 Q3 Q4 Q5 uneven across countries, and many chal- East Asia and Paci c Europe and Central Asia lenges still remain. Latin America and the Caribbean Middle East and North Africa Women in East Asia experience relatively high levels of agency at the household level South Asia Sub-Saharan Africa compared with other developing regions. Cross-country data suggest that women Source: World Development Report 2012 team estimates based on Demographic and Health Surveys of 40 countries, 2003–09. across the wealth distribution in East Asia have a greater say in decisions regarding large household purchases and also experi- ence greater freedom to visit family and rela- tives without husbands’ permission than do women in other developing regions. Women in East Asia also have as great as or greater FIGURE 1.23  Female representation in parliament in control over their own earnings compared East Asian and Pacific countries has hardly changed since with women in other developing regions the 1990s ­ (figure 1.22). Reductions in fertility rates and unwanted fertility, defined as gaps between percentage of parliamentary actual and desired fertility, observed in most 25 seats held by women East Asian and Pacific countries suggest that 20 women may have gained greater control over their reproductive decisions. 15 percent Women’s voice in the public domain 10 remains weak despite economic development. 5 Women have relatively low levels of represen- tation in political assemblies, whether at the 0 national or local levels. For example, women c sia a ia ca an ric As ri ci be lA Af Af Pa make up only one-fifth of national parlia- h rib ra ut rth n d nt ra Ca So an No Ce ha mentarians worldwide. The share of female e ia th Sa nd nd As b- d a ta parliamentarians in East Asian and Pacific an st Su pe Ea as ica ro eE countries is just below the global average, Eu er dl Am id at approximately 18 percent in 2011 (figure M tin 1997 2002 2007 2011 La 1.23). Female representation in the region has not increased over time with economic Source: Inter-Parliamentary Union data extracted from World Bank 2011c. 5 0    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE 1.24  Women’s representation in parliament is low, especially in the Pacific percentage of seats held by women in the lower or single house, 2011 45 40 35 30 25 percent 20 15 10 5 0 . P rea an G s  rsh an ea T ar m b ia Sin Chi a ilip re La nam P a mo P r u a N Isla u ng a Sa olia Tu a Kir alu lay i em Ko Japsia ’s . on d ga n a Vie ines rd r-L R co este wo es rld N a ts. an s Is u Th Rep. ple ep Ma ibat M y l and mo i Mo ong pu n ala No Timo o PD all uat Ind ilan e w nd od Ca es m P h po u tri S Ma V uin v eo , R d. p un t a Fe, sia ic ne lo cro So Mi ,D rea Ko Source: PARLINE database, 2011 data. FIGURE 1.25  Men, and in some cases women, believe that men growth and development; the overall share make better political leaders than women of female parliamentarians in East Asian and Pacific countries has barely changed since percentage of men and women aged over 30 who think men the late 1990s. This trend stands in contrast 80 make better political leaders to other developing regions, where levels of 70 female political representation have tended to 60 increase, at least since 2000.11 50 The share of women in national parlia- ments varies tremendously across the region. percent 40 In December 2011, the highest levels of 30 female representation were found in Timor- 20 Leste (32.3 percent), Lao PDR (25.2 per- 10 cent), Vietnam (24.4 percent), the Philippines 0 (22.9 percent), and China (21.3 percent) ­ (figure 1.24). Female parliamentary represen- a sia ia d m at d ay in an St ite ys na rw ne Ch tation in the Pacific is systematically lower. es ala ail Un et No do Th Vi M In male female Four countries in the Pacific—the Federated States of Micronesia, Nauru, Palau, and Source: World Values Survey, 2006 and 2007 data. the Solomon Islands—do not have female representation in parliament (PARLINE database, 2011 data). Countries that have enacted temporary special measures to T H E S T A T E O F G E N D E R E Q U A L I T Y I N E A S T A S I A A N D T H E P A C I F I C    51 promote female participation in political participation in management in the region leadership have achieved higher levels of and all other regions is lower than participa- female representation in parliament, with tion in ownership, however, indicating that an average of 27 percent in 2009, compared women have a more limited voice in running with 14 percent for those countries without a firm and making decisions, despite being temporary special measures. represented in the ownership of the firm. Barriers to political representation in the The incidence of physical, sexual, psy- region are likely to reflect perceptions held chological, and emotional violence against by both men and women that female politi- women is high throughout the region, par- cians make less competent political leaders ticularly in the Pacific, where the prevalence than their male counterparts (figure 1.25). of domestic violence is among the highest in Evidence from India suggests that people’s the world (figure 1.27). Human trafficking perceptions of women as political leaders is also a growing concern throughout the improve with exposure to elected female offi- region. An increase in female migration in the cials (Beaman et al. 2009). However, these past decade has brought about increased eco- perceptions may take time to evolve, reinforc- nomic opportunities as well as increased risk ing the case for concerted action to support of being trafficked. The ILO estimates that increased political participation on the part Asia and the Pacific account for over half of of women. Ensuring that women are repre- all trafficked victims worldwide, with an esti- sented, whether at the local or national level, mated 1.36 million victims, most of whom is a first step to ensuring that their views and are women and girls (ILO 2008). preferences are expressed in public policies. Gender-based violence and human traf- Women from East Asian and Pacific ficking of women and girls are often enabled countries have experienced improvements by the lack of and weak enforcement of rel- in their voice and ability to actively partici- evant legislation. Although the factors that pate in civil society and grassroots move- enable gender-based violence are multiple ments. Civil rights groups tackle a variety and complex, the phenomenon is exacerbated of issues to improve gender equality in their by a lack of adequate legal protections in community, country, and region. PEKKA, a many countries, most notably in the Pacific program in Indonesia, was created to address islands. A recent United Nations Develop- the needs of widows of conflict in Aceh and ment Programme study (UNDP 2010) found now provides training for village paralegals that although more than three-quarters of that focuses on domestic violence and family the countries in East Asia have strength- law. The program also holds district forums ened legislation on domestic abuse in recent to bring together judges, prosecutors, police, years, more than 60 percent of countries in nongovernmental organizations, and govern- the Pacific still lacked relevant legislation on ment officials to raise awareness of gender domestic violence. Police forces in several issues (PEKKA 2012; World Bank 2011b). countries in the region—including Indonesia, As another example, the Fiji Women’s Crisis M alaysia, and Thailand—have been criti- ­ Center lobbied for a nationally representa- cized for ­limited commitment in investigating tive quantitative survey on violence against trafficking and enforcing antitrafficking laws women, to be released in 2012. (U.S. D­ epartment of State 2011). Women’s representation in top manage- ment and participation in ownership is high relative to other developing regions and the Fostering new opportunities, world average. In the East Asia and Pacific managing emerging risks region, women are represented among the As the region continues to develop, sev- owners in over 50 percent of small, medium, eral emerging trends will present both new and large firms, higher than in any other opportunities and new risks to achieving developing region (figure 1.26). Female gender equality. As noted earlier, East Asian 5 2    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE 1.26  More women hold top management positions in (e) rising availability and use of information East Asia and the Pacific than in other developing regions and communication technologies (ICTs). Though not commonly accounted for by policy makers, these trends are likely to have East Asia and the Paci c important gender dimensions and, thus, affect the evolution of gender equality in the region. Europe and Central Asia They will create new challenges for public policy, in terms of fostering new opportunities Latin America and the Caribbean and managing new risks to gender equality. world Greater global economic integration brings with it substantial economic opportunities Middle East and North Africa for women, but also potentially greater expo- sure to economic shocks. For example, evi- Sub-Saharan Africa dence indicates that greater global economic South Asia integration can contribute to greater access to economic opportunities for women and 0 10 20 30 40 50 60 reduce gender wage gaps by raising demand firms with female top manager (%) for female workers in export-oriented enter- firms with female participation prises (Boserup 1970; Schultz 2006). At the in ownership (%) same time, increasing global integration also increases risk and uncertainty, as shocks are Source: Enterprise Surveys database, World Bank. more quickly transmitted across integrated markets. A number of studies find that FIGURE 1.27  Violence against women is high in East Asia and although shocks do not necessarily have more the Pacific adverse effects on women than men, they do have gender-differentiated effects on outcomes percentage of ever-married women ages 15–49 who have as diverse as employment, earnings, labor force experienced sexual and/or physical intimate-partner violence participation, education investment, health, 80 70 and nutrition (see, for example, Bruni et al. 60 2011; Rodgers and Menon forthcoming). 50 Opportunities gained by migration are bal- percent 40 anced by new risks. High economic growth 30 and increased economic integration in the 20 region over the past three decades has spurred 10 0 significant migration across the region. Women make up nearly half of all migrants Thailand Cambodia Philippines Japan Kiribati Marshall Islands Kenya Peru Nepal Moldova Samoa Solomon Islands in East Asia and the Pacific and in some coun- tries represent the majority of new migrants. Migration can provide both women and men with access to new economic opportunities, East Asia Paci c other developing which commonly differ by gender because of countries employment segregation in receiving areas’ labor markets. At the same time, migration Source: Demographic and health surveys, various countries and years, and government surveys. brings with it important gender-specific risks. For example, many female migrants find and Pacific countries are at the forefront of employment as domestic workers, a sector several global trends: (a) increasing global with particularly weak worker protection economic integration, (b) increasing domes- in most countries. Female migrants are also tic and cross-border migration , (c) rapid disproportionately susceptible to risks associ- urbanization , (d) population aging, and ated with human trafficking. T H E S T A T E O F G E N D E R E Q U A L I T Y I N E A S T A S I A A N D T H E P A C I F I C    53 Unprecedented levels of urbanization are Advances in ICTs are revolutionizing the affecting all aspects of life: from the nature ways in which both men and women in the of employment to the availability of services region are exposed to ideas, share knowl- to individuals’ ability to rely on extended edge, and networks. Existing evidence sug- family and community networks. Many gests, however, that women in the region still East Asian and Pacific countries are experi- have lower access to information technologies encing unprecedented levels of urbanization than men. New and emerging technologies, as migrants move to urban areas in search of if accessible, can help to empower women economic opportunity. Between 2000 and by opening new economic opportunities, 2015, Indonesia, China, and Cambodia are breaking down information barriers, help- predicted to see an increase in the percentage ing women in isolated communities engage in of the population residing in urban areas by distance learning or commerce, or enabling 17, 13, and 9 percentage points respectively women to take collective action. As in the (World Bank 2009). These changes almost case of other productive resources, such as certainly have gender-­ s pecific impacts. land, machinery, or credit, growing gender Although urban residence can open up a gaps in access to ICTs could lead to widening wide range of economic opportunities for gaps in access to economic opportunities as both men and women, women’s ability to well as in voice and influence in society. take advantage of new opportunities is likely to depend fundamentally on the nature and availability of urban services—for example, Gender equality in East Asia and whether transport systems facilitate safe the Pacific: A roadmap to the travel of women to job sites or if affordable report child care can compensate for the loss of This chapter has provided a basic profile on extended family networks. the status of gender equality in East Asia Aging populations will represent another challenge for women. The high-income economies in East Asia are experiencing FIGURE 1.28  The dependency ratio is increasing in most East rapid population aging.12 Most emerging Asian countries countries in the region have also begun this process; dependency ratios are already increasing in many middle-income coun- 0.45 old-age dependency ratio tries in East Asia and the Pacific. Old-age 0.40 dependency is expected to increase even 0.35 more quickly in the coming decades (fig- 0.30 ure 1.28). Population aging is likely to have 0.25 ratio gender-differentiated impacts at all age lev- 0.20 els. Gender differences in the time devoted 0.15 to caring for the elderly imply that in the 0.10 absence of institutionalized care services, 0.05 women are likely to bear the brunt of the 0.00 1990 2000 2010 2020 2030 2040 2050 increased demand for elder care (Dwyer and Coward 1992; Ofstedal, Knodel, and China Fiji Indonesia Chayovan 1999). Moreover, while women Malaysia Mongolia Papua New Guinea tend to live longer than men, gender dif- Philippines Thailand Vietnam ferences in education and labor force par- ticipation imply that women are less likely to be vested in formal pension systems and Sources: World Bank 2010b (1990–2010 data); HNP Stats—Population Projection (projections for 2020–50). may have fewer assets to ensure a basic Note: Old-age dependency ratio is defined as the ratio of the population 65 years of age and older level of well-being in old age. over working-age population (15–64 years of age). 5 4    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C and Pacific countries, taking into account in Pacific countries experience among the several important factors that have charac- highest rates of violence against women in terized development in the region. Specifi- the world; they also have among the low- cally, the region has experienced significant est levels of female representation in poli- growth, poverty reduction, and economic tics. Within countries, gender frequently structural transformation over the past sev- interacts with other socioeconomic char- eral decades. The chapter has shown that acteristics, such as ethnicity or geographic rapid growth and development have been remoteness, resulting in specific subgroups accompanied by reduced gender inequali- of countries’ populations facing a double ties in several key dimensions, most nota- disadvantage. bly in education and several key aspects of Because many aspects of gender inequality health. In East Asian countries, women’s do not disappear automatically with growth voice and influence, whether in the home and development, and because persistent gen- or in the economy, are relatively strong. In der inequalities impose high costs on women many ways, women in the region are better and girls and on societies more broadly, a positioned than ever before to participate case can be made for public policy to promote in, contribute to, and benefit from their gender equality. In this context, an impor- economies and societies. tant contribution of this report will be to (a) At the same time, the evidence demon- clarify empirically the relationship between strates clearly that economic growth and gender equality and development, (b) ana- development alone are not enough to attain lyze the factors that contribute to or impede gender equality in all its dimensions. While gender equality in its different dimensions, gender outcomes in education and health and (c) identify effective avenues for public have been responsive to growth, other areas action to promote gender equality and, thus, have proved “stickier�; significant gender more effective development in East Asian and inequalities persist in a number of important Pacific countries. areas despite development. Women still have To achieve these objectives, the three chap- less access than men to a range of productive ters that follow focus on providing a deeper assets and services. There remains substan- understanding of the factors affecting gender tial employment segregation, by gender. And equality in endowments, economic oppor- despite closing of education gaps, women tunity, and agency. Specifically, chapter 2 continue to earn less than men. Moreover, examines in more depth the evidence on gen- women in the region still have weaker voice der dimensions of human and physical capital and influence than men, whether in house- accumulation. Chapter 3 analyzes access to hold decision making, in the private sector, in economic opportunity, including the ­ factors civil society, or in politics. And women across affecting female labor force participation, the region remain vulnerable to gender-based employment segregation across o ­ ccupations violence. and industries, and persistent gender gaps in Progress toward gender equality has wages and earnings. And chapter 4 focuses been uneven across and within countries. on factors that enhance or constrain women’s Despite widespread progress, a few, mostly voice and influence in society, both in the low-income, countries have yet to close private and public domains. In carrying out gender gaps in basic education. In China their analyses, each of these chapters seeks to (and to a lesser extent Vietnam), significant frame an agenda for effective public action imbalances occur in the sex ratio at birth, moving forward. ­ r eflecting strong son preference in those The report also analyzes the gender dimen- societies. At the same time, a number of sions of several important emerging trends Pacific island countries face particular chal- in the region. Specifically, chapter 5 exam- lenges with respect to promoting women’s ines globalization and economic integration, voice and influence. Specifically, women increasing migration, rapid urbanization, T H E S T A T E O F G E N D E R E Q U A L I T Y I N E A S T A S I A A N D T H E P A C I F I C    55 population aging, and enhanced access to men are more likely to be found among math- ICTs, while identifying both the emerging ematics and computer science graduates than opportunities and the emerging risks to gen- women and have also been found to outperform der equality that these phenomena entail. women in mathematics (Schleicher 2008). The maternal mortality ratio (MMR) in Lao   7.  Building on the in-depth analyses presented PDR is a modeled estimate, to make it com- in chapters 2 through 5, chapter 6 then parable to MMR estimates in other countries. outlines directions for public policy to pro- Lao PDR’s national estimate in 2005 was mote gender equality and, thus, more effec- lower, at 405 deaths per 100,000 births (WDI tive development in East Asian and Pacific database). countries. The report concludes by framing For data on land holdings by gender in China,   8.  a forward-looking agenda for analysis and see de Brauw et al. (2011). Data on other action—to continue to fill knowledge gaps countries are based on World Bank staff cal- and strengthen public policy responses to culations, using household survey data. promoting gender equality in the region. Evidence from other parts of Indonesia sug-   9.  gests that land ownership patterns, by gender, can differ in important ways, depending on Notes local norms and customs. In the matrilineal region of West Sumatra, Indonesia, for exam-  1. Sen (1999) defines freedoms and “unfree- ple, at the time of marriage, husbands com- doms� in five categories: (a) political freedoms, monly own more forest land than their wives, (b) economic facilities, (c) social opportunities, and wives commonly own more paddy land (d) transparency guarantees, and (e) protective (Quisumbing and Maluccio 2003). security. 10. Self-reported information on credit con-  2. Adopted in 1979 by the United Nations straints, by gender, should be interpreted with General Assembly, CEDAW is often referred caution. As discussed below, female-led enter- to as the international bill of rights for prises are smaller, use less capital, and operate women. The convention defines what con- in difference sectors than male-led enterprises, stitutes discrimination against women and making direct comparisons of self-reported provides an agenda for national action to end credit constraints, by gender, difficult. such discrimination. To date, it has been rati- 11.  In the Europe and Central Asia region, female fied by 187 countries worldwide (http://www representation in national assemblies fell sub- .un.org/womenwatch/daw/cedaw/). stantially following the dissolution of the These studies must be interpreted with caution,   3.  Soviet Union, although levels increased again given the difficulty of establishing a causal rela- between 2000 and 2008. tionship between gender equality in education 12.  For example, the share of the elderly (age 65 and growth in cross-country studies. or above) in Hong Kong SAR, China; Japan; Evidence from Africa and Latin America, for   4.  Korea; Singapore; and Taiwan, China, is example, suggests that ensuring equal access above 10 percent and is expected to increase to productive assets and technologies could substantially in the next two decades. significantly raise agricultural production and household income (Goldstein and Udry 2008; Quisumbing 1995; Udry 1996).   5.  A number of studies show that gender-based References violence itself imposes significant costs on the Agarwal, Bina. 2010a. “Does Women’s Propor­ economies of developing countries, for exam- tional Strength Affect Their Participation? ple, through lower worker productivity and Governing Local Forests in South Asia.� World incomes, lower human capital investments, Development 38 (1): 98–112. and weaker accumulation of social capital ———. 2010b. Gender and Green Governance: (see Morrison, Ellsberg, and Bott 2007, for a The Political Economy of Women’s Presence review of key findings). For related evidence Within and Beyond Community Forestry. from the United States, see National Center New York: Oxford University Press. for Injury Prevention and Control 2003. Asia Foundation, ADB (Asian Development Bank),  6. Similarly, in Organisation for Economic Co- CIDA (Canadian International Development operation and Development (OECD) countries, Agency), NDI (National Democratic Institute), 5 6    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C and World Bank. 2006. Indonesia: Country of Agriculture with Chinese Characteristics.� Gender Assessment. Manila, Philippines: Asia Background paper for World Development Foundation, ADB, CIDA, NDI, and World Report 2012, World Bank, Washington, DC. Bank. Dhanani, Shafiq, and Iyanatul Islam. 2004. Asian Development Bank and World Bank. 2005. Indonesian Wage Structure and Trends, 1976– “Mongolia: Country Gender Assessment.� 2000. International Labour Organization: Countr y Planning Docu ment, A sian Geneva. Development Bank, Manila, Philippines. Do, Quy-Toan, Andrei Levchenko, and Claudio AusAID. 2008. Making Land Work: Vol. 1. Raddatz. 2011. “Engendering Trade.� Policy Reconciling Customary Land and Development Research Working Paper 5777, World Bank, in the Pacific. Canberra: Australian Agency for Washington, DC. International Development. Duf lo, Esther. 20 03. “Grandmothers and Baulch, Bob, Truong Thi Kim Chuyen, Dominique Granddaughters: Old-Age Pensions and Haughton, and Jonathan Haughton. 2002. Intrahousehold Allocation in South Africa.� “Ethnic Minority Development in Vietnam: World Bank Economic Review 17 (1): 1–25. A S o cio econom ic Persp ec t ive.� Pol ic y Dwyer, Jeffrey, and Raymond Coward. 1992. Research Working Paper 2836, World Bank, “Gender and Family Care of the Elderly: Washington, DC. Research Gaps and Opportunities.� In Gender, Beaman, Lori, Raghabendra Chattopadhyay, Families, and Elder Care, edited by Jeffrey Esther Duf lo, Rohini Pande, and Petia W. Dwyer and Raymond T. Coward, 151–62. Topalova. 2009. “Powerful Women: Does London: Sage. Exposure Reduce Bias?� Quarterly Journal of Edlund, Lena, and Rohini Pande. 2001. “Why Have Economics 124 (4): 1497–540. Women Become Left-Wing? The Political Gender Beegle, Kathleen, Elizabeth Frankenberg, and Gap and the Decline in Marriage.� Quarterly Duncan Thomas. 2001. “Bargaining Power Journal of Economics 117 (3): 917–61. within Couples and Use of Prenatal and Enterprise Surveys (database). International Delivery Care in Indonesia.� Studies in Family Finance Corporation and the World Bank. Planning 32 (2): 130–46. http://www.enterprisesurveys.org/. Boserup, Ester. 1970. Woman’s Role in Economic Esteve-Volart, Berta. 2004. “Gender Discrimination Development. London: Earthscan. and Growth: Theory and Evidence from India.� Bruni, Lucilla, Andrew Mason, Laura Pabon, and Development Economics Discussion Paper Carrie Turk. 2011. “Gender Impacts of the Series 42, Suntory and Toyota International Global Financial Crisis in Cambodia.� World Centres for Economics and Related Disciplines, Bank, Washington, DC. London School of Economics and Political Butzer, Rita, Yair Mundlak, and Donald Larson. Science, London. 2003. “Intersectoral Migration in Southeast FAO (Food and Agriculture Organization)/Sida Asia: Evidence from Indonesia, Thailand and Partnership Cooperation. 2010. National the Philippines.� Policy Research Working Gender Profile of Agricultural Households, Paper 2949, World Bank, Washington, DC. 2010: Report Based on the Lao Expenditure Chattopadhyay, Raghabendra, and Esther Duflo. a n d C o n s u m pt i o n S u r ve y s , N a t i o n a l 2004. “Women as Policy Makers: Evidence Agricultural Census and Population Census. from a Randomized Policy Experiment in Vientiane, Lao PDR: FAO. India.� Econometrica 72 (5): 1409–43. Fehringer, Jessica, and Michelle J. Hindin. 2009. Constable, Nicole. 2009. “Migrant Workers and “Like Parent, Like Child: Intergenerational the Many States of Protest in Hong Kong.� Transmission of Partner Violence in Cebu, the Critical Asian Studies 41 (1): 143–64. Philippines.� Journal of Adolescent Health 44 Cuberes, David, and Marc Teignier-Baqué. 2011. (4): 363–71. “Gender Inequality and Economic Growth.� G old i n , C laud ia. 1995. “T he U - S h ap ed Background paper for World Development Female Labor Force Function in Economic Report 2012, World Bank, Washington, DC. Development and Economic History.� In Daley-Harris, Sam. 2009. “State of the Microcredit Investment in Women’s Human Capital, Summit Campaign Report 2009.� Microcredit edited by T. Paul Schultz. Chicago: University Summit Campaign, Washington, DC. of Chicago Press. de Brauw, Alan, Jikun Huang, Linxiu Zhang, Goldstein, Markus, and Christopher Udry. and Scott Rozelle. 2011. “The Feminization 2008. “The Profits of Power: Land Rights and T H E S T A T E O F G E N D E R E Q U A L I T Y I N E A S T A S I A A N D T H E P A C I F I C    57 Agricultural Investment in Ghana.� Journal of Scope of Government?� Journal of Political Political Economy 116 (6): 981–1022. Economy 107 (6): 1163–98. General Statistics Office of Vietnam. 2009. Mammen, Kristin, and Christina Paxson. 2000. R e s u l t o f t he S u r v e y o n H o u s e ho l d “Women’s Work and Economic Development.� Living Standards 2008. Hanoi: Statistical Journal of Economic Perspectives 14 (4): Publishing House. http://www.gso.gov.vn 141–64. /default_en.aspx?tabid=515&idmid=5&Item Mason, Karen Oppenheim. 2005. “Measuring ID=9647. Women’s Empowerment: Learning from H N P Stats (Health Nutrition Population Cross-National Research.� In Measuring Statistics) database. World Bank, Washington, Empowerment: Cross-Disciplinar y DC. http://data.worldbank.org/data-catalog/ Perspectives, edited by D. Narayan, 89–102. health-nutrition-population-statistics. Washington, DC: World Bank. Ilahi, Nadeem. 2000. “The Intra-household McElroy, Marjorie. 1990. “The Empirical Allocation of Time and Tasks: What Have We Content of Nash-Bargained Household Learnt from the Empirical Literature?� Working Behavior.� Journal of Human Resources 25 Paper 13, World Bank Development Research (4): 559–83. Group, World Bank, Washington, DC. Morrison, Andrew, Mary Ellsberg, and Sarah ILO (International Labour Organization). 2008. Bot t. 20 07. “Addressing G ender-Based Women’s Entrepreneurship Development Violence: A Critical Review of Interventions.� Aceh: Gender and Entrepreneurship Together World Bank Observer 22 (1): 25–51. (GET Ahead) Training Implementation. Naudeau, Sophie, Naoko Kataoka, Alexandria Impact Assessment. Jakarta, Indonesia: ILO Valerio, Michelle Neuman, and Leslie Elder. Jakarta Office. 2011. Investing in Young Children: An Early Indonesia National Socioeconomic Survey Childhood Development Guide for Policy (SUSENAS). 2009. Badan Pusat Statistik, Dialogue and Project Preparation. Washington, Jakarta, Indonesia. http://dds.bps.go.id/eng/ DC: World Bank. aboutus.php?id_subyek=29&tabel=1&fl=3. National Center for Injury Prevention and Johnson, Kiersten, Sovanratnak Sao, and Darith Control. 2003. Costs of Intimate Partner Hor. 2000. Cambodia 2000 Demographic Violence against Women in the United States. and Health Survey: Key Findings. Calverton, Atlanta, GA: Centers for Disease Control and MD: ORC Macro. Prevention. Judd, Ellen R. 2010. “Family Strategies: Fluidities NIPH (National Institute of Public Health of Gender, Community and Mobility in Rural [Cambodia]), N IS (National Institute of West China.� China Quarterly 204: 921–38. Statistics [Cambodia]), and ORC Macro. 2006. Kabeer, Naila. 1999. “Resources, Agency, Cambodia Demographic and Health Survey Achievements: Reflections on the Measurement 2005. Phnom Penh, Cambodia, and Calverton, of Women’s Empowerment.� Development Maryland, U.S.: National Institute of Public and Change 30 (3): 435–64. Health, National Institute of Statistics, and Klasen, Stephan. 2002. “Low Schooling for ORC Macro. Girls, Slower Growth for All? Cross-Country NSD (National Statistics Directorate [Timor- Evidence on the Effect of Gender Inequality in Leste]), Ministry of Finance [Timor-Leste], and Education on Economic Development.� World ICF Macro. 2010. Timor-Leste Demographic Bank Economic Review 16 (3): 345–73. and Health Survey 2009–10. Dili, Timor- Knowles, Stephen, A. K. Lorgelly, and Dorian Leste: NSD and ICF Macro. Owen. 2002. “Are Educational Gender Gaps Ofstedal, Mary Beth, John E. Knodel, and a Brake on Economic Development? Some Napaporn Chayovan. 1999. “Intergenerational Cross-Country Empirical Evidence.� Oxford Support and Gender: A Comparison of Four Economic Papers 54 (1): 118–49. Asian Countries.� Southeast Asian Journal of Lao Statistics Bureau. 2008. Lao Expenditure Social Sciences 27 (2): 21–41. and Consumption Survey. Vientane, Lao PDR. PARLINE database on national parliaments. http://www.nsc.gov.la/index.php?option=com_ Inter-Parliamentary Union, Geneva. http:// content&view=article&id=50&Itemid=73& www.ipu.org/parline-e/parlinesearch.asp and lang=. http://www.ipu.org/wmn-e/world.htm. Lott, John R., and Lawrence W. Kenny. 1999. PEKKA (Perempuan Kepala Keluarga [Women “Did Women’s Suffrage Change the Size and Headed Household Empowerment Program]). 5 8    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C Accessed 2012. http://www.pekka.or.id/8/ Development Report. Washington, DC: World index.php. Bank. Pham, Hung T., and Barry Reilly. 2007. “The Schleicher, Andreas. 2008. “Student Learning Gender Pay Gap in Vietnam, 1993–2002: A Outcomes in Mathematics from a Gender Quantile Regression Approach.� Journal of Perspective: What Does the International PISA Asian Economics 18 (5): 775–808. Assessment Tell Us?� In Girls’ Education in Pierre, Gaelle. 2012. “Recent Labor Market the 21st Century, edited by M. Tembon and L. Performance in Vietnam through a Gender Fort. Washington, DC: World Bank. Lens.� Policy Research Working Paper 6056, Schultz, Paul. 2006. “Does the Liberalization of World Bank, Washington, DC. Trade Advance Gender Equality in Schooling PovcalNet: An Online Poverty Analysis Tool. and Health?� Discussion Paper 935, Economic World Bank, Washington, DC. http://iresearch Growth Center, Yale University, New Haven, .worldbank.org/PovcalNet/index.htm. CT. Qian, Nancy. 2008. “Missing Women and the Sen, Amartya. 1992. “Missing Women.� British Price of Tea in China: The Effect of Sex-Specific Medical Journal 304: 587–88. Earnings on Sex Imbalance.� Quarterly ———. 1999. Development as Freedom. Oxford, Journal of Economics 123 (3): 1251–85. U.K.: Oxford University Press. Quisumbing, Agnes. 1995. “Gender Differences SPC (Secretariat of the Pacific Community), in Agricultural Productivity: A Survey of Ministry of Internal and Social Affairs Empirical Evidence.� FCND Discussion Paper [Republic of Kiribati], and Statistics Division No. 5, International Food Policy Research [Republic of Kiribati]. 2010. Kiribati Family Institute, Washington, DC. Health and Support Study: A Study on Quisumbing, Agnes R., and John A. Maluccio. Violence against Women and Children. 2003. “Resources at Marriage and Intra- Nouméa, New Caledonia: SPC. hou s ehold A l lo c at ion : Ev idenc e f rom SPC (Secretariat of the Pacific Community) and Bangladesh, Ethiopia, Indonesia, and South NSO (National Statistics Office) [Solomon Africa.� Oxford Bulletin of Economics and Islands]. 2009. Solomon Islands Family Statistics 65: 283–327. Health and Safety Study: A Study on Violence Quota Project: Global Database of Quotas for against Women and Children. Nouméa, New Women. Institute for Democracy and Electoral Caledonia: SPC. Assistance and Stockholm University. “About Thomas, Duncan. 1990. “Intra-household Quotas.� ht tp: //w w w.quotaproject.org / Resource Allocation: An Inferential Approach.� aboutQuotas.cfm. Journal of Human Resources 25 (4): 635–64. Rand, John, and Finn Tarp. 2011. “Does Gender ———. 1995. “Like Father, Like Son, Like Influence the Provision of Fringe Benefits? Mother, Like Daughter, Parental Resources Evidence from Vietnamese SMEs.� Feminist and Child Height.� Papers 95-01, RAND Economics 17 (1): 59–87. Reprint Series. Santa Monica, CA: RAND Rodgers, Yana, and Nidhiya Menon. Forthcoming. Corporation. “Impact of the 2008–2009 Twin Economic Udry, Christopher. 1996. “Gender, Agricultural Crises on the Philippine Labor Market.� World Production, and the Theory of the Household.� Development. Journ al of Politic al Economy 104 (5): Ross, Michael. 2008. “Oil, Islam and Women.� 1010–14. American Political Science Review 102 (1): UIS (UNESCO Institute for Statistics) Data 107–23. Centre. Montreal, Canada. http://www.uis Rozelle, Scott, Xiao-yuan Dong, Linxiu Zhang, .unesco.org/Pages/default.aspx. and Amelia Hughart. 2002. “Opportunities UNDP (United Nations Development Programme). and Barriers in Reform China: Gender, Work, 2010. Asia-Pacific Human Development and Wages in the Rural Economy.� Pacific Report. Power, Voice and Rights: A Turning Economic Review 7 (1). Point for Gender Equality in Asia and the Sakellariou, Chris. 2011. “Determinants of the Pacific . Colombo, Sri Lanka: Macmillan Gender Wage Gap and Female Labor Force Publishers India Ltd. for UNDP. Participation in EAP.� Paper commissioned UNESCAP (United Nations Economic and Social for Toward Gender Equality in East Asia Commission for Asia and the Pacific), Statistics and the Pacific: A Companion to the World Division. 2003. Integrating Unpaid Work into T H E S T A T E O F G E N D E R E Q U A L I T Y I N E A S T A S I A A N D T H E P A C I F I C    59 National Economics. Bangkok, Thailand: World Bank. 2001. Engendering Development: UNESCAP. through Gender Equality in Rights, Resources, UNESCO (United Nations Educational, Scientific and Voice. Washington, DC: World Bank. and Cultural Organization) Institute for ———. 2009. “Gender Equality in East Asia: Statistics (UIS). 2009. Global Education Progress and the Challenges of Economic Digest. Montreal, Quebec: UIS. Growth and Political Change, East Asia ———. 2011. Global Education Digest. Montreal, Update.� World Bank, Washington, DC. Quebec: UIS. ———. 2010a. “Indonesia—Reconstruction of USCDC (United States Centers for Disease Aceh Land Administration System Project.� Control and Prevention). 2003. “Costs of World Bank, Washington, DC. Intimate Partner Violence against Women in ———. 2010b. World Development Indicators the United States.� USCDC, National Center 2010. Washington, DC: World Bank. for Injury Prevention and Control, Atlanta, ———. 2011a. “Vietnam Country Gender Georgia. Assessment.� World Bank, Washington, DC. U.S. Department of State. 2011. Trafficking in http://documents.worldbank.org/curated/en/ Persons Report 2011. Washington, DC: U.S. 2011/11/15470188/vietnam-country-gender- Department of State. assessment. VWC (Vanuatu Women’s Centre) in partnership ———. 2011b. World Development Report with the Vanuatu National Statistics Office. 2012: Gender Equality and Development. 2011. Vanuatu National Survey on Women’s Washington, DC: World Bank. Lives and Family Relationships. Port Vila, ———. 2011c. World Development Indicators Vanuatu: VWC. 2011. Washington, DC: World Bank. WDI (World Development Indicators) database. ———. 2012a. Capabilities, Opportunities World Ban k, Wash ing ton, DC . ht t p: // and Participation: Gender Equality and data.worldbank.org /data-catalog /world- Development in the Middle East and North development-indicators. Africa Region. Washington, DC: World Bank. WHO (World Health Organization) Global Health World Values Survey. World Values Survey Observatory Data Repository. WHO, Geneva, Association, Stockholm, Sweden. http://www Switzerland. http://apps.who.int/ghodata/. .worldvaluessurvey.org/. Gender and Endowments: Access to Human Capital and 2 Productive Assets A ccess to human capital and produc- endowments—specifically, education, health, tive assets allows individuals to live and assets—and their ties with economic healthy and productive lives. Imbal- opportunities and agency. ances between such opportunities for men Education and health are areas in which and women are costly to individual welfare, gender equality has generally been the most to society, and to development. The East responsive to growth and development in the Asia and Pacific region has experienced region. remarkable economic growth and poverty Girls’ enrollment has recently caught up •   reduction, combined with the spread of edu- with that of boys, except in several coun- cation and progress in health, from the latter tries and subpopulations experiencing half of the 20th century to the first decade of slower progress in education overall. the 21st century. Therefore, East Asian and H ealth indicators such as infant and •   Pacific countries are well placed to promote maternal mortality have also had impres- strong improvements in access to human sive gains, except in several places with capital and productive assets for many men slower economic progress. and women in those countries. However, the region has a large number •   This chapter analyzes gender differ- of “missing girls,� a persistent issue that ences in endowments—defined here as the appears not to be mitigated by growth human and productive capital that enables and development. opportunities to improve welfare—in the G ender equality in assets has been less •   region and lays the foundations for discuss- responsive to development than that in ing countries’ policy priorities in chapter 6. education and health and is constrained It examines factors underlying those gender by complex legal, social, and economic differences and identifies what drives prog- factors. ress toward gender equality or impedes it. The analysis relies on the framework of the These messages, stemming from the analy- interactions among households, markets, and sis in this chapter, help shape thinking about institutions to understand gender outcomes policies to promote gender equality in endow- in endowments. The chapter also acknowl- ments. The end of the chapter sets the stage edges the links among the different types of for that purpose, and a policy discussion 61 6 2    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C placing priorities in the endowment domain Recognizing the importance of educating in the broader development context will fol- girls, many East Asian and Pacific countries low in chapter 6. have made great strides toward equal enroll- ment between girls and boys, as highlighted in chapter 1. The discussion on education in Substantial progress toward this chapter will analyze these patterns in gender equality in education more depth to understand their dynamics and Closing gender gaps in education is benefi- their determinants. These gains in enroll- cial.1 Several cross-country analyses find a ment have responded to changes in both positive relationship between female educa- supply-side and demand-side factors that tion and growth in gross domestic product enable better education outcomes. Where (GDP) (Klasen and Lamanna 2009; Knowles, the education system as a whole is lagging, Lorgelly, and Owen 2002). Looking at house- progress on the gender front has also been holds within countries, an extensive litera- l imited. However, even with equal enroll- ­ ture has found clear evidence of correlation ment, quality of education and choice of between mothers’ education and children’s education streams still affect girls and boys education and health, particularly children’s differently and have strong implications for health and nutrition status (Schultz 1993; school-to-work transition. young people’s ­ Thomas and Strauss 1992).2 Women are usu- ally the primary child rearers, and a mother Closing gender gaps in enrollment with more education is likely to provide bet- ter child care. In East Asian and Pacific coun- Most countries in the East Asia and Pacific tries, analyses of national demographic and region have seen narrowing gender gaps in health surveys (DHS) show that Cambodian school enrollment and completion over the women with little education are less likely past two decades. Girls’ and boys’ enrollment than educated women to receive antenatal rates are now roughly on par at all levels, care and assistance from trained health per- including tertiary education. The female-to- sonnel during birth deliveries (Johnson, Sao, male enrollment ratio in secondary school and Hor 2000). Similarly, in Timor-Leste, has approached parity in most countries. highly educated mothers are most likely to Tertiary enrollment ratios between females have their births delivered by skilled atten- and males are more dispersed, but mostly on dants (88 percent), as are mothers in the an upward trend. The East Asia and Pacific wealthiest households (69 percent) (NSD, region has performed better than any other Ministry of Finance, and ICF Macro 2010). region at increasing both enrollment levels Gender equality in endowments can feed and the female-to-male enrollment ratio. In into development indirectly through links to 2010, the region had the highest primary gender equality in economic opportunities school female-to-male ratio of enrollment and agency. Promoting equal access to educa- among the developing regions and the second tion through investing in girls’ education can highest secondary school ratio.3 broaden girls’ economic opportunities and The narrowing of the gender gap in edu- raise their income. Better economic oppor- cation since the 1990s is observed not only tunities and higher income, in turn, have at the aggregate level, but also for the poor positive intergenerational effects: income in and nonpoor alike. Figure 2.1 shows the ratio women’s hands is likely to improve children’s of female-to-male enrollment rates in upper health (Duflo 2003; Fiszbein and Schady secondary schools for children in the poor- 2009; Thomas 1995). More education and est quintile. In most of the countries depicted, more income also empower women and female and male enrollment rates have been provide them with more bargaining power, converging among the poor. Similar patterns voice, and representation, as discussed in are also observed for primary and lower sec- chapter 4. ondary education. G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    63 Another way to see the significant prog- FIGURE 2.1  Enrollment for both genders has been converging ress in education is through the smaller gen- even among the poorest populations der gaps in youth literacy compared to gaps in adult literacy. Figure 2.2 shows Indonesia female-to-male gross upper-secondary enrollment ratios and Cambodia as illustrative examples, but in the poorest expenditure quintile the observations are similar in many East Asian and Pacific countries. Younger genera- Mongolia, 2008 tions are more likely to be literate. Although Mongolia, 2002 the gender gaps in adult literacy can be stark, Thailand, 2009 such as in the case of Cambodia, youth liter- Thailand, 2000 acy tends to be more equal across genders in Vietnam, 2006 both urban and rural areas, as well as across Vietnam, 1993 gender parity income quintiles. The gender gaps in literacy Indonesia, 2009 continue to close over time. Indonesia, 1999 Given this trend, some countries are even Lao PDR, 2008 starting to experience a reverse gender gap in Lao PDR, 2002 education: girls’ secondary enrollment exceeds Cambodia, 2008 that of boys in China, Fiji, Malaysia, Mongo- Cambodia, 2004 ­ hailand. The lia, the Philippines, Samoa, and T Timor-Leste, 2007 reverse gender gaps at the tertiary level are Timor-Leste, 2001 often even starker: in Thailand, 122 females were enrolled for every 100 males in 2008. In 0 0.5 1.0 1.5 2.0 Samoa, for example, boys underperform rela- ratio tive to girls in both enrollment and academic achievement. Secondary education enrollment Sources: World Bank estimates using country household income and expenditure surveys: Cambodia Socio-Economic Survey (CSES) (NIS Cambodia), 2004, 2008 data; Indonesia National rates have consistently been higher for girls, Socioeconomic Survey (SUSENAS) (BPS Indonesia), 1999, 2009 data; Lao Expenditure and Consump- by a large margin that is widening over time. tion Survey (LECS) (LSB Lao PDR), 2002, 2008 data; Living Standards Measurement Survey (LSMS) (NSO Mongolia), 2002, 2007–08 data; Thailand Socio-Economic Survey (SES) (NSO Thailand), 2000, In tests taken in year 4 and year 6 (the first 2009 data; Timor-Leste Demographic and Health Survey (DHS) (NSD Timor-Leste), 2001, 2007 data; year of education starts at age 5), girls sig- Vietnam Household Living Standards Surveys (VHLSS) (GSO Vietnam), 1993, 2006 data. nificantly outperform boys in all three tested subjects: English, Samoan, and numeracy. A FIGURE 2.2  Gender gaps in youth literacy are smaller than significantly higher proportion of boys than gender gaps in adult literacy girls have been at risk of not achieving mini- mum competencies in these years. In year 8, youth (ages 15–24) versus adult (ages 25+) literacy rates, girls still outshine their male counterparts in by gender all subjects, including science and mathemat- male-female adult literacy gap ics, but the gender gap is smaller than it was 100 in the earlier years (Jha and Kelleher 2006). Box 2.1 discusses the reverse gender gap in 80 literacy rate, % education in further detail. 60 40 Persistent gender gaps in some 20 countries and subpopulations 0 youth adult youth adult Despite progress in narrowing the enrollment gaps between genders, several countries and Indonesia, 2009 Cambodia, 2008 disadvantaged populations within countries female male-female gap still experience more visible gender disparities than elsewhere in the region.4 Girls’ enroll- Source: World Bank estimates using CSES (NIS Cambodia), 2008 data; SUSENAS (BPS Indonesia), ment has not caught up with that of boys in 2009 data. 6 4    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C BOX 2.1  Various parts of the world experience reverse gender gaps in education Although the phenomenon of reverse gender gaps in stereotyping and how masculinity values within education is relatively new in the East Asia and Pacific society are reflected in the classroom have been region, it has long been documented in parts of Europe argued to contribute to high dropout rates among and Central Asia, Latin America and the Caribbean, boys in the Caribbean (Bailey and Bernard 2003; and Organisation for Economic Co-operation and Davis 2002; Figueroa 2000). Marks (2001) shows Development (OECD) countries. The range of possible that, by age 14, girls in the United Kingdom start reasons varies, including returns in the labor market, to substantially outperform boys in English. Boys’ norms and gender identity, and the school environ- lower performance in the United Kingdom has been ment itself. Interactions between households, mar- attributed to the use of more “female-oriented� kets, and institutions influence households’ decisions reading materials, with suggestions that the inclu- about education investments; therefore, explanations sion of more factual, male-oriented works could for gender gaps in education are also context-specific. increase male performance. Other authors argue Given the differential payoffs in the labor market, by that teachers have low expectations of boys’ behav- which men tend to earn more (discussed further in ior and academic effectiveness, which contributes chapter 3), men have an incentive to drop out earlier to the levels of boys’ underachievement in Latin to join the labor force. Social norms perceiving men America and the Caribbean (Davis 2002; Figueroa as the breadwinner and stressing masculinity values 2000; Martino and Berrill 2003). The experimental may reinforce this incentive. In many cases, such as literature on test grading suggests that there can be Mongolia or the Philippines, male underachievement gender bias, but it is context-specific. For example, in education is most stark among the poor. The fig- Lavy (2008) finds that male high school students ure below shows the biggest enrollment disadvantage in Israel face discrimination in teachers’ test grad- for boys in the poorest quintile. A U.K. Department ing, but Hanna and Linden (2009) do not find such for International Development study in Botswana bias in their study in India. The United Nations and Ghana also shows a similar relationship between Children’s Fund (UNICEF 2004) outlines the role economic disadvantages and boys’ underperformance that poverty has to play in boys’ underachieve- (Dunne and Leach 2005). The study identifies reasons ment in the Caribbean and Latin America, where for boys dropping out in the studied areas of Ghana governments have become increasingly aware that as related to employment opportunities. boys and young men are more likely to be alienated The school environment itself may perpetuate from school if they come from poor socioeconomic this set of norms and gendered identity. Gender backgrounds. BOX FIGURE 2.1.1  The biggest enrollment disadvantage for boys in the Phillippines is among the poor Gross enrollment rates in secondary schools, by expenditure quintile, gender, and region, Philippines, 2006 120 100 80 percent 60 40 20 0 –20 poorest richest poorest richest poorest richest 20% 20% 20% 20% 20% 20% national rural urban female male-female gap Source: World Bank estimates using Family Income and Expenditures Survey 2006 (NSCB Philippines 2006). G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    65 those places where the overall education level, school as ­Lao-Tai (ethnic majority) boys and regardless of gender, is also low. (figure 2.3). Even in Vietnam, where girls ­ Gender gaps in enrollment are still high girls do not lag boys at the aggregate level, in Papua New Guinea and low-income coun- the Hmong and Dao populations have stark tries such as Cambodia and the Lao People’s gender gaps in secondary school enrollment: Democratic Republic. Papua New Guinea’s about one girl to two boys is enrolled in sec- education enrollment rates are among the ondary school (figure 2.4). lowest in the region, although they have been rising moderately in recent years. In 2007, FIGURE 2.3  Girls in some ethnic minority groups in Lao PDR lag primary education (elementary prepara- even further in enrollment tion through grade 8) gross enrollment rates were 73 percent for males and 66 percent for gross enrollment rates in lower-secondary school, by ethnicity and females (Papua New Guinea Department of gender, Lao PDR, 2002 and 2008 Education 2009). Cambodia and Lao PDR 100 have notable gender gaps in enrollment at the secondary level, and more extreme gaps 80 at the tertiary level. The secondary educa- 60 tion enrollment ratio is about 8 females to 10 percent males in Cambodia and Lao PDR. House- 40 hold survey data from 2008 indicate that 20 although the ratio in enrollment between girls and boys is about equal up to around 0 age 14 in Cambodia, it diverges significantly above that age. In 2008, at the tertiary level, –20 40 percent of urban males in Cambodia were 2002 2008 2002 2008 2002 2008 2002 2008 Lao-Tai/Tai-Kadai Mon-Khmer Chinese-Tibet Hmong enrolled, whereas only about 20 percent of urban females were. Enrollment rates in female male-female gap these countries are relatively low regardless of gender. Cambodia’s adult literacy rates are Source: World Bank estimates using LECS (LSB Lao PDR), 2002, 2008 data. also among the lowest in the region, with the biggest gender gaps. About 60 percent of Lao FIGURE 2.4  Girls in some ethnic minority groups in Vietnam lag women are literate, as opposed to 80 percent even further in enrollment of Lao men and more than 90 percent of peo- ple in most other countries in the region. gross enrollment rates in lower-secondary school, by ethnicity and In some cases, girls in disadvantaged pop- 120 gender, Vietnam, 1993 and 2006 ulations have the lowest education outcomes 100 when gender interacts with other forms of vulnerability, such as poverty and ethnicity. 80 Household survey data indicate that girls in 60 percent the poorest quintile in rural areas in Cambo- 40 dia and Lao PDR have the lowest secondary school enrollment rates in these countries. 5 20 In Lao PDR, gender gaps in school enroll- 0 ment can be particularly stark among the –20 Hmong population. Girls in the Hmong and 1993 2006 1993 2006 1993 2006 1993 2006 1993 2006 Mon-Khmer groups have a lower chance of Kinh/ Tay/Thai/ Central Hmong/ Khmer being in school than boys of the same ethnic- Chinese Muong/Nung ethnic Dao ity, a disparity that has been slow to change female male-female gap over the past decade. They are only half as likely to be enrolled in lower-secondary Source: World Bank estimates using VHLSS (GSO Vietnam), 1993, 2006 data. 6 6    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C Explaining progress and pending terms of passing grades per year of schooling. challenges in education Cash transfers that alleviate households’ bud- get constraints have also been shown in many The observed gains in enrollment in East developing countries to increase children’s Asian and Pacific countries have been a result school enrollment. In the East Asia and Pacific of changes in supply-side and demand-side region, a program providing scholarships to factors that enable better education out- poor girls in lower secondary schools in Cam- comes. Explanations for progress (or lack of bodia had large effects on girls’ enrollment, progress) in reducing gender gaps in educa- an increase of 20 percentage points. A related tion are context-specific because interactions program for both boys and girls had similar between households, markets, and institu- positive impacts on enrollment and attendance tions influence households’ d ­ ecisions regard- for boys and girls (Filmer and Schady 2008, ing education investments. Responding to 2009). Economic development in the region, market returns, or payoff, to investment which brings more stable income to house- choices, household d ­ ecisions reflect individ- holds, also helps protect girls’ education. In ual preferences, constraints, and the relative times of income shocks, families with girls are bargaining power of members. The returns more likely to reduce education expenditure, to education, as determined in the labor mar- as shown, for example, in the case of Indone- ket, play a role. Costs or prices in the form sia (Cameron and Worswick 2001). of direct costs (fees and uniforms), indirect In addition, norms and preferences also costs (distance to schools), and opportunity affect households’ demand for schooling by costs (wages earned outside of school) also gender. Changing norms in some contexts matter and can be shaped by markets and have led to changing girls’ status relative to institutions. Households have preferences, boys. Qualitative research through focus which may be influenced by cultural norms. group discussions in six communities in Fiji They face budget and possible credit con- suggests that parents value girls’ education straints. With this framework in mind, the more now than in previous generations (Chat- following discusses how changing demand- tier 2011). In Indonesia today, compared to a side constraints for households (for example, few decades ago, women exhibit a stronger household income), institutional factors that preference for fewer children and stronger affect the supply side (for example, reducing emphasis on children’s health and educa- the cost/prices of education), and returns to tion. Evidence suggests that this preference educating girls each and together have led to change is associated with decreased prefer- more gender equality in school enrollment in ence for sons over daughters compared to the most East Asian and Pacific countries. past (Niehof 2003). Indonesian parents now First, factors affecting households’ demand appear to intrinsically value daughters no less for education matter. Household survey data than sons (Kevane and Levine 2003). in East Asian and Pacific countries indicate Second, changes in formal institutions, that enrollment is always higher for children such as better service delivery and easier in richer families. Poor households in devel- access to schools, have improved the supply oping countries tend to face borrowing con- of education services and lowered the cost straints, and, under limited budgets, they of education. Economic development in East tend to invest more in sons than in daughters. Asian and Pacific countries has been associ- Income gains are thus likely to raise school ated with better infrastructure and service participation relatively more for girls than delivery, either through the public or private for boys, as empirically documented in vari- sector. The massive school construction pro- ous countries (World Bank 2001). Behrman gram in Indonesia in the 1970s led to signifi- and Knowles (1999) showed that, in Vietnam, cant increases in education attainment and the income elasticity of demand for education earnings, presumably through reducing costs was 6 percent lower for boys than for girls in in terms of distance to schools (Duflo 2000). G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    67 Although expanded service delivery might FIGURE 2.5  Even girls in wealthier households in Cambodia and not specifically target girls, the benefits to Lao PDR lag behind boys them may be disproportionately high because distance and safety tend to be larger barriers gross enrollment rates in upper-secondary schools of the poorest 20% and for girls than for boys. richest 20% of population, by gender, 2008 Third, in some cases, better employment 100 opportunities and returns to education for females could have encouraged parents to 80 educate girls. Cross-country evidence shows that trade liberalization, which has led to the 60 percent expansion of nonagricultural jobs for women, is positively linked to greater human capital 40 and more gender equality in human capital 20 investment (Schultz 2006). The literature also shows that education investments do respond 0 to expected returns as long as information on poorest 20% richest 20% poorest 20% richest 20% these returns is available (Jensen 2010; Oster and Millet 2010). Longer expected durations Cambodia, 2008 Lao PDR, 2008 of receiving payoff from daughters, as a result female male-female gap of declining maternal mortality, can also Source: World Bank estimates using CSES (NIS Cambodia), 2008 data; LECS (LSB Lao PDR), 2008 data. induce parents to invest more in girls’ educa- tion (Jayachandran and Lleras-Muney 2009). However, this component might have played and become household heads. Females are a relatively small role in the East Asia and expected to submit to their husbands and be Pacific region, because the gender wage gaps caregivers and homemakers; therefore, par- and gender gaps in expected returns to educa- ents value boys’ education over girls’ (Tara- tion are still large, as discussed in chapter 3.6 ria 2011). In Lao PDR, social norms about What constrains progress? Where the gender roles within the family may mean that education system as a whole is lagging, prog- girls face higher opportunity costs of school- ress on the gender front has also been lim- ing as a result of their socially defined value ited. Although the explanations can be very in home production. Poor rural girls spend context-specific, factors related to both the ­ the fewest hours in school but spend almost supply and demand sides of education in cer- three hours a day fetching water, collecting tain places constrain progress in education out- firewood, and caring for other household comes in general as well as outcomes for girls. members (King and van der Walle 2007). Low household income coupled with high Families’ perception of lower benefits from costs of education can limit the demand educating girls is also expected to impede for schooling, but income is usually not the girls’ enrollment in Cambodia, particularly whole story. Families in Cambodia have after puberty (Velasco 2001). raised concerns about bearing the high direct The supply side of service delivery also cost of education (Velasco 2001). However, matters for the overall school enrollment as in both Cambodia and Lao PDR, gender gaps well as for girls’ enrollment. Low enrollment in secondary school enrollment exist even in rates in Papua New Guinea are the result of the top quintile, as shown in figure 2.5, even limited access and high dropout rates. Apart though gaps across quintiles are still much from demand-side factors such as affordabil- larger. Aside from income, norms also have a ity of school fees and low perceived value of strong influence on household decisions. Par- education, elementary schools have not been ticipants of focus group discussions in a quali- established in many communities. Even urban tative research exercise in Papua New Guinea areas may have inadequate capacity to admit report that males will carry the family name all children wanting to enroll. Poor learning 6 8    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C environments (often due to lack of educa- Asia and Pacific region, despite the closing tion materials), poor school infrastructure, of gaps in enrollment and attainment, male poor teacher attitude and attendance, lack of and female students still differ in their choice teachers in remote areas, and negative pupil of education streams. Factors that influence behavior all contribute to poor enrollment gender streaming in education are important overall (Papua New Guinea Department because such streaming affects the occupa- of Education 2009). In Cambodia and Lao tions and sectors that men and women engage PDR, distance to schools is an important bar- in and the income that they subsequently rier. The unequal provision of schools in Lao earn. The persistence of these patterns PDR makes schooling more costly for girls implies a gender-differentiated school-to- than for boys (King and van der Walle 2007). work behavior and ultimately sustains gender In Cambodia, given the long distance to inequalities in job placement and earnings. schools, boys can live in wats (temples) while Earning gaps across genders largely reflect attending secondary school, but girls have no differences across occupations and sectors of comparable system of accommodation. Anec- employment. Chapter 3 discusses further the dotal evidence suggests that the lack of toilets importance of education and labor market at school also makes it more difficult for girls streaming to labor market outcomes. than boys to attend (Velasco 2001). Evidence from the East Asia and Pacific Limitations of the education system in region shows clear patterns of specialization catering to the rural poor and ethnic minori- by field among men and women, implying ties mean that the gender disadvantage can differences in skill profiles when they enter interact with these forms of vulnerability. In the labor market. Figure 2.6 shows the frac- East Asian and Pacific countries, gender gaps tion of females in each field of study in Indo- within a subpopulation are usually smaller nesia, Thailand, and Vietnam. The fields of than the enrollment gaps across income engineering and law are heavily dominated quintiles or between major and minor eth- by males, whereas the fields of education, nicities. In Vietnam, though the ethnic- health, and business administration are dom- ity gap in primary enrollment had almost inated by females. These gender differences in closed between 1993 and 2006, enrollment the choice of field are larger than the gender at the secondary level among the Kinh/Chi- differences in enrollment or completion rate nese majorities is still substantially higher in secondary or tertiary education in these than that among ethnic minority popula- countries. The positive relationship between tions. With the exception of the Hmong and being male and choosing science, engineer- Dao, the gender gaps within an ethnicity in ing, or law is statistically significant in mul- ­ V ietnam are smaller than the gaps across tinomial logit analysis that also accounts for ethnicities (figure 2.4). In Cambodia, the characteristics such as parental education gap in secondary school enrollment across and area of residence (Sakellariou 2011). quintiles also exceeds gender gaps within Education streaming can be slow to change any given quintile. Similar patterns occur over time and is less responsive to economic in many other East Asian and Pacific coun- growth than enrollments, at least as evidence tries. These large enrollment gaps across indicates for Indonesia from 1997 to 2009. subpopulations suggest that making general Such gender streaming in education is not improvements in the education system to unique to East Asian and Pacific countries, reach v­ ulnerable subpopulations could con- but is evident in many Organisation for tribute to closing the gender gap as well. Economic Co-operation and Development (OECD) countries as well. Flabbi (2011) documents that in many OECD countries, Gender streaming in education relatively more women enter the social sci- Education investment does not end at enroll- ences and business, whereas more men enter ment. In middle-income countries in the East the fields of engineering and architecture. G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    69 There could be multiple reasons for men FIGURE 2.6  Women are concentrated in certain fields of study, and women’s choice of education streams. such as education and health, but are underrepresented in law The relative payoff of the different streams— and engineering the expected returns from labor markets—is percentage of students who are women, by eld of study likely to influence this decision. However, evi- 90 dence shows that women do not necessarily 80 take up fields with the highest premiums in 70 60 the labor market wage. Table 2.1 shows the percent 50 estimated returns to selected fields of study in 40 Indonesia. The returns to studying engineer- 30 ing are high for females and are much higher 20 than the returns to studying education. For 10 example, among the female adults surveyed 0 n h law g n h y s a ienc ial in 2009, the premium for having studied og io alt in tio at c at er dm es m o ol he ra s uc ne hn nd ist engineering (rather than religion) was 72 per- nd ed gi in ec a ,a c en ce lt s rts cent higher earnings, whereas the premium ien ra ,a tu sc es es ul for having studied education was 17 per- iti sin ric an ag bu m cent. This phenomenon has been the case hu for over a decade, since the late 1990s. Still, Indonesia, 2009 Thailand, 2006 Vietnam, 2006 females prefer education to engineering. As Source: Sakellariou 2011. a note of caution, the data available here do not account for differences in unobservable characteristics, such as ability, among those TABLE 2.1  Labor market returns to studying engineering are high relative to studying education choosing the different fields. They capture Indonesia: Estimated return to field of study within tertiary education, % change only monetary returns, whereas nonmon- in wage etary aspects—such as values, attitudes, and social expectations about women as mothers Field of study 1997 2006 2009 (compare to and homemakers or caregivers—also play an field: Religion) Males Females Males Females Males Females important role in influencing the decisions. Nonetheless, the statistics indicate what Law 19.7 20.9 32.3 28.4 47.7 23.4 Health 39.1 10.5 32.3 20.9 56.8 39.1 appear to be persistent patterns that also Engineering 58.4 78.6 44.8 68.2 60.0 71.6 translate into patterns in the labor market, as Education 13.9 10.5 13.9 15.0 23.4 17.4 discussed in chapter 3. Source: Sakellariou 2011. An alternative explanation for the gen- Note: The estimates were based on a log wage regression, accounting for characteristics such as der pattern of subject choice might be that experience and experience squared, marital status, and urban/rural residence. All coefficients were significant at the 5 percent level. females and males have different compara- tive advantages in a particular field based on their academic performance. The avail- women in math and computer science (Schle- able evidence from the East Asia and Pacific icher 2008). Moreover, female students out- region does not support this hypothesis. perform their male counterparts in reading Female students do not systematically per- assessments, according to the Programme form worse than males in key subjects for International Student Assessment (PISA) related to male-dominant fields of study. As test scores. shown in figure 2.7, results of the Trends In addition, norms and expectations in International Mathematics and Science may influence preferences and, therefore, Study (TIMSS) indicate no evident female the choice of education streams. They can disadvantage in math and science scores in do so, for example, through shaping paren- Indonesia, Malaysia, the Philippines, and tal expectations, shaping role models in the Thailand, unlike the findings for OECD labor market, or shaping how gender roles are countries that men tend to outperform portrayed in school curricula. Social norms 7 0    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE 2.7  Girls tend to outperform boys in several subjects in school. Gender stereotyping also appears to affect school performance of boys and female-to-male ratio of test scores, by subject 1.10 girls in key subjects (science, mathematics, 1.08 and computer science), ultimately contribut- ing to varying occupational choices (Blum- 1.06 berg 2007; Eccles and Blumenfeld 1985; 1.04 Shel 2007). This literature also explores the ratio 1.02 importance of the gender roles conveyed 1.00 through teaching materials in shaping boys’ 0.98 and girls’ aspirations. School, in particular primary school, is believed to be a key ele- 0.96 ment of socialization for children and a site 0.94 where key social values are transmitted. Indonesia Thailand Malaysia Philippines Teaching materials and teachers’ feedback math science reading are the key instruments through which ideas about what are appropriate areas of study Source: World Bank staff calculations based on TIMSS and PISA data. and acceptable professions for men and Note: Mathematics and science scales for eighth graders are from Trends in International Mathemat- ics and Science Study (TIMSS). The most recent year of data is 2003 for the Philippines and 2007 women are conveyed to children (Blumberg for all other countries. The reading scale is from the Programme for International Student 2007; Nielsen and Davies 2010). Assessment (PISA). In East Asia and the Pacific, the results of a 2008 review of the Education for All (EFA) about the role of females as “homemakers� initiative indicated that teaching materials translate into expectations about appropriate in the region included stereotypical portray- jobs for men and women. In fact, women in als of boys and girls (UNICEF 2009). The the region work more hours and devote more findings, in line with those of other regions, time to caregiving and housework activi- stress that boys appear more than girls in the ties than do men. Given these expectations, learning materials, that they are portrayed as female students may be inclined to choose a more active (and girls as more passive), and field of study that will lead to a job with suffi- that they are shown more frequently in lead- cient time flexibility, such as teaching, to bal- ership roles.7 The report also found a lack ance their home and labor market work. The of female role models in teaching materials. absence of female role models in the labor Women in the school textbooks reviewed market has also been identified as a contrib- were portrayed as secretaries, assistants, uting factor in influencing aspiration and nurses, and teachers more frequently than career choice and in limiting women’s access men, who often appeared as doctors, politi- to nontraditional careers. These topics will cians, or police officers. be explored further in chapter 3. The rest of Qualitative research involving both the this section focuses on how gender roles are review of teaching materials and classroom portrayed in school curricula, which pertains observation, which allows for an analysis of directly to the education system. teacher-student interaction, has highlighted School curricula are important in shap- how gender stereotyping frequently exists in ing choices of what to study, because the curricula in a number of East Asian countries. way in which gender roles are portrayed in Table 2.2 provides selected examples of the school curricula is believed to affect chil- types of issues typically encountered in cur- dren’s performance and aspirations. Cur- riculum analysis. Male characters tend to be riculum analysis in developed and developing portrayed as dominant in the public sphere. countries alike indicates that gender stereo- Similar stereotyping is also observed in other typing in curricula as well as differences in countries throughout the region. The depic- teacher interactions with male and female tion of women in China’s history and social students affect the probability of girls staying sciences manuals illustrates this point. Guo G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    71 TABLE 2.2  School curricula in a number of East Asian countries have gender stereotyping China Vietnam Korea, Rep. Overall The proportion of male characters In grade 4 and grade 5 texts analyzed, visibility rises from 48 percent in books for most authors mentioned and quoted of male four-year-olds to 61 percent in books were men (74 of 85 and 77 of 84, and female for six-year-olds. Female characters respectively). characters appear most commonly in reading In mathematics textbooks analyzed, or authors materials for very young children. female characters were found to in teaching appear in illustrations more often than materials men (some grade levels); however, they were associated with less challenging activities. Stereotypical Portrayal of male and female Stereotypical depictions of men/ Curricular materials presented traditional portrayals characters follows gender role boys and women/girls are present gender roles with women depicted of men and stereotypes in both mathematics and in a range of textbook illustrations doing housework versus office work for women in social sciences manuals. An analysis of and exercises. A detailed analysis men, male characters leading activities textbooks mathematics texts indicated that male of materials used in grades 1 to 5 and female characters assisting. characters were 74 percent of those highlights the following depictions of in stimulating activities, and female men and boys: (a) heroes/courageous; characters represented 70 percent (b) strong/able to do complicated of those in passive roles. Men and and physically challenging jobs; (c) boys are therefore typically portrayed knowledgeable/smart; (d) naughty; (e) as courageous, independent, and creative; and (f) leaders. ambitious, in contrast to “passive, In contrast, women and girls are obedient, neat, cooperative girls.� typical portrayed as (a) “nice and lovely�; (b) caring (as teachers, nurses); (c) clean and ordered; and (d) weak/ emotional. Interestingly, men and boys were also more frequently associated with “forbidden� or dangerous activities. Student- Teacher-pupil and peer interaction Explanations and examples reflecting teacher were observed to follow the same gender role stereotypes were used interaction stereotypes, to girls’ disadvantage. frequently, as was gender-discriminative language. Male students had more teacher-student interaction opportunities as well as more social contact with teachers. Teachers were found to discipline male students more severely than female students. Sources: Jung and Chung 2005 (Korea); Ross and Shi 2003; Shi and Ross 2002 (China); UNESCO/Vietnam Ministry of Education and Training 2011 (Vietnam). and Zhao’s (2002) analysis of elementary lan- Improvements and remaining guage textbooks highlights that only about gender issues in health a fifth of the historical characters portrayed are female. When they are depicted, they also Promoting better health is important for tend to be portrayed in stereotypical roles. For enhancing welfare. Most societies recog- example, an influential female leader of the nize that girls and boys should have equal Chinese Communist Party is depicted twice: access to the elements of a healthy life and once mending Premier Zhou Enlai’s clothes, that maternal health is important. Mothers’ and another time bringing an umbrella to a health and nutrition affect their children’s guard on a rainy day. physical health as well as cognitive abilities. 7 2    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C Research shows that delays in cognitive and men (Eskes and Haanen 2006). Male mortal- overall development from the time a child is ity is naturally higher than female mortality conceived up to age six, a sensitive period for the first six months of life (Waldron 1998). for brain formation, have long-lasting con- Men and women are also susceptible to dif- sequences that are difficult to compensate ferent diseases, such as different types of can- for later on in life (Naudeau et al. 2011). In cer. Aside from biological differences, health addition, improved maternal health has been outcomes are also affected by differences in shown to enable women to reconcile work behaviors and health investments that could and motherhood, playing a role in raising disadvantage one particular gender. Thus, married women’s labor force participation many unobservable factors affect morbidity in the United States (Albanesi and Olivetti and mortality for men and women. 2009). Research has also shown that with This chapter focuses on mortality risks declining maternal mortality—that is, lon- throughout the life cycle. For early child- ger life expectancy for women—parents hood and childbearing periods, most East can expect a longer duration of payoff from Asian and Pacific countries have experienced daughters. In Sri Lanka, for example, par- impressive progress in narrowing the gender ents respond with more investment in girls’ gaps in the infant mortality rate and reducing education (Jayachandran and Lleras-Muney maternal mortality. However, as discussed 2009). later, the period before birth is a concern, To explain progress or the lack thereof with male-skewed sex ratios at birth in sev- in gender equality in health in East Asian eral parts of the region. and Pacific countries, this section focuses Many East Asian and Pacific countries on several key indicators, such as fertility, have significantly improved several health child and maternal mortality, and sex ratios outcomes during the past two decades. Fertil- at birth. The chapter argues that many, ity rates went down sharply. Infant and child but not all, health outcomes for males and mortality rates for both boys and girls have females in the East Asia and Pacific region declined substantially since 1990, closing have improved with development. Fertil- the gender gaps in the infant mortality rate. ity and child and maternal mortality have The maternal mortality rate (MMR) has also had impressive gains, except in a few places been declining, and, along with the female with slower economic progress, such as infant mortality rate in most of East Asia and Cambodia, Lao PDR, and Timor-Leste. the Pacific, is now low relative to the region’s However, the region still has a large num- income level. Figures 2.8 and 2.9 illustrate ber of missing girls, a persistent issue not this point: in the cross-country graphs of easily mitigated by growth and develop- these health indicators and income measured ment. The chapter also sheds light on two in terms of purchasing power parity, most behavioral health issues associated with East Asian and Pacific countries lie below the high adult mortality risks for men in all downward-sloping curve representing this East Asian and Pacific countries: excessive relationship. smoking and drinking. As in the case of education, progress has not been uniform across the region. First, China differs from other East Asian Declines in fertility and child and and Pacific countries in having a high rate maternal mortality rates of female child mortality relative to that of Gender differences in health outcomes reflect males. Figure 2.10 graphs the ratio of female- biological differences as well as gender- to-male child mortality as well as the ratio differential behaviors, which are difficult to ­ for infant mortality against GDP per capita. separate in what we observe. As a result of Although in most countries, male infant biological factors determining life expectancy, and child mortality rates are slightly higher for example, women tend to live longer than than female rates, consistent with biological G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    73 FIGURE 2.8  Maternal mortality is lower in higher-income countries maternal mortality rate (modeled estimate, per 100,000 live births) against income levels, 2008 1,500 maternal deaths per 100,000 live births 1,000 Lao PDR 500 Timor-Leste Cambodia Indonesia Papua New Guinea Solomon Islands Philippines Thailand New Zealand Fiji Korea, Rep. Singapore 0 Vietnam Mongolia China Malaysia Japan Australia 6 8 10 12 log GDP per capita (constant 2005 internaltional $) rest of world East Asia and Paci c Source: World Bank estimates using World Development Indicators (WDI) database, 2011 data. FIGURE 2.9  Female infant mortality is lower in higher-income countries female infant mortality rate (number of deaths between birth and age one, per 1,000 live births) against income levels, 2009 150 number of deaths per 1,000 live births 100 KHM PNG 50 TMP LAO FSM KIR IDN CHN PHL SLB TON VNM MNG KOR AUS FJI 0 WSM THA NZL JPN MYS SGP 6 7 8 9 10 11 log GDP per capita (constant 2005 international $) rest of world East Asia and Paci c Source: World Bank estimates using WDI database, 2011 data. 7 4    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C factors, China experiences the opposite pat- Second, the maternal mortality rate is terns. Females face higher mortality risks still a serious concern in a number of places during infancy than males, and even more so in the region. As shown in figure 2.8, Lao before birth, which is related to the “missing PDR experienced more than 500 maternal girls at birth� topic, discussed later. deaths per 100,000 births in 2008, a rate FIGURE 2.10  Most East Asia and Pacific region countries do not have female-skewed under-five mortality and infant mortality, except China a. female-to-male ratio of under- ve mortality, 2009 1.5 CHN SLB VUT KOR 1.0 KIR FSM PNG TON VNM PHL NZL AUS IDN ratio KHM LAO TMP FJI THA JPN MNG MYS SGP 0.5 WSM 0 6 7 8 9 10 11 log GDP per capita (constant 2005 international $) b. female-to-male ratio of infant mortality, 2009 1.5 CHN VNM VUT 1.0 PNG SLB FSM KOR ratio KIR FJI NZL AUS KHM PHL TMP LAO IDN THA MYS SGP MNG JPN 0.5 WSM 0 6 7 8 9 10 11 log GDP per capita (constant 2005 international $) rest of world East Asia and Paci c Source: World Bank estimates using WDI database, 2011 data. G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    75 much higher than other countries of similar T hailand before 1980 has been fertility in ­ income level. The maternal mortality rate attributed to government subsidies to public remains high—above 240 maternal deaths and private family planning systems and to per 100,000 births—in several other coun- the rapid increase in female education (Schultz tries, such as Cambodia, Papua New Guinea, 1997). Breierova and Duflo (2004) showed and Timor-Leste, despite progress over the that higher education among females in Indo- past two decades. Indonesia’s maternal mor- nesia led them to have fewer children early tality rate remains high compared to other on, and the increased education of mothers countries in the region at similar levels of and fathers led to lower child mortality. As development. Even for Vietnam, a country mentioned earlier, in Indonesia today, com- that has successfully reduced this rate at the pared to a few decades ago, women exhibit a national level, maternal health outcomes still stronger preference for fewer children and for lag in rural areas and among ethnic minor- more per-child investment in health and edu- ity groups (World Bank 2011b). cation. This preference change is attributed to decreased preference for sons over daughters and to other social changes (Niehof 2003). Growth, development, and Growth and development in the East Asia improvements in health outcomes and Pacific region have contributed to the Maternal health and child health outcomes region’s progress in improving maternal and are a result of many factors, including ser- reducing child mortality through a combina- vice delivery related to the public and private tion of demand-side and supply-side factors. health systems, prices and the availability of In fact, living in a high-income country is insurance mechanisms, and the demand side generally associated with lower risk of mater- of households’ fertility and health-seeking nal death and of female infant death (as well behavior. Social norms regarding childbirth as male infant death, not shown). Figures 2.8 practices also have an important influence in and 2.9 show a negative relationship between many contexts. a country’s maternal mortality rate or female The East Asia and Pacific region has expe- infant mortality rate with its GDP per capita rienced a substantial reduction in total fertil- level. ity, and this decline has reduced the risk of For the household, rising income in the maternal health complications and death. region appears to have positive impacts on The share of women using modern contra- health outcomes. A rise in income may loosen ception, and thus are presumably more able the incentives to differentiate health invest- to control fertility, has been on the rise in ments across boys and girls. Evidence from a many East Asian and Pacific countries. Gov- large data set of developing countries shows ernment policies such as family planning that, on average, a one-unit increase in log programs and China’s one-child policy were GDP per capita is associated with a decrease expected to control fertility. Schultz and Yi in mortality of between 18 and 44 infants per (1997) argued that institutional reforms, 1,000 births. This negative relationship holds such as the replacement of the collective pro- true even when various factors—such as the duction team with the household responsi- mother’s characteristics, weather shocks, bility system starting in 1979 in China, may conflicts, and the quality of institutions— have contributed to the decline in fertility in are accounted for. Female infant mortality is the long term for a variety of reasons. For more sensitive to changes in economic condi- example, the intensified market competition tions than male mortality (Baird, Friedman, encouraged parents to educate children and and Schady 2007).8 In addition, households focus on the “quality� rather than quantity with higher income can afford more health of children. Increased mobility and migration services, such as the use of prenatal care and for better economic opportunities were also hospitals during births. Figure 2.11 shows linked to delayed childbearing. The declining that in every East Asian and Pacific country 7 6    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE 2.11  Women in wealthier households are more likely to height of children in Mexico (Atkin 2010). have births assisted by trained medical staff In Indonesia, women with a higher share of household assets tend to use more prenatal percentage of women having birth deliveries assisted by trained and delivery care (Beegle, Frankenberg, and medical sta ; comparison of the poorest and richest 20% 120 Thomas 2001). richest 20% Improvements in health technologies as 100 95.6 99.7 well as improvements in the institutions sup- 92.3 82.0 plying health care have led to lower costs of 80 health services and better health outcomes. percent 60 60.5 58.1 Medical progress with the medicalization 56.0 and hospitalization of childbirth contributed 40 42.8 national average to the substantial decline in maternal mortal- 25.1 21.3 ity in the United States in the first half of the 20 poorest 20% 20th century (Albanesi and Olivetti 2009). 0 These technologies, when adopted in devel- Timor-Leste, Philippines, Cambodia, Indonesia, Vietnam, oping countries in the region, were likely to 2007 2003 2008 2009 2002 have similar effects. In addition, the share of Sources: World Bank staff estimates using household income and expenditure surveys of various women using modern contraception has been countries and years; World Bank Health, Nutrition, and Population Statistics (HNPStats) database. on the rise in many East Asian and Pacific countries. Contraception use allows families examined, women living in richer house- to control fertility and avoid extremely short holds are more likely to have birth deliver- periods between pregnancies, which tend to ies assisted by trained health professionals. pose higher health risks. Countries shown in In some cases, such as Cambodia, Indone- figure 2.12 with a high rate of contraceptive sia, and the Philippines, the gap between the use (for example, the high and increasing rate richest quintile and the bottom one can be in Vietnam) are also those with a low mater- fourfold. Over time, as the economy grows, nal mortality rate. the rate of professional birth attendance The functioning of the health system, increases. Lastly, the availability of insurance including infrastructure, medical facilities, mechanisms has also made health care more and equipment and staffing, is also key to affordable. The expansion of health insur- improving health outcomes. Experience in ance coverage in Vietnam, from 25 percent the United States illustrates the importance of the population in 2004 to 40 percent in of public health investments: two-thirds of 2006, and serious efforts by the government the decline in overall infant mortality and the to extend health insurance to the poor and entire decline in excess female infant mortal- ethnic minorities in recent years have had ity in the early 20th century were attribut- positive impacts on access to health services able to clean water and sanitation (Cutler and (World Bank 2011b). Miller 2005). Vietnam’s remarkable achieve- Rising women’s income, in particular, is ments in bringing down child and maternal likely to have contributed to this progress mortality have been attributed to a general in health outcomes. Many countries in the strengthening of the health system (World East Asia and Pacific region have experienced Bank 2011b). The gains in the share of births recent increases in female labor force par- attended by professionals from 2000 to 2008 ticipation. Evidence shows that income in the were very impressive in the East Asia and hands of women positively affects children’s Pacific region, larger than any other develop- health, particularly girls’ (Thomas 1995 for ing region. Bringing better services closer to Brazil, Ghana, and the United States; Duflo women can change their patterns of use, and 2003 for South Africa). Higher female labor positively affects health outcomes. Franken- force participation has been shown to improve berg and Thomas (2001) analyzed the Indo- their bargaining power and to increase the nesia Family Life Survey panel data using G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    77 community-level fixed effects to measure FIGURE 2.12  Contraceptive prevalence varies across East Asian the impacts of a major expansion in mid- and Pacific countries wifery services between 1990 and 1998 on health and pregnancy outcomes for women contraceptive prevalence (share of women ages 15–49) of reproductive age. The authors showed China, 2006 Thailand, 2006 that the addition of a village midwife to Vietnam, 2006 communities between 1993 and 1997 was Mongolia, 2006 associated with a significant increase in body Indonesia, 2007 mass index (BMI) for women of reproductive Philippines, 2008 Marshall Islands, 2007 age, as well as an increase in birthweight of Fiji, 2007 newborns. Frankenberg et al. (2009) further Cambodia, 2005 investigated the reasons behind this impact Vanuatu, 2007 Lao PDR, 2005 on outcomes. The presence of village mid- Papua New Guinea, 2006 wives appears to have increased women’s Solomon Islands, 2007 receipt of iron tablets and influenced their Tonga, 2006 Micronesia, Fed. Sts., 2004 choice of childbirth practice away from reli- Kiribati, 2004 ance on traditional birth attendants toward Timor-Leste, 2007 delivery attended by skilled professionals. 0 20 40 60 80 100 percent Ongoing challenges: Poor service Source: WDI database, Gender Statistics. delivery and social norms East Asian and Pacific countries with high FIGURE 2.13  The percentage of births attended by skilled maternal mortality rates—Cambodia, Indo- professionals varies across East Asian and Pacific countries nesia, Lao PDR, Papua New Guinea, and Timor-Leste—are precisely those with low births attended by skilled health sta (share of total) rates of contraceptive use and low rates of Mongolia, 2008 births delivered by professionals (figure 2.12 China, 2007 and figure 2.13). As shown in figure 2.12, Malaysia, 2005 Pacific Island countries tend to have low Thailand, 2006 contraceptive prevalence compared to East Vietnam, 2006 Asia. In Timor-Leste, for example, only 20 Indonesia, 2007 percent of women ages 15–49 use contracep- Philippines, 2008 tion, leading to high fertility rates and very Papua New Guinea, 2006 short periods between pregnancies. Less than Cambodia, 2005 20 percent of the births in Timor-Leste are Lao PDR, 2006 assisted by professionals. Delivery at home Timor-Leste, 2003 without professional help and without easy 0 20 40 60 80 100 access to a functioning referral center poses percent high risks, particularly in case of compli- Source: WDI database, Gender Statistics. cations. The absence of the factors that explain progress elsewhere is at play in these countries. Demand-side constraints from rural and remote areas, places women at households, including social norms about a high risk for maternal death. Rural areas pregnancy and birthing practices, and poor tend to be less well served by the health sys- supply-side provision of care explain the poor tem; figure 2.14 shows that rural residents health outcomes in these countries. The exact have substantially less access to birth deliver- reasons can vary from context to context. ies attended by trained staff than urban resi- On the supply side, poor access to quality dents. Delivery in the home is of particular obstetric health services, particularly among concern for poor, rural women because they 7 8    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE 2.14  Women in rural areas are less likely to have births in Lao PDR for whom both distance and assisted by trained medical staff infrastructure are of concern (GRID 2005). Chine-Tibet women appear to be most at risk percentage of women that had births assisted by trained medical sta , by location during childbirth since they are very likely to 120 give birth outside of a hospital, and a large urban majority of Chine-Tibet villages (76 percent) share of women having birth, % 100 99.0 lack safe water, let alone access to other 88.9 80 85.9 82.2 sanitary measures (GRID 2005). In Cambo- 79.0 dia, low quality health care, the poor state 64.7 64.6 63.7 60 of rural roads, lack of transport, and poor 40.8 access to a clean water supply have all been 40 national average shown to impede progress in bringing down 31.0 rural 20 maternal mortality (UNIFEM, World Bank, ADB, UNDP, and DFID/UK 2004). 0 On the demand side, tight household bud- Timor-Leste, Philippines, Cambodia, Indonesia, Vietnam, gets can constrain the use of health care ser- 2007 2003 2008 2009 2002 vices. Women from poor families may not be able to afford health costs, which are the Source: World Bank estimates using household income and expenditure surveys of various countries and years, and the HNPStats database (see figure 2.1 country sources). major barrier to seeking health care in many developing countries. As shown earlier in fig- ure 2.12, in every country examined, women lack the basic sanitary conditions needed for a living in poor households are less likely safe delivery. In Lao PDR, almost 90 percent to have birth deliveries assisted by trained of rural women deliver at home, compared to health professionals. In Indonesia, the com- approximately a quarter of urban women. As munity health insurance scheme, Jamkesmas, a result, a large disparity in maternal mor- has had little effect on facility-based deliver- tality rates exists between urban and rural ies because some delivery costs such as trans- areas in Lao PDR: 170 versus 580 mater- port and costs for family members were not nal deaths per 100,000 births, respectively covered (World Bank 2010). Even in a coun- (GRID 2005). In Timor-Leste, 59 percent of try with health insurance subsidies for the urban births are assisted by skilled providers, poor, only 60 percent of Vietnamese women compared with 21 percent of births in rural in the poorest quintile had births attended by areas. This rate is 69 percent in the capital trained medical staff in 2002 (figure 2.11). city, Dili, but less than 10 percent in the Women’s access to reproductive health Oecussi region (NSD, Ministry of Finance, care could be constrained by norms. Culture and ICF Macro 2010). Moreover, the qual- and tradition play an important role in the ity of prenatal care in Timor-Leste, is limited: choice of health practices, such as the location in 2001–02, only 41 percent of those giving of childbirth, the use of birth attendants, and birth were protected against neonatal tetanus, sterilization practices. For instance, follow- a major cause of neonatal death (ADB 2005). ing a traditional practice, a number of Mon- Substantial disparities also exist across prov- Khmer women in Lao PDR deliver neither in inces in Indonesia: Jakarta has 97 percent of the home nor in a medical center, but rather births attended by a skilled provider, whereas in the forest (GRID 2005). In Cambodia, cul- Maluku has only 33 percent (World Bank tural beliefs that pregnancy and childbirth 2010). are part of the natural process lead families Long distance to the nearest health cen- to perceive that women do not need prenatal ter and the poor infrastructure available can care or delivery supported by skilled atten- both impose high costs of access. Although dants. Thus, many women continue heavy high in all rural areas, the rate of deliveries physical labor and long work hours during at home is highest among highland women pregnancy and immediately after childbirth G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    79 (UNIFEM, World Bank, ADB, UNDP, and 106.4 in 2008, now close to the biological DFID/UK 2004). ratio (figure 2.15).9 The evidence suggests that improving Sex ratios at higher order births (that is, service delivery is key to reducing gender second children or above) are usually worse disparities and improving health outcomes. than the average ratios. In China, although Given the central influence of social norms in the sex ratio at birth in the 1980s was within birthing practice, service delivery could and the normal range for the first birth, it became should be strengthened by providing services unbalanced at higher orders—1.3 for the in a culturally acceptable way. Policy implica- fourth or later child in 1989. The sex ratios tions and recommendations are discussed at for higher order births, conditional on earlier the end of this chapter and in chapter 6. female births, were even more starkly skewed toward males (Zeng et al. 1993). Chung and Das Gupta (2007) used data from Korea’s Missing girls at birth 2003 fertility survey to show that the sex One concerning issue that persists in the East ratio at birth after the first birth was 129 if all Asia and Pacific region despite tremendous the previous births were girls, and it is 112 if growth and development is the phenomenon at least one previous birth was a boy. This dif- of missing girls, particularly at birth. The ference was even starker among those women term “missing women� was first coined by who stated that having a son was imperative. Sen (1992) to refer to the phenomenon that More recently in Korea, even though the aver- many low-income countries have far fewer age sex ratio at birth is close to the normal women than men, relative to what is observed range, the ratios for the third birth and for in developed countries. At birth, the biologi- the fourth or higher births were still 116 and cal norm is approximately 105 boys born for 124, respectively, in 2008 (figure 2.15). In every 100 girls. Yet, the male-female ratio at Vietnam, the pattern is unusual in that the birth in East Asian and Pacific countries far sex ratio for the first birth is already higher exceeds that of other regions, mainly driven than that for the second birth and similar to by China’s ratio. that for the third or later births. However, The United Nations Department of according to Vietnam’s 2006 population ­ E conomic and Social Affairs, Population survey, the sex ratio at birth for third-order Division, database provides estimates of cross-country sex ratios at birth over time. FIGURE 2.15  Sex ratios at higher order births are still of concern, These estimates are based on projections from even though the overall sex ratio at birth has approached the national census data and a fertility modeling normal range in the Republic of Korea exercise. Estimates from national statistical offices can sometimes stem from more recent Korea: sex ratio at birth, by birth order data and, as a result, differ from the United 170 number of male live births per 100 female live births Nations (UN) projections. According to the 160 UN projections, over the 2005–10 period, 150 120 boys in China were born for every 100 140 girls. Outside of China, new concerns are 130 emerging regarding Vietnam’s rising sex ratio 120 at birth. According to the Vietnam General 110 Statistical Office’s Annual Population Change 100 Surveys, the sex ratio at birth increased regu- 1998 2000 2002 2004 2006 2008 larly from 2004 and crossed the 110 thresh- sex ratio the rst child old in 2005 (UNFPA 2009). Yet, trends in the second child the third child sex ratios in the East Asia and Pacific region the fourth child and higher are not all bad news. Korea has seen declin- ing sex ratios at birth, from 110.2 in 1998 to Source: Korea National Statistical Office. http://kostat.go.kr. 8 0    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C births decreases with the previous number of female-to-male child mortality ratio. How- sons born between 2000 and 2006: 110.3 for ever, the most worrisome period of missing no sons, 103.5 for one son, and 102.2 for two girls in China is at birth. The number of sons (UNFPA 2009). missing girls at birth increased from 890,000 As a consequence, in comparison to other in 1990 to 1,092,000 in 2008. Missing girls developing regions, the number of missing as a fraction of the total number of female girls at birth in the East Asia and Pacific births increased from 8.6 percent in 1990 to region, particularly in China, dominates the 13.3 percent in 2008. excess mortality risks for females after birth. The reason for the missing girls phenom- The World Development Report 2012 uses enon has been attributed to son preference the following methodology to calculate the (Das Gupta 2005). Parents’ preferential number of missing girls at birth and excess choices to keep and care for boys over girls female deaths in other parts of the life cycle. can depend on social norms and values, dif- The number of missing girls at birth is cal- ferent economic opportunities by gender, and culated by comparing the sex ratio at birth what benefits parents expect from a son or a in a particular country to the ratio in high- daughter. As an example of how economic income countries (105.9 boys for 100 girls). opportunities influence parental choice over Throughout the age distribution, excess the gender of their child, Qian (2008) showed female mortality is calculated by compar- that the sex ratio at birth is responsive to ing the mortality risks of females relative to returns in the labor markets for women in males in a particular age group in a coun- rural China. An increase in women’s income try with the mortality risks in a reference relative to men’s led to higher survival rates for group of high-income countries.10 Table 2.3 girls. Another example of how parental choice shows the number of missing girls at birth responds to changing economic conditions is and excess female deaths per year, calculated the rise of marriage migration—cross-border using this methodology. Excess female mor- marriages between women from Southeast tality in infancy and during the reproductive Asia and men from East Asia. Through a years have substantially decreased in China. 2007 survey of three migrant-sending com- An estimated 71,000 girls under age five munities in southern Vietnam, Bélanger and were missing in China in 2008, consistent Tran (2011) documented an enhanced status with figure 2.10, which shows China’s high of emigrating daughters sending remittances TABLE 2.3  The East Asia and Pacific region, mainly driven by China, is characterized by its large number of missing girls at birth Missing girls at birth and excess female deaths throughout the age distribution (1,000s per year) At birth Under 5 5–14 15–49 50–59 Total (under 60) 1990 2008 1990 2008 1990 2008 1990 2008 1990 2008 1990 2008 China 890 1,092 259 71 21 5 208 56 92 30 1,470 1,254 India 265 257 428 251 94 45 388 228 81 75 1,255 856 Sub-Saharan Africa 42 53 183 203 61 77 302 751 50 99 639 1,182 South Asia (excluding India) 0 1 99 72 32 20 176 161 37 51 346 305 East Asia and Pacific (excluding China) 3 4 14 7 14 9 137 113 48 46 216 179 Middle East and North Africa 5 6 13 7 4 1 43 24 15 15 80 52 East and Central Asia 7 14 3 1 0 0 12 4 4 3 27 23 Latin America and the Caribbean 0 0 11 5 3 1 20 10 17 17 51 33 Total 1,212 1,427 1,010 617 230 158 1,286 1,347 343 334 4,082 3,882 Source: World Bank 2011c. Note: Estimates are based on data from World Health Organization (WHO) 2010 and United Nations Department of Economics and Social Affairs, Population Division (UN DESA 2009). G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    81 back home, and, consequently, a change in their fetus rose from 60 percent in 2003 to families’ preference for having girls. 73 percent in 2007 (UNFPA 2009). Most societies have some mild degree of With prior knowledge of the sex of the preference for sons (Williamson 1976), but fetus, families can discriminate through less the manifestation of extreme sex ratios comes prenatal investment or even through abor- from rather extreme son preferences. The tion. In China, as well as other countries with interplay of culture, the state, and political prevalent son preference, mothers are 5 per- processes appears to generate extreme patri- cent more likely to acquire prenatal care and lineality and highly skewed child sex ratios visit an antenatal clinic 10 percent more fre- such as in the case of China, northwest India, quently when pregnant with a boy (Bharad- and Korea (Das Gupta 2009). Chung and Das waj and Nelson 2010). Bélanger and Khuat Gupta (2007) argued that son preference in (2009) examined the timing of abortion Korea is correlated with factors such as lower among 885 married women in an obstetric socioeconomic status, rural area residence, hospital in Hanoi, Vietnam, in 2003 to study higher parental control in terms of arranged sex-selective abortions, which generally hap- marriage and co-residence with the parents, pen during the second trimester of pregnancy. and lower education of the woman. They found that women with more daughters In addition, the manifestation of son and without a son were more likely to have preference is also influenced by public a second-trimester than a first-trimester abor- policies and the availability of technology. tion. Their estimates suggest that 2 percent of China’s one-child policy and Vietnam’s all abortions by women with at least one prior two-child policy, though intended to reduce child were intended to avoid a female birth. fertility, may have put additional pressure Given the factors discussed earlier, on the incentives to have a son and inten- the literature shows mixed evidence on sify the skewed sex ratios. In fact, Eben- whether development mitigates or wors- stein (2010) showed evidence of a positive ens son preference and sex ratios at birth correlation between the fines imposed by in Asia (Chung and Das Gupta 2007). China’s one-child policy and the sex ratio. Development can bring about substantial With development and the introduction normative changes within the entire society of prenatal sex determination technology together with improvements in individu- (ultrasound) in the early 1980s, male-to- als’ socioeconomic situations, as argued in female sex ratios became unnaturally very the case of Korea by Chung and Das Gupta high in a few East Asian countries. Li and (2007). However, cross-country evidence Zheng (2009) found a strong impact of the shows that modernization does not appear B-ultrasound technology on the sex ratio to bring down son preference. In South of second-order births for rural mothers in Asia, son preference is greater for women Fujian province, China, but no effect among with more education and is increasing over first-born children. The recent increase in time (Filmer, Friedman, and Schady 2008). Vietnam’s sex ratio at birth may be related Unbalanced sex ratios at birth could be to supply-side factors, that is, access to worsened by economic development, as quality sex determination technology, sometimes argued in the literature, since rather than to an increasing preference for highly educated and wealthier women tend sons. Ultrasound technology first started to to have better access to technologies. Viet- appear in major hospitals in Vietnam dur- nam’s 2006 population survey shows that ing the mid-1990s and was subsequently the sex ratio at birth is high for women who offered through the private sector (Bélanger have a graduate education (113), have high- et al. 2003), but the quality and availability est grade of 10 and above (111), work in a of medical equipment have improved during foreign organization (117), and have previ- the past 10 years. Thus, the proportion of ous knowledge of the baby’s sex (111). Of mothers with prior knowledge of the sex of women with a graduate degree, 87 percent 8 2    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C knew the gender of their child, whereas no suggests that active measures to influence more than 28 percent of illiterate women social norms and facilitate the spread of new had prior knowledge of the sex of their values may be very important, in addition to child. The sex ratio at birth increases with relying on raising female education and labor the level of education, rising from 103 for force participation alone (box 2.2). illiterate women to 113 for women with a graduate education (UNFPA 2009). Risky behaviors in men The collection of evidence has several impli- cations for approaches to address the unbal- Men, as well as women, experience gender- anced sex ratios at birth, and China has taken specific health risks. Men tend to bear the active measures in this direction. General pol- burden of higher morbidity and premature icies to promote economic development might mortality related to substance abuse, war and play a role, but Korea’s recent experience conflict, and violence. The latter tends to be BOX 2.2  Recent improvement in the sex ratio at birth in the Republic of Korea Since the 1970s, Korea has experienced significant in larger part to changing social norms (changes in industrialization and urbanization, coupled with son preference within all education and urban/rural increases in women’s education and labor force par- population groups) and in smaller part to increased ticipation. It is also the first Asian country to regis- urbanization and education (changes due to move- ter a decline in the proportion of missing girls, from ments between education and urban/rural popula- the most male-skewed sex ratios at birth in the mid- tion groups). Another methodological approach 1990s to ratios within the normal range by 2008. using simulations of an economic model also implies As the sex ratios at birth are usually argued to be the impact of development: as the Korean society a manifestation of son preference, Chung and Das becomes richer, households that initially selected Gupta (2007) used data from fertility surveys to boys will select girls because of increasing bride measure trends in son preference directly. They doc- price and declining marginal benefits from unmar- umented a continuous decline from 1985 to 2003 ried sons (Edlund and Lee 2009). in the fraction of Korean women who reported that Second, though the role of Korean public poli- they must have a son, from almost 50 percent to less cies in this process is a mixed story, its experience than 20 percent. Women with similar characteristics suggests that interventions to influence social norms demonstrated lower son preference over time. and facilitate the spread of new values may be very Both the process of development and public important, as opposed to reliance on raising female policies since the 1950s have influenced the fac- education and labor force participation alone. Active tors underlying son preference in Korea. First, the policies to promote rapid economic development in impacts of development were expected to work in Korea played a role early on in breaking down previ- many ways, as argued by Chung and Das Gupta ous norms of son preference as well as raising female (2007): (a) higher earning prospects increased indi- education and labor force participation. And chang- viduals’ independence of family lineage; (b) retire- ing social norms contributed relatively more to the ment savings reduced financial dependence on chil- decrease in son preference. Reforms to policies that dren in old age; (c) urban life setting reduced the directly constrain women’s status, such as the Fam- focus on traditional filial duty and promoted female- ily Law established in 1958, which stipulated male inclusive social networks; (d) females’ greater eco- family headship and inheritance only through the nomic and physical mobility enhanced the value of male line, were slow to follow. With several women’s daughters; and (e) urban life, with assets associated movements demanding greater gender equity since with nonfarm activities and less pressure from cus- the establishment of democracy in 1987, this law tomary laws, facilitated gender equity in inheritance. went through major reforms in 1990, but the system Through a decomposition exercise, the authors of male household headship was not officially abol- attributed the observed reduction in son preference ished until 2005 (Chung and Das Gupta 2007). G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    83 context-specific rather than an issue that pre- FIGURE 2.16  Men are more likely to smoke than women vails in the region. However, two behavioral health issues—smoking and drinking—are prevalence of currently smoking any tobacco product more concerning among men than women in among adults (≥15 years) (%), male vs. female, 2006 all East Asian and Pacific countries, as well 70 as globally.11 Figures 2.16 and 2.17 show the IDN LAO prevalence of smoking and drinking among 60 PRKCHN TON WSM males compared with the prevalence among MYS KOR PHL TUV VUT females. All East Asian and Pacific countries 50 VNM KHM JPN MMR lie above the 45-degree line, implying a much 40 THA MNG PLW males, % higher rate among males than females. SGP MHL FSM The gender difference in the incidence 30 of tobacco use is higher in the East Asia FJI 20 and Pacific region than in other develop- ing regions, even though male dominance 10 in smoking and drinking is a global phe- nomenon. The data for China in 2006 show 0 that tobacco use is 60 percent for men and 0 10 20 30 40 50 60 70 4 percent for women. Similar gender gaps females, % exist throughout the region. Indonesia (62 Source: World Health Organization (WHO) Global InfoBase. percent versus 5 percent), Korea (53 percent versus 6 percent), Lao PDR (64 percent ver- sus 15 percent), and Tonga (62 percent ver- FIGURE 2.17  Men are more likely to drink than women sus 15 percent) lead the region in the largest gender differentials in tobacco prevalence prevalence of drinking among adults (≥15 years) (%), past (figure 2.16). 12 months, male vs. female, 2002–06 KOR JPN Alcohol consumption can vary from occa- 90 sional drinking to heavy episodic drink- 80 CHN MNG ing, and the gender differential in the latter 70 is particularly stark. Although women are 60 PHL less likely to report drinking at all in many KIR LAO SGP SLB countries in the East Asia and Pacific region, 50 VNM males, % THA FSM they do come close to men in places such as WSM 40 FJI Japan and Mongolia (figure 2.17). In terms VUT 30 of heavy episodic drinking (binge drinking) TON and chronic heavy drinking, data show a 20 MMR large gender gap: overall in the region, men 10 IDN MYS are more than twice as likely to be heavy 0 episodic drinkers. According to the World 0 10 20 30 40 50 60 70 80 90 Health Organization 2011 Global InfoBase females, % database, the countries in the region with Source: WHO Global InfoBase. the largest gender gaps in heavy drinking are Kiribati (22 percent among males versus 1 percent among females), Samoa (22 percent in Tonga (38 liters per capita consumption), versus 1 percent), Lao PDR (22 percent ver- Malaysia (32 liters), Thailand (29 liters), and sus 5 percent), Japan (18 percent versus 3 per- Korea (29 liters). cent), Mongolia (14 percent versus 0 percent), These behaviors pose substantial risks to and Micronesia (13 percent versus 1 percent). men’s health and can translate into high costs The level of annual per capita consumption for productivity and economic growth. Glob- of pure alcohol is especially high for males ally, 6 percent of all male deaths are related 8 4    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C to alcohol, compared to 1 percent of female factors may also play a role, such as obesity deaths (WHO 2011a). Half of today’s smok- and heart conditions, which are particularly ers are likely to die from tobacco-related high in the Pacific countries). In China, the causes. A simple cross-country relationship rising risks of noncommunicable diseases, (shown in figures 2.18 and 2.19) between partly linked to men’s smoking and drinking these behaviors and morbidity and prema- behaviors, put greater pressure on the size of ture mortality related to smoking and alcohol the working-age population, already a con- use also indicates positive correlations (other cern because of its aging population (World Bank 2011a). These behaviors are influenced by norms FIGURE 2.18  Tobacco use is positively correlated with mortality about masculinity, cultural beliefs about due to lung cancer health, and the surrounding environment, and they can be slow to change. For men, percentage of current tobacco use among smoking and drinking alcohol are commonly 80 male adults (≥15 years), 2006 viewed as masculine behaviors, and studies 70 TON LAOIDN show that men and boys feel substantial pres- 60 WSM PRK CHN sure to accept gender stereotypes that they PHL KOR VUT MYS should be strong and tough, and the opposite 50 KHM MNG VNM for women. A recent national survey in Viet- percent 40 PLW THA JPN FSM nam found that the primary reason women 30 MHL SGP did not use tobacco was the belief that women 20 FJI should not smoke. In a country where 50 per- 10 cent of men but just over 3 percent of women 0 smoke, 76 percent of the 2,020 young urban 0 20 40 60 80 age-standardized deaths, trachea, bronchus, and lung Vietnamese women surveyed said that this cancers per 100,000 inhabitants, 2008 low female prevalence could be attributed to world East Asia and Paci c linear (world) gender norms (that is, social disapproval of women who smoke). Only 20 ­ percent said Source: WHO Global InfoBase. that the low prevalence was due to health concerns (WHO 2003). In addition, these patterns of smoking behaviors stayed very FIGURE 2.19  Alcohol consumption is positively correlated with stable in Vietnam over the period from 1993 mortality due to alcohol use disorders to 2006. The poor tend to be slightly more likely to drinking prevalence in the past 12 months among report “ever smoking,� but the relationship 120 male adults (≥15 years), 1997–2009 is not strong among those countries where 100 such data is available, except in Cambodia. JPN As shown in figure 2.20, individuals ages 15 80 MNG and above in the poorest quintile in Cam- CHN bodia were almost twice as likely to engage percent 60 SLB PHL FSM LAO in smoking cigarettes or chewing tobacco as WSM VNM 40 FJI those in the richest quintile. Among those TON who smoke, the intensity can vary because 20 VUT MMR of the affordability of cigarettes. In Cam- MYS IDN bodia, particularly in rural areas, the richer 0 0 2 4 6 8 10 12 14 16 18 20 smokers consume more cigarettes per day age standardized deaths, males, alcohol use than the poorer smokers. However, no clear disorders per 100,000 inhabitants, 2008 world East Asia and Paci c linear (world) pattern between income and the intensity of smoking is observed in Mongolia and Source: WHO Global InfoBase. Vietnam. G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    85 Gender equality in productive FIGURE 2.20  The poor are slightly more likely to engage in assets: An unfinished agenda smoking Promoting gender equality in the control percentage of adults (ages 15+) that ever smoked cigarettes or chewed of productive assets (such as land, financial tobacco, comparing the poorest and richest 20% of population 40 capital, social capital, and information and 35 technology) is likely to enhance development, 30 through both economic and empowerment 25 percent benefits. Asset ownership can influence men’s 20 and women’s income and their voice and 15 influence within the household and within 10 5 society. This effect can happen through 0 strengthening their ability to take advantage 1993 1998 2006 2004 2002 of economic opportunities; for example, Vietnam Cambodia Mongolia evidence shows that clear land-ownership poorest 20% richest 20% rights have positive effects on agricultural productivity and access to credit (Deininger Source: World Bank estimates using household income and expenditure surveys of various countries and years. 2003). In the agricultural sector, evidence from Africa and Latin America suggests that ensuring equal access to productive assets and technologies raises agricultural produc- still persist around the world and in some tion (Goldstein and Udry 2008; Quisumbing parts of the East Asia and Pacific region. 1995; Udry 1996). As discussed throughout For example, household surveys in Bangla- this report, income in the hands of women desh, Ethiopia, Indonesia, and South Africa has been shown to positively affect chil- indicate that women bring fewer assets dren’s education and health outcomes (Duflo into marriage (Quisumbing and Maluccio 2003; Lundberg, Pollak, and Wales 1997). 2003). Evidence from the Philippines and Women’s assets prior to marriage have been other developing countries shows that the shown to have positive effects on education husband-wife asset difference at the time expenditures on children in Indonesia and of marriage has not changed over time and other countries (Quisumbing and Maluc- favors the husband, even though gaps in age cio 2003). Beegle et al. (2001) showed that, and education have been closing (Quisumb- in Indonesia, women with a higher share of ing and Hallman 2005). household assets made more use of prenatal This section focuses mainly on land care. Asset ownership can lead to women’s and credit for the following two reasons. empowerment, such as reducing vulnerabil- First, they are the major types of assets ity to domestic violence in India (Panda and that strongly influence well-being. Land Agarwal 2005). and property are usually the most valu- Gender equality in assets has been less able assets for a poor person. Aside from responsive to growth and development than itself being a productive asset, land can be has equality in education and health. Over used as collateral to acquire credit. Access time, economic growth can promote access to financing is also very important because to financing that benefits women and men it usually presents a major barrier to real- as well as improves economic opportunities izing economic opportunities in developing and women’s income. However, development countries. Second, data and rigorous quan- impacts on gender equality in this domain titative evidence on gender and assets in the are constrained by the complex legal, social, East Asia and Pacific region are very lim- and economic factors that shape the control ited, particularly for assets other than land of productive assets. Gender disparities in and credit. Individual-level data of asset access to and control of productive assets ownership are sometimes available for land 8 6    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C holdings but rarely for household durables, even within the same region, and data are since many are considered jointly owned. In available for only a few countries in each the face of scarce individual-level data on region. asset ownership, this section often resorts Evidence on individual land ownership to comparing female-headed households suggests different ownership rates and dif- and male-headed ones.12 This section also ferent ownership composition by gender, draws on qualitative evidence and research depending on the context. In post-tsunami from outside the region to help complement Aceh, Indonesia, women have fewer land the limited quantitative evidence from East holdings than men (Bell 2010). Vietnamese Asian and Pacific countries. men owned more agricultural land plots than women in 2008. Figure 2.23 demon- strates the gender composition of ownership Persistent gender disparities in access among the agricultural land plots with iden- to productive assets tified owners on long-term user right certifi- Analysis of the East Asian and Pacific coun- cates. Overall, and among those plots owned tries shows persistent gender disparities in by rural households, less than 17 percent of productive assets that result in women hav- the plots are owned by a woman, but more ing lower rates of ownership of and access to than 65 percent are owned by a man. Plots land, fewer agricultural inputs such as live- jointly owned by a male and a female repre- stock and less access to extension services, sent a nonnegligible share, in part because of and limited access to credit in some countries Vietnam’s recent land reform (discussed later and subregions. in the chapter). However, women are still clearly at a disadvantage in terms of having Gender-differentiated ownership of land their name on the land title. The same 2008 Female-headed households tend to own less household survey data suggest that the gen- land than male-headed households. Even der gap is even starker among ethnic minori- though the probability of owning land ties. In some contexts, the allocation of land is not substantially lower among female- assets between men and women can depend headed households (figure 2.21), figure 2.22 on the type of land. For example, in matri- indicates that male-headed households own lineal parts of West Sumatra, Indonesia, much more land in terms of land size. This wives own more paddy land, and husbands gap exists even among richer households. own more forest land (Quisumbing and Over time, there have been small improve- Maluccio 2003). ments in women’s land ownership as well as men’s, although the gender gap has not Gender disparities in agricultural inputs necessarily narrowed or disappeared except Female-headed households tend to own for the case of Vietnam in 2006. Similarly less livestock, and female-run farms tend in China, most male-headed and female- to have less access to extension services. In headed households had access to rural land many countries, at least among rural house- in 2008, but the amount of land per capita holds, livestock is one of the most valuable in female-headed households was roughly agricultural assets. It represents a source of 70 percent of that in male-headed house- income, wealth accumulation, and buffer holds (de Brauw et al. 2011). The pattern against shocks. Yet, female-headed house- that female-headed households tend to holds tend to own less livestock across own less land is similar to other developing developing regions (FAO 2011). Figure countries (Agarwal 1994; Deere and Leon 2.24 shows livestock ownership of female- 2003; FAO 2011). Exact comparisons of headed and male-headed households in five the gender gaps across regions are difficult, countries in the region, for households of however, since the incidence of land own- different wealth quintiles. The gender gap ership varies drastically across countries, in Lao PDR appears irrespective of income, G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    87 but this observation does not apply in other FIGURE 2.21  The probability of owning land is not substantially contexts. In places such as Vietnam or lower for female-headed households than for male-headed Timor-Leste, gender gaps are more visible households among poorer households. In many coun- percentage of households that own land, tries, this same pattern is observed in urban 100 by gender of the household head areas as well as rural areas. The data show 90 that gaps also persist over time. 80 70 Extension service provision remains low 60 percent in developing countries, and women tend to 50 have less access to extension services than 40 men (FAO 2011). In Cambodia, few women 30 benefit from agricultural extension services or 20 10 credit made available to rural people, despite 0 the fact that they make up the majority of 8 8 07 6 8 00 00 00 00 farmers and informal sector workers. Agricul- 20 ,2 2 ,2 ,2 a, , lia te m DR di tural research and extension efforts usually do es na go oP bo -L et on m La or Vi not consider women’s activities—seed prepa- M Ca Tim ration and planting—or take into account the female-headed households male-female gap fact that men and women tend to specialize in Source: World Bank estimates using household income and expenditure surveys of various countries different rural tasks. Distance to the point of and years. service provision, lack of female agents, and insensitivity to illiterate customers (the major- FIGURE 2.22  Female-headed households own less land in terms of ity of whom are often women) are other rea- land size sons for this lower access of female farmers to extension services (UNIFEM, World Bank, average size of land among households that own land, ADB, UNDP, and DFID/UK 2004). comparing the poorest and richest 20% 3.2 Gender gaps in access to credit 2.7 Evidence on gender differentials in access to 2.2 credit is mixed. Female-headed households 1.7 hectares are slightly less likely to borrow from finan- 1.2 cial institutions, as illustrated by figure 2.25. 0.7 In most of the East Asian and Pacific coun- 0.2 tries for which data are available, the gap –0.3 between households headed by females and males varies widely both across and within –0.8 poorest 20% richest 20% poorest 20% richest 20% poorest 20% richest 20% poorest 20% richest 20% poorest 20% richest 20% countries. The most recent data available indicate that gaps have been small except for Timor-Leste, rural Lao PDR, rural Cambo- Cambodia, Lao PDR, Mongolia, Timor- Vietnam, dia, and urban Mongolia. Using survey data 2008 2008 2008 Leste, 2007 2006 from 2000, de Brauw et al. (2011) found no female-headed households male-female gap difference between female-managed farms and male-managed farms in China in terms Source: World Bank estimates using household income and expenditure surveys of various countries and years. of access to credit. As discussed in chapter 3, female-run and female-owned firms in East Asia, at least those in the formal sector, do headed households borrowed less, had less not appear to be systematically more con- access to formal credit, and paid higher inter- strained in accessing finance than male-run est on loans than male-headed households. firms. However, a joint study by FAO/UNDP Individual-level data on access to financ- (2002) in Vietnam revealed that female- ing, by gender, suggest that the gender gap 8 8    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE 2.23  Vietnamese men owned more agricultural land plots countries surveyed, women were as likely than did women in 2008 as men to report having an account in a formal financial institution in Cambodia percentage of land plots with identi ed owners on and Thailand; less likely to have an account the certi cates, Vietnam, 2008 in China, Indonesia, Lao PDR, Malaysia, total 16.9 65.1 18.0 and Vietnam (although the difference is statistically significant only in Malaysia and Vietnam); and more likely to have an urban 19.7 61.2 19.1 account in Mongolia and the Philippines. A small-scale study of Vietnam’s rural credit market in 2002 indicated that credit ration- rural 16.8 65.3 18.0 ing depended on education and credit his- tory, but found no evidence of bias against women (Barslund and Tarp 2003). The 0 20 40 60 80 100 percent Thailand 2005 Household Socio-Economic only a female owner only a male owner joint owners Panel data show a similar rate between men and women holding a savings account in a Source: World Bank estimates using VHLSS (GSO Vietnam), 2008 data. financial institution (46 percent and 49 per- cent, respectively). Although this similarity is observed in Thailand’s urban as well as FIGURE 2.24  Female-headed households are less likely to own rural areas, the situation may vary between livestock rural and urban areas of other countries and might be very different in the Pacific, ownership of livestock by poorest and richest 20% of for which individual data are lacking. 100 population, by gender of household head share of households that own 90 80 70 Limited effects of economic growth on gender gaps in assets livestock, % 60 50 40 E conomic grow th could, in principle, 30 20 increase women’s asset holdings by increas- 10 ing income, but this factor alone is insuf- 0 –10 ficient to close the gender gap in asset hold- ings. Market transactions are an important poorest, 20% richest, 20% poorest, 20% richest, 20% poorest, 20% richest, 20% poorest, 20% richest, 20% poorest, 20% richest, 20% way to accumulate assets, and evidence from Latin America and the Caribbean sug- gests that, after inheritance, markets are the Cambodia, Indonesia, Lao PDR, Timor-Leste, Vietnam, 2008 1999 2008 2007 1998 second most important channel for women female-headed households male-female gap in that region to acquire land. In that sense, income plays an important role. However, Source: World Bank estimates using household income and expenditure surveys of various countries as shown in chapter 3, substantial gender and years. gaps in income still persist in the East Asia and Pacific region, despite economic growth. Other complex legal and social factors also in access to formal finance is likely to be make it very challenging to close the gender small in East Asia. Evidence from G ­ allup gap in asset holdings with economic growth surveys conducted since 2011 suggests alone. Evidence from South Asia indicates that men and women have similar access that better employment opportunities and to formal finance in East Asian countries progressive legislation do not necessar- (Demirgüç-Kunt and Klapper 2012). As ily lead to gender equality in access to and figure 2.26 shows, of the nine East Asian control of land, because of social factors G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    89 (Agarwal 1994). Actually, wealthier house- FIGURE 2.25  Female-headed households are slightly less likely to holds with more valuable assets are not nec- borrow from a financial institution essarily willing to give women more owner- ship rights. Analysis of the Vietnam 2008 percentage of households that borrowed from a nancial institution household survey data by World Bank staff (among those that borrowed money) by gender of the household head shows that richer households are less likely 100 to have a female name in the title of their 90 agricultural land plots, accounting for fac- 80 70 tors such as land size and basic household 60 percent characteristics. 50 40 As part of the development process, the 30 expansion of microfinance coverage has 20 been argued to improve women’s access to 10 0 credit. In terms of coverage, microfinance has 8 7 09 8 8 6 greatly expanded all over the world, reach- 00 00 00 00 00 20 2 ,2 ,2 ,2 ,2 a, d, te lia ing many poor clients (Daley-Harris 2009). DR m di an es na go bo oP ail -L et on m In China, microcredit has been used in vari- or La Th Vi Ca M Tim ous instances to support women. The Tianjin female-headed households male-female gap Women’s Association for Business Develop- ment and Promotion, the Guangxi Provincial Source: World Bank estimates using household income and expenditure surveys of various countries and years. Women’s Federation, and Liuzhou Municipal Women’s Federation are examples of micro- credit schemes that target poor, laid-off, and FIGURE 2.26  Women are slightly less likely than men to report unemployed women (ADB 2006a). Not all having an account at a formal financial institution countries provide such targeted support. In Indonesia, although women are considered percentage of men and women reporting having an account, by to be an important market for microfinance, themselves or with someone else, at a nancial institution or post o ce 90 the Indonesian microfinance industry has 80 never made targeting of women a hallmark 70 of their business. The average proportion of 60 percent female clients in Indonesia served by major 50 40 microfinance institutions has remained fairly 30 constant over the past 20 years (Asia Foun- 20 dation, ADB, CIDA, NDI, and World Bank 10 0 2006). However, rapid expansion of microfinance * In dia nd es * R a lia sia ci d sia m in PD Pa an in go la ne Ca c na Ch bo ay ilip ai o ia on does not necessarily imply a de facto control do et m al La Th As Ph Vi M M st of resources. Goetz and Gupta (1996) show Ea that women’s access to microcredit has not all men women been matched by an increase in their control Sources: World Bank Global Financial Inclusion (Global Findex) database; Demirgüç-Kunt and of these funds. Microcredit facilities have Klapper 2012. been established for women in Lao PDR, * Denotes a statistically significant difference between men and women at the 1 percent level. but women’s role in decision making has not improved (GRID 2005). Although women Legal and social constraints to make up a high proportion of membership in equalizing access to assets credit schemes in Cambodia, they tend to be excluded from the decision-making processes To understand the observed gender dispari- and receive smaller amounts of credit than ties in asset holdings, one must understand men (UNIFEM, World Bank, ADB, UNDP, how assets are accumulated and the factors and DFID/UK 2004). determining asset accumulation. Individuals 9 0    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C can accumulate, or lose, assets in several These countries—Cambodia, China, Lao ways. First, individuals can make market PDR, Mongolia, Thailand, and Vietnam— transactions—such as buying land or live- do not have plural legal systems (for example, stock, or acquiring a bank loan. Second, customary or religious laws), which means assets are acquired through inheritance or that all citizens adhere to civil law. For exam- through allocation or acquisition by the state, ple, in China, many advances toward ensur- for example, through land redistribution. ing equal treatment and the protection of the Inheritance is one of the main mechanisms rights of women under the law occurred as for asset accumulation (Deere and Doss early as the 1950s. Women’s property and 2006), and equality in asset endowments inheritance rights were protected through the can be limited by differences in the right to enactment of the 1982 constitution, which inherit property. Third, life-cycle events such protects the right of citizens to inherit private as marriage, including marriage payments, property (Article 13),14 and the 1985 Law of or separation also alter asset holdings. Succession of the People’s Republic of China, Complex legal, social, and economic fac- which states that males and females are equal tors determine or constrain asset accumula- in their right to inheritance (Article 9).15 The tion: (a) formal institutions—particularly the 1992 Law of the People’s Republic of China legal framework for property and inheritance on the Protection of Rights and Interests of rights, family laws, and law enforcement; Women further promotes gender equality; (b) informal institutions—social norms and Article 28 declares that the state shall guar- customary laws affecting women’s prefer- antee that women enjoy the equal right, with ences and ability to acquire and accumulate men, to property.16 assets; and (c) human capital and other eco- In several of the countries examined, nomic factors, such as income and the rate of males and females are not treated equally, returns on productive assets. The impact of mostly under inheritance laws. Plural legal economic growth on promoting more equal systems exist in Indonesia, Malaysia, the asset holdings is limited, as discussed earlier, Philippines, and Singapore. Muslim laws because legal and social barriers often act as govern the majority of the populations in binding constraints. Despite positive changes Indonesia and Malaysia and a small minor- in the legal framework in the East Asia and ity of the population in the Philippines and Pacific region, the interactions between for- Singapore. For example, according to the mal and informal institutions still leave Islamic Law Compilation in Indonesia, when women at a disadvantage with respect to the a married person dies, each son is entitled to control of assets.13 The discussion below first receive a share twice as large as each daugh- describes the legal framework and then high- ter (Asia Foundation, ADB, CIDA, NDI, and lights the challenges in practice due to weak World Bank 2006). The autonomous Muslim implementation and the influence of norms region of Mindanao in the Philippines can and customary laws. independently promulgate its own legisla- tion following Islamic law (as allowed in the The legal framework constitution). Although the Philippines is a The majority of countries in East Asia no lon- community property regime,17 the Muslim ger differentiate by gender in statutory law. In family code reflects the husband having the that respect, the East Asia and Pacific region final say concerning the handling of joint differs from some other regions: inheritance property.18 Among Pacific Island countries, rights are still unequal in the Middle East and Kiribati and Tuvalu have unequal statutory North African countries and half of the coun- legislation. Equal inheritance laws exist in tries in South Asia (World Bank 2011c). As Fiji, Papua New Guinea, Samoa, the Solomon shown in table 2.4, most East Asian countries Islands, and Vanuatu; however, customary have legislation for property and inheritance law in relation to land has constitutional rights with no discrimination against women. status in these countries and may ­ lawfully G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    91 TABLE 2.4  Most East Asian and Pacific countries do not differentiate by gender in inheritance and property laws Question Country Answer Law Do sons and daughters have Cambodia Yes The Inheritance Law equal inheritance rights? China Yes Law of Succession (Article 9) Indonesia No Islamic Law Compilation, Book II Lao PDR Yes The Inheritance Law Malaysia No Constitution Mongolia Yes Civil Code (Part V) Philippines Noa Presidential Decree No. 1083 Singapore Noa Intestate Succession Act; Administration of Muslim Law Act Thailand Yes Civil and Commercial Code (Sections 1599–1710) Vietnam Yes Civil Code (Art. 635) Fiji Yesb Succession, Probate and Administration Act [Cap 60] 1970 Kiribati No Laws of Kiribati Act 1989, Schedule 4 Do men and women have equal Cambodia Yes Constitution (Arts. 31 and 45); Law on Marriage and Family (Arts. 29, 32–37) rights over property? China Yes Law on Protection of Women’s Rights (Arts. 30 and 47) Indonesia Yes Marriage Law No. 1 of 1974 Lao PDR Yes Law of Property (Arts. 20, 26) Malaysia Yes Constitution Mongolia Yes Civil Code (Ch. 12) Philippines Noa Presidential Decree No. 1083 Singapore Yes Women’s Charter (Arts. 51, 52, 56) Thailand Yes Constitution (Section 30); Civil and Commercial Code (Book IV) Vietnam Yes Land Law; Civil Code (Sec. 8) Fiji No Constitution (Amendment) Act 1997 Laws of Kiribati Act 1989, Schedule 4; Magistrates Court Act; Gilbert and Kiribati No Phoenix Islands Land Code Sources: Women, Business and the Law database; Jivan and Forster 2007. a. Implies that unequal legal systems apply only to a minority population. b. Denotes that despite equal inheritance in legislation, Fijian custom in relation to land has constitutional status and may lawfully discriminate against women. discriminate against women (Jivan and women to access loans, empowers women in Forster 2007). case of disputes, and leads to higher mutual Beyond the protection of equal inheritance decision making (World Bank 2008). rights, several countries in the region have recently adopted legal changes that actively Weak implementation and enforcement promote better gender equality in access to of the law land. Since concerns have been raised about Although women and men may be equal promoting gender equity in land titling pro- under the law, these legal rights do not always grams, places such as Indonesia, Lao PDR, translate into equal access to land in practice. and Vietnam have recently adopted gender- Progress toward gender equality in assets is sensitive reforms in land titling. Since the still limited because of weak implementation 2004 Land Law in Vietnam, all new land and enforcement of the law. tenure certificates must include the names of Implementation of the laws may be imper- both spouses, a provision intended to reduce fect and ineffective as a result of challenges gender inequality in access to land, protect in incorporating existing cultural norms and families against unilateral decisions by one practices. Even where the legal system sup- spouse, and protect women in case of divorce ports equal access to land, traditional values or disputes. Qualitative analysis of impacts and norms create difficulty with respect to in three provinces suggests that joint titling enforcing the legislation. After the 2004 Land improves procedures and opportunities for Law requiring joint titling was introduced in 9 2    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C Vietnam, the results have been varied across governing land ownership. They are also ethnic majority and minority groups. Analy- less likely to know about land registration sis of the Vietnam Household Living Stan- requirements (including whether the land dards Survey 2008 shows that ethnic minor- owned by their households is registered and ity households are much less likely to have a whether their names are included in land female name in the title of their agricultural ­ ocuments). Dissemination activi- titling d land plots, even when accounting for factors ties on rights and entitlements can exclude such as land size and basic household char- women, especially in rural areas.19 Even acteristics. Studies of existing kinship sys- if they have information about their legal tems in China and Vietnam show resistance land rights, women tend not to pursue for- to endowing daughters with land, as land is mal complaints in cases of land grabbing or considered lost when daughters get married disputes regarding inheritance or division of (Bélanger and Li 2009). In Northern Liaon- property. A 2008 AusAID report on land ing, China, women are more likely to become tenure in the Pacific region stresses the fol- landless at marriage because the population lowing factors constraining women’s access control and land tenure policies reinforce to the formal legal system to resolve disputes traditional forms of gender bias: a daughter regarding land ownership and use: (a) the is expected to marry out, whereas a son is system was culturally unfamiliar and based expected to reside with or near his parents on “adversarial� methods rather than on the (Chen and Summerfield 2007). Therefore, consensus building usually preferred by com- during the redistribution of farmland to munities; (b) women also faced more “prac- households based on household size, town- tical� obstacles to accessing courts (limited ship and village officials allocated a larger access to transport, lack of time, income); (c) share of land for each son in the household the “technical nature� of the procedures and than for each daughter. inadequate support from court staff were also The Lao PDR land titling program noted as important barriers (AusAID 2008). 1997–2010 offered useful implementation lessons to account for context-specific social Social norms and practices norms. An early assessment during the first In some contexts, customary practices, rather phase of the program noted that traditional than statutory laws, directly govern land own- family roles dedicated the man to handling ership and land use. Most land in the Pacific taxes and, thus, to having only his name on region is under customary authority (approxi- land-related tax documents. Mostly men mately 80 percent of total land area) (AusAID interacted with government officials and par- 2008). Although the rules vary widely within ticipated in information meetings and titling the Pacific region, usually in customary sys- activities. As a response to this assessment, tems, (a) land can only be transferred within Lao PDR introduced a stronger gender inclu- networks of social/political relationships, sion program, engaging the Lao Women’s (b) land use is governed by reciprocal rela- Union, to raise community awareness of land tions within the kinship or customary land titling, to include special training for women groups, and (c) social hierarchies and status on their rights, and to ensure their partici- are important factors determining one’s rights pation during titling. Following this active to land. In these customary systems, women engagement, a higher number of titles went have access to land primarily through their to women than to men (Lao Land Titling kinship relations with men. Women also have Project II ICR 2010). less voice in public decisions about the use Qualitative research, more broadly, has of land (AusAID 2008). For example, Fijian stressed that a lack of information about women in most parts of the Fiji Islands are legal entitlement is a key barrier to enforcing excluded from inheritance rights in custom- women’s land rights throughout the region. ary land and have no rights to land other than Women are generally less aware of the laws those permitted by their fathers or husbands. G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    93 Nor do they normally receive land rents. Most have an impact. What follows is an initial dis- Indo-Fijians with land also practice father- cussion about policy priorities for promoting to-son inheritance (ADB 2006b). gender equality in human capital: (a) closing Practices within the region and within persistent gender gaps in human develop- countries vary considerably. Inheritance ment, (b) reducing gender streaming in edu- practices based on norms can be patrilineal, cation, (c) promoting balanced sex ratios at matrilineal, or bilateral and, as such, are not birth, and (d) addressing male-specific gender always tilted against women. In the matrilin- issues. Policy approaches to close gender gaps eal society of Sumatra, Indonesia, together in assets are important for gender equality in with the shift from communal to individ- endowments, and they also enable more equal ual tenure, the inheritance system became economic opportunity and agency. These pol- more egalitarian in that sons and daughters icies are discussed within the general frame- inherit the type of land that is most inten- work of policies to promote gender equality sive in their own work effort, and gender in economic opportunity, which is discussed bias in land inheritance is either nonexistent in chapter 3 and chapter 6. A more detailed or small (Quisumbing and Otsuka 2001). In discussion will follow in chapter 6 on policies the Philippines, sons are preferred in land to promote overall gender equality and more inheritance but daughters are favored in edu- effective development. cation investments (Estudillo, Quisumbing and Otsuka 2001). In Lao PDR, land inheri- Closing persistent gender gaps in tance and ownership are important elements human development of women’s autonomy in lowland areas, with daughters customarily inheriting land. How- For countries with overall low and unequal ever, women in the midlands and highlands, gender outcomes in education and health, such as those in the Khmu and Hmong ethnic the priority remains to improve these out- groups, face important barriers to controlling comes. Actions to strengthen the education land (Ireson-Doolittle 1999). Understanding and health systems are called for to improve the specifics of a particular context is thus overall outcomes, in addition to any focus on important for designing appropriate policies gender. Interventions may be needed at the and interventions. national level and may yield high economic In summary, given the limited evidence in and social returns. For countries with local- the East Asia and Pacific region, understand- ized gender disparities among certain ethnic ing gender inequality in assets is an important groups or low-income regions, interventions research agenda. Gender disparities in access may be targeted to these groups. Though the to productive assets can hinder women’s abil- exact constraints vary by country context, ity to participate and benefit from economic the analysis in this chapter has shown that opportunities as well as constrain women’s both demand-side and supply-side factors voice and representation in the society. Fur- are responsible for these poor human capi- ther research and better data are called for tal outcomes. Policies can have an impact to disentangle the complex mechanisms influ- through improving service delivery (infra- encing men’s and women’s control of assets structure, staffing, incentives, use of infor- and to shed light on policies. mation and communication technology) and demand-side interventions (conditional cash transfers, information campaigns, account- Policies to promote gender ability). For education, policies to improve equality in endowments education outcomes in general are expected The analysis thus far has identified the fac- to also improve gender equality. For health, tors influencing or constraining gender equal- the slow improvements in health outcomes ity in endowments. This analysis also sheds in several East Asian and Pacific countries light on where policies may be used and can underscore the importance of improving 9 4    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C service delivery in general, including efforts hamper growth and development. First, to account for gender norms that affect ser- the initial signs of the reversed gender gap vice utilization and effectiveness. in education need to be monitored closely where applicable. Second, the excessive Reducing gender streaming in tobacco and alcohol consumption among education males in many parts of the East Asia and Pacific region deserves policy attention For many countries in the East Asia because the social costs, passed on as exter- and Pacific region, addressing education nalities to other members of the society, are ­ quality—specifically, gender streaming in usually higher than private costs. Possible education—requires policy attention. Many measures to tackle this challenge include aspects of gender issues regarding basic access providing information about the health in education and health have abated with risks of excessive tobacco and alcohol con- growth and development. However, con- sumption, taxation, regulatory measures on certed efforts in education and labor market advertisement, and restrictions on smoking policies are needed to break the traditional in public sites. patterns of females going into certain fields of study and, consequently, jobs in lower- paying occupations and sectors. Within the Notes education system, possible approaches in this   1. See Malhotra, Pande, and Grown (2003) on agenda include both curriculum reforms to impacts of investments in female education reduce gender stereotyping and active inter- on gender equality. ventions—financial and nonfinancial incen-  2. See Engendering Development (World Bank tives as well as information campaigns—to 2001) and World Development Report 2012 promote entrance into nontraditional fields. (World Bank 2011c) for similar literature. However, few studies rigorously identify the causal effects as opposed to simple Promoting balanced sex ratios at birth correlation. And female education or income In the few countries with “missing girls� at might not always have dominant, widespread birth, rooted in the prevalence of son prefer- benefits over male education as commonly ence, continuing efforts are needed. A promis- perceived. For example, controlling for household average education, Breierova and ing strategy is to adopt policy approaches that Duflo (2004) found no impact of female aim to enhance the relative value of daugh- education on infant mortality in Indonesia. ters as perceived by families. General policies Edmonds (2006) found that in South Africa, to promote economic development may play pension money going to grandmothers a role, but Korea’s recent experience suggests improved children’s health while that going to that interventions to influence social norms grandfathers improved children’s schooling. and facilitate the spread of new values may be   3. Enrollment in different types of education very important, rather than simply relying on also shows gender differences that vary raising female education and labor force par- across countries in the region. Females’ ticipation. Information campaigns, financial completion rate in vocational training has incentives, and improved social security for been increasing in Thailand and Vietnam. In recent years, this rate among females is still the elderly are worthwhile efforts. China has lower than that among males in Indonesia, been adopting many of these programs, and Thailand, Vietnam, and Mongolia, even they can be expected to reduce the imbalance though the reverse tends to be true for in the sex ratio at birth. completion of general secondary education. Cambodia experiences the opposite patterns, Addressing male-specific gender issues that is, women have lower general secondary completion rates but higher vocational Attention to male gender issues is crucial in completion rates than men (Sakellariou many country contexts since they may also 2011). G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    95   4.  Conflict and postconflict areas also suffer Looking at gender of the household head poor education outcomes for both boys and as an approximate measure presumes that girls, for different reasons, such as the risk females in female-headed households own for boys of being taken out of school to join and control most of the assets while females the military and the risk of safety for girls at in male-headed households control and own the school. relatively few assets. This measure could   5. Girls do not always lag boys in poor rural be misleading in cases where male heads of areas. Evidence from China’s Gansu province households are temporarily absent. shows no significant gender disadvantage 13. Recent analysis of women and land in (Hannum and Adams 2002). Association of Southeast Asian Nations   6. The returns to education for women have (ASEAN) countries identifies the following not increased uniformly: the returns for supply constraints to women’s equal access women increased relative to those of men to land in the region: unfavorable legal in Vietnam but decreased in Indonesia and framework, pro-male customs, limited Thailand since the late 1990s (Sakellariou opportunities, and lack of data (ASEAN 2011). 2008).   7. See Kobia (2009) and Lloyd, Mensch, and 14. http://english.people.com.cn/constitution/ Clark (1998), for example, for analysis of constitution.html. similar gender stereotyping in Kenya. 15. http://www.chinaembassycanada.org/eng/   8. Prior empirical evidence on the extent to lsfw/Relevant%20Chinese%20Laws%20 which income causally affects health status and%20Regulations/t37737.htm. has been controversial, partly owing to 16. h t t p : / / w w w.w o m e n o f c h i n a . c n / h t m l / methodological challenges such as possible report/515-1.htm. omitted variables and the reversed feedback 17. In a community property jurisdiction, most from health to income (Deaton 2006; Filmer property acquired during the marriage and Pritchett 1999; Pritchett and Summers (except for gifts or inheritances) is owned 1996). jointly by both spouses and is divided upon   9. Das Gupta, Chung, and Shuzhuo (2009) divorce, annulment, or death. argue that the recent provincial sex ratios 18. h t t p : // w w w. l aw. e m o r y. e du / i f l / l e g a l / in China suggest an incipient turnaround philippines.htm. of the “missing girls� phenomenon in East 19.  In China, for example, it is a common Asia. However, the concern is far from over practice for new policies, regulations, and because child sex ratios in China are still programs to be discussed in village meetings high. in which heads of households (primarily 10. See World Bank 2011c, World Development men) participate. Information on changes Report 2012, chapter 3 technical annex for to land laws, for example, might reach detailed methodology. women later (or partially, as they may rely 11. Prevalence estimates of current smoking of on other household members to convey the any tobacco product result from the latest information) (Liaw 2008). In Indonesia, adult tobacco use surveys, which have been land acquired during marriage tends to adjusted according to the WHO regression be registered primarily under the name of method for standardizing described in the the male head of household because most Method of Estimation. “Tobacco smoking� landowners are not aware of the possibility includes cigarettes, cigars, pipes, or any of registering land in more than one person’s other smoked tobacco products. “Current name (Brown 2003). smoking� includes both daily and nondaily or occasional smoking. 12. Households headed by women and those References headed by men may be very different as a ADB (Asian Development Bank). 2005. Country result of unobserved factors. The observation Gender Assessment Timor-Leste. Pacific that female-headed households have fewer Regional Department and Regional and assets does not necessarily mean that Sustainable Development Department, Asian women have less access to assets in general Development Bank East and South East Asia (many are widows or single-parent families Regional Office United Nations Development that face economic difficulties anyway). Fund for Women. Manila, Philippines: ADB. 9 6    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C —— —. 2006a. ADB Country Gender Assessment: Behrman, Jere Richard, and James C. Knowles. Peopl e’s R e p ublic of C hin a . M a n i la , 1999. “Household Income and Child Schooling Philippines: ADB. in Vietnam.� World Bank Economic Review ———. 2006b. Republic of the Fiji Islands: 13: 211–56. C ountr y Ge nde r A ssessme nt . Manila, Bélanger, Danièle, and T. H. Oanh Khuat. 2009. Philippines: ADB. Second-Trimester Abortions and Sex-Selection Agarwal, Bina. 1994. A Field of One’s Own: of Children in Hanoi, Vietnam. Population Gender and Land Rights in South Asia. Studies 63 (2): 163–71. Cambridge, U.K.: Cambridge University Bélanger, Danièle, and Xu Li. 2009. “Agricultural Press. Land, Gender and Kinship in Rural China and Albanesi, Stephania, and Claudia Olivetti. Vietnam: A Comparison of Two Villages.� 2009. “Production, Market Production Journal of Agrarian Change 9 (2): 204–30. and the G ender Wage Gap: I ncentives Bélanger, Danièle, and Giang Linh Tran. 2011. and Expectations.� Review of Economic “The Impact of Transnational Migration of Dynamics 12 (1): 80–107. Gender and Marriage in Sendign Communities ASEAN (Association of Southeast Asian Nations) of Vietnam.� Current Sociology 59 (1): 59–77. 2008. “ASEA N Continues to Empower Bélanger, Danièle, Thi Hai Oanh Khuat, Jianye Women.� http://www.aseansec.org/Bulletin- Liu, Thanh Thuy Le, and Viet Thanh Pham. Feb-08.htm#Article-2. 2003. “Are Sex Ratios at Birth Increasing in Asia Foundation, ADB (Asian Development Bank), Vietnam?� Population (English Edition) 58 CIDA (Canadian International Development (2): 231–50. Agency), NDI (National Democratic Institute), Bell, Keith Clifford. 2010. “Study on Gender and World Bank. 2006. Indonesia: Country Impacts of Land Titling in Post-Tsunami Gender Assessment. Manila, Philippines: Asia Aceh, Indonesia.� Photocopy, World Bank, Foundation, ADB, CIDA, NDI, and World Washington, DC. Bank. Bharadwaj, Prashant, and Leah K. Nelson. 2010. Atkin, David. 2010. “Endogenous Skill Acquisition “Discrimination Begins in the Womb: Evidence a nd M a nu fac t u ri ng i n Mex ico.� Yale of Sex-Selective Prenatal Investments.� University, Department of Economics, New Photocopy, University of California, San Haven, CT. http://www.econ.yale.edu/~da334/ Diego. Endogenous_Skill_Acquisition_Mexico.pdf. Blumberg, Rae Lesser. 2007. “Gender Bias in AusAID. 2008. Making Land Work: Vol. 1. Textbooks: A Hidden Obstacle on the Road Reconciling Customary Land and Development to Gender Equality in Education.� Paper in the Pacific. Canberra: Australian Agency for commissioned for the Education for All Global International Development. Monitoring Report 2008, Education for All Bailey, B., and M. Bernard. 2003. “Establishing a by 2015: Will We Make It? http://unesdoc Database of Gender Differentials in Enrolment .unesco.org/images/0015/001555/155509e.pdf. and Performance at the Secondary and Tertiary BPS (Badan Pusat Statistik). Indonesia National Levels of the Caribbean Education Systems.� Socioeconomic Survey (SUSENAS). Jakarta, Canada-Caribbean Gender Equality Fund Indonesia. Programme II (CARICOM), Georgetown, Breierova, Lucia, and Esther Duflo. 2004. “The Guyana. Impact of Education on Fertility and Child Baird, Sarah, Jed Friedman, and Norbert Schady. Mortality: Do Fathers Really Matter Less 2007. “Aggregate Income Shocks and Infant Than Mothers?� NBER Working Paper 10513, Mortality in the Developing World.� Policy National Bureau of Economic Research, Research Working Paper 4346, World Bank, Cambridge, MA. Washington, DC. Brown, Jennifer. 2003. “Rural Women’s Rights in Barslund, Mikkel, and Finn Tarp. 2003. “Rural Java, Indonesia: Strengthened by Family Law Credit in Vietnam.� Discussion Paper, but Weakened by Land Registration.� Pacific University of Copenhagen, Denmark. Rim Law and Policy Journal 12 (631): 51. Beegle, Kathleen, Elizabeth Frankenberg, and Cambodia Development Resource Institute. Duncan Thomas. 2001. “Bargaining Power 20 01. “S ocial A ssessment of L a nd i n within Couples and Use of Prenatal and Cambodia—A Field Study.� Cambodia Delivery Care in Indonesia.� Studies in Family Development Resource Institute, Phnom Planning 32 (2): 130–46. Penh, Cambodia. G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    97 Cameron, L., and C. Worswick. 2001. “Education Know and Why Does It Matter?� Feminist Expenditure Responses to Crop Loss in Economics 12 (1–2): 1–50. Indonesia: A Gender Bias.� Economic Deere, Carmen Diana, and Magdalena Leon. Development and Cultural Change 49 (2): 2003. “The Gender Asset Gap: Land in 351–63. Latin America.� World Development 31 (6): Chattier, Priya. 2011. “Gender and Economic 925–47. Choice.� Background paper for the World Deininger, Klaus. 2003. Land Policies for Growth Development Report 2012. University of the and Poverty Reduction. Washington, DC: South Pacific Laucala Campus, Suva, Fiji. World Bank. Chen, Junjie, and Gale Summerfield. 2007. Demirgüç-Kunt, Asli, and L eora K lapper. “Gender and Rural Reforms in China: A Case 2012. “Measuring Financial Inclusion: The Study of Population Control and Land Rights Global Findex Database.� Policy Research Policies in Northern Liaoning.� Feminist Wo r k i n g P a p e r 6 0 2 5 , Wo r l d B a n k , Economics 13: 63–92. Washington, DC. Chung, W., and Monica Das Gupta. 2007. “The Duflo, Esther. 2000. “Child Health and Household Decline of Son Preference in South Korea: Resources in South Africa: Evidence from The Roles of Development and Public Policy.� the Old Age Pension Program.� American Population and Development Review 33 (4): Economic Review 90 (2): 393–98. 757–83. —— —. 2003. “Grandmothers and Granddaughters: Cutler, D., and G. Miller. 2005. “The Role Old-Age Pensions and I nt rahousehold of Public Health Improvements in Health Allocation in South Africa.� World Bank Advances: The Twentieth Century United Economic Review 17 (1): 1–25. States.� Demography 42 (1): 1–2. Dunne, M., and F. Leach. 2005. “Gendered Daley-Harris, Sam. 2009. State of the Microcredit School Experiences: The Impact on Retention Summit Campaign Report 2009. Washington, and Achievement in Botswana and Ghana.� DC: Microcredit Summit Campaign. Researching the Issues 56, Department for Das Gupta, Monica. 2005. Explaining Asia’s International Development, London. “Missing Women�: A New Look at the Data. Ebenstein, A. 2010. “The Missing Girls of China Population and Development Review 31 (3): and the Unintended Consequences of the One 529–35. Child Policy.� Journal of Human Resources ——— . 20 09. “Fa m i ly System s , Pol it ic a l 45 (1): 87–115. Systems, and Asia’s ‘Missing Girls.’� Policy Eccles, J. S., and P. Blumenfeld. 1985. Classroom Research Working Paper 5148, World Bank, Experiences and Student Gender: Are There Washington, DC. Differences and Do They Matter? In Gender Das Gupta, Monica, W. Chung, and L. Shuzhuo. Influences in Classroom Interaction, edited by 2009. “Is There an Incipient Turnaround in L. C. Wilkinson and C.B. Marrett. Orlando, Asia’s ‘Missing Girls’ Phenomenon?� Policy FL: Academic Press. Research Working Paper 4846, World Bank, Edlund, L., and C. Lee. 2009. “Son Preference, Sex Washington, DC. Selection and Economic Development: Theory Davis, James. 2002. “Boys to Men: Masculine and Evidence from South Korea.� Discussion Diversity and Schooling.� Paper presented at Paper 0910 - 0 4, C olu mbia Un iversit y, the School Leadership Centre of Trinidad and Department of Economics, Chicago, IL. Tobago, Port of Spain, Trinidad, August 6–8. Edmonds, E. 2006. “Child Labor and Schooling Deaton, Angus. 2006. “Global Patterns of Income Responses to Anticipated Income in South and Health: Facts, Interpretations, and Africa.� Journal of Development Economics Policies.� WIDER Annual Lecture, Helsinki, 81 (2): 386–414. September 29. Eskes, T., and C. Haanen. 2006. “Why Do Women d e B r au w, A l a n , J i k u n H u a n g , L i n x iu Live Longer Than Men?� European Journal Zhang, and S cot t Rozelle. 2011. “T he of Obstetrics Gynecology and Reproductive Feminization of Agriculture with Chinese Biology 133 (2): 126–33. Characteristics.� Background paper for the Estudillo, J., Agnes Quisumbing, and Keijiro World Development Report 2012 , World Otsuka. 2001. “Gender Differences in Land Bank, Washington, DC. Inheritance, Schooling and Lifetime Income: Deere, Carmen Diana, and Cheryl Doss. 2006. Evidence from the Rural Philippines.� Journal “The Gender Asset Gap: W hat Do We of Development Studies 37 (4): 23–48. 9 8    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FAO (Food and Agriculture Organization)/UNDP Goetz, A. M., and R. Gupta. 1996. “Who Takes the (United Nations Development Programme). Credit? Gender, Power, and Control over Loan 2002. “Gender Differences in the Transitional Use in Rural Credit Programs in Bangladesh. Economy of Viet Nam.� Hanoi. http://www World Development 24 (1): 45–63. .fao.org/docrep/005/ac685e/ac685e00.htm. Goldstein, Markus, and Christopher Udry. FAO. 2011. The State of Food and Agriculture 2008. “The Profits of Power: Land Rights 2010 –11: Women in Agriculture: Closing and Agricultural Investment in Ghana.� the Gender Gap for Development. Food and Journ al of Politic al Economy 116 (6): Agriculture Organization, Rome. 981–1022. Figueroa, M. 2000. “Making Sense of the GR I D (Gender Resource Information and Male Experience: The Case of Academic Development Center). 2005. “Lao PDR Gender Underachievement in the English-Speaking Profile.� GRID, with support of the World Caribbean.� IDS Bulletin 31 (2). Bank, Washington, DC. http://siteresources Filmer, Deon, Jed Friedman, and Norbert Schady. .worldbank.org/INTLAOPRD/Resources/ 2008. “Development, Modernization, and Lao-Gender-Report-2005.pdf. Son Preference in Fertility Decisions.� Policy GSO (General Statistics Office) Vietnam. Research Working Paper 4716, World Bank, Household Living Standards Surveys, http:// Washington, DC. www.gso.gov.vn. Filmer, Deon, and Lant Pritchett. 1999. “The Guo, L., and Z. Zhao. 2002. “Children, Gender, Impact of Public Spending on Health: Does and Language Teaching Materials.� Chinese Money Matter?� Social Science and Medicine Education and Society 35 (5): 34–52. 49 (10): 1309–23. H a n na, Rema, a nd L eig h L i nden. 20 09. Filmer, Deon, and Norbert Schady. 2008. “Measuring Discrimination in Education.� “Getting Girls into School: Evidence from NBER Working Paper 15057, National Bureau a Scholarship Program in Cambodia.� of Economic Research, Cambridge, MA. Economic Development and Cultural Change Hannum, E., and J. Adams. 2002. “Girls in 56 (3): 581–617. Gansu, China: Expectations and Aspirations ———. 2009. “School Enrollment, Selection and for Secondary Schooling.� In Exclusion, Test Scores.� Policy Research Working Paper Gender and Education, edited by M. Lewis 4998, World Bank, Washington, DC. and M. Lockheed, 71–96. Washington, DC: Fiszbein, Ariel, and Norbert Schady. 2009. Center for Global Development. “Conditional Cash Transfers: Reducing HNPStats (Health Nutrition Population Statistics) Present and Future Poverty.� Policy Research database. World DataBank, World Bank, Series, World Bank, Washington, DC. Washington, DC. http://data.worldbank.org/ Flabbi, Luca. 2011. “Gender Differences In data-catalog/health-nutrition-population- Education, Career Choices and Labor Market statistics. Outcomes on a Sample of OECD Countries.� IDN CGA. (2006). “Indonesia: Country Gender OE C D B ac kg rou nd Pap er for Wo rl d Assessment.� Asian Development Bank, Development Report 2012 , World Bank, Canadian International Development Agency, Washington, DC. National Democratic Institute, and the World Frankenberg, Elizabeth, A. Buttenheim, B. Sikoki, Bank. and W. Suriastini. 2009. “Do Women Increase Ireson-Doolittle, C. 1999. “Gender and Changing Their Use of Reproductive Health Care When Property Rights in Laos.� In Women’s Rights It Becomes More Available?� Evidence from to House and Land: China, Laos, Vietnam, Indonesia, Studies in Family Planning 40 (1): edited by I. Tinker and G. Summerfield, 145– 27–38. 52. Boulder, Colorado: Lynne Rienner. Frankenberg, Elizabeth, and Duncan Thomas. Jayachandran, S., and A. Lleras-Muney. 2009. 2001. “Women’s Health and Pregnancy “Life Expectancy and Human Capital Outcomes: Do Services Make a Difference?� I nvest ments: Ev idence f rom M ater nal Demography 38 (2): 253–65. Mortality Declines.� Quarterly Journal of Global Findex (Global Financial Inclusion Economics 124 (1): 349–97. Database). World Bank, Washington, DC. Jensen, R. 2010. “Economic Opportunities http://data.worldbank.org /data-catalog / and Gender Differences in Human Capital: financial_inclusion. Experimental Evidence for India.� NBER G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    99 Working Paper 16021, National Bureau of of Sdr 10.8 Million (US$14.82 Million Economic Research, Cambridge, MA. Equivalent) to the Lao People’s Democratic Jha, Jyotsna, and Fatimah Kelleher. 2006. Republic for the Second Land Titling Project.� Boys’ Underachievement in Education: An World Bank, Washington, DC. Exploration in Selected Commonwealth Lavy, Victor. 2008. “Do Gender Stereotypes Countries. London, U.K.: Commonwealth Reduce Girls’ or Boys’ Human Capital S ecretariat , G ender S ec tion, Social O utcome s? Ev idenc e f rom a Nat u ra l Transformation Programmes Division, and Experiment,� Journal of Public Economics. 92 the Commonwealth of Learning. (10–11): 2083–2105. Jivan, Vedna, and Christine Forster. 2007. Li, H., and H. Zheng. 2009. “Ultrasonography Translating CEDAW into Law: CEDAW and Sex Ratios in China.� Asian Economic Legislative Compliance in Nine Pacific Policy Review 4: 121–37. Countries. Suva, Fiji: UNDP Pacific Centre Liaw, H. Ray. 2008. “Women’s Land Rights in and UNIFEM Pacific Regional Office. Rural China: Transforming Existing Laws Johnson, Kiersten, Sovanratnak Sao, and Darith into a Source of Property Rights.� Pacific Rim Hor. 2000. Cambodia 2000 Demographic Law and Policy Journal 17 (1): 237–64. and Health Survey: Key Findings. Calverton, Lloyd, C., B. Mensch, and W. Clark. 1998. “The MD: ORC Macro. Effects of Primary School Quality on the Jung, Kyung, and Haesook Chung. 2005. Educational Participation and Attainment of Gender Equality in Classroom Instruction: Kenyan Girls and Boys.� Population Council Introducing Gender Training for Teachers in Paper 116. http://www.popcouncil.org/pdfs/ the Republic of Korea. Bangkok, Thailand: wp/16.pdf. UNESCO. L SB (Lao Statistics Bureau). E xpenditure Kevane, M., and D. Levine. 2003. “Changing and Consumption Survey, LAO People’s Status of Daughters in Indonesia.� Working D emo crat ic Republ ic. ht t p: // w w w.ns c Paper C03-126, University of California, .gov.la/. B e r k e l e y, C e n t e r f o r I n t e r n a t i o n a l Lundberg, S. J., R. Pollak, and T. J. Wales. Development, Berkeley, CA. 1997. “Do Husbands and Wives Pool Their King, E., and D. van der Walle. 2007. Girls Resources? Evidence from the United Kingdom in Lao PDR: Ethnic Affiliation, Poverty Child Benefit.� Journal of Human Resources and Location . In Exclusion, Gender and 32 (3): 463–80. Education: Case Studies from the Developing Malhotra, A., R. Pande, and C. Grown. 2003. World, edited by M. Lewis and M. Lockheed, “Impact of Investments in Female Education 31–70. Washington, DC: Center for Global on Gender Equality.� Commissioned by the Development. World Bank Gender and Development Group Klasen, Stephan, and Francesca Lamanna. 2009. and International Center for Research on “The Impact of Gender Inequality in Education Women, Washington, DC. and Employment on Economic Growth: New Marks, J. 2001. “Girls Know Better: Educational Evidence for a Panel of Countries.� Feminist Attainment of Boys and Girls.� CIVITAS— Economics 15 (3): 91–132. The Institute for the Study of Civil Society, Knowles, Stephen, A. K. Lorgelly, and Dorian London, U.K. Owen. 2002. “Are Educational Gender Gaps Martino, W., and D. Berrill. 2003. “Boys, a Brake on Economic Development? Some Schooling and Masculinities: Interrogating the Cross-Country Empirical Evidence.� Oxford ‘Right’ Way to Educate Boys.� Educational Economic Papers 54 (1): 118–49. Review 55 (2): 99–117. Kobia, J. 2009. “Femininity and Masculinity in Naudeau, Sophie, Naoko Kataoka, Alexandria English Language Textbooks in Kenya.� The Valerio, Michelle Neuman, and Leslie Elder. International Journal of Language Society 2011. Investing in Young Children: An Early and Culture, 28: 57–71. Childhood Development Guide for Policy Korea National Statistical Office. Daejeon, Dialogue and Project Preparation. Washington, Republic of Korea. http://kostat.go.kr. DC: World Bank. L ao L a nd Tit l i ng Projec t I I IC R . 2010. Niehof, A. 2003. “Women and the Social Context “Implementation Completion and Results of Fertility Under the New Order.� In Two Report (Ida-38010) on a Credit in the Amount is Enough: Family Planning in Indonesia 1 0 0    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C Under the New Order 1968–1998 , edited PISA (Programme for International Student by A. Niehof and F. Lubis, 163–183. Leiden: Assessment). Organisation for Economic KITLV Press. Co-operation and Development, Paris, France. Nielsen, Harriet Bjerrum, and Bronwyn Davies. www.pisa.oecd.org/. 2010. “Discourse and the Construction Pritchett, Lant, and Lawrence H. Summers. 1996. of Gendered Identities in Education.� In “Wealthier Is Healthier.� Journal of Human Encyclopedia of Language and Education: Resources 31 (4): 841–68. Volume 3, Discourse and Education, edited Qian, Nancy. 2008. “Missing Women and the by Marilyn Martin-Jones, Anne-Marie de Price of Tea in China: The Effect of Sex-Specific Mejía, and Nancy H. Hornberger. Berlin: Earnings on Sex Imbalance.� Quarterly Springer. Journal of Economics 123 (3): 1251–85. NIS (National Institute of Statistics) Cambodia. Quisumbing, Agnes. 1995. “Gender Differences Cambodia Socio-Economic Survey (CSES). in Agricultural Productivity: A Survey of National Institute of Statistics, Phnom Penh, Empirical Evidence.� FCN D Discussion Cambodia. http://www.nis.gov.kh/index.php/ Paper 5, International Food Policy Research social-statistics/cses. Institute, Washington, DC. NSD (National Statistics Directorate) Timor- Quisumbing, Agnes, and Kelly Hallman. 2005. Leste. Timor-Leste Survey of Living Standards. “Marriage in Transition: Evidence on Age, Ministry of Planning and Finance, National Education, and Assets from Six Developing Statistics Directorate, Dili, Timor-Leste. http:// Countries.� In The Transition to Adulthood in dne.mof.gov.tl/TLSLS/AboutTLSLS/index Developing Countries: Selected Studies, edited .htm. by Jere Behrman, Barney Cohen, Cynthia B. NSD (National Statistics Directorate) Timor- Lloyd, Nelly Stromquist . Washington, DC: Leste, Ministry of Finance, and ICF Macro. National Academies Press. 2010. Timor-Leste Demographic and Health Quisumbing, Agnes R., and John Maluccio. 2003. Survey 2009–10. Dili, Timor-Leste: NSD and “Resources at Marriage and Intra-household ICF Macro. Allocation: Evidence from Bangladesh, NSO Mongolia (National Statistical Office of Ethiopia, Indonesia, and South Africa.� Mongolia). Living Standards Measurement Oxford Bulletin of Economics and Statistics Survey (LSMS). Ulaanbaatar, Mongolia: 65: 283–327. National Statistical Office. http://www.nso Quisumbing, Agnes R., and Keijiro Otsuka. .mn. 2001. “Land Inheritance and Schooling in NSO Philippines (National Statistics Office, Matrilineal Societies: Evidence from Sumatra.� Republic of the Philippines). 2006. Family World Development 29 (12): 2093–2110. Income and Expenditure Survey 20 06 . Ross, Heidi, and Jinghuan Shi, eds. 2003. “Entering Manila, the Philippines: National Statistics the Gendered World of Teaching Materials, Office. Part II.� Chinese Education and Society NSO Thailand (National Statistical Office (May–June) 36: 3. Thailand). Thailand Socio-Economic Survey Sakellariou, Chris. 2011. “Determinants of the (SES). Bangkok, Thailand: National Statistical Gender Wage Gap and Female Labor Force Office. http://web.nso.go.th/tnso.htm. Participation in EAP.� Paper commissioned Oster, Emily, and Bryce Millet. 2010. “Do Call for Toward Gender Equality in East Asia Centers Promote School Enrollment? Evidence and the Pacific: A Companion to the World from India.� Draft, University of Chicago Development Report. Washington, DC: World Booth School of Business. Bank. Panda, Pradeep, and Bina Agarwal. 2005. Schleicher, A. 2008. “Student Learning Outcomes “Marital Violence, Human Development and in Mathematics from a Gender Perspective: Women’s Property Status in India.� World What Does the International PISA Assessment Development 33 (5): 823–50. Tell Us?� In Girls’ Education in the 21st Papua New Guinea Department of Education. Century, edited by M. Tembon and L. Fort. 2009. “Achieving Universal Education for a Washington, DC: World Bank. Better Future: Universal Basic Education Plan Schultz, T. P. 1993. “Returns to Women’s 2010 –2019.� National Executive Council, S chooling.� I n Women’s E duc ation in Port Moresby, Papua New Guinea. Developing Countries: Barriers, Benefits G E N D E R A N D E N D O W M E N T S : A C C E S S T O H U M A N C A P I T A L A N D P R O D U C T I V E A S S E T S    101 and Policy. Baltimore, MD: Johns Hopkins from a G ender Perspective: Repor t of University Press. Findings.� UNESCO, Hanoi, Vietnam. ———. 1997. “The Demand for Children in U N FPA (United Nations Population Fund). Low-Income Countries.� In Handbook of 2009. “Recent Change in the Sex Ratio Population and Family Economics: Vol. 1A, at Birth in Viet Nam.� UNFPA Viet Nam, edited by M. R. Rosenzweig and O. Stark. Hanoi. Amsterdam: North-Holland. UNICEF (United Nations Children’s Fund). 2004. ———. 2006. “Does the Liberalization of Trade “What About the Boys?� In The State of the Advance Gender Equality in Schooling World’s Children 2004, ed. UNICEF, 58–69. and Health?� IZA Discussion Papers 2140, New York: UNICEF. Institute for the Study of Labor, Bonn. ———. 2009. “Education for All Mid-Decade Schultz, T. P., and Z. Yi. 1997. “The Impact of Assessment, East Asia and Pacific.� UNICEF Institutional Reform from 1979 to 1987 East Asia and Pacific Regional Office, Bangkok, on Fertility in Rural China.� Photocopy. Thailand. August 29. UNIFEM, World Bank, ADB, UNDP, and DFID/ Sen, Amartya. 1992. “Missing Women.� British UK. 2004. A Fair Share for Women: Cambodia Medical Journal 304: 587–88. Gender Assessment. UNIFEM, WB, ADB, Shel, T. A. 2007. “Gender and Inequity in UNDP, DFID/UK. Phnom Penh. Education Literature Review.� Background U N DESA (United Nations Department of paper prepared for the Education for All Economic and Social Affairs), Population Global Monitoring Report 2008 Education for D iv i sion . 2 0 09. Wo rl d U rba n i z a t i o n All by 2015: Will We Make It? http://unesdoc Prospects: The 2009 Revision. New York: .unsco.org/imagers/0015/001555/155580e. United Nations. http://esa.un.org/unpd/wup/ pdf. index.html. Shi, Jinghuan, and Heidi Ross, eds. 2002. Velasco, E. 2001. “Why Are Girls Not in School: “Entering the Gendered World of Teaching Perceptions, Realities and Contradictions in Materials, Part I.� Chinese Education and Changing Cambodia.� UNICEF/Cambodia, Society 35 (5): 3–13. Phnom Penh. Tararia, A. 2011. “WDR Rapid Qualitative Waldron, Ingrid. 1998. “Factors Determining Assessment on Gender and Economic Choice.� the Sex Ratio at Birth.� In Too Young to Die: Country Synthesis Report, Papua New Guinea. Genes or Gender?, 53–63. New York: United World Bank, Washington, DC. Nations. Thomas, Duncan. 1995. “Like Father, Like W HO (World Health Organization) Global Son, Like Mother, Like Daughter, Parental Infobase. World Health Organization, Geneva, Resources and Child Height.� Papers 95-01, Switzerland. https://apps.who.int/infobase/ RAND Reprint Series, RAND corporation, report.aspx. Santa Marica, CA. ———. 20 03, November. “Gender, Health Thomas, Duncan, and John Strauss. 1992. “Prices, and Tobacco.� Gender and Health. http:// Infrastructure, Household Characteristics www.who.int/gender/documents/Gender_ and Child Height.� Journal of Development Tobacco_2.pdf. Economics 39 (2): 301–33. ———. 2010. World Health Statistics. Geneva: TIMSS (Trends in International Mathematics and WHO. Science Study). International Association for ———. 2011a. “Action Needed to Reduce the Evaluation of Educational Achievement, Health Impact of Harmful Alcohol Use.� Amsterdam, the Netherlands. http://rms.iea- News Release, February 11. http://www dpc.org/. .who.int/mediacentre/news/releases/2011/ Udry, Christopher. 1996. “Gender, Agricultural alcohol_20110211/en/. Production, and the Theory of the Household.� ———. 2011b. Global Status Report on Alcohol Journal of Political Economy 104 (5): and Health. Geneva: W HO. http://www 1010–14. .who.int /substance _ abuse /publications / UNESCO (United Nations Educational, Scientific global_alcohol_report/en/. and Cultural Organization) and the Vietnam Williamson, N. 1976. Sons or Daughters: A Ministry of Education and Training. 2011. Cross-Cultural Survey of Parental Preferences. “National Textbook Review and Analysis Beverley Hills, CA: Sage. 1 0 2    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C Women, Business and the Law (database). World ———. 2011b. Vie tn am C ountr y Ge n d e r Bank, Washington, DC. http://wbl.worldbank Assessment. Washington, DC: World Bank. .org/data. http://www-wds.worldbank.org/external/ World Bank. 2001. Engendering Development: default/WDSContentServer/WDSP/IB/2011 t h ro ugh G e n d e r E q u a l i t y i n R ight s , / 11 / 14 / 0 0 0333 03 8 _ 2 0111114 0 03 42 0 / Resources, and Voice. Washington, DC: Rendered/PDF/655010WP0P12270sessment World Bank. .0Eng.0Final.pdf. ———. 2008. “Analysis of the Impact of Land ———. 2011c. World Development Report Tenure Certificates with Both the Names of 2012: Gender Equality and Development. Wife and Husband in Vietnam.� World Bank, Washington, DC: World Bank. Washington, DC. https://openknowledge World Development Indicators (WDI) database. .worldbank.org/handle/10986/7810. World DataBank, World Bank, Washington, ———. 2010. “. . . and Then She Died�: Indonesia DC. http://data.worldbank.org/data-catalog/ Maternal Health Assessment. Washington, world-development-indicators. DC: World Bank. https://openknowledge Zeng, Yi, Tu Ping, Gu Baochang, Xu Yi, Li .worldbank.org/handle/10986/2837. Bohua, and Li Yongqing. 1993. “Causes ——— . 2 011a . “ Towa rd a H e a lt hy a nd and Implications of the Recent Increase in Harmonious Life in China: Stemming the the Reported Sex Ratio at Birth in China.� Rising Tide of Non-Communicable Diseases. Population and Development Review 19 (2): Human Development, East Asia and the 283–302. Pacific.� World Bank, Washington, DC. Gender and Economic Opportunity 3 I n the dynamic East Asia and Pacific division of labor within the household imply region, many countries have undergone that women work longer hours than men, structural transformations that have although fewer of those hours are devoted shifted the balance of economic opportu- to remunerated activities. nities away from rural areas and toward Reducing gender inequalities in eco- urban areas. These growth processes have nomic opportunities can improve economic opened up nonfarm economic opportunities outcomes in multiple ways. First, reducing for men and women in the region, particu- employment segregation (the unequal dis- larly among younger cohorts. The educa- tribution of male and female workers across tional attainment and health outcomes of occupations and sectors) will reduce efficiency women, particularly younger women, have losses associated with the misallocation of been catching up to those of men during this talent (Anker 1998; M ­ orrison, Raju, and period of growth, as discussed in chapter 2. Sinha 2007). Men and women often choose Along some dimensions of gender equal- occupations on the basis of norms, gender ity in economic opportunities, there has stereotypes, and sometimes prejudice, rather been substantial progress over the last two than on the basis of earnings or job match decades. For example, the evidence suggests (Klasen and Lamanna 2009). Encouraging that the labor market participation deci- workers and employers to make labor choices sions of younger women resemble those of on the basis of their skills, competencies, and their male counterparts in the region. How- inherent ability is likely to raise productivity ever, many other indicators demonstrate the and may have a positive impact on economic substantial challenges to be overcome to growth by increasing the size of the labor close the gender gap in access to economic force as well as by expanding the pool of man- opportunities. The type of work women do agerial and innovative talent in the economy remains very different from that of men, and (Esteve-Volart 2004). As noted in chapter 1, their remuneration for these tasks is lower. estimates for East Asian and Pacific countries Women of all ages are more likely than men suggest that output per worker could be 7 to to be in poorly remunerated occupations 18 percent higher if female entrepreneurs and and sectors, women are paid less than men workers were to work in the same sectors, for similar work, and gender norms in the types of jobs, and activities as men (Cuberes 103 1 0 4    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C and Teignier-Baqué 2011). Second, empiri- Three main messages for the East Asia and cal evidence from other developing regions Pacific region emerge in this chapter: suggests that reducing gender inequalities in access to productive inputs can increase over- I n some dimensions, such as labor force •   all production by increasing the productivity participation, gender inequalities in of female-run farms and enterprises (Gold- the economic sphere in the region have stein and Udry 2008; Quisumbing 1995; improved in recent decades and are nar- Udry 1996). rower than those in other regions; how- This chapter examines differences in pro- ever, multiple dimensions of inequalities ductive opportunities for men and women remain that will require concerted effort within and across countries in the East Asia to change, such as persistent gender wage and Pacific region and explores the eco- and productivity gaps. In several areas, nomic and institutional factors that have policies and public investment can help determined how those opportunities have to ease the constraints on women and evolved over time. Because education enroll- support them in their multiple roles as ment and attainment are growing and labor entrepreneurs, farmers, wage and salaried markets are changing, this chapter differen- workers, mothers, and caregivers. tiates between the experiences of older and T he constraints faced by women vary •   younger generations. It also uses examples across sectors of the economy and also from both high-income and low-income by country according to the institutional countries across the globe to understand environment. Some common themes these trends. emerge, however. In the agricultural sec- Gender differences in the economic tor, female farmers’ access to productive sphere manifest themselves in several indi- capital, technologies, and governmental cators, including differences in labor force services is lower than that of male farm- participation, in the time spent on produc- ers; improving access to these inputs is tive and reproductive activities, in the sec- tantamount to increasing productivity. In tors in which men and women work, in the the nonfarm sector, female-run enterprises tasks and occupations that they do within are smaller and in different sectors than those sectors, and in the types of firms that male-run enterprises. The constraints and employ them. Differences in these indica- productivity differences of female enter- tors contribute to and are themselves deter- prises are predominantly attributable to mined by gaps in the earnings of men and their small size and to the sectors in which women. Women are often paid less than they are found. Constraints in all countries men for the same work, and female-run are likely to be greatest at start-up, when enterprises and farms typically produce less access to finance and entrepreneurial skills than those of men. are likely to be important ­ determinants of Gender inequalities in economic oppor- sector and initial scale of enterprise. In the tunities are driven by multiple interacting labor market, gender-based employment factors. In this chapter we examine how segmentation—or sorting across types of gender differences in access to human and firms, industries and sectors—affects both physical capital, technologies, and gov- the wages women earn as well as produc- ernment services; gender stereotypes; and tivity in the economy, particularly when ­ gender roles explain gender inequality in men and women sort into occupations on economic opportunities. These factors are, the basis of gender rather than skill. Poli- in themselves, determined by the house- cies that encourage both men and women hold, market, and institutional environment to think outside of gender-based occupa- in which preferences and gender roles are tional norms will be productivity enhanc- learned and in which education and time ing and likely have positive repercussions allocation choices are determined. for female empowerment. G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    105 A s in other parts of the world, women •   determining gender wage gaps and, indeed, in the region have multiple roles and will that these factors interact with the develop- require more support to manage com- ment process to determine the degree to peting demands for their time from pro- which growth narrows gender inequalities ductive, reproductive, and community (Meng 1996).1 ­ m anagement activities as development In high-income countries, the age profile proceeds and greater nonfarm sector of women’s economic participation and the opportunities emerge. Nearly all coun- sensitivity of their participation to life-cycle tries in the region see declines in the factors have changed with development. female labor force participation of young Between 1950 and 2010, female labor force mothers. Furthermore, to manage their participation in Hong Kong SAR, China; dual roles, women are often obliged to Japan; the Republic of Korea; and Singapore enter into different occupations and work increased substantially across all age cohorts. fewer hours than men, both of which are For example, in Korea, women’s labor force found to have negative implications on participation has increased monotonically their wages and earnings. Policies that over time (figure 3.1). Similar changes were support women in juggling the compet- seen in the United States, particularly among ing demands of home and market work married women (Juhn and Potter 2006). Fur- will be required, particularly as women thermore, the decline in labor force participa- start moving into “male� occupations tion by women in their early 30s has become that have not traditionally allowed them less severe in these high income countries. the flexibility to lead their dual lives. Rising female labor force participation with development has been attributed to a number The chapter is structured as follows. The of demographic factors, including later mar- first section explores whether growth is suf- riage and childbirth and lower fertility rates. ficient to reduce gender inequalities in pro- However, income growth in these coun- ductive activities, drawing upon evidence tries has not been enough to eliminate gender from high-income countries within and out- side the region. The second section describes the current situation with regard to gender inequalities in economic opportunities in FIGURE 3.1  Female labor force participation in the Republic of East Asian and Pacific countries. The third Korea rose for women of all ages between 1960 and 2005 section examines the determinants of the most persistent gender inequalities, and the female labor force participation rate (%) 70 fourth section concludes by briefly exam- ining policy directions, a discussion that is 60 taken up again in chapter 6. 50 40 percent Limited effects of growth 30 on gender gaps in economic opportunity 20 10 The empirical literature suggests that eco- nomic development alone is not sufficient 0 15–19 20–24 25–29 30–34 35–39 40–44 45–49 50–54 55–59 60–64 to narrow gender differences in earnings age group (Blau and Kahn 2003; Hertz et al. 2009; 1960 1970 1975 1980 1985 Tzannatos 1999). Evidence from East Asian 1990 1995 2000 2005 and Pacific countries suggests instead that social, political, and cultural factors are Source: ILO Key Indicators of the Labour Market (KILM) database: Korea Labor Force Survey and as important as economic development in Population Census. 1 0 6    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C inequalities in all dimensions of economic reforms to level the playing field for women opportunities. Despite a substantial increase in the labor market have been difficult to in young women’s labor force participation, enforce in Japan, however (Lam 1992; Liu sharp gender differences in employment and Boyle 2001; Miller 2003). ­ status, occupational status, and wage rates continue to exist. In Japan, women are more likely than men to be in nonregular employ- Gender differences in economic ment, including temporary work, contract activity work, and part-time work (Hill 1996; Yu Gender differences in the economic sphere 2002). Women continue to leave the work- manifest themselves in a number of indi- force in substantial numbers during their cators. This section focuses on examining childbearing years, even if only temporarily. gender differences in labor force participa- The greatest drop in female labor force par- tion, in earnings, and in the labor market ticipation rates comes for married women segregation of men and women. in their mid-20s to early 30s when they start having children and assuming greater Labor force participation family responsibilities (Hill 1996; Lee, Jang, and Sarkar 2008; Miller 1998, 2003; The East Asia and Pacific region is character- Sasaki 2002). ized by high female labor force participation Welfare, child care, and tax reforms; leg- on average, but also by substantial variation islative changes; and women’s movements across and within countries. Labor force par- have helped to narrow gender differences ticipation is defined as all productive work, in economic opportunities in high-income whether as wage workers or as unpaid family countries in the East Asia and Pacific region workers. Female labor force participation in and across the world. The United States the East Asia and Pacific region is the high- provides a good example of how the com- est in the developing world—70 percent of bination of economic forces, social changes, females were participating in labor market and legal reforms has transformed women’s activities in 2008 (see figure 1.14 in chap- labor force participation. Explanations of ter 1). The participation gap (the difference changes in female labor force participation between the participation of men and women include demand-side factors that shifted in paid or unpaid market-oriented work) women’s market wage as well as supply- was approximately 15 percentage points in side factors that reduced women’s oppor- the region. In comparison, the female labor tunity costs of working (Fang and Keane force participation rate in the Latin America 2004; Galor and Weil 1996; Greenwood, and the Caribbean region and Europe and Seshadri, and Yorukoglu 2005; Weinberg Central Asia region was 55 and 58 percent 2000). Since the 1980s, Korean labor and in 2008, respectively, and their gender gaps family law has sought to improve women’s in labor force participation were 27 and 16 status within marriage and the family and percentage points, respectively. Box 3.1 dis- to eliminate gender bias in other areas, cusses the ways in which high female labor including in labor law (Kim 2005). In addi- force participation may, or sometimes may tion, the Korean government has undertaken not, be an indicator of enhanced welfare several measures to support married women for women. with their child-care responsibilities, such The average rates of female labor force as reforming maternity and paternity leave participation vary substantially within the and expanding expenditures on child-care region (see figure 1.15 in chapter 1). Par- facilities. Women’s and labor organizations ticipation and gender gaps in some parts have played an important role in advocating of the region are among the highest in the for legislation that reduces gender discrimi- world, whereas in others they are among natory practices (Kim 2005). Legislative the lowest. For example, in Fiji, Samoa, G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    107 BOX 3.1  Is higher female labor force participation always a good thing? Studies from across the world indicate that increased to be devoted to children’s education, health, and female labor force participation and control by nutrition (Thomas 1990; Thomas and Strauss 1997). women over household resources are associated with Women’s borrowing from microfinance programs their enhanced well-being and status. For example, a also appears to improve child welfare more substan- study on India shows that female mortality is lower tially than does borrowing by men, including signifi- where female labor force participation rates and cant improvements in children’s nutritional levels and earnings are higher (Murthi, Guio, and Drèze 1995). increases in the school enrollment levels of both boys At the same time, although higher household income and girls. However, labor force participation is not is associated with higher levels of welfare for house- always empowering and may be a consequence of hold members, the marginal impacts are considerably poverty, which pushes women into low-paying jobs greater when income is in the hands of the mother. with poor working conditions and job insecurity. As women gain more control over expenditure deci- Jobs may also be exploitative and may reinforce tra- sions, a larger share of household resources tends ditional gender roles (Elson and Pearson 1981). and Malaysia, participation rates in 2008 to supplement household income is neces- were an order of magnitude lower than the sary (Pierre 2011). average for the region, with approximately Even in countries with high overall partici- 55 percent of working-age females in these pation rates, female labor force participation countries participating in the labor force. rates decline during childbearing years and Female labor force participation is high- old age. Average female labor force participa- est in countries where the state has put a tion rates hide important variation over wom- priority on gender equality, for example, en’s life cycles due to factors such as marriage in socialist or formerly socialist countries, and childbearing, whereas male participa- as well as in agrarian and rural economies tion rates remain fairly constant throughout where high rates of participation may be a their life cycles. The gaps between men and consequence of poverty. women in labor force participation, earn- Substantial gender differentials in par- ings, and job composition increase after mar- ticipation have also been noted within riage and childbearing in many East Asian countries in the region. National averages and Pacific countries, as they do globally. In of labor force participation fail to cap- the United States, for example, young single ture important differences in participa- men and women have more similar labor tion across rural and urban areas, as well participation, earnings, and career profiles as regional variation within countries. In than do married men and women. In many all countries within the region, rural labor East Asian and Pacific countries, life-cycle force participation is higher than urban patterns of labor force participation by birth labor force participation, for both males cohorts reveal specific trends that are not and females (figure 3.2). Participation gaps apparent in the average rates of participation, in all countries, apart from Mongolia and especially during periods of rapid growth and the Philippines, are substantially greater structural transformation. Rapid changes in urban areas than in rural areas. In Viet- in education levels across cohorts, the avail- nam, higher levels of female participation ability of new economic opportunities in and lower gender differentials are found in growing sectors, and urbanization mean poorer and often low-productivity areas, that young women are more likely than older where female participation in employment women to participate in the labor force, and 1 0 8    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE 3.2  Labor force participation is greater in rural areas than in urban areas for both males and females 100 labor force participation of males and females age 20–64 in rural and urban areas, by country 90 80 70 60 percent 50 40 30 20 10 0 6 08 7 8 06 8 8 09 00 00 00 00 00 20 20 20 ,2 ,2 ,2 ,2 ,2 d, a, s, m ste lia sia DR ne di an na go ne oP Le bo pi ail et on do ilip m or La Th Vi M Ca In Tim Ph female urban female rural male urban male rural Source: Data generated using country level household data, with the exception of Indonesia which uses labor force data. Note: The countries are sorted by the gap between females in rural and urban areas. to participate in different sectors (Mammen are older. A “single-peaked� pattern exhibits and Paxson 2000). higher rates of participation at younger ages, Female labor force participation is more which then fall after the peak. This is a more sensitive to life-cycle factors, such as mar- extreme version of the double-peaked pattern riage and childbearing, in some countries and characterizes a labor market in which than others. The birth cohort life-cycle few women return to work after marriage participation patterns of women vary sub- and childbearing. In the region, this pattern stantially within the East Asia and Pacific can be seen only in Malaysia, where it con- region and have also exhibited considerable tinues despite increases in female participa- change over time. Three patterns can be tion between 1980 and 2010 (figure 3.3). The distinguished in the region (Horton 1996). 2 plateau pattern, in contrast, describes a situ- The “plateau� pattern exhibits relatively flat ation in which female labor force participa- female labor force participation until women tion is less sensitive to life-cycle effects than reach their early 50s, and then declines into in the single- and double-peaked patterns. old age. The “double-peaked� or “M� pat- The plateau pattern can be seen in countries tern is generally observed in more industri- with political regimes that encouraged wom- alized countries: it is characterized by high en’s economic independence, such as China, participation in the labor market prior to Mongolia, and Vietnam. marriage and childbearing, with a subse- Labor force participation decisions of quent return to the labor force once children rural and urban women in many countries G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    109 FIGURE 3.3  Female labor force participation rates in Malaysia have risen over time among 20- to 55-year- olds, but continue to decline during child-bearing years female labor force participation (%), by age cohort 70 60 50 40 percent 30 20 10 0 15–19 20–24 25–29 30–34 35–39 40–44 45–49 50–54 55–59 60–64 age groups 1980 1990 2000 2010 Source: LABORSTA Internet (1980, 1990, and 2000 data), CEIC Data (2010 data). display different responses to life-cycle East Asian and Pacific countries with changes, and their sensitivity to life-cycle higher rates of female labor force participa- changes may evolve differently over time. tion overall have smaller declines in partici- In Indonesia, for example, urban women pation during childbearing years, although are more likely than rural women to leave ­ p redominantly in urban settings. In all the labor force when they have children, countries in the region, women with young and they also leave the labor force for lon- children are substantially less likely to par- ger. Patterns of participation have changed ticipate in the labor market than men with substantially since 1990, however, and young children and than women without have evolved differently in rural and urban young children. In some countries, such areas. Although urban women with young as Mongolia and Vietnam, a reduction in children in 2008 are still less likely to work labor market participation is seen in either than their rural counterparts, urban women urban or rural areas alone, whereas in oth- display less sensitivity to childbearing deci- ers, such as Indonesia and the Philippines, it sions than they did in 1990. In contrast, is seen in both rural and urban areas. Data rural women in 2008 are more likely to from the 2009 Vietnam Labor Force Survey reduce their labor supply when having chil- (GSO [Vietnam]) show that rural Vietnamese dren and to leave the labor force for longer women between ages 25 and 35 with a child than rural women did in 1990. under the age of 2 are a third less likely to 1 1 0    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C participate in the labor market than men of the ­Philippines, in marked contrast to the the same age with a child. In comparison, the examples of M ­ ongolia and Vietnam above, participation of women between ages 25 and female-male participation gaps are substan- 35 without a child is fairly similar to that of tial even among women who do not have men in this age group (figure 3.4). In Vietnam, children, and the gaps widen with age. the effect of children on labor force participa- tion is smaller and of shorter duration than Earnings in Mongolia. In Vietnam, the participation gap disappears by the time the child reaches Gender gaps in wages and earnings are school age, whereas in Mongolia the partici- found in all countries in the region and pation gap continues to be substantial even in all sectors, with rare exceptions. Male for older children. and female wages differ in multiple sec- In countries where female labor force tors and settings across the region, with participation is low on average, it also tends female agricultural wage workers earning to be highly sensitive to marriage and child- less than male agricultural wage workers, bearing years. Indonesia, Japan, Korea, and female urban wage workers earning and Malaysia have historically displayed less than their male counterparts. 3 Female a single- or double-peaked pattern; that entrepreneurs and farmers often display is, women participate in the labor market lower revenues and profits than their male prior to marriage and childbearing and, in counterparts. the case of a double-peaked pattern, even- A substantial body of evidence has accu- tually return to the labor force once the mulated over the past three decades to sug- children are older. These countries also gest that women are paid less for similar work have lower average female labor force par- across the world as well as in the region. A ticipation than countries in the region that cross-country comparison of wages indicates exhibit a plateau pattern. In Indonesia and that women earn between 70 and 80 percent of the wages men receive for similar work (Hausmann, Tyson, and Zahidi 2010). 4 Evidence from across the world suggests FIGURE 3.4  Women in rural Vietnam with children under age 2 that economic development is not sufficient are substantially less likely to participate in the labor market than to reduce the gender earnings gaps (Blau those without young children and Khan 2003). Even in high-income coun- tries in the East Asia and Pacific region— male-female labor force participation gap 0.25 rural male-female labor force participation gap, by notably Japan and Korea—the average age cohort and child status woman earns less than half the wage of 0.2 the average man. Lower-income countries, 0.15 including ­ M ongolia, Lao People’s Demo- cratic Republic, Papua New Guinea, and 0.1 Vietnam, have lower gender wage gaps, on average, than many richer countries. 0.05 Within the agricultural, manufacturing, service, and government sectors, women 0 earn less than men on average (figure 3.5), although the ratio of male-to-female earn- –0.05 ings varies substantially across sectors as 15–19 20–24 25–29 30–34 35–39 40–44 age groups well as across countries. Female-to-male without a child under 10 with a child under 2 wage ratios tend to be lower in the agricul- with a child between 2 and 10 tural and manufacturing sectors and higher in the service sector and in government, with Source: World Bank staff estimates using Vietnam Labor Force Survey (GSO Vietnam), 2009 data. some exceptions. G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    111 Gender wage gaps in the East Asia and FIGURE 3.5  Women earn less than men in the majority of East Pacific region are greatest between men and Asian and Pacific countries and in all sectors of the economy women with the lowest educational endow- ments and in the lowest paying occupations. average female-to-male hourly wage ratio, by sector Several studies have indicated that the gap is Cambodia, 2008 wider at the bottom end of the wage distribu- tion than at the top (Chi and Li 2007; Li and Thailand, 2009 Song 2011; Sakellariou 2011). In contrast, in Organisation for Economic Co-operation Philippines, 2006 and Development (OECD) countries, gen- Vietnam, 2006 der wage gaps tend to be wider at the top than at the bottom of the wage distribution Timor-Leste, 2007 (Albrecht, Björklund, and Vroman 2003; Arulampalam, Booth, and Bryan 2007; de la Lao PDR, 2008 Rica, Dolado, and Llorens 2005). Indonesia, 2009 In the nonfarm sector, the performance of male and female firms around the world 0.0 0.2 0.4 0.6 0.8 1.0 1.2 varies substantially, as measured by total ratio factor productivity, labor productivity, prof- government services industry agriculture itability, and capital intensity (Sabarwal, Terrell, and Bardasi 2011). In the East Asia Source: World Bank estimates using household income and expenditure surveys. Cambodia ­Socio-Economic Survey (CSES) (NIS Cambodia), 2008 data; Indonesia National Socioeconomic and Pacific region, formal sector firms with Survey (SUSENAS) (BPS Indonesia), 2009 data; Lao Expenditure and Consumption Survey (LSB at least one female owner do not have sig- [Lao PDR]), 2008 data; Philippines Family Income and Expenditures Survey (NSCB 2006); Thailand ­Socio-Economic Survey (SES) (NSO Thailand), 2009 data; Timor-Leste Survey of Living Standards nificantly lower sales than those with no (NSD Timor-Leste), 2007 data; Vietnam Household Living Standards Surveys (VHLSS) (GSO Vietnam), female owners, with the exception of firms 2006 data. in the Philippines ( ­figure 3.6). However, in the informal sector, output per worker in female-owned e ­ nterprises is substantially lower than that in male-owned enterprises. Data from Indonesia and Vietnam suggest FIGURE 3.6  Male- and female-owned firms in the formal sector do that in the informal sector, gender earnings not display substantial differences in productivity gaps are more pronounced than in the for- mal sector: female-owned enterprises with ratio of female-to-male value added per worker fewer than five employees generate only 1.4 approximately 60 to 70 percent of the out- 1.2 put per worker generated by male-owned enterprises (figure 3.7). 1.0 0.8 Labor market segregation ratio 0.6 Multiple studies find that men and women work in different sectors, industries, occu- 0.4 pations, and types of firms across the world (Anker 1998; Boserup 1970). As can be 0.2 seen in figures 3.8 and 3.9, across the region 0 women are overrepresented in unpaid fam- Lao PDR Indonesia Vietnam Philippines ily work, particularly in rural areas, and are slightly more likely to be employed in Source: World Bank estimates using Enterprise Surveys, 2006–11 data. Note: Data are for small (19 employees or less), medium (20–99 employees), and large (100 and more the informal sector. 5,6 Within the infor- employees) firms in the formal sector. Female-owned firms are defined as those with at least one mal sector, women are more likely to be female owner. Value added is measured as revenues minus material expenses. 1 1 2    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE 3.7  In the informal sector, gender-based differences in Across the world, women in the nonagri- productivity are more pronounced than in the formal sector cultural workforce are more likely to work in social and communal services (such as ratio of female-to-male value added per worker education and health) and in commerce and 1.2 restaurants, whereas men are more likely 1.0 to work in transport, construction, public administration, and manufacturing (ILO 0.8 2010). Similar trends are found in East Asian and Pacific countries. For example, ratio 0.6 men in Cambodia and Indonesia are dispro- 0.4 portionately found in the manufacturing, 0.2 transport, construction, and public admin- istration sectors, and women are found in 0 manufacturing, education, health and social Indonesia Vietnam Indonesia Vietnam services, and commerce (figure 3.10). Esti- informal formal mates from household survey data suggest that, in Fiji and Vietnam, men are more Source: Data for informal firms are based on Indonesia Family Life Survey 2007/2008; VHLSS likely to work as professionals and man- (GSO Vietnam), 2008 data. Data on formal firms are from Enterprise Surveys. agers and as plant and machine workers, Note: Based on a sample of firms with fewer than five employees. whereas women are more likely to work as technicians, sales workers, and clerks, and in elementary occupations.7 Women also FIGURE 3.8  Women are more likely than men to work as unpaid make up a smaller fraction of the public family workers sector workforce than men in Cambodia, Thailand, Timor-Leste, and Vietnam. The percentage of employment, by gender and location Philippines, however, displays much higher 60 levels of occupational segregation and also 50 has a higher fraction of women in the public 40 sector than men, a trend that is likely to be percent 30 related to the relatively high investment in 20 education by ­ Filipino women. 10 Within the manufacturing sector, women 0 are more likely to be found in industries male female male female male female male female such as textiles and food processing, and rural urban rural urban are also found in large and export-oriented firms. Data from country enterprise surveys Cambodia, 2008 Indonesia, 2009 (2002–05) indicate that, in all developing wage and salaried worker self employed unpaid family worker regions, the fraction of full-time female workers is greater in export-oriented firms. The East Asia and Pacific region has the Source: World Bank estimates using socioeconomic surveys: CSES (NIS Cambodia), 2008 data; ­Indonesia SUSENAS (BPS Indonesia), 2009 data. second-largest fraction of full-time female workers, after Europe and Central Asia. However, within export-­ o riented firms, own-account workers (self-employed workers women are also more likely than men to be working by themselves) and subcontracted temporary workers in Cambodia, China, workers, whereas men are more likely to be and Indonesia, although in Thailand men employers or paid employees of informal are more likely to be temporary workers in enterprises (World Bank 2011g). Globally, all firms (see figure 1.18 in chapter 1). women are more likely than men to work Women are underrepresented in manage- part-time (ILO 2010). rial positions and positions of power in all G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    113 sectors, from government to manufactur- FIGURE 3.9  Women are slightly more likely to be employed in the ing (Anker 1998). The East Asia and Pacific informal sector than men. Rural-urban differences in informality region performs relatively well compared are greater than gender differences within rural or urban areas to other regions in terms of the fraction of firms with a top female manager (see figure percentage of workers in the informal sector, by gender and location 1.16 in chapter 1). However, women remain 100 less likely than men to serve as managers 90 and directors. The share of female directors 80 ranges from 10 percent in the Philippines 70 and 7 percent in China and Thailand to 60 percent 50 5 percent in Indonesia and under 2 percent 40 in Korea and Japan (CWDI 2010). Globally, 30 only one country in the world has succeeded 20 in having more than 30 percent female rep- 10 resentation on corporate boards (namely, 0 Norway), and one-third of countries have rural urban rural urban rural urban rural urban female board representation over 10 percent Timor-Leste Mongolia Thailand Indonesia (CWDI 2010). male female Segregation goes beyond the wage labor market, with smaller female-owned and Source: World Bank estimates using Indonesia SUSENAS (BPS Indonesia), 2009 data; Mongolia Living Standards Measurement Survey (LSMS) (NSO Mongolia), 2007–08 data; Thailand SES (NSO Thailand), female-managed firms located in less- 2009 data; Timor-Leste Survey of Living Standards (NSD Timor-Leste), 2007 data. capital-intensive sectors than male-owned Note: The informal sector is defined using information on an individual’s occupation and sector of employment. and male-managed firms. Female-owned and female-run enterprises are, on average, smaller than male-run enterprises in terms services and trade, than in capital-intensive of the number of employees hired, sales, and sectors (Klapper and Parker 2010). For exam- profits (Aterido and Hallward-Driemeier ple, a study commissioned by this report finds 2009; Badiani and Posadas 2011; Costa and that women are more likely to be found in ­ R ijkers 2011; Sabarwal, Terrell, and Bar- manufacturing, food sales, and food prepara- dasi 2011). Among formal sector firms in tion in Indonesia and less likely to be found the East Asia and Pacific region, enterprise in transport, construction, and other services survey data suggest that small firms are (Badiani and Posadas 2011). Similar pat- more likely than medium and large firms to terns have been observed in other countries have a top female manager (see figure 1.19 in the region, including Lao PDR (Davies and in chapter 1). Evidence by ownership dis- Record 2010), Mongolia (World Bank 2011c), plays a more mixed picture. In Indonesia, and Vietnam (Bjerge and Rand 2011). Lao PDR, Mongolia, Timor-Leste, Tonga, The enterprises that women work in are and Vanuatu, this pattern still holds, but also less productive and capital intensive. In evidence from the Philippines, Samoa, and Indonesia, the sectors that women are more Vietnam suggests that female owners are likely to be employed in—food and garments not disproportionately represented among production—are among the least capital small firms (figure 3.11).8 This pattern may intensive and productive sectors (figure 3.12). be the result of more open cultural norms By contrast, the transportation and other ser- regarding women’s role in business. How- vice sectors—where male entrepreneurs are ever, the lower levels of female management most likely to be found—has higher produc- relative to ownership suggest that women tivity and capital intensity. may still have less control or representation Having a female presence in management within firms. may have positive implications for workers, Female-run enterprises are also more likely however, even if productivity per worker is to be found in labor-intensive sectors, such as lower. Female-run firms have been found to 1 1 4    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE 3.10  Men and women work in different sectors throughout and Ramos 2011; Sakellariou 2011). Among the East Asia and Pacific region entrepreneurs, differences in industry and size of enterprise have been found to explain percentage of workers a substantial fraction of the raw performance gaps in profits and revenues among male and a. Indonesia, male b. Indonesia, female female firms.11 What determines gender inequality in economic opportunities? Men and women differ in their economic activities in a number of respects. These dif- ferences in activities may represent differ- ences in choices, but they may also represent differences in the constraints that men and c. Cambodia, male d. Cambodia, female women face in their working lives. This sec- tion explores the economic, demographic, and cultural factors that determine current gender inequalities in access to economic opportunities in the East Asia and Pacific region. Determinants of labor market participation Female labor force participation is affected by growth processes and development. As commerce manufacturing construction development occurs, changes in household transport public administration education, health and social work income, education, wages, marital, and fer- tility choices affect participation. The evolu- Source: CSES (NIS Cambodia), 2008 data; World Bank estimates using Indonesia SUSENAS (BPS ­Indonesia), 2009 data. tion of market opportunities alters the types of jobs present in the economy, as well as the relative demand for skills. The institutional influence factors beyond productivity—from framework of society affects and is affected the provision of benefits to gender-sensitive by the economic participation of women dur- policies.9 In Vietnam, workers in female-run ing the process of development; that is, gender small and medium enterprises are more likely norms, expectations, and perceptions within to receive fringe benefits in addition to wage the household and society affect female labor compensation (Rand and Tarp 2011). Female force participation and are also likely to be owners are more likely to provide health and affected by its evolution. social insurance and to allow sick, vacation, Female labor force participation and its and maternity leave with pay. determinants vary over the life cycle. Early Labor market sorting has been found in their careers, women are similar to men to contribute to the gender wage and earn- in terms of their decisions to join the labor ings gaps.10 Differences in occupational and market, but their participation begins to dif- industrial sorting explain a greater fraction fer as their domestic responsibilities increase. of gender wage gaps in the East Asia and This change is partly due to the increased Pacific region and across the world than time that women devote to household activi- differences in human capital (Ñopo, Daza, ties, such as housework and child rearing, as G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    115 their marital and familial status changes, but FIGURE 3.11  The pattern of female ownership by firm size varies it is also a reflection of differences in the roles across countries of and expectations of married and single percentage of rms owned by women women. Younger women in the region are 90 investing more in education and hence have delayed their entry into the labor market, in 80 both rural and urban areas. Married women 70 appear to take substantial time out of the 60 workforce for raising children, but the same 50 percent is not true for men. Furthermore, large labor force participation gaps open up toward the 40 end of women’s careers, in part as a result 30 of labor market regulations such as gender- 20 differentiated retirement policies. 10 The stage of development of countries and 0 sia DR lia es oa ste a u m ng at their institutions na in go m ne oP Le nu To ilip Sa et on do or Va La Vi Evidence from across the world suggests that, Ph M In Tim as countries develop, female labor force par- Small Medium Large ticipation displays a U-shaped trajectory.12 Source: World Bank estimates using Enterprise Surveys (database), 2006–11 data. Female labor force participation usually Note: Survey data are for small (less than 20 employees), medium (20–99 employees), and large declines as incomes rise and opportunities (100 and more employees) firms in the formal sector. Female-owned firms are defined as those with females among the owners. in the labor market become less attractive to female workers; it then increases again when more attractive employment opportunities emerge (Bloom et al. 2009; Chaudhuri 2009; Goldin 1995; Sinha 1967, cited in ­ Mammen FIGURE 3.12  In Indonesia, female-led enterprises are clustered and Paxson 2000; Tam 2011). In poor, agri- in lower-productivity and capital-intensive industries cultural economies, female participation tends to be high because agricultural work value added (in thousand Indonesia Rupiah) 50 and family responsibilities can easily be com- 45 bined. However, in middle-income countries 40 dominated by the manufacturing and ser- 35 vice sectors, female participation declines in part because most new jobs are difficult to 30 combine with family responsibilities. Female 25 participation rates are higher in high-income 20 countries that have large service sectors and 15 a highly educated workforce. This finding 10 holds both across and within countries over 5 time (Fatima and Sultana 2009; Fuwa 2004; 0 od g ail rt es Juhn and Ureta 2003; Tansel 2001).13 rin po vic t fo re tu ns er c The stylized U-shaped relationship bet­ tra rs fa u he an ot ween female labor force participation and m economic development holds for countries in female-dominated sector male-dominated sector the East Asia and Pacific region, as well as productivity per worker assets per worker globally. Figure 3.13 depicts the relationship between economic development (as captured Source: World Bank staff estimates using Indonesia Family Life Survey 2007/2008. by income per capita) and female labor force Note: The graph shows productivity and assets per worker in five industries for firms with fewer than participation across the globe between 1980 five workers. Productivity is measured by value added. 1 1 6    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE 3.13  Female labor force participation is high by global Gender norms and expectations standards but also varies substantially across the region An important question is why labor force participation varies so much between coun- female labor force participation rate, % tries with similar per capita income levels, 90 Lao PDR as can be seen in figure 3.13. Gender norms 80 Cambodia China strongly influence labor force participation Papua New Guinea Vietnam Australia rates and help to explain the variation in 70 Mongolia Thailand participation across countries that is not Japan Timor-Leste explained by the level of economic devel- percent 60 Tonga Singapore opment. Societal p ­ erceptions of women in Indonesia Korea Rep. 50 Philippines the workplace and gender norms strongly influence labor force participation deci- Malaysia 40 sions (Antecol 2000; Fernández 2010; Fer- Fiji nández and Fogli 2005; Fernández, Fogli, 30 and Olivetti 2004). Countries in which strong socioreligious views exist about women’s role 0 5 6 7 8 9 10 11 12 13 in the public sphere, including the workplace, log GDP per capita (2005 PPP $) have been found to have lower female labor global relationship,1980 global relationship, 2009 force participation rates (Psacharopoulos and Tzannatos 1989). Source: World Bank estimates using labor force data from the ILO KILM database and purchasing power parity (PPP)-adjusted GDP per capita from the Penn World Table. Whether a woman is entitled to make Note: GDP = gross domestic product, PPP = purchasing power parity. The data shown for each her own choices in the economic sphere, country reflect the 2009 data, and the estimated U-shaped relationships use data from across the world. such as entering the workforce or starting a business, varies substantially from coun- try to country. For example, in Vietnam, young single women are increasingly choos- and 2009. The position of the East Asian ing to migrate for employment and to com- and Pacific countries relative to the rest of mute long distances (World Bank 2011a). the world is depicted in red. Two key fea- However, in rural Morobe Province in tures emerge. First, the U-shape pattern holds Papua New Guinea, young women are not across countries for both periods of time. allowed by their families to migrate out of Second, the participation rate associated the village because of the fear of early mar- with each level of development has increased riages, which are regarded as taboo (World between 1980 and 2009, as can be seen by Bank 2011a). the upward shift in the U-shaped pattern In countries with large male-female partic- over time. In the East Asia and Pacific region, ipation gaps throughout the life cycle—such Vietnam and China have substantially higher as Indonesia, Malaysia, and the Philippines— levels of female labor force participation than female labor force participation is considered the world average relative to their income lev- socially and culturally acceptable as long as it els, whereas participation is near the world does not interfere with women’s primary role average in countries such as Indonesia and as wives and mothers. For example, public the Philippines. The pattern of female labor gender discourse in Indonesia and Malaysia force participation seen in parts of the Pacific places particular emphasis on motherhood partly reflects economic structure, whereby and child care as a “woman’s true vocation� labor force participation is higher in agricul- (Blackburn 2001, 2004; Stivens 2006; White ture-based economies. For example, although 2006). Political and institutional ­ evolution Fiji has lower rates of participation relative to also affects the scope for women’s partici- its income level, the rates of participation in pation in education and in the workplace. Papua New Guinea and Vanuatu are substan- Women’s labor force participation is sub- tially higher. stantially higher in socialist and ex-socialist G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    117 ­ ountries, such as China, Mongolia, and c Vietnam has declined from almost 89 percent Vietnam, than in others. in 1998 (similar to men) to about 81 percent in Legislation can codify social norms into 2008 (versus 84 percent for men). Most of the discriminatory labor practices. In Korea, decline occurs in rural areas and is accounted marriage bars to private and public employ- for by rising participation in education. Given ment were common until the 1980s (Hill the corresponding though smaller decline in 1996). In Mongolia, women retire approx- male participation, the overall male-female imately 10 years earlier than men; this participation gap increased from approximately practice is, in part, attributable to a lower 1.5 percentage points in 1998 to 4.5 percent- retirement age for women (World Bank age points in 2008 (Sakellariou 2011). Simi- 2011c). In Mongolia, these differences in larly, in China, the population census indicates retirement ages have contributed to the that the labor force participation of 15- to female-male participation gap rising by 22-year-old urban males and females dropped approximately 20 percentage points between from 70.6 and 72.7 percent, respectively, in the ages of 50 and 60. 1982 to 43.8 and 46.4 percent, respectively, in However, gender roles and relations 2000, largely because of an increase in educa- within households do change over time, par- tional enrollment (Hughes, Maurer-Fazio, and ticularly in evolving environments. In China, Zhang 2005). In rural areas, participation of the economic reforms of recent decades have the female population ages 15 to 22 declined increased the range of opportunities avail- from 84.3 percent to 67.3 percent between able to both men and women in paid employ- 1982 and 2000 for the same reason. Where ment, as the structure of the economy has education acquisition varies by gender, a gen- moved away from predominantly agrarian der gap in participation can be observed. with a capital-intensive heavy industry sec- Although changes in education have not tor toward labor-intensive light industry and explained a large fraction of changes in services (Hughes, Maurer-Fazio, and Zhang participation, variation in education con- 2007). However, this transition has been tributes to explaining differences in female argued to have created new obstacles for labor force participation within a country. women: the state has retreated from its for- In Indonesia, women with higher levels mer commitments to gender equality and to of education are more likely to enter the strong enforcement of workplace protections labor market, particularly in urban areas, for women, thus allowing the reemergence of which may reflect their higher wage premi- traditional patriarchal values (Entwisle and ums and higher opportunity cost of being Henderson 2000). inactive (Ogawa and Akter 2007; World Bank 2010a). Literacy in Indonesia is also Individual and household-level factors: strongly associated with both participation Income, education, marriage and children in the labor market and occupational seg- Changes in the characteristics of women and regation among women; thus, being illit- households—notably, changes in educational erate poses a double barrier to labor force attainment, changes in the demographic pro- outcomes (Gallaway and Bernasek 2004). file of the population, and growth in house- In Vietnam, women with no educational hold incomes—do not, for the most part, qualifications are more likely than their explain changes in the male-female gap in male counterparts to be inactive, those with labor force participation in many countries in primary or secondary education are slightly the East Asia and Pacific region during the less likely than their male counterparts to past decade (Sakellariou 2011). be working, and those with higher levels of Rising education levels among girls in coun- education are more likely than their male tries across the region have led to a substantial counterparts to be working (Pierre 2011). decline in girls’ workforce participation rates. Declines in fertility have been found to For example, women’s participation in rural exert a large positive effect on the labor 1 1 8    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C force participation rate across the world. Retirement policies The effect is strongest for women ages In nearly all countries in the region, the male- 20 to 39, with an additional child being asso- female participation gap rises after the age ciated with a reduction of approximately of 50, indicating that women retire earlier four years of paid work over a woman’s than men. These differences can be partly lifetime (Bloom et al. 2009). In Korea, the attributed to gender-differentiated retire- reduction in the total fertility rate, from 5.6 ment policies. In 4 of 12 countries studied in children per woman in 1962 to 1.2 in 2002, the region (see table 3.1), the statutory age of has been suggested to have increased per retirement for women in the private sector is capita income by approximately 36 percent five years earlier than that of men, although over the period, because of an increase in only in China is it mandatory to retire at the the size of the workforce and higher female statutory age. In China, gender differences labor force participation as well as a longer- in mandatory retirement ages influence the term increase in the capital-to-labor ratio prospective value of hiring older men and (Bloom et al. 2009). women, implying that the expected return of What will stop us from getting a job is having hiring an older man is greater than that of plenty of children and having nobody to mind hiring an otherwise identical older woman them. (Giles, Wang, and Cai 2011). Differences in Young woman, Indonesia retirement prospects across rural and urban (World Bank 2011a) areas also contribute to the explanation of differences in participation across rural and Child-care costs have a negative impact on urban areas among older workers. In urban female labor force participation, particularly areas, where most long-term residents have in families without the support of familial had formal wage employment, residents networks.14 Reductions in child-care provi- retire at a relatively young age and receive sions in China and Mongolia have had a neg- substantial pension support. In contrast, ative impact on labor force participation. In rural residents often lack pension support China, participation of women of childbear- and hence make their labor supply decisions ing age has declined in urban areas, a trend in the absence of pension availability and the that is partly attributable to higher child-care constraint of mandatory retirement (Giles, costs now that child care is no longer sub- Wang, and Cai 2011). Since women over 60 sidized by the state (Chi and Li 2007; Du in rural areas are less likely to be vested in and Dong 2010; Li and Song 2011; Maurer- formal labor markets, they are less likely Fazio et al. 2011). Similar patterns have been than other elderly to be covered by the pen- observed in Mongolia, where state-funded sion system—only 1 percent of rural women early child care and education were rolled over 60 report that pension income is their back in the 1990s (World Bank 2011c; World most significant source of financial support, Bank and ADB 2005). compared with 8 percent of rural men over The labor force participation of older 60 (World Bank 2012). women is also affected by the presence of Differences in retirement ages are likely children, since they are often responsible for to affect outcomes beyond labor force par- caring for younger household members. In ticipation, including educational invest- Mongolia and rural Vietnam, the evidence ment, the number of women in positions of suggests that a “grandmother effect� is pres- power, and the risk of poverty for elderly ent. Older women in households with chil- women. Because women anticipate having dren under the age of 10 have a 15 percentage a shorter working life than men, differences point larger participation gap than women in in retirement ages affect education and households with no children under the age of occupation choices. Gender differences in 10, a difference that may be attributable to retirement ages also imply that women are older women staying at home to look after less likely to rise to the top of occupational their grandchildren. G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    119 ladders, because they have less experience TABLE 3.1  Legal retirement ages in the East Asia than men toward the end of their careers. and Pacific region Fewer women in Mongolia reach higher- Retirement age Men Women level managerial positions, despite women China 60 55 having one of the highest levels of education Hong Kong SAR, China 65 65 on average in the region. Furthermore, early Indonesia 55 55 retirement reduces pension payouts, which Lao PDR 60 60 depend on the length of tenure. This disad- Malaysia 55 50 vantage in pension receipt can increase the Mongolia 60 55 risk of poverty for low-income households Papua New Guinea 55 55 headed by females—women accounted for Philippines 65 65 approximately 70 percent of elderly single– Singapore 67 67 headed households in Mongolia in 2010 Taiwan, China 65 65 (World Bank 2011c). Thailand 55 55 Vietnam 60 55 Source: World Bank 2011f. Determinants of gender gaps in earnings contrast, in the Philippines, the wage gap wid- Earnings gaps between male ened between 2000 and 2009, a change that and female workers is partly attributable to growing differences The gender earnings and wage gap may reflect between men and women in terms of their differences between men and women in edu- returns to education and other characteristics. cation and other characteristics, as well as Indonesia has the widest earnings gap among differences in the returns to these characteris- the six countries examined, and the study tics by gender. The literature separates gender found substantial differences between rural gaps into parts that are “explained� by mea- and urban areas, in the size of the wage gap sured factors, such as education, age, experi- and the factors contributing to it. ence, and marital status, and those that are Average differences between the character- “unexplained,� often regarded as a measure of istics of men and women, such as education, discrimination against female workers. Care experience, and the sector and occupation must be taken in interpreting the remaining of employment, explain up to one-third of component of wages as discrimination, how- the male-female wage differentials across the ever, since this interpretation assumes that all region and across the world.16 Occupational relevant factors have been considered.15 and industrial sorting have been found to con- The level and changes in the gender wage tribute more to explaining the gender wage gap over time in the East Asia and Pacific gaps than do differences in human capital region vary by country and across the income across the world, and the same holds true for distribution, suggesting that the forces that the East Asia and Pacific region (Ñopo, Daza, lead to narrowing wage gaps over time are and Ramos 2011; Sakellariou 2011). idiosyncratic. A study commissioned for this The narrowing of gender wage gaps over report found that the evolution of gender time can be partially attributed to the con- wage gaps over time has followed different verging of educational attainment of men and paths in Cambodia, Indonesia, Mongolia, women. Between 1985 and 2005, the average the Philippines, Thailand, and Vietnam characteristics of the female wage and sala- (Sakellariou 2011). In Thailand, gender ried workforce in Thailand improved over wage gaps throughout the wage distribu- time relative to those of males. Nakavachara tion decreased substantially between 1996 (2010) finds that the substantial increase in and 2006. Changes in experience, education, the education of females between 1985 and and return to experience were found to have 2005 was the major source of the narrow- contributed significantly to this reduction. In ing of the gender earnings gap in Thailand 1 2 0    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C during this period. Dhanani and Islam (2004) gender wage gap attributable to returns to find that, although females earned on average characteristics is greater than the gender wage about 30 percent less than men between 1976 gap attributable to characteristics across the and 2000 in Indonesia, overall wage inequal- distribution. On average, differences in char- ity varies with industrial affiliation as well as acteristics explain just over 35 percent of the education and age. The wage gap in ­ Indonesia gross gap in 2009. Differences in labor market increases substantially with age, reflecting in experience and returns to labor market expe- part growing differences in education and rience constituted the major contributor to the experience between older men and women. characteristics component in both years. Since women have higher levels of educa- There is substantial variation in the share tion on average than men in some countries of gender wage gaps explained by differ- in the region, estimates suggest that women ences in characteristics between men and would have earned more than men had they women. In Vietnam, the fraction of the gap faced the same returns to their education and explained by differences in characteristics other characteristics as men. In the Philippines is only 11 percent in urban areas, whereas and Mongolia, women’s levels of human capi- in rural areas the characteristics of females tal are higher than those of men on average. would suggest that, on average, the wages of In Mongolia, taking into account the different women should be higher than those of men characteristics of men and women, women (Sakellariou 2011). In China, the mean gen- should have earned 22 percent more than men der pay gap increased substantially between in 2006 (Pastore 2009). Similarly, in the Phil- 1987 and 2007, from 18 percent in 1995 (Chi ippines, education and other characteristics of and Li 2007) to nearly 30 percent in 2007 (Li women would suggest that, on average, the and Song 2011). Over this period, the major- wages of women should be higher than those ity of the increase was not attributable to dif- of men (Sakellariou 2011). ferences in characteristics but rather was due The bulk of the gender wage gap within to unexplained differences in the returns to the region is due to differences in the labor male and female characteristics. market value of male and female char- Marriage and childbearing have a larger acteristics.17 In the East Asia and Pacific negative effect on the wages of females than region, the fraction of the gender wage gap on the wages of males. The trade-off for explained by differences in characteristics women between career, earnings growth, and (such as education and experience)—the family does not appear to exist for men. For explained component—is smaller than the some women, this trade-off is associated with fraction of the wage gap attributable to dif- increasingly stark choices. Anecdotal evidence ferences in returns—the unexplained com- suggests that women in richer parts of East ponent (Sakellariou 2011). Figure 3.14 shows Asia are increasingly less likely to marry and the percentage difference between male and marry later when they do, in part because of female wages at deciles of the wage distribu- the perceived incompatibility of marriage and tion in Indonesia in 2009, and the difference career (Economist 2011). The negative effect in male and female wages that is attributable of childbearing on earnings and employment to differences in characteristics and returns has been found across the world as well as on those characteristics.18 Two noteworthy within countries in the East Asia and Pacific points emerge from figure 3.14. First, the region.19 In Cebu, in the Philippines, children gender wage gap is wider at the bottom than have a strong negative effect on a woman’s at the top of the wage distribution, pointing likelihood of participating in the labor force toward the phenomenon of “sticky floors.� and, once in the labor force, on her earnings Second, differences in returns to character- over time. The negative effect of children on istics between men and women are able to women’s earnings represents both a reduc- account for a greater share of the gender wage tion in the number of hours worked and a gap at all points in the wage distribution—the shift to lower-paying and often less secure G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    121 jobs that are more compatible with child- FIGURE 3.14  The fraction of the gender wage gap explained by bearing responsibilities (Adair et al. 2002). differences in characteristics is smaller than the fraction of the In China, married women have substantially wage gap attributable to differences in returns in Indonesia, 2009 larger gender wage gaps than their unmar- percentage di erence between male and female workers ried counterparts, and the wage gap between 0.45 married men and women grew between 1995 and 2007 while it closed for single men and 0.40 women (Hughes and Maurer-Fazio 2002; Li 0.35 and Song 2011). Furthermore, the proportion 0.30 of the gender wage gap unexplained by dif- 0.25 percent ferences in the productive characteristics of 0.20 men and women is greater for married than for single women, although the gap is nar- 0.15 rower in the case of more educated married 0.10 women. These results may be attributable to 0.05 the greater responsibilities borne by Chinese 0 women for household chores and child care. 1st 2nd 3rd 4th 5th 6th 7th 8th 9th Gender wage paths over careers may income decile reflect differences in male and female work- wage gap returns component characteristics component place behavior, as well as differences in the types of characteristics that men and women Source: Sakellariou 2011. Note: The graph uses the approach of Firpo, Fortin, and Lemieux (2009), based on Indonesia value in a job. Evidence from the organiza- National Labor Force Survey (BPS Sakernas), 1997 and 2009 data. tional and human resources literature shows females as being less competitive, more risk averse (Croson and Gneezy 2009), less likely reflecting their respective responses to these to push for pay rises (Babcock and Laschever life-cycle transitions. 2003), and more likely to be content with lower starting salaries. This tendency may Earnings gaps between male and female be due in part to women having different entrepreneurs demands for job characteristics and expecta- Female-run firms have fewer employees, tions of labor market outcomes than men. For lower sales, and lower capital stock than example, Bender, Donohue, and Heywood those run by men, as evidence from Europe (2005) find that flexibility is an important and Central Asia, Africa, and Latin America job characteristic that appeals to women and and the Caribbean has shown (Amin 2010; may play a contributing role in gender-based Bardasi, Blackden, and Guzman 2007; Bruhn labor market segregation. 2009; Costa and Rijkers 2011; Sabarwal and In sum, men earn more than women across Terell 2008). Various conditions explain why countries in the East Asia and Pacific region, female-run firms produce less per worker although the reasons for this vary from coun- than male-run firms and, in particular, why try to country. Education differences between women tend to head smaller firms in less men and women are not able to explain the capital-intensive sectors than men. The three bulk of gender gaps in labor market wages. predominant explanations examined here are In some countries, such as Mongolia and sectoral segregation, skills, and constraints to the Philippines, education differences in fact business development. imply that women should earn more than Sectoral segregation. Firm size and indus- men. The most important differences between trial and occupational sorting along gender men and women are those related to occupa- lines explain a large part of the differences tional choice and family life. Marriage and in returns to capital between firms run by childbearing are likely to affect men’s and men and those run by women. In a number women’s earnings in different ways, partly of studies comparing the productivity of 1 2 2    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C male- and female-run enterprises around the microenterprises make up 98.9, 98.2, and 92.3 world, productivity gaps diminish substan- percent of all micro, small, and medium enter- tially once the industry or sector of employ- prises (MSMEs), respectively, and they make ment is taken into account (Aterido and up 79.3 percent of all enterprises in Malaysia Hallward-Driemeier 2009; Bruhn 2009; (Kushnir, Mirmulstein, and Ramalho 2010). Costa and Rijkers 2011; de Mel, McKenzie, The East Asia and Pacific region has the and Woodruff 2009a, 2009b). This ­ fi nding highest ratio of MSME employment to total also holds for the East Asia and Pacific employment of all regions, mainly driven by region. In Mongolia, substantial differences China, where MSMEs account for 80 percent in capital and sales exist between male- of total employment. and female-owned firms. These are largely Skills. Evidence on whether female entre- explained by the sector of employment and preneurs have a lower level of managerial firm size (World Bank 2011c). In Lao PDR, skills than male entrepreneurs is limited in differences in the productivity between the region. However, gender differences in male- and female-owned firms in the formal both education and soft skills are likely to sector disappear once differences in sector, affect men’s and women’s respective decisions capital, size, and other factors that affect to participate in self-employment and in their productivity are considered (Davies and sector of choice (Brush 1992). Furthermore, Record 2010). In Vietnam, a study commis- differences in entrepreneurial and managerial sioned for this report found that there were skills (such as being able to identify market no gender differences in short-term revenue niches and do bookkeeping) between men growth or survival probability among in and women may explain differences in the firms in the formal sector between 1997 size of the firms that they establish and earn- and 2009 (Bjerge and Rand 2011). 20 In the ings gaps (Bruhn, Karlan, and Schoar 2010). informal sector in Indonesia, the sector of Although evidence from studies of entre- operation and the firm’s size accounts for preneurial skills in the region is limited, the almost 90 percent of the observed gender evidence from outside the region is mixed gap in productivity in rural areas (Costa with regard to whether a lack of skills is an and Rijkers 2011), although they account important constraint for entrepreneurs. A for substantially less of the gap in urban study from Peru suggests that giving busi- Indonesia (Badiani and Posadas 2011). ness training to female clients of a microcre- Informality is likely to be a greater deter- dit program did not lead to higher profits or minant of an enterprise’s success than the revenues on average, although the clients did gender of the manager or owner. Evidence adopt some of the activities taught in the pro- from Indonesia and Vietnam suggests that gram, including thinking proactively about differences in profits between female-headed new markets and profit-making opportuni- firms and male-headed firms are substantially ties (Karlan and Valdivia 2011). In Pakistan, smaller than the differences between female- business training was found to increase the headed firms in the formal sector and those survival and profitability of male-run firms in the informal sector (figure 3.15). This mir- but had no effect on female-run firms (Giné rors evidence from Africa (Aterido and Hall- and Mansuri 2011). However, in India, a ward-Driemeier 2009). two-day training program for female clients Evidence on gender constraints in informal of an Indian microfinance institution was enterprises within the East Asia and Pacific found to increase both the amount that they region is more limited than evidence from the borrowed and the likelihood of the clients’ formal sector, however. More data are urgently receiving labor income (Field, Jayachandran, needed on the informal sector, as microen- and Pande 2010). terprises constitute a substantial fraction of Within the region, female entrepreneurs enterprises and employment in the region. themselves consider that their skill limitations In Indonesia, Vietnam, and the Philippines, are a barrier to their success. In Vietnam, a G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    123 survey of 500 female owners of enterprises FIGURE 3.15  Differences in productivity across informal and revealed that female entrepreneurs felt the formal firms in Vietnam are larger than differences across need to improve their skills, particularly male- and female-led enterprises in business management and leadership, through training and education (IFC and 60 productivity measured as value MPDF 2006). These findings are supported added (Dong, thousands) Female informal by another survey that indicated that females 50 rms earn have lower general training levels than men 15,268 dong per (VCCI 2006). worker less 40 than formal rms Constraints to business development. In value added the formal sector, the constraints faced by 30 female-run firms do not appear to be uni- formly greater than those faced by male-run 20 Female rms firms (Davies and Record 2010; IFC 2011; earn 5,809 IFC/NORC Indonesia 2010; World Bank 10 dong per worker 2011b). 21 In the Philippines, qualitative and less than male quantitative research has found little evi- 0 rms unlicensed licensed dence of gender differentiation in lending or female male borrowing in small and medium enterprises (IFC 2011). In Indonesia, insufficient finance Source: World Bank estimates based on Vietnam Household Living Standards Surveys (VHLSS) and financial management were found to be (GSO Vietnam), 2009 data. less of a concern for women (IFC/NORC Note: unlicensed = informal. ­Indonesia 2010). FIGURE 3.16  Self-reporting shows little difference in access to Data from enterprise surveys carried out finance between male- and female-run firms, with the exception of in five Pacific and four East and Southeast Timor-Leste and Tonga Asian countries in 2009 show that entrepre- neurs, regardless of gender, named compe- percentage of formal sector rms that reported tition, finance, and electricity as their top 35 di culty in accessing credit three constraints in five of the nine countries. Male-led and female-led firms reported the 30 same constraints as being the most important 25 in all countries, except Tonga, Vanuatu, and Vietnam. The difference between ­ self-reported 20 percent access to credit constraints for male- and 15 female-run formal sector firms is small in all 10 countries, with the exception of Timor-Leste and Tonga (figure 3.16). 5 However, credit constraints are greater 0 in the smallest firms and in the informal te u R a oa m ia s i Fij ne ng PD at es es na m sector where female-run firms are concen- nu pi To n -L Sa et o do ilip Va La or Vi In Tim trated. Constraints to female entrepreneurs Ph female male may arise in both formal and informal insti- tutional structures, for example, because of Source: World Bank estimates using Enterprise Surveys (database), 2006–11 data. difficulties interacting with male officials Note: Finance is considered to be an obstacle if it is reported as a major or severe constraint. who adhere to cultural norms of female pro- priety. In Indonesia, access to capital is the run informal firms. Female-led microen- most important constraint reported in both terprises also have lower levels of start-up male- and female-run informal firms (figure capital than male-run microenterprises. 22 3.17), and female-run informal firms appear Access to capital may be more limited—and to have substantially less capital than male- more expensive—for female entrepreneurs 1 2 4    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE 3.17  Male- and female-led informal firms report similar Lao PDR over 70 percent of the workforce is constraints in Indonesia employed in agriculture. As the countries across the world have 35 percentage of rms reporting constraints developed, women have been playing a greater role in agricultural activities as 30 men move to the cities in search of non- 25 farm work (Deere and Leon 2003; Ganguly 2003; Mu and van de Walle 2009). This 20 percent “feminization� of agriculture can be seen in 15 multiple countries in the region. In China, the number of households in which women 10 participated in agricultural activities and 5 performed all the farm work rose from 14 percent in 1991 to just under 30 percent 0 in 2009, whereas the fraction of households capital marketing raw transport energy paying of materials wages in which men did all the work remained male female fairly stable (de Brauw et al. 2011; ­ R awski and Mead 1998). In C ­ ambodia, women Source: World Bank estimates using SUSI Integrated (Survey of Cottage and Small-Scale Firms) 2002. make up the majority of farmers—1.4 mil- lion female farmers compared to 1.2 million male farmers (UNIFEM, World Bank, ADB, because they lack collateral (World Bank UNDP, and DFID/UK 2004). In Vietnam, 2011g). According to calculations from the agriculture accounted for 64 percent of work- World Bank Enterprise Surveys database of ing women in rural areas in 2008 compared 2006–11, in all countries in the region for to 53 percent of working men (World Bank which data were available, land and build- 2011e). ings were the predominant forms of collat- Despite the important role played in agri- eral used to acquire a loan for production culture by women in the region, little evidence purposes. This practice puts female entre- is available regarding gender differences in preneurs at a disadvantage since they have productivity, access to inputs, or agricultural fewer of these assets as well as less secure services. Remedying this information gap is access to land or other immobile assets. likely to become increasingly important if the However, the most important determi- feminization of agriculture continues within nant of access to credit is the size of the the region. firm (Beck et al. 2006). Credit institutions The evidence from other regions suggests tend to regard small firms as a bigger risk that female farmers earn less than male farm- than large firms, and, since women tend ers. 23 For example, evidence from Africa to manage smaller firms, this is likely to suggests that, within a household, yields lead to female-run businesses being more on male-run plots are higher than those on constrained in the credit market than busi- female-run plots, predominantly because of nesses that are run by men (Simavi, Manuel, increased male access to fertilizer and labor and Blackden 2010). inputs (Goldstein and Udry 2008; Udry 1996). In comparison, the available evidence Earnings gaps between male and female from China suggests that female farmers are farmers just as productive and efficient as male farm- The agricultural sector continues to be the ers, despite differences in mechanization and major sector of employment in many coun- fertilizer use (de Brauw et al. 2011). Although tries in the region. In Vanuatu, 80 percent female farmers are able to produce similar of the population rely upon small-scale agri- amounts per hectare to men, they do not pro- culture (IFC 2010), and in Cambodia and duce as much overall owing to differences in G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    125 land and nonland inputs between male and inputs. In Cambodia, access to agricultural female farmers. extension services was found to be substan- Agricultural production varies in part tially lower for female farmers than for male between male and female farmers because of farmers in the early 2000s, despite the fact differences in access to land. Female farmers that there were more female farmers than across the world own and operate less land, male farmers (UNIFEM, World Bank, ADB, and often have lower-quality land than male UNDP, and DFID/UK 2004). 25 Reasons for farmers (Deere and Doss 2006; Quisumbing the lower access to extension services among 1998; Quisumbing, Estudillo, and Otsuka female farmers include the extension services’ 2004). 24 Evidence from the East Asia and focus on activities normally conducted by Pacific region suggests that there are sub- men, the focus on literate farmers, and the stantial differences in access to land between need to travel several kilometers to access ser- male and female farmers, as discussed in vices (UNIFEM, World Bank, ADB, UNDP, chapter 2. In China, female-headed house- and DFID/UK 2004). Efforts to reduce these holds own 30 percent less land per capita constraints in the Cambodian context are than male-headed households (de Brauw discussed in chapter 6. et al. 2011). In Lao PDR, male-headed agri- Finally, gender-based differences in social cultural households own approximately and political capital have been found in a 16 percent more land than female-headed number of rural areas, reducing women’s households on average, although in the access to information on farming techniques north of the country men have access to and their ability to protect and regulate local 28 percent more land (FAO/Sida Partnership resources and their marketing channels. In Cooperation 2010). A World Bank study the Philippines, men and women do not differ of women’s landholdings in post-tsunami in their level of participation in local groups, Aceh found that women’s landholdings but they do differ in the types of groups that were considerably less than men’s (World they join. Men are more likely than women Bank 2010b). to be members of production-based groups, Female farmers have also been found to whereas women are more likely to be mem- have less access to technological inputs such as bers of civic groups, which include women’s fertilizers and high quality seeds ­ (Peterman, groups, village youth associations, school Behrman, and Quisumbing 2010). An impor- committees, and village officials (Godquin tant theme in the literature on the use of and Quisumbing 2008). In Indonesia, men productivity-enhancing technology is that have been found to participate in civil soci- access to inputs, not the propensity to use ety organizations related to community-level inputs or productivity once the inputs have governance, physical infrastructure, envi- been used, is the key factor for many female ronmental improvements, and neighborhood farmers. In a review of 24 studies of techno- security, whereas women participate in orga- logical input use, access, and adoption of fer- nizations that focus on family welfare, eco- tilizer, seed varieties, tools, and pesticides in nomics, and health (Beard 2005). Africa and South Asia, Peterman, Behrman, and Quisumbing (2010) found that men have Determinants of labor market greater access to technological resources in segregation the majority of cases. To know whether the same constraints are pertinent in the region The determinants of persistent gender requires further evidence on this topic. employment segregation across countries and Gender differences in access to exten- over time lie at all levels of the economy and sion services are also likely to contribute to society, from gender norms within households ­ differences in the adoption of new technolo- and ­communities to economic signals in the gies and farming practices and to perpetu- marketplace and ideological predilections at ate existing gender inequalities in access to a societal level. Factors that influence labor 1 2 6    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C market segregation along gender lines include only a small fraction of men and women (a) differences between men and women in continue to higher education. Therefore, the level and composition of their education the nexus of education choice and occupa- and experience, (b) differences in their pref- tional segregation is not currently a primary erences for types of work and job character- driving force for labor market segregation istics, (c) their comparative advantages due in the region. However, it is likely to play to their differences in physical endowments, a larger role among younger cohorts and (d) employers’ perceptions about male and future generations as education continues female employees (sometimes to the extent of to expand. For example, in Taiwan, China, discrimination), and (e) labor market insti- gender wage gaps declined substantially tutions that limit or restrict occupational between 1979 and 1995, a period during choices. These factors are, in themselves, which there was a rapid increase in average determined by the economic signals, institu- educational attainment and a shift from aca- tions, and norms facing individuals, house- demic general-curriculum education toward holds, and communities. vocational training. Among university gradu- This section examines the effects of three ates, differences in degree type between men key factors: (a) gender differences in educa- and women are able to explain between tion levels and labor market skills; (b) gender 20 percent and 50 percent of the gender wage differences in time-use patterns and gender gap, although the link between occupation roles within households, both of which affect and degree type is lower among women than the characteristics and types of work women men (Baraka 1999). are able to do; and (c) labor market and insti- tutional barriers to women’s choices. Time use and gender roles within households Education and skills Differences in the ways men and women Education levels and the type of education spend their time are informative in illuminat- acquired (for example, vocational versus gen- ing why they may invest differently in human eral education) affect the skills that individu- capital, how gender differences in labor mar- als bring into the labor market and hence are ket participation may arise, and what their likely to contribute to their sectoral and occu- occupational and industry choices will be. pational choices.26 As discussed in chapter 2, Time is a valuable resource that is endowed male and female education has been converg- equally across men and women—everyone ing over the past two decades in the East Asia has just 24 hours in a day. Time can be and Pacific region, and in some countries, devoted to a number of uses: labor market such as Mongolia and the Philippines, edu- work; unpaid work within the household, cation acquisition by women has overtaken such as conducting domestic chores and car- that of men. At the tertiary level of educa- ing for children and the elderly; and personal tion (universities, technical training schools, activities, including sleeping, eating, and lei- and so forth), a student’s field of specializa- sure (Becker 1965, 1981). tion was likely to be closely related to his or Women across the world work more than her subsequent sectoral choices. Any gender men. This stylized fact holds true in multiple differences in education choices are therefore low- and middle-income countries across the likely to be mirrored in sectoral dissimilarities world (Berniell and Sanchez-Páramo 2011; among men and women. If women tend to World Bank 2011g), as well as in several choose different fields than men, sectoral and OECD countries (Burda, Hamermesh, and occupational segregation is likely to increase Weil 2007; Slootmaekers-Miranda 2011). as the number of men and women continuing Gender differences in time-use patterns exists to higher and tertiary education rises. at all ages. Whereas men are able to focus pre- In the lower- and middle-income coun- dominantly on their single productive role and tries in the East Asia and Pacific region, conduct their other roles sequentially, women G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    127 are more likely to play these roles simultane- FIGURE 3.18  In Cambodia, women—particularly those with young ously and have to balance competing uses of children—balance household work commitment with market work their time (Blackden and Wodon 2006). However, time differences in hours worked hours spent on activities per day 10 by men and women diminish when growth 9 in gross domestic product (GDP) is combined 8 with gender-neutral social norms. A study 7 of 25 countries across the world finds that 6 men and women did the same amount of hours 5 total work in rich northern countries (Burda, 4 Hamermesh, and Weil 2007). The gender 3 difference in total working time is close to 2 zero for countries with relatively high female 1 employment and with more gender-neutral 0 social norms. without a child with a child without a child with a child In the East Asia and Pacific region, women work more hours and devote more time to male female caregiving and housework than men, and men specialize in market-oriented activities. market work housework and caring Since women’s share of unpaid work is higher Source: World Bank estimates using CSES (NIS Cambodia), 2004 data. than that of men, this translates into shorter time in paid work. Cambodia and 1 hour extra per day in Lao Women indeed work harder. We go to the field PDR (figure 1.20 and 3.18). and return home at the same time as our hus- Third, gender differences in the time band. Afterwards, we still have to cook, do laundry and do other household chores. The devoted to household activities are smaller men after they return home from the field do in high-income households, although this not want to work anymore. represents a reduction in female working Adult woman in rural Batu Palano, time rather than an increase in male work- Indonesia (World Bank 2011a) ing time. In Timor-Leste, rural women in the highest income quintile spend more time on A number of stylized facts emerge from domestic activities than men (figure 3.19). time-use patterns by gender in the region. The difference between men and women First, women work more hours than men; this narrows with rising income. The narrow- holds true at all ages. In Lao PDR, women ing of time-use differences, however, arises work on average 2 hours extra per day than from a reduction in the time devoted by men, and in Cambodia, they work on average women to these activities, thus reflecting 1.2 hours more per day. increased access to technologies and house- Second, gender differences in time-use hold help rather than an increase in male patterns are starker during the childbearing participation. years. In Cambodia, men and women spend Finally, gender differences in the time similar amounts of time in market work until devoted to household activities start at an they are 20 years old, but after that point, early age. In the Philippines, Pörtner (2009) women devote less time to market work studied time use of children between the ages than men and more time to domestic activi- of 7 and 16 and found that, although par- ties (CSES [NIS Cambodia], 2004 data). The ticipation rates and time spent in school by greatest difference between male and female boys and girls were similar, time spent on hours worked is during childbearing years. work and household chores is closely related In households with young children, women to the gender of the child. Boys participate work on average 2 hours extra per day in in market-related activities more than twice 1 2 8    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C FIGURE 3.19  Women in Timor-Leste spend more time on domestic to be found in the formal sector, whereas activities than men, and these differences are found among richer married women with children are found as well as poorer households in the informal sector, particularly in self- employment. The evidence suggests that hours spent on domestic work per day the movement of married women with chil- 2.5 dren into self-employment is a choice rather than an effect of discrimination (Bosch and 2.0 Maloney 2011). Differences in the time-use patterns of 1.5 men and women also reflect society’s norms hours regarding gender roles and “appropriate 1.0 work.� These social norms and customs affect the ability of both men and women to par- 0.5 ticipate in the economic, social, or political domain. In China, despite strong governmen- 0 Q1 Q2 Q3 Q4 Q5 tal support for gender equality on multiple expenditure quintile levels in the workplace, cultural norms have female male still emphasized the role of women as xianqi liangmu (a good wife and a good mother) Source: World Bank estimates using Timor-Leste Survey of Living Standards (NSD [Timor-Leste] and as the bearer of responsibility for house- 2007 data). hold work (Chen 2005; Honig 2000). These norms persist even as development progresses as often as girls, whereas from the age of 14, and women’s share of income in household girls spend on average twice as much time on economies rises. In Beijing, even when the household chores than boys. wife’s income represents a larger share of Spending greater time on household activi- the couple’s combined income, women still ties has implications for women’s labor mar- do the majority of the household work (Zuo ket outcomes. Because women do a greater and Bian 2001). Both men and women justi- proportion of household work and child care fied this in terms of their gender roles in the than men, they have less time for productive household. 27 Similar patterns are found in activities. This pattern makes them less likely Nanjing (Kim et al. 2010). to enter the labor market and implies that, Social norms may be reinforced at the when they do, they are more likely to choose level of the nation-state, if the role of women occupations that give them the flexibility to as homemakers and mothers and of men as conduct their domestic responsibilities (Anker breadwinners is underlined in political dis- 1998; Becker 1965). course. In Malaysia, the Nation of Character Women may regard the informal sec- project focused on 25 key values important tor as more compatible with childbearing for the development of good character in chil- and household responsibilities than the dren. This project very clearly put forward formal sector, since it offers greater flex- that women’s most important tasks were ibility in hours, although flexibility comes related to the home and to strengthening of at a price—less security, lower status, and the family. The division of labor between a lack of nonwage benefits such as social men and women in the public and private security, health insurance, and paid sick sphere is further emphasized from an early leave. Female entrepreneurs may similarly age in parts of the region (White 2006). prefer to keep their businesses small because In Cambodia, a book of moral codes—the doing so allows them to combine household Chba’p srey — emphasizes the “proper� responsibilities with work (Sabarwal, Ter- behavior and conduct of women, and limits rell, and Bardasi 2011). In Thailand, single their opportunities outside the household women without children are more likely (Dasgupta and Williams 2010). G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    129 Labor regulations and informal more restrictions on industry choices than institutions other men and women (World Bank 2010f). Labor laws and regulations can directly These protective restrictions may increase affect the demand for female workers and the cost of employing women. For example, constrain these workers in their choice of in Kiribati, Papua New Guinea, and the occupation. In many countries, restrictions Philippines, women are not allowed to work on women’s working hours or industries the same night hours as men. In the Philip- were introduced as measures to protect the pines, the law states that women are unable health of women (particularly those who to work between 10 at night and 6 in the are pregnant or lactating) or women in morning of the following day. The restric- potentially hazardous jobs. However, with tions on women’s work have been criticized improvements in labor market conditions in as particularly restrictive by the call center dangerous industries and with the passing industry, which employs a large propor- of employment legislation designed to pro- tion of women who are required to work tect the health and safety of workers, many at night (Keitel 2009). Paid maternity leave of these restrictions may no longer be rel- also increases the cost to employers of hir- evant and could be changed. Measures that ing female workers rather than male work- limit women to work only daytime hours or ers, particularly in countries where paternity restrict their work to a subset of industries leave policies are not in place. In contexts limit their employment options and also where employers bear the burden of this drive employers to hire only men for jobs cost, this is particularly likely to raise the that women might otherwise do. cost of hiring female workers. A study on labor laws commissioned Discriminatory laws in the area of family for this report found that protective legis- and marriage also affect women’s economic lation prohibits women from working in opportunities. Laws relating to family, mar- industries and occupations in 9 of the 12 riage, and inheritance play a key role in influ- low- and ­ m iddle-income countries exam- encing women’s economic rights, including ined in the region (China, Fiji, Kiribati, access to land, housing, and other forms of Korea, M ­ alaysia, Papua New Guinea, the property (Ellis, Kirkwood, and Malhotra ­ Philippines, T­ hailand, and Vietnam) (World 2010). The law of succession in Tonga’s con- Bank 2010f). In Thailand, women are pro- stitution, for example, allows only males to hibited from working in certain occupations, inherit. In Indonesia, the civil code prevents including mining; working on a scaffold women from entering into contracts on their more than 10 meters high; and producing or own behalf, whether to sell or buy property, transporting explosives or flammable materi- which enables men to influence women’s als. In Vietnam, the Labor Code prohibits access to collateral. Access to land is depen- assigning female workers to “heavy or dan- dent on a woman’s married status, and her gerous work, or work requiring contact with control and ownership can be lost upon toxic substances, which has adverse effects divorce, widowhood, migration, or desertion on her ability to bear and raise a child.� The by her husband (IFC 2010). code also provides that women, regardless of Women also have less access to informa- age, cannot be employed in mines or in deep tion on opportunities and job networks. water. No similar provisions are applicable Personal connections are recognized as to men. In Mongolia, until 2008, extensive facilitating job search, but women’s domestic labor market regulations limited female par- responsibilities make it difficult for them to ticipation in multiple sectors of the economy, forge useful social connections (Timberlake including mining, transport, and construc- 2005). In Nanjing city, China, a case study tion (World Bank 2011c). Pregnant women suggests that women secured less-­ attractive are even more limited in their choices: in jobs through their own networks than 6 out of 10 countries in the region, they have through those of their husbands. Women 1 3 0    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C who have fewer social contacts also were and among genders (World Bank 2011a). See more vulnerable to redundancy and expe- box 3.2 for more detail. rienced more hardship in finding a new job after having been displaced from their old job (Hiroko, Liu, and Tamashita 2011). Policies to promote gender Finally, occupational segregation may equality in economic be related to persistent stereotypes about opportunity what are appropriate occupations for The analysis in this chapter has identified men and women. Stereotypes of women as several indicators of gender inequalities in the homemakers and men as breadwinners can productive sphere where challenges remain translate into perceptions about their skills and where policy may be able to reduce per- in the labor market and, therefore, the type sistent gender-based differences in economic of market work they do (Anker 1998). For opportunities and outcomes. A detailed dis- example, positive stereotypes of women— cussion of policy examples and their impacts having a caring nature, manual dexterity, will follow in chapter 6. skill at conducting household work, and greater honesty—could lead to the hypoth- Mitigating trade-offs between women’s esis that women would be better qualified household and market roles to work as nurses, teachers, clerks, and sales assistants, among other occupations (Anker Promoting gender equality in economic 1998). In qualitative work undertaken in opportunity requires policies and investments Fiji, Indonesia, Papua New Guinea, and that address the trade-offs between women’s Vietnam, gendered beliefs about appropri- household and market roles. Women often ate employment for men and women were face stark time trade-offs between house- remarkably consistent across these countries hold chores and market work, particularly in BOX 3.2  Gender-related beliefs on appropriate employment in Fiji, Papua New Guinea, Indonesia, and Vietnam Gender-related beliefs about appropriate employ- a woman with a knife in her hand.� Differing skill ment for men and women were remarkably consis- levels could be conflated with conceptions of heavy tent across East Asian and Pacific countries where and light work, for example, when one young male qualitative fieldwork was undertaken (Fiji, Papua respondent said that men do heavy work such as that New Guinea, and Vietnam), and among both men of an auto mechanic, while women do light work, and women, although some change in gender norms “such as accounting.� A focus group discussion with was observed among younger cohorts. Comments adult women in the urban National Capital District largely revolved around the concept of “heavy� of Papua New Guinea revealed further gender ste- (physically demanding) versus “light� work, with reotypes and fears held by these women about what the former ascribed to men and the latter to women. they considered to be gender-specific employment. In Vietnam, an adult man from Hanoi said that “In They said that secretaries should be women (so that general, men are better with heavy jobs like taxi they are not “aggressive to their bosses�), and trad- driver. Women are better with lighter jobs like sew- ing stores should be run by men as they have the ing, and selling things. . . . Men are better with big “ideas, mindset and business knowledge� and would business because it is hard work—men are more not be “targeted by criminals.� decisive than women. Job[s] like killing pigs, in prin- ciple women can do, but it would be strange to see Source: World Bank 2011a. G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    131 rural areas. In such contexts, programs and also lead by example. Even if women enter policies targeted at reducing women’s time more “male� occupations, stereotypes are on chores—for example, through investment likely to persist with respect to women as in infrastructure—are likely to increase their leaders and managers. The public sector is ability to engage in market-based, income- in a unique position to establish good prac- earning opportunities. Policies that support tice in this regard by encouraging women’s women in balancing their caregiving and professional advancement, either through market roles are also critical to strengthening direct measures such as targets or quotas, their access to economic opportunity. Access or through specialized training programs. In to affordable and accessible child care can be this context, the government of Malaysia put critical in this regard. Community child-care in place a system of quotas for female man- centers, particularly those targeted at low- agers in the public sector; the approach has income neighborhoods, have been found to recently been extended to private sector firms increase maternal employment in a number to encourage women to assume leadership of Latin American countries. roles. In Mexico, the government initiated a Parental and paternity leave can promote system of grants to firms to improve gender- greater parity between the sexes by facilitat- based employment issues in their workplace ing a more equitable division of child-rearing and also to improve the gender distribution responsibilities, thus allowing women to have in management. the same opportunities as men for advanc- ing their careers in the formal sector. Within the region, only Cambodia, Indonesia, and Eliminating resource constraints on the Philippines currently have provisions for female-led farms and enterprises paternity leave. Evidence from the OECD Despite progress, the existing evidence sug- on the take-up of paternity leave is mixed, gests that women continue to have less access however, suggesting that providing paternity to a range of productive resources than do leave alone is not sufficient to change the cur- men, as a function of their gender as opposed rent gender division of child-rearing respon- to their innate productive capabilities. Pub- sibilities within households. Rather, it needs lic policies may thus have an important to be combined with approaches to breaking role to play in promoting gender equality in down gender norms regarding household the control of productive inputs—whether caregiving roles. land, agricultural extension, technology, or financial capital. Policies aimed at promot- Breaking down gender silos in the labor ing equal access to assets, particularly land, market requires careful thought, since complex legal, social, and economic factors are at play and A critical element of breaking down gender the evidence base is thin. Levelling the legal silos in the labor market involves investing in playing field is usually a good start, but it is skills on the basis of productivity rather than even more important to work with the infor- on gender norms and perceptions of “appro- mal institutions and take into account norms priate� occupations. Beyond efforts to reduce and customary practices in order to remove gender streaming in education, discussed barriers in that domain. above, programs that encourage both women Improving women’s access to productive and men to think outside of gender stereo- assets can play an important role in rais- types in the job market will likely improve ing enterprise productivity, whether in the the allocation of talent toward jobs in a way farm or nonfarm sector, as in the following that enhances both equality of economic examples: opportunity and productivity. Breaking down social norms and percep- Several countries in the region have made •   tions is an area where the public sector can headway in recent years in increasing 1 3 2    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C o ­ wnership and control of land. In response Creating an enabling environment for to concerns about persistent ­ gender inequal- gender equality in employment ities in land, several countries—including Creating an enabling environment is a key Indonesia, Lao PDR, and Vietnam— component of efforts to promote gender recently adopted gender-sensitive reforms equality in economic opportunity in the in land titling. Because the reasons for long term in East Asia and the Pacific. An women’s lower access to land differ across important starting point for promoting equal the region—from unfavorable legal frame- opportunity in employment is to ensure that works to cultural norms and practices that women and men face a level legal playing deem land a male asset—effective policies field with respect to jobs and sectors. Labor to increase female landholdings need to regulations that result in asymmetries in account for context-specific constraints in the costs of hiring male and female workers developing context-specific solutions. can be found in countries across the region. I ncreasing women’s access to information •   Ostensibly protective legislation—in the form and training, extension services, and other of restrictions on women working at night, productive inputs can also play a key role working overtime, and working in so-called in enhancing the productivity of female- dangerous sectors—in practice inhibits wom- led enterprises, both within and outside en’s economic participation. Priority should agriculture. be given to reducing labor market restrictions A lthough evidence on access to finance in •   that limit women’s employment options. East Asian and Pacific countries is mixed, Where the original concerns motivating these women do face particular challenges in policies continue to be valid—for example, accessing credit, given their weaker access health and safety concerns—measures should to land, which is an important source of be taken to ensure that these concerns are collateral. addressed more directly and for both men Where evidence is thin, greater resources and women, whether through workplace should be invested in uncovering the greatest safety codes or through the provision of safe constraints. For example, investments could and reliable transport infrastructure. target the collection of information, by gen- Although formal sector employment is still der, on access to inputs in the agricultural small as a share of total employment in most sector in the region, as well as to inputs and East Asian and Pacific countries, an impor- productivity in the informal sector. tant role for public policy is to strengthen As in the case of education and health, the enabling environment for gender equal- broader systemic weaknesses—whether in the ity in economic opportunity. Active labor form of cumbersome registration procedures, market policies are one means of overcom- weak systems of financial intermediation, or ing gender differences in access for formal lack of electricity—affect both female- and employment. For example, wage subsidies male-led enterprises. Evidence suggests that may allow individuals, albeit temporarily, to such constraints may be more onerous among signal their abilities to future employers and small and informal firms than among larger make it cheaper for employers to hire female firms and, as such, may constrain female-led workers whom they may not otherwise have enterprises disproportionately. Thus, inter- considered. This approach provides the ventions that focus on improving the overall opportunity to reduce stereotypes through investment climate and, in particular, pro- directly observing their skills, and also gives moting small business development will be a women valuable labor market experience. critical part of a strategy to promote gender Skills training programs may enable women equality in access to economic opportunity. and men to move into professions outside of In sum, strategies to promote gender equality gender silos, particularly when paired with in economic opportunity should address sys- apprenticeship opportunities. Although evi- temic as well as gender-specific constraints. dence on the impact of active labor market G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    133 policies in East Asia and the Pacific is limited, differences in human capital and experience, studies from Latin America and the Middle occupational and sectoral selection, underly- East suggest that well-designed active labor ing ability, selection into the labor market, and market policies can help to improve women’s discrimination. 5. Household surveys in Thailand and Vietnam employment outcomes. corroborate this. In urban areas, women Affirmative action policies have also been are approximately 10 percentage points less used to overcome gender-specific barriers to likely to receive benefits than men in Vietnam, employment, whether those barriers are due whereas they are 3 percentage points less to implicit or overt discrimination in hir- likely to do so in Thailand. In the postreform ing and promotion. Although the literature period in China, a growing number of women continues to debate the benefits and costs of and urban workers have been pushed into affirmative action, the evidence (largely from temporary, part-time, insecure, or low-paying high-income countries) suggests that carefully work in the informal sector (Yuan and Cook designed policies can help break down bar- 2010). riers to female employment with few or no 6. Notably, an individual is defined as working in the informal sector if he or she works in adverse effects on firm productivity (World agriculture or is self-employed, is working in Bank 2011g). Affirmative action hiring and the household enterprise, or is working as an promotion in the public sector can also have unpaid family worker. However, an individual important demonstration effects. For exam- is classified as working in the formal sector if ple, in 2004, the government of Malaysia he or she works as a legislator or manager, introduced a public sector gender quota of professional, technician or associate profes- 30 percent female representation across all sional, or plant machine operator or assem- decision-making levels, including positions bler, or is in the armed forces. from department head to secretary general 7. Elementary occupations consist of simple (ASEAN 2008). and routine tasks that mainly require the use of hand-held tools and often some physi- cal effort. For a more detailed explanation, please refer to the International Standard Notes Classification of Occupations (ISCO) by the 1. Using cross-country data for selected Asian International Labour Organization. economies, Meng (1996) found no significant 8. Care should be taken when interpreting the relationship between economic development data from Samoa, Timor-Leste, Tonga, and and the relative earnings of men and women. Vanuatu because of very small sample sizes. In fact, gender inequality in earnings within 9. Evidence from across the world suggests that the East Asia and Pacific region was worse firms with greater female representation in in high-income countries such as Japan and management display lower levels of gender the Republic of Korea than in the low- and inequalities, including wage gaps and inequal- middle-income countries. ities within firms (Cohen and Huffman 2007; 2. Unfortunately, data constraints prevent look- Graves and Powell 1995; Huffman, Cohen, ing at birth age-cohort patterns within the and Pearlman 2010). majority of countries in the region. 10. Occupational and industrial segregation by 3. Gender wage gaps do not capture earnings dif- gender is detrimental for labor market effi- ferences among all men and women of work- ciency and welfare for four principal reasons ing age. First, they miss a large fraction of the (Anker 1998). First, men and women are not workforce, notably those in unpaid work or working in occupations or industrial sec- self-employed workers. Second, since there tors to which they are best suited and most may be differential selection between males productive but are rather choosing their work and females into the labor force—and into based on other factors. This trend reduces wage employment rather than entrepreneur- overall incomes and aggregate productivity. ship and agriculture—gender wage gaps are Second, gender-based segregation increases also likely to reflect these selection decisions. labor market rigidity and reduces the abil- 4. Aggregate relative wage data should, however, ity of labor markets to respond to economic be treated with caution since it confounds reforms and labor market shocks, such as those 1 3 4    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C related to globalization. Third, segregation to explain gender earnings gaps, as well as reinforces and perpetuates negative gender other unobserved characteristics that may stereotypes, consequently reducing women’s vary between males and females, such as status, income, education, and skills. Finally, the intensity of work conducted, workplace segregation of the current generation has neg- characteristics, and unobserved measures of ative effects on future generations. ability. Furthermore, if the occupation choice 11. In Indonesia, controlling for sector of operation or tenure trajectories within a firm reflect the and firm size accounts for 17 percent of differ- impact of discrimination, then measuring dis- ences in profits in rural areas, and 50 percent in crimination as the unexplained component of urban areas (Badiani and Posades 2011), and wage gaps will underestimate its true extent. industry accounts for 9 to 14 percent of earn- 16. In the OECD, Blau and Kahn (2003) find ings among self-employed individuals in the that 38 percent of the total gender wage gap United States (Hundley 2001). is unexplained by differences in occupational 12. Several hypotheses suggest why female labor and sectoral choice, education, and experi- force participation first falls before rising with ence. In addition, women have less labor economic development. Boserup (1970) sug- market experience—in OECD countries as gests that men’s greater access to education well as developing countries (Goldin, Katz, and technologies implies that they displace and Kuziemko 2006)—in part because they women from the labor force during the early are more likely both to take time out from the stages of a country’s development. As devel- work force and to work part-time. opment continues and women gain more 17. The evidence suggests a similar pattern in the access to education and technologies, female United States. Bertrand, Goldin, and Katz labor force participation increases. Another (2010) find that female master of business well-established hypothesis for this phenom- administration (MBA) graduates earn less, even enon focuses on income and substitution if they choose similar professional paths as men. effects (Goldin 1995; Mammen and Paxson Bayard et al. (2003) find that a large portion of 2000). As development occurs, households’ the gender wage gap in the United States can unearned incomes rise, reducing the incen- be explained by pay differences between males tive of women to work outside the home. and females within narrowly defined occupa- The negative impact of rising incomes on tions and within ­establishments. women’s labor force participation is termed 18. The gender wage gap is decomposed at differ- the “income effect,� since greater household ent points in the earnings distribution using income implies that households are able to the decomposition method based on uncondi- afford more female leisure time. The substitu- tional quantile regression as outlined in Firpo, tion effect works in the opposite direction—as Fortin, and Lemieux (2009). In this method, female wages rise, more women are enticed to the estimates from the unconditional quantile enter the workforce (Goldin 1995; Mammen regression constitute average partial effects and Paxson 2000). of a small location shift of an independent 13. In the East Asia and Pacific region, evidence variable on the unconditional quantile of the of declining female labor force participation dependent variable. as incomes rise has been found in Thailand 19. Evidence from across the world suggests that (Mammen and Paxson 2000). marriage and childbearing have a large impact 14. In the OECD, a number of studies find that on the gender wage gap. In the United States, childcare costs negatively impact rates of marital status and young children account for female labor force participation and that the approximately half the gender wage gap faced provision of subsidized child care raises par- by young women. A large component of the ticipation (Anderson and Levine 1999; Blau gender gap in earnings is attributable to women and Currie 2006; Chevalier and Viitanen 2002; having more career interruptions and shorter Del Boca 2002; Gelbach 2002; Gustafsson work hours, including more work in part-time and Stafford 1992). positions and self-employment (Becker 1981; 15. For example, even if differences in average Bertrand, Goldin, and Katz 2010; Korenman human capital have been considered, the and Neumark 1992; Mincer and Polachek residual is likely to still contain differences 1974; Sasser 2005; Wood, Corcoran, and between the composition and quality of edu- Courant 1993). These estimates may, how- cation of males and females that may help ever, underestimate the effect of parenting and G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    135 gender divisions of household labor on wage that women are more risk averse than men gaps, since the demands placed on women (Eckel and Grossman 2008). at home can influence education, career, and 27. Women tended to consider men’s “over engage- work choices (O’Neill 2003). ment� in household chores as “unmanly� and 20. Indeed, household survey data in Vietnam, “non ambitious.� Women often explained the which also cover microenterprises, suggest fact that they carried out most domestic work that female-run household enterprises were (while having a full-time job) as fulfilling their more likely to survive compared to male- obligation to care for their families. In addition, run enterprises between 1993 and 1998 the view was expressed that women with extreme (Vijverberg and Haughton 2004). The higher career ambitions—who did not assume much rates of female-run firms’ survival are linked household responsibilities—would be criticized to women’s predominance in certain sectors. as “selfish� or “non feminine.� Interestingly, 21. Evidence from the enterprise surveys conducted wives with higher income and “occupational in 2009 in Lao PDR suggests that female-run prestige� than their husbands often retained the firms consistently report business environment primary responsibility for taking care of domes- constraints to be less severe than male-run tic work. This was seen as a “counter balance� firms (Davies and Record 2010). Similarly, to their violation of cultural values about who in rural Indonesia, there are no differences in should be the primary b ­ readwinner. the severity of reported business constraints between female- and male-run household References enterprises (Costa and Rijkers 2011). 22. In Indonesia, start-up capital is 3.4 times Adair, Linda, David Guilkey, Eilene Bisgrove, smaller in female-managed nano- and micro­ and Socorro Gultiano. 2002. “Effect of enterprises than in male-managed enterprises Childbearing on Filipino Women’s Work (Badiani and Posades 2011). Hours and Earnings.� Journal of Population 23. Data on female-managed or female-owned Economics 15 (4): 625–45. plots and on female farmers are extremely Albrecht, James, Anders Björklund, and Susan limited in the world and, in particular, in the Vroman. 2003. “Is There a Glass Ceiling in East Asia and Pacific region. Therefore, wher- Sweden?� Journal of Labor Economics 21 (1): ever necessary, this report examines differences 145–77. between farms run by female-headed house- Amin, Mohammad. 2010. “Gender and Firm-Size: holds and those run by male-headed house- Evidence from Africa.� Economics Bulletin 30 holds. Because female-headed households are (1): 663–68. different from male-headed households in a Anderson, Patricia M., and Phillip B. Levine. number of respects, this comparison is likely 1999. “Child Care and Mothers’ Employment to exaggerate gender differences in agricultural Decisions.� NBER Working Paper 7058, productivity and access to inputs because it will National Bureau of Economic Research, confound differences in socioeconomic charac- Cambridge, MA. teristics with differences in p ­ roductivity. Anker, Richard. 1998. Gender and Jobs: Sex 24. In Latin America, women constitute between Segregation of Occupations in the World . 13 and 27 percent of landowners in Mexico, Geneva: International Labour Organization. Nicaragua, Paraguay, and Peru and own Antecol, Heather. 2000. “An Examination of smaller plots of land than men in all of these Cross-Country Differences in the Gender Gap countries (Deere and Doss 2006). in Labor Force Participation Rates.� Labour 25. It should be noted that the fraction of farmers Economics 7: 409–26. overall receiving agricultural extension ser- Arulampalam, Wiji, Alison L. Booth, and Mark vices is very low. L. Bryan. 2007. “Is There a Glass Ceiling over 26. Gender differences in “soft skills,� which are Europe? Exploring the Gender Pay Gap across acquired within and outside of school, also the Wage Distribution.� Industrial and Labor contribute to gender differences in occupa- Relations Review 60 (2): 163–86. tions. Personality traits have wage returns A SE A N (A ssociation of S out heast A sian that are both occupation and gender specific Nations). 20 08. “ASE A N Continues to (Cobb-Clark and Tan 2011). Women have E mp ower Women .� A S E A N B u ll e t i n . been found to be employed in safer or lower- http://www.aseansec.org / Bulletin-Feb- 08 risk jobs, which is consistent with evidence .htm#Article-2. 1 3 6    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C Aterido, Reyes, and Mary Hallward-Driemeier. Bertrand, Marianne, Claudia Goldin, and Lawrence 2009. “Whose Business Is It Anyway?� Policy F. Katz. 2010. “Dynamics of the Gender Gap Research Working Paper 5571, World Bank, for Young Professionals in the Financial and Washington, DC. Corporate Sectors.� American Economic Babcock, Linda, and Sara Laschever. 2003. Journal: Applied Economics 2 (3): 228–55. Women Don’t Ask: Negotiation and the Bjerge, Benedikte, and J. Rand. 2011. “Gender Gender Divide . Princeton, NJ: Princeton Differences in the Vietnamese Business University Press. Sector: Evidence from the SME Survey.� Paper Badiani, Reena, and Josefina Posadas. 2011. commissioned for Toward Gender Equality in “Gender and Entrepreneurship in Household East Asia and the Pacific: A Companion to the Firms in Indonesia: Characteristics and World Development Report. Washington, DC: Performance.� Paper commissioned for Toward World Bank. Gender Equality in East Asia and the Pacific: Blackburn, Susan. 2001. “Gender Relations in A Companion to the World Development Indonesia: What Women Want.� In Indonesia Report. Washington, DC: World Bank. Today, Challenges of History, edited by Baraka, Jessica. 1999. “Does Type of Degree Grayson Lloyd and Shannon Smith, 270–82. Explain Taiwan’s Gender Gap?� Research Pasir Panjang, Singapore: Institute of Southeast Program in Development Studies, Princeton Asian Studies. University, Princeton, NJ. ———. 2004. Women and the State in Modern Bardasi Elena, Mark Blackden, and Juan Carlos Indonesia . Cambridge, U.K.: Cambridge Guzman. 2007. “Gender, Entrepreneurship, University Press, Cambridge. and Competitiveness in Africa.� Africa Blackden, C . Mark, and Quentin Wodon, Competitiveness Report 20 07. Geneva, eds. 2006. Gender, Time Use, and Poverty Switzerland: World Economic Forum. in Sub-Saharan Africa. Washington, DC: Bayard, Kimberly, Judith Hellerstein, David World Bank. Neu mark, and Ken neth Troske. 20 03. Blau, David, and Janet Currie. 2006. “Preschool, “New Evidence on Sex Segregation and Sex Day Care, and Afterschool Care: Who’s Differences in Wages from Matched Employee- Minding the Kids?� In Handbook of the Employer Data.� Journal of Labor Economics. Economics of Education, Volume 2, edited by 21 (4): 887–922. Eric A. Hanushek and Finis Welch. Amsterdam: B e a rd , V i c t o r i a . A . 2 0 05. “ I n d i v i du a l North-Holland. Determinants of Participation in Community Blau, Francine, and Lawrence M. Kahn. 2003. Development in Indonesia.� Environment “Understanding International Differences and Planning C: Government and Policy 23 in the Gender Pay Gap.� Journal of Labor (1): 21–39. Economics 21 (1): 106–44. Beck, Thorsten., Aslı Demirgüç-Kunt, Luc Bloom, David, David Canning, Günther Fink, and Laeven, and Vojislav Maksimovic. 2006. “The Jocelyn Finley. 2009. “Fertility, Female Labor Determinants of Financing Obstacles.� Journal Force Participation and the Demographic of International Money and Finance 25 (6): Dividend.� Journal of Economic Growth 14: 932–52. 79–101. Becker, Gary S. 1965. “A Theory of the Allocation Bosch, Mariano, and William Maloney. 2011. of Time.� Economic Journ al 75 (299): “Sectoral Choice and Family Formation: 493–517. Evidence from Labor Market Transitions ———. 1981. A Treatise on the Family. Cambridge, in Ghana, Mexico, Serbia, and Thailand.� MA: Harvard University Press. Background paper for the World Development Bender, Keith, Susan Donohue, and John Report 2012 and commissioned for Toward Heywood. 2005. “Job Satisfaction and Gender Gender Equality in East Asia and the Pacific: Segregation.� Oxford Economic Papers 57 (3): A Companion to the World Development 479–96. Report. Washington, DC: World Bank. Berniell, M. I., and Carolina Sanchez-Páramo. Boserup, Ester. 1970. Woman’s Role in Economic 2011. “Closing the Access Gap: Recent Development. London: Earthscan. Advances in Female Labor Force Participation.� BPS (Badan Pusat Statistik). Indonesia National Background paper for the World Development Labor Force Survey (SAKERNAS). Jakarta, Report 2012, World Bank, Washington, DC. Indonesia. G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    137 ———.BPS (Badan Pusat Statistik). Indonesia Cuberes, D., and Marc Teignier-Baqué. 2011. National Socioeconomic Survey (SUSENAS). “Does Gender Inequality Hinder Growth? Jakarta, Indonesia. http://dds.bps.go.id/. The Evidence and Its Policy Implications.� Bruhn, Miriam 2009. “Female-Owned Firms Background paper for the World Development in Latin America. Characteristics, Perfor- Report 2012, World Bank, Washington, DC. mance, and Obstacles to Growth.� Policy CWDI (Corporate Women Directors International). Research Working Paper 5122, World Bank, 2010. “CWDI/IFC 2010 Report: Accelerating Washington, DC. Board Diversity.� CWDI and International Bruhn, Miriam, Dean Karlan, and Antoinette Finance Corporation, Washington, DC. Schoar. 2010. “What Capital Is Missing in Dasgupta, Sukti, and David Williams. 2010. Developing Countries?� American Economic “Women Facing the Economic Crisis—The Review: Papers & Proceedings 100 (2): 629–33. Garment Sector in Cambodia.� In Poverty Brush, Candida G. 1992. “Research on Women: and Sustainable Development in Asia: Impact Past Trends, a New Perspective and Future and Responses to the Global Economic Crisis, Directions.� Entrepreneurship Theory and edited by A. Bauer and M. Thant. Manila: Practice 16 (4): 5–30. Asian Development Bank. Burda, Michael, Daniel S. Hamermesh, and Davies, Simon, and Richard Record. 2010. Philippe Weil. 2007. “Total Work, Gender “Background Paper for the Second Lao PDR and Social Norms.� NBER Discussion Paper Investment Climate Assessment: Gender and 2705, National Bureau of Economic Research, Entrepreneurship in the Lao PDR. Evidence Cambridge, MA. from the 2009 Enterprise Survey.� World CEIC Data. New York. http://www.ceicdata. Bank., Washington, DC. com/. de Brauw, Alan, Jikun Huang, Linxiu Zhang, C h aud hu r i , S a nju k t a. 2 0 09. “E conom ic and Scott Rozelle. 2011. “The Feminization Development and Women’s Empowerment.� of Agriculture with Chinese Characteristics.� Working paper, University of Wisconsin–Eau Background paper for the World Development Claire. Report 2012, World Bank, Washington, DC. Chen, Feinian. 2005. “Employment Transitions Deere, Carmen Diana, and Cheryl Doss. 2006. and the Household Division of Labor in “The Gender Asset Gap: What Do We Know China.� Social Forces 84 (2): 831–51. and Why Does It Matter?� Feminist Economics Chevalier, Arnaud, and Tarja Viitanen. 2002. 12 (1–2): 1–50. “The Causality between Female Labour Force Deere, Carmen Diana, and Magdalena Leon. Participation and the Supply of Childcare.� 2003. “The Gender Asset Gap: Land in Applied Economics Letters 9: 915–18. Latin America.� World Development 31 (6): Chi, Wei, and Bo Li. 2007. “Glass Ceiling or 925–47. Sticky Floor? Examining the Gender Pay Gap de la Rica, Sara, Juan J. Dolado, and Vanesa across the Wage Distribution in Urban China, Llorens. 2005. “Ceiling and Floors: Gender 1987–2004.� MPRA Paper 3544, University Wage Gaps by Education in Spain.� IZA Library of Munich, Germany. Discussion Paper 1483, Institute for the Study Cobb-Clark, Deborah A., and Michelle Tan. of Labor (IZA), Bonn, Germany. 2011. “Noncognitive Skills, Occupational Del Boca, Daniela. 2002. “The Effect of Child Care Attainment, and Relative Wages.� Labour and Part Time Opportunities on Participation Economics 18 (1): 1–13. and Fertility Decisions in Italy.� Journal of Cohen, Philip, and Matt Huffman. 2007. Population Economics 15 (3): 549–73. “Working for the Woman? Female Managers de Mel, Suresh, David McKenzie, and Chris and the Gender Wage Gap.� Americ an Woodruff. 2009a. “Are Women More Credit Sociological Review 72 (5): 681–704. Constrained? Experimental Evidence on Gender Costa, Rita, and Bob Rijkers. 2011. “Gender and Microenterprise Returns.� American and Rural Non-Farm Entrepreneurship.� Economic Journal: Applied Economics 1 (3): Background paper for the World Development 1–32. Report 2012, World Bank, Washington, DC. ———. 2009b. “Measuring Microenterprise Croson, Rachel, and Uri Gneezy. 2009. “Gender Profits: Must We Ask How the Sausage Is Differences in Preferences.� Journ al of Made?�Journal of Development Economics Economic Literature 47(2): 448–74. 88 (1): 19–31. 1 3 8    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C Dhanani, Shafiq, and Iyanatul Islam. 2004. Fatima, Ambreen, and Humera Sultana. 2009. Indonesian Wage Structure and Trends, “Tracing Out the U-Shape Relationship 1976 –2000. Geneva: International Labour between Female Labor Force Participation Rate Organization. and Economic Development for Pakistan.� Du, Fenglian, and Xiao-Yuan Dong. 2010. International Journal of Social Economics 36 “Women’s Labor Force Participation and (1/2): 182–98. Childcare Choices in Urban China During Fernández, Raquel. 2010. “Does Culture Matter?� the Economic Transition.� World Paper 2010- NBER Working Paper 16277, National Bureau 04, University of Winnipeg, Department of of Economic Research, Cambridge, MA. Economics, Manitoba. Fernández, Raquel, and Alessandra Fogli. 2005. Eckel, Catherine C., and Philip J. Grossman. Fertility: The Role of Culture and Family 2008. “Men, Women and Risk Aversion: Experience. NBER Working Paper 11569, Experimental Evidence.� In Handbook of National Bureau of Economic Research, Experimental Economic Results, edited by Cambridge, MA. C. Plott and V. Smith. Amsterdam: Elsevier. Fernández, Raquel, Alessandra Fogli, and Economist. 2011. “Asian Demography: The Flight Claudia Olivetti. 2004. “Mothers and Sons: from Marriage.� August 20. Preference Formation and Female Labor Force Ellis, Amanda, Daniel Kirkwood, and Dhruv Dynamics.� Quarterly Journal of Economics Malhotra. 2010. Economic Opportunities for 119 (4): 1249–99. Women in the East Asia and Pacific Region. Field, Erica, Seema Jayachandran, and Rohini Directions in Development Series. Washington, Pande. 2010. “Do Traditional Institutions DC: World Bank. Constrain Female Entrepreneurship? A Field E lson , Dia ne , a nd Rut h Pe a rson. 1981. Experiment on Business Training in India.� “The Subordination of Women and the American Economic Review Papers and Internationalisation of Factory Production.� Proceedings 100 (2): 125–29. In Of Marriage and the Market: Women’s Firpo, S., N. M. Fortin, and T. Lemieux. 2009. Subordination in International Perspective, “Unconditional Quantile Regressions.� edited by Kate Young, Carol Wolkowitz, and Econometrica 77 (3): 953–73. Roslyn McCullagh. London: CSE Books. Fuwa, Makiko. 2004. “Macro Level Gender Enterprise Surveys (database). World Bank, Inequality and the Division of Household Washington, DC. http://data.worldbank.org/ Labor in 22 Countries.� American Sociological data-catalog/enterprise-surveys Review 69 (6): 751–67. Entwisle, Barbara, and Gail E. Henderson, Gallaway, Julie H., and Alexandra Bernasek. eds. 2000. Re-Drawing Boundaries: Work, 2004. “Literacy and Women’s Empowerment Household, and Gender in China. Berkeley: in Indonesia: Implications for Policy.� Journal University of California Press. of Economic Issues 38 (2): 519–25. Esteve-Volart, Berta. 2004. “Gender Discrimination Galor, Oded, and David N. Weil. 1996. “The and Growth: Theory and Evidence from India.� Gender Gap, Fertility, and Growth.� American Development Economics Discussion Paper Economic Review 86: 374–87. Series 42, Suntory and Toyota International Ganguly, A nuradha, and Talwar Swapan. Centres for Economics and Related Disciplines, 2003. “Feminization of India’s Agricultural London School of Economics and Political Work force.� L abour E duc ation (2 –3): Science, London. 29–33. Fang, Hanming, and Michael P. Keane. 2004. Gelbach, Jonah B. 2002. “Public Schooling for “Assessing the Impact of Welfare Reform Young Children and Maternal Labor Supply.� on Single Mothers.� Brookings Papers on American Economic Review 92 (1): 307–22. Economic Activity 35 (1): 1–116. Gender Statistics (database). World DataBank, FAO (Food and Agricultural Organization)/Sida World Bank, Washington, DC. http://data Partnership Coooperation. 2010. National .worldbank.org/data-catalog/gender-statistics. Gender Profile of Agricultural Households, Giles, John, and Firman Witoelar Kartaadipoetra. 2010: Report based on the Lao Expenditure and 2011. “Education, Segregation and Labor Consumption Surveys, National Agricultural Markets in Indonesia.� Background paper for Census and the National Population Census. the World Development Report 2012, World Vientiane, Lao PDR: FAO. Bank, Washington, DC. G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    139 Giles, John, Dewen Wang, and Wei Cai. Hausmann, Ricardo, Laura D. Tyson, and Saadia 2011. “The Labor Supply and Retirement Zahidi, eds. 2010. The Global Gender Gap Behavior of China’s Older Workers and Report 2010 . Geneva: World Economic Elderly in Comparative Perspective.� Policy Forum. Research Working Paper 5853, World Bank, Hertz, Thomas, Anna P. de la O Campos, Esteban Washington, DC. J. Quiñones, Benjamin Davis, Carlo Azzari, Giné, Xavier, and Ghazala Mansuri. 2011. “Money and Alberto Zezza. 2009. “Wage Inequality in or Ideas? A Field Experiment on Constraints International Perspective: Effects of Location, to Entrepreneurship in Rural Pakistan.� Sector, and Gender.� Paper presented at the Photocopy, Development Economics Research FAO-IFAD-ILO Workshop on “Gaps, Trends Group, World Bank, Washington, DC. and Current Research in Gender Dimensions Godquin, Marie, and Agnes Quisumbing. of Agricultural and Rural Employment: 2008. “Separate but Equal? The Gendered Differentiated Pathways out of Poverty,� Rome, Nature of Social Capital in Rural Philippine March 31–April 2, 2009. Communities.� Journal of International Hill, Anne M. 1996. “Women in the Japanese Development 20 (1): 13–33. Economy.� In Women and Industrialization G old i n , C l aud i a. 1995. “T he U - S h ap e d in Asia , edited by Susan Horton. London: Female Labor Force Function in Economic Routledge. Development and Economic History.� In Hiroko, Takeda, Jieyu Liu, and Junko Tamashita. Investment in Women’s Human Capital, 2011. “Economic Restructuring and Changing edited by T. Paul Schultz. Chicago: University Work/Family Life: The Cases of Japan and of Chicago Press. China.� In New Welfare States in East Asia: Goldin, Claudia, Lawrence F. Katz, and Ilyana Global Challenges and Restructuring, edited Kuziemko. 2006. “The Homecoming of by Gyu-Jin Hwang. Cheltenham, U.K.: Edward American College Women: The Reversal of the Elgar. Gender Gap in College.� Journal of Economic Honig, Emily. 2000. “Iron Girls Revisited: Perspectives 20 (Fall): 133–56.  Gender and the Politics of Work in the Cultural Goldstein, Markus, and Christopher Udry. Revolution.� In Re-Drawing the Boundaries 2008. “The Profits of Power: Land Rights and of Work, Households, and Gender, edited Agricultural Investment in Ghana.� Journal of by Barbara Gutwisle and Gail Henderson. Political Economy 116 (6): 981–1022. Berkeley: University of California Press. Graves, Laura, and Gary Powell. 1995. “The Horton, Susan. 1996. “Women and Industrial- Effect of Sex Similarity on Recruiters’ ization in Asia: Overview.� In Women and Evaluations of Actual Applicants: A Test of the Industrialization in Asia, edited by in Susan Similarity-Attraction Paradigm.� Personnel Horton. London: Routledge. Psychology 48 (1): 85–98. Huffman, Matt L., Philip Cohen, and Jessica Greenwood, Jeremy, Ananth Seshadri, and Pearlman. 2010. “Engendering Change: Mehmet Yorukoglu. 2005. “Engines of Organizational Dynamics and Workplace Liberation.� Review of Economic Studies G e n d e r S e g r e g a t i o n , 1 9 75 – 2 0 0 5 .� 72 (1): 109–33. Administrative Science Quarterly 55 (2): GSO (General Statistics Office) [Vietnam]. 255–77. Vietnam Household Living Standards Surveys. Hughes, James, and Margaret Maurer-Fazio. Hanoi, Vietnam. http://www.gso.gov.vn. 2002. “The Effects of Market Liberalization ———. Vietnam Labor Force Survey. Hanoi, on the Relative Earnings of Chinese Women.� Vietnam. http://www.gso.gov.vn. Journal of Comparative Economics 30 (4): Gustafsson, Siv, and Frank Stafford. 1992. 709–31. “Child Care Subsidies and Labor Supply in Hughes, James, Margaret Maurer-Fazio, and Sweden.� Journal of Human Resources 27 (1): Dandan Zhang. 2005. “Economic Reform and 204–30. Changing Patterns of Labor Force Participation Hallward-Driemeier, Mary. 2006. “Improving in Urban and Rural China.� William Davidson the Legal Investment Climate for Women in Institute Working Paper 787, University of Sub- Saharan Africa.� Paper prepared for the Michigan, Ann Arbor. NEPAD – OECD Conference, Brazzaville, ———. 2007. “An Ocean Formed from One December 12, 2006. Hundred Rivers: The Effects of Ethnicity, 1 4 0    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C Gender, Marriage, and Location on Labor Conference World Bank–KERIS High-Level Force Participation in Urban China.� Feminist Seminar on ICT in Education. Economics 13 (3–4): 125–53. Kim, Elim. 2005. “Korean Women’s Activities for Hundley, Greg. 2001. “Why Women Earn Less Legislation to Guarantee Gender Equality in than Men in Self-Employment.� Journal of Employment. A Symposium: Legal Feminism in Labor Research 22 (4): 817–29. Korea.� Journal of Korean Law 5 (2): 49–65. IFC (International Finance Corporation). 2010. Kim, Sung won, Vanessa Fong, Hirokazu “Economic Opportunities for Women in the Yoshikawa, Niobe Way, Xinyin Chen, Huihua Pacific.� International Financial Corporation, Deng, and Zuhong Lu. 2010. “Income, Work Washington, DC. Preferences and Gender Roles among Parents ———. 2011. Women’s Entrepreneurship in the of Infants in Urban China: A Mixed Method Phillippines, Vol. 1, Washington, DC: IFC. Study from Nanjing.� China Quarterly 204: IFC (International Finance Corporation) and 939–59. MPDF (Mekong Private Sector Development Klapper, Leora F., and Simon C. Parker. 2010. Facility). 2006. “Women Business Owners “Gender and the Business Environment for in Vietnam: A National Survey.� IFC Gender New Firm Creation.� World Bank Research Entrepreneurship Markets Initiative and Observer 26 (2): 237–57. MPDF, Washington, DC. Klasen, Stephan, and Francesca Lamanna. 2009. I FC (I nternational Finance Corporation)/ “The Impact of Gender Inequality in Education NORC Indonesia. 2010. “Indonesia Small and Employment on Economic Growth: New and Medium Enterprise Banking Survey.� Evidence for a Panel of Countries.� Feminist I nt e r n at ion a l Fi n a n c i a l C or p orat ion , Economics 15 (3): 91–132. Washington DC; NORC, Chicago. http:// Korenman, Sanders, and David Neumark. 1992. w w w.norc.org / Research / Projects/ Pages/ “Marriage, Motherhood, and Wages.� Journal ifc-indonesia-small-and-medium-enterprise- of Human Resources 27 (2): 233–55. banking-survey.aspx. Kushnir, Khrystyna, Melina Laura Mirmulstein, ILO (International Labour Organization). 2010. and Rita Ramalho. 2010. “Micro, Small “Women in Labor Markets: Measuring and Medium Enterprises Around the World: Progress and Identifying Challenges.� ILO, How Many Are There, and What Affects the Geneva. Count?� MSME Country Indicators, World ILO Key Indicators of the Labour Market (KILM) Bank/IFC, Washington, DC. database. International Labour Office. Korea LABORSTA Internet (database). International Labor Force Survey and Population Census. Labour Organization, Geneva, Switzerland. http://kilm.ilo.org/kilmnet/. http://laborsta.ilo.org/. Indonesia Family Life Survey. 2007/2008. RAND L a m , A lice. 1992 . “T he Japa nese E qual Family Life Surveys, IFLS-4. http://www.rand Employment Opportunity Law: Its Effects .org/labor/FLS/IFLS/ifls4.html. on Personnel Management and Policies Juhn, Chinhui, and Simon Potter. 2006. “Changes and Women’s Attitudes.� MPRA Working in Labor Force Participation in the United Paper 11559, University Library of Munich, States.� Journal of Economic Perspectives 20 Germany. (3): 27–46. Lee, Bun Song, Soomyung Jang, and Jayanta Sarkar. Juhn, Chinhui, and Manuelita Ureta. 2003. 2008. “Women’s Labor Force Participation “E mploy ment of M a rried Women and and Marriage: The Case of Korea.� Journal of E conomic Development: Evidence from Asian Economics 19 (2): 138–54. Latin American Countries.� Society of Labor Li, Shi, and Jin Song. 2011. “Changes in Economists Meeting, Toronto. Gender Wage Gap in Urban China During Karlan, Dean, and Martin Valdivia. 2011. 1996–2007.� CIBC Working Paper 2011-20, “Teaching Entrepreneurship: Impact of University of Western Ontario, Department of Business Training on Microfinance Clients Economics, London, Ontario. and Institutions.� Review of Economics and Liu, Dongxiao, and Elizabeth Heger Boyle. Statistics 93 (2): 510–27. 2001. “Making the Case: The Women’s Keitel, Robert S. 2009. “Innovation in Borderless Convention and Gender Discrimination in Distance Learning of English.� Paper presented Japan.� International Journal of Comparative at the 13th UNESCO-APEID International Sociology 42: 389–404. G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    141 LSB (Lao Statistics Bureau). Lao Expenditure World.� Background paper for the World and Consumption Survey. Vientane, Lao Development Report 2012, World Bank, PDR. http://w w w.nsc.gov.la /index.php? Washington, DC. option=com_content&view=article&id=50&I NSCB (National Statistical Coordination Board) temid=73&lang=. [Philippines]. 2006. Family Income and Mammen, Kristin, and Christina Paxson. 2000. Expenditures Survey (FIES) 2006. National “Women’s Work and Economic Development.� Statistical Coordination Board, Makati City, Journal of Economic Perspectives 14 (4): Philippines. http://www.nscb.gov.ph/fies/ 141–64. default.asp. Maurer-Fazio, Margaret, Rachel Connelly, Lan NSD (National Statistics Directorate) [Timor- Chen, and Lixin Tang. 2011. “Childcare, Leste]. 2007. Timor-Leste Survey of Living Eldercare, and Labor Force Participation Standards. Dili, Timor-Leste: Ministry of of Married Women in Urban China, 1982– Planning and Finance, National Statistics 2000.� Journal of Human Resources 46 (2): Directorate. http://dne.mof.gov.tl/ TLSLS/ 261–94. AboutTLSLS/index.htm. Meng, Xin 1996. “The Economic Position of NSO Mongolia (National Statistical Office of Women in Asia� Asian-Pacific Economic Mongolia). Living Standards Measurement Literature 10 (1): 23–41. Survey (LSMS). National Statistical Office, Miller, Robbie Louise. 1998. “Women’s Job Ulaanbaatar, Mongolia. http://www.nso.mn. Hunting in the ‘Ice Age’: Frozen Opportunities NSO Thailand (National Statistical Office in Japan.� Wisconsin Journal of Law, Gender, Thailand). Thailand Socio-Economic Survey and Society 13 (2): 228. (SES). Bangkok, Thailand: National Statistical ———. 2003. “The Quiet Revolution: Japanese Office. http://web.nso.go.th/tnso.htm. Women Working Around the Law.� Harvard Ogawa, Keiichi, and Masuma Akter. 2007. Journal of Law and Gender 26: 163–215. “Female L abor Force Pa r t icipat ion i n Mincer, Jacob, and Solomon Polachek. 1974. I ndone si a .� Jo u r n a l of I n t e r n a t i o n a l “Family Investment in Human Capital: Cooperation Studies 14 (3): 71–108. Earnings of Women.� Journal of Political O’Neill, June. 2003. “The Gender Gap in Wages, Economy 82 (2): S76–S108. Circa 2000.� American Economic Review Morrison, Andrew, Dhushyanth Raju, and Nistha May. Sinha. 2007. “Gender Equality, Poverty, and Pastore, Fransesco 2009. “The Gender Gap in Economic Growth.� Policy Research Working Early Career in Mongolia.� Discussion Paper Paper 4349. World Bank, Washington, DC. No. 4480, Institute for the Study of Labor Mu, Ren, and Dominique van de Walle. 2009. (IZA), Bonn, Germany. “Left Behind to Farm? Women’s Labor Penn World Table 7.0. Alan Heston, Robert Re-A llocation in Rural China.� Policy Summers and Bettina Aten, Penn World Research Working Paper 5107, World Bank, Table Version 7.0, Center for International Washington, DC. Comparisons of Production, Income and Prices Murthi, Mamta, Anne-Catherine Guio, and Jean at the University of Pennsylvania, May 2011. Drèze. 1995. “Mortality, Fertility, and Gender Peterman, Amber, Julie Behrman, and Agnes Bias in India: A District-Level Analysis.� Quisumbing. 2010. “A Review of Empirical Population and Development Review 21 (4): Evidence on Gender Differences in Nonland 745–82. Agricultural Inputs, Technology, and Service Nakavachara, Voraprapa. 2010. “Superior Female in Developing Countries.� IFPRI Discussion Education: Explaining the Gender Earnings Paper 975, International Food Policy Research Gap Trend in Thailand.� Journal of Asian Institute, Washington, DC. Economics 21 (2): 198–218. Pier re , G . 2011. “Recent L abor M a rket NIS (National Institute of Statistics) [Cambodia]. Performance in Vietnam through a Gender Cambodia Socio-Economic Survey (CSES). Lens.� Background paper for the “Vietnam National Institute of Statistics, Phnom Penh, Country Gender Assessment.� World Bank, Cambodia. http://www.nis.gov.kh/index.php/ Washington, DC. http://documents.worldbank social-statistics/cses. .org/curated/en/2011/11/15470188/vietnam- Ñopo, Hugo, Na nc y Da za, a nd Joha n na country-gender-assessment.Vietnam Gender Ramos. 2011. “Gender Earnings Gaps in the Assessment. 1 4 2    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C Pörtner, Claus C. 2009. “Children’s Time Simavi, Sevi, Clare Manuel, and Mark Blackden. Allocation, Heterogeneity and Simultaneous 2010. Gender Dimensions of Investment Decisions.� Working Paper UWEC-2009-15, Climate Reform: A Guide for Policy Makers University of Washington, Department of and Practitioners. Washington, DC: World Economics, Seattle. Bank. Psacharopoulos, George, and Zafiris Tzannatos. Sinha, J. N. 1967. “Dynamics of Female 1989. “Female Labor Force Participation: Participation in Economic Activity in a An International Perspective.� World Bank Developing Economy.� In Proceedings of the Research Observer 4 (2): 187–201. World Population Conference, Belgrade, 1965; Quisumbing, Agnes R. 1995. “Gender Differences Vol. 4. Migration, Urbanization, Economic in Agricultural Productivity.� FCND Discussion Development , 336–37. New York: United Paper 5, International Food Policy Research Nations. Institute, Washington, DC. Slootmaekers-Miranda, Veerle. 2011. “Cooking, — — —. 1998. “Women in Agricultural Systems.� Caring and Volunteering: Unpaid Work In Women in the Third World: An Ency- around the World.� Social, Employment and clopedia of Contemporary Issues, edited by Migration Working Papers, Organisation for N. Stromquist, 261–72. New York: Garland. Economic Co-operation and Development, Quisumbing, Agnes R., Jonna P. Estudillo, and Paris. Keijiro Otsuka. 2004. Land and Schooling: Stivens, Maila. 2006. “Family Values and Islamic Transferring Wealth Across Generations . Revival: Gender, Rights and the State Moral Baltimore, MD: Johns Hopkins University Project.� Women’s Studies International Press. Forum 29 (4): 354–67. Rand, John, and Finn Tarp. 2011. “Does Gender SUSI (Survei Usaha Terintegrasi). 2002. Integrated Influence the Provision of Fringe Benefits? Survey of Cottage and Small-Scale Firms. Evidence from Vietnamese SMEs.� Feminist Badan Pusat Statistik, Jakarta, Indonesia. Economics 17 (1): 59–87. Tam, Henry. 2011. “U-Shaped Female Labor Force Rawski, Thomas, and Robert W. Mead. 1998. Participation with Economic Development: “On the Trail of China’s Phantom Farmers.� Some Panel Data Evidence.� Economic Letters World Development 26 (5): 767–81. 110 (2): 140–42. Sabarwal, Shwetlena, and Katherine Terrell. 2008. Tansel, Aysit. 2001. “Economic Development and “Does Gender Matter for Firm Performance? Female Labor Force Participation in Turkey: Evidence from Eastern Europe and Central Time-Series Evidence and Cross-Province Asia.� Policy Research Working Paper 4705, Estimates.� ERC Working Paper 0105, Middle World Bank, Washington, DC. East Technical University, Economic Research Sabarwal, Shwetlena, Katherine Terrell, and Elena Center, Ankara, Turkey. Bardasi. 2011. “How Do Female Entrepreneurs NSO Thailand (National Statistical Office Perform? Evidence from Three Developing Thailand). Thailand Socio-Economic Survey Regions.� Small Business Economics 37 (4): (SES). Bangkok, Thailand. http://web.nso.go.th/ 417–41. tnso.htm. Sakellariou, Chris. 2011. “Determinants of the Thomas, Duncan. 1990. “Intra-household Gender Wage Gap and Female Labor Force Resource Allocation: An Inferential Approach.� Participation in EAP.� Paper commissioned Journal of Human Resources 25 (4): 635–64. for Toward Gender Equality in East Asia Thomas, Duncan, and John Strauss. 1997. “Health and the Pacific: A Companion to the World and Wages: Evidence on Men and Women in Development Report. Washington, DC: World Urban Brazil.� Journal of Econometrics 77 (1): Bank. 159–85. Sasaki, Masaru. 2002. “The Causal Effect of Timberlake, Sharon. 2005. “Social Capital Family Structure on Labor Force Participation and Gender in the Workplace.� Journal of among Japanese Married Women.� Journal of Management Development 24 (1): 34–44. Human Resources 37 (2): 429–40. Tzannatos, Zafris. 1999. “Women and Labor Sasser, Alicia C. 2005. “Gender Differences in Market Changes in the Global Economy: Physician Pay: Tradeoffs Between Career and Growth Helps, Inequalities Hurt and Public Family.� Journal of Human Resources 40 (2): Policy Matters.� World Development 27 (3): 477–504. 551–69. G E N D E R A N D E C O N O M I C O P P O R T U N I T Y    143 Udry, Christopher. 1996. “Gender, Agricultural Around the World (qualitative assessment Production, and the Theory of the Household.� dataset). World Bank, Washington, DC. http:// Journ al of Politic al Economy 104 (5): go.worldbank.org/CD8RN24BP0. 1010–46. ———. 2011b. “Female E nt repreneu rsh ip UNIFEM, World Bank, ADB, UNDP, and DFID/ in Mongolia.� Background paper for the UK (United Nations Development Fund for Mongolia Gender Assessment. World Bank, Women, World Bank, Asian Development Washington, DC. Ban k, United Nations Development —— —. 2011c. “Gender Disparities in Mongolian Programme, and Department for International Labor Markets and Policy Suggestions.� World Development of the United Kingdom). 2004. Bank, Washington, DC. “A Fair Share for Women: Cambodia Gender ———. 2011d. “Gender Equality and Health.� Assessment.� UNIFEM, WB, ADB, UNDP, I ndonesia Policy Brief 2 , World Bank, and DFID/UK, Phnom Penh, Cambodia. Washington, DC. VCCI (Vietnam Chamber of Commerce and ———. 2011e. “Vietnam Countr y G ender Industry). 2006. “Targeted Policies That Assessment.� World Bank, Washington, DC. Support Women’s Entrepreneurship Can Boost —— —. 2011f. “Women, Business and the Law in Vietnam’s Economic Growth.� Business Issues East Asia and Pacific Region.� A background Bulletin 13 (16). paper commissioned for Toward Gender Vijverberg, W., a nd J. H aug hton. 20 0 4. Equality in East Asia and the Pacific: A “Household Enterprises in Vietnam: Survival, Companion to the World Development Report. Growth, and Living Standards.� In Economic Washington, DC: World Bank. Growth, Poverty, and Household Welfare in ———. 2011g. World Development Report Vietnam, edited by David Dollar and Paul 2012: Gender Equality and Development. Glewwe. Washington, DC: World Bank. Washington, DC: World Bank. Weinberg, Bruce. 2000. “Computer Use and the ———. 2012. The Elderly and Old Age Support Demand for Female Workers.� Industrial in Rural China: Challenges and Prospects. and Labor Relations Review 53 (January): Washington, DC: World Bank. 290–308. World Bank and ADB (Asian Development White, Sally. 2006. “Gender and the Family.� In Bank). 2005. “Mongolia, Country Gender Voices of Islam in Southeast Asia, edited by Assessment.� World Bank: Washington DC. Greg Fealy and Virginia Hooker, 273–352. World Development Indicators (WDI) database. Singapore: Institute of Southeast Asian Studies. World DataBank, World Bank, Washington, Women, Business and the Law (database). World DC. http://databank.worldbank.org. Bank, Washington, DC. http://wbl.worldbank Yu, Wei-hsin. 2002. “Jobs for Mothers: Married .org/data. Women’s Labor Force Reentry and Part- Wood, Robert, Mary Corcoran, and Paul time, Temporary Employment in Japan.� Courant. 1993. “Pay Differences Among the Sociological Forum 17 (3): 493–523. Highly Paid: Male-Female Earnings Gap in Yuan, Ni, and Sarah Cook. 2010. “Gender Law Salaries .� Journal of Labor Economics Patterns of Informal Employment in Urban 11 (3): 417–44. China.� In Gender Equality and China’s World Bank. 2010a. “Indonesia Jobs Report: E conomic an d Soci al Tran sfor m ation: Towards Better Jobs and Security for All.� Informal Employment and Care Provision , World Bank, Washington, DC. edited by Xiao-Yuan Dong and Sarah Cook. ———. 2010b. Indonesia—Reconstruction of Beijing: Economic Science Press. Aceh Land Administration System Project. Zuo, Jiping, and Yanjie Bian. 2001. “Gendered World Bank, Washington, DC. Resources, Division of Housework and ———. 2011a. Defining Gender in the XXI Perceived Firness: A Case in Urban China.� Century: Conversations with Men and Women Journal of Marriage and Family 63: 1122–33. Agency: Voice and Influence within the Home and in Society 4 I n much of the world, women have a family resources. Similarly, at the commu- more limited voice and influence than nity or societal level, the relative power of men in decision making in their homes, individuals and groups affects their abil- their communities, and society. Women are ity to act on their preferences and influ- also more likely to be victims of gender- ence outcomes in the economic, social, and based violence. The inability of women to political domains. The relative power of voice and act on their preferences negatively different members of society, which often affects their own welfare and is detrimental differs systematically by gender, reflects to development. a complex combination of one’s personal This chapter analyzes women’s agency characteristics, prevailing social norms, in East Asian and Pacific countries. The and the broader legal and institutional ­c hapter defines agency as the ability of environments. individuals or groups to give voice to and act on their preferences and to influence o utcomes that affect them and others in ­ Agency—important for gender society. Agency is affected by and also equality and development affects individuals’ ownership of and con- The ability to act on one’s preferences, trol over endowments and their access to regardless of one’s gender, and to translate economic opportunities (Kabeer 1999). those preferences into desired outcomes is a The discussion of agency in this chapter development objective in its own right. As also includes the ability of countries to discussed in chapter 1, development not only ensure the safety and security of women in involves raising incomes or reducing pov- their homes and in society, because gender- erty, but also involves a process of expand- based violence and trafficking of women ing freedoms and choices available to all reflects the extreme deprivation of women’s people (Sen 1999). Agency is a measure of a agency in society. person’s well-being, reflecting both the abil- Within a household or partnership, ity to achieve as well as actual achievements one’s relative power affects the strength (Sen 1992). of one’s voice and influence in household Women’s agency enhances development. decisions, such as how to spend or invest When women are free to make choices, that 145 1 4 6    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C freedom positively affects all levels of soci- sphere.1 To the extent that development is ety. Increasing women’s voice and influence accompanied by stronger legal and judicial in the home has been found to improve chil- systems, more-developed societies provide dren’s education, health, and welfare (Duflo women (and men) with better access to jus- 2003; Fiszbein and Schady 2009; Haddad, tice, which strengthens their voice and pro- Hoddinott, and Alderman 1997; Thomas tects them against the extreme deprivation 1995), as discussed in chapter 2. Increasing of agency. women’s representation in firm ownership Growth and development alone are not and management, and on corporate boards, enough to enhance women’s agency in all its also increases gender equality within firms dimensions. As shown in previous chapters, and increases the provision of nonwage women have made positive strides toward benefits to workers (Cohen and Huffman gender equality in education and health, yet 2007; Ely 1995; Hultin and Szulkin 2003; gender gaps remain in access to assets and Rand and Tarp 2011). Increasing women’s economic opportunity. Increasing women’s representation in elected office not only ability to earn and accumulate assets is criti- ensures that decisions are more representa- cal for strengthening their voice and influ- tive of the voting population, but also can ence in society and making them less vulner- lead to increased provision of public goods, able to domestic violence and other types of better natural resource management, and abuse. Moreover, the relationship between increased reporting of crimes against women economic development and women’s politi- (Agarwal 2009; Ban and Rao 2009; Bea- cal representation, an important pathway man et al. 2012; Chattopadhyay and Duflo toward agency in society, is unclear (figure 2004). Global evidence shows that violence 4.2). Although development can contribute against women has lasting negative effects to strengthening women’s agency in some on economic development in addition to dimensions, data show that improvements in causing significant social, psychological, a number of other areas are not automatic. and physical harm to those who experi- Thus, governments need to develop policies ence and witness it (Morrison and Orlando that actively raise women’s agency if they 2004; Morrison, Ellsberg, and Bott 2007). are to induce meaningful changes toward Reducing gender-based violence thus results gender equality in agency. in healthier workers and higher economic This chapter aims to strengthen under- productivity, with dynamic benefits across standing of gender and agency in East Asian generations. and Pacific countries, and to lay the foun- Economic growth and development can, dation for identifying policy priorities to in turn, contribute to strengthening wom- strengthen women’s voice and influence. The en’s agency in some areas. As discussed in analysis in the chapter focuses on agency in earlier chapters, growth and development the following three domains: result in better education and health out- comes for women—and better human capi- Agency in the household and in individual •   tal outcomes for women contribute directly decisions is examined through household to stronger voice and influence, whether in decision making, control of resources, and the home, in the economy, or in society. reproductive decisions. Economic development, as measured by Agency in the public sphere is examined •   gross domestic product (GDP) per capita, through women’s participation and repre- is also associated with higher levels of civic sentation in the private sector, civil soci- activism, including on issues related to gen- ety, politics, and public institutions. der equality (figure 4.1). Civic activism is S afety and security in expressing one’s •   a measure of “collective agency�; that is, agency are examined through the preva- it is the space for both male and female lence of gender-based violence, an extreme citizens to express their voice in the public deprivation of agency. A G E N C Y : V O I C E A N D I N F L U E N C E W I T H I N T H E H O M E A N D I N S O C I E T Y    147 The form of agency that is most frequently FIGURE 4.1  There is a positive relationship between economic measured is the decision-making power development and civic activism of men and women (Kabeer 1999; Mason levels of civic activism across the world 2005; McElroy 1990). Agency may be more 0.8 explicitly measured by examining women’s mobility in the public domain, their partici- 0.7 civic activism (index) pation in public action, and the incidence of AUS JPN gender-based violence (Kabeer 1999). Some 0.6 HKG researchers have assessed gender differences WSM KOR SGP in bargaining power within a household PHL FJI MYS TWN 0.5 IDN THA by examining the extent to which people’s VNM CHN MNG TON choices change when factors affecting their LAO SLB VUT 0.4 KHM PNG bargaining power, such as education, relative FSM earnings, or asset holdings, change (Duflo 0 4 6 8 10 12 2003; Quisumbing and Maluccio 2003; log GDP per capita (constant 2000 $) Thomas 1990, 1992). rest of world East Asia and Pacific global relationship Agency at the household level is difficult to measure, since negotiations for decision- Source: World Development Indicators (WDI) database; Indices of Social Development (ISD) m aking power often occur within private ­ ­database. Note: GDP = gross domestic product. spaces in the household. Furthermore, since indicators of agency are of relatively new ­ i nterest to the international develop- FIGURE 4.2  There is no clear relationship between economic ment community, many have not yet been development and women’s representation in parliament measured over time. Where possible, the share of parliamentary seats held by women chapter will present information on how 60 women’s agency has evolved over time in the region. When data on different dimen- sions of agency are not available over time, however, the chapter will present the most 40 recent evidence available—both qualitative percent TMP and quantitative. VNM AUS LAO PHL SGP The remainder of the chapter is struc- 20 KHM CHN tured as follows. The next section analyzes IDN THA KOR the state of agency in the region. The third WSM MYS JPN section analyzes the factors that influence KIR VUT MNG MHL PNG TUV TON agency, and the fourth section identifies key 0 SLB FSM PLW policy priorities for promoting gender equal- 4 6 8 10 12 log GDP per capita (constant 2000 $) ity in voice and influence in the region. These rest of world East Asia and Paci c directions for policy are discussed in further detail in chapter 6. Source: WDI database; ISD database. Despite the geographic proximity between Note: GDP = gross domestic product. East Asian countries and the Pacific Islands, their development experiences and paths toward gender equality have been different, The state of women’s agency in especially with respect to women’s voice and East Asia and the Pacific influence. For this reason, this chapter distin- The East Asia and Pacific region has experi- guishes, where possible, between the progress enced uneven gains in women’s agency over made and the challenges faced by East Asian the past two decades. Although women now and Pacific countries in the different domains have more household decision-making power, of agency. more influence and voice in the public sphere, 1 4 8    TO WA R D G E N D E R E Q UA L I T Y I N E A S T A S I A A N D T H E PAC I F I C and new laws to protect their choices and Women in the Pacific have relatively less interests in society, progress has been uneven control over their own earnings. Between across countries, and many challenges still 13 and 15 percent of women in Marshall remain. Women in East Asia had the high- Islands, Samoa, and Tuvalu report that their est representation in national parliaments in husbands have control over their wives’ cash the developing world in the 1990s; yet, the earnings (figure 4.3). Studies in different share of women in parliaments has reached countries in the Pacific also find high levels a plateau and fallen behind other regions in of financial control by men in the house- the past decade. Women have little represen- hold. In Kiribati, for example, 19 percent tation among parliamentarians in the Pacific. of women report that their partners do not Women’s participation in the private sector allow them to make any financial decisions has increased, but women remain a minority for household expenditures, and 12 percent among firm owners and on corporate boards. of women report that they cannot e ­ xercise The sections that follow examine evidence on control of their own incomes because their women’s voice and influence in the home and partners take it away. When intimate partner in the public domain, as well as evidence on violence occurs, women have an even weaker gender-based violence in the region. voice over household financial decisions. Twenty-three percent of women who expe- rience ­intimate partner violence report that Agency in the household and in their partners do not allow them to make personal decisions any financial decisions related to household Women’s household decision-making power expenditure, and 22 percent report having no in the region is relatively high. Women from control of their own income. all wealth quintiles in East Asia and the Most women in the region have the free- Pacific are more likely to have control over dom to make other household decisions, large purchases and over decisions to visit whether related to their own health care, family and relatives than women in other household purchases, or visits to family and developing regions, and they are among the relatives. Women in the Pacific have slightly most likely to have control over their own less ability to make decisions on their own earnings. However, a higher share of women health care and household purchases, on in East Asia than in the Pacific control their average, than women in East Asia. In the own income and have a say in other house- Philippines, 94 percent of women make their hold decisions, reflecting a great intraregional own health care decisions solely or jointly variation. with their spouse, whereas only 70 percent Women in East Asia, in particular, of women in the Marshall Islands do so have high levels of autonomy. According (figure 4.4). Similar patterns occur in how to Demographic and Health Survey (DHS) much say wives have in decisions regarding data, nearly 70 percent of married women visits to her family and relatives. However, ages 15 to 49 in Cambodia and Indonesia in some areas in East Asia, women also report that they control their own earnings; have relatively little decision-making power 31 and 28 percent, respectively, report joint with regard to certain choices. For instance, control of their earnings with their h­ usbands. 18 percent of women surveyed in Indonesia Only 1 percent of women in Cambodia, and report that men have the final say in making 3 percent of women in Indonesia, report decisions on household purchases. Women in that their husbands decide how their earn- Tuvalu and the Marshall Islands report hav- ings are used (figure 4.3). Women in Cam- ing the lowest control over their own health bodia and Timor-Leste report high levels care decisions and household purchases of control not only over their own income, ­ (figure 4.4; DHS data, various years). but also over their husband’s income (DHS Where DHS data are not available, country data, various years). 2 surveys from Kiribati, the Solomon Islands, A G E N C Y : V O I C E A N D I N F L U E N C E W I T H I N T H E H O M E A N D I N S O C I E T Y    149 and Vanuatu suggests that women have rela- FIGURE 4.3  Who decides how wives’ cash earnings are used varies tively low levels of autonomy with respect to across the region household decision making. This data indi- cates a high prevalence of controlling behav- person who decides how the wife’s cash earnings are used (%) ior by husbands and male partners over 100 household, financial, and mobility decisions. Of partnered women, 58 percent and 69 per- 90 cent reported the experience of some sort of 80 controlling behavior by their partners in the 70 Solomon Islands and Vanuatu, respectively. 60 percent This sort of behavior includes preventing her 50 from seeing family, wanting to know where 40 she is at all times, forbidding contact with 30 other men, and controlling access to health 20 care. Controlling behaviors by husbands are 10 often correlated with a lack of agency in 0 other dimensions, as well as with personal a oa sia s ds lu te ne di va es an characteristics that hinder agency, such as m ne bo pi -L Tu Sa Isl do ilip m or all Ca In poor endowments and economic opportuni- Tim Ph sh ar ties, discussed in more detail below. In both M mainly wife jointly mainly husband the Solomon Islands and Vanuatu, women with little to no education are more likely Source: DHS data, various years. to have partners who exhibit controlling Note: Among currently married women ages 15–49 who receive cash earnings for employment. behaviors than women with higher educa- FIGURE 4.4  A majority of wives control decisions regarding their tional attainment. Women who have expe- own health care and household purchases rienced intimate partner violence are also significantly more likely to experience con- trolling behaviors than women who have not percentage of currently married women ages 15–49 who control health care and purchasing decisions solely or with husband experienced violence (SPC and NSO 2009; 100 VWC and NSO 2011). In that sense, the fac- 90 tors that inhibit women’s agency are likely to 80 be mutually reinforcing. 70 Globally, women living in wealthier households are likely to have more decision- 60 percent making power than women living in poorer 50 households. This pattern bears out in Pacific 40 countries, but less so in East Asia. Evidence 30 from countries in other regions of the world 20 illustrates that women in wealthier house- 10 holds have a wider set of choices and face 0 a lu s