Feminist Economics and the World Bank: History, Theory and Policy (Routledge Iaffe Advances in Feminist Economics)

Feminist Economics and the World BankThe past decade has witnessed a paradigm shift at the World Bank from a focus on structural adjustment to a focus on poverty reduction. As evidenced by the Bank’s 2001 report, Engendering Development: Through Gender Equality in Rights, Resources, and Voice, an increased attention to gender issues has been an important part of this process. The premise of the report is that economic growth and development cannot be effectively addressed when gender inequality is not taken into account, because poverty increases gender inequalities and gender inequalities hinder economic development. This book brings together a range of responses from feminist economists and other social researchers on the issues raised in this report. With contributions from highly esteemed scholars such as Violet Eudine Barriteau, Diane Elson, Gale Summerﬁeld, and Zaﬁris Tzannatos, this anthology critically examines the relationships between gender, growth, development, and the World Bank. It develops a history of the World Bank’s perspectives on gender, empirically evaluates the impacts of the Bank’s policies on three different regions of the world, explores the ideological and methodological commitments of the report from a variety of feminist and interdisciplinary social science perspectives, and inquires into future directions for feminist economic research. The book shows the importance and challenge of taking gender into account in development theory and policy. Its complex and nuanced analyses of the social relations of gender in a global context will be an important resource for policymakers, activists, and scholars. Edith Kuiper is Researcher at the Department of Economics and Econometrics at the Universiteit van Amsterdam, The Netherlands. Drucilla K. Barker is Professor of Economics and Women’s Studies at Hollins University, Virginia, USA.

Routledge IAFFE Advances in Feminist Economics IAFFE aims to increase the visibility and range of economic research on gender; facilitate communication among scholars, policy-makers, and activists concerned with women’s well-being and empowerment; promote discussions among policy-makers about interventions which serve women’s needs; educate economists, policy-makers, and the general public about feminist perspectives on economic issues; foster feminist evaluations of economics as a discipline; expose the gender blindness characteristic of much social science and the ways in which this impoverishes all research, even research that does not explicitly concern women’s issues; help expand opportunities for women, especially women from underrepresented groups, within economics; and encourage the inclusion of feminist perspectives in the teaching of economics. The IAFFE book series pursues the aims of the organization by providing a forum in which scholars have space to develop their ideas at length and in detail. The series exemplifies the value of feminist research and the high standard of IAFFE sponsored scholarship. 1 Living Wages, Equal Wages Gender and labor market policies in the United States Deborah M. Figart, Ellen Mutari and Marilyn Power 2 Family Time The social organization of care Edited by Nancy Folbre and Michael Bittman 3 Feminist Economics and the World Bank History, theory and policy Edited by Edith Kuiper and Drucilla K. Barker

Feminist Economics and the World Bank History, theory and policy

Edited by Edith Kuiper and Drucilla K. Barker

First published 2006 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN Simultaneously published in the USA and Canada by Routledge 270 Madison Ave, New York, NY 10016 Routledge is an imprint of the Taylor & Francis Group © 2006 Selection and editorial matter, Edith Kuiper and Drucilla K. Barker; individual chapters, the contributors

This edition published in the Taylor & Francis e-Library, 2006.

“To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data A catalog record for this book has been requested ISBN10: 0–415–70064–7 ISBN10: 0–415–76381–9

ISBN13: 9–78–0–415–70064–1 (hbk) ISBN13: 9–78–0–415–76381–3 (pbk)

Contents

List of ﬁgures and tables Notes on contributors Acknowledgments

1 Feminist economics and the World Bank: an introduction

ix xi xvii

DRUCILLA K. BARKER AND EDITH KUIPER

PART I

Gender and the World Bank: an institutional history and assessment

2 The World Bank, development, adjustment and gender equality

ZAFIRIS TZANNATOS

3 An assessment of efforts to promote gender equality at the World Bank

CAROLYN M. LONG

4 Rhetoric and realities: a comment

SAKUNTALA NARASIMHAN

PART II

Policy evaluations 5 Engendering Development: a critique

63 65

ROSE-MARIE AVIN

6 Engendering agricultural technology for Africa’s farmers: promises and pitfalls CHERYL DOSS

vi Contents

7 Taking gender differences in bargaining power seriously: equity, labor standards, and living wages

STEPHANIE SEGUINO

8 World Bank discourse and World Bank policy in Engendering Development: a comment

117

KARIN SCHOENPFLUG

PART III

Disciplinary paradigms/development paradigms 9 Colonizing knowledge: economics and interdisciplinarity in Engendering Development

125

127

SUZANNE BERGERON

10 Adjustment with a woman’s face: gender and macroeconomic policy at the World Bank

142

CYNTHIA A. WOOD

11 Gender and intrahousehold decision-making: international migration and other frontiers for development policy

162

AIDA ORGOCKA AND GALE SUMMERFIELD

12 Engendering development or gender main-streaming? A critical assessment from the Commonwealth Caribbean

176

VIOLET EUDINE BARRITEAU

13 “Disciplining” and “engendering” the World Bank: a comment

199

LAURA PARISI

PART IV

Explorations: future directions of feminist economic research

207

14 A seat at the table: feminist economists negotiate development

209

DRUCILLA K. BARKER

15 Why feminist economists should pay more attention to the coherence between the World Bank and the WTO MARIAMA WILLIAMS

218

Contents vii

16 Engendering the German Parliamentary Commission Report on “Globalization of the World Economy”

232

BRIGITTE YOUNG

17 Women’s rights and Engendering Development

243

DIANE ELSON

Index

256

Figures and tables

Figures 2.1 Projects rated as achieving satisfactory outcomes, 1974–2001 7.1 Total foreign direct investment as a percentage of gross ﬁxed capital formation 7.2 Wage-induced productivity growth 7.3 Average productivity growth and total FDI as percentage of gross ﬁxed capital formation, 1972–1990

19 100 104 106

Tables 2.1 Evolution of the World Bank’s attention to gender and development issues 2.2 Gender policies of multilateral donor agencies 2.3 Gender analysis in Bank projects by sector 1988–1999 7.1 Trends in FDI in developing countries 7.2 Indices of labor productivity in selected South Korean manufacturing industries, 1976–1990 12.1 World rank of Caribbean countries on the Human Development Index, 1995–2001 12.2 Comparative rank of Caribbean countries on the Gender Development Index, 1995–2001 12.3 Earned income share, 1995–1998 12.4 Real GDP per capita (US$), 1999–2001 12.5 Women’s political participation in Caribbean countries, 2001

22 28 29 98 105 184 188 189 190 191

Contributors

Rose-Marie Avin is Professor of Economics at the University of WisconsinEau Claire. She received her Ph.D. in Economics from the University of Maryland-College Park, USA. She teaches courses on the economic development of the Third World and of Latin America, and on women and economic development. Her research interests focus on women in the Third World, especially in Nicaragua. She has written articles on gender issues in Latin America, more speciﬁcally on economic restructuring, the empowerment of Nicaraguan women, free-trade zones, the changing economic role of women in Brazil, and the role of international organizations in economic development. Drucilla K. Barker is Professor of Economics and Women’s Studies at Hollins University, Virginia, USA. She received her Ph.D. in Economics from the University of Illinois at Champaign-Urbana and her B.A. in Philosophy from Sonoma State University. A founding member of the International Association for Feminist Economics (IAFFE), she is the co-author (with Susan Feiner) of Liberating Economics: Feminist Perspectives on the Families, Work and Globalization, and the co-editor (with Edith Kuiper) of Toward a Feminist Philosophy of Economics (also published by Routledge). Violet Eudine Barriteau is Senior Lecturer and Head of the Centre for Gender and Development Studies, University of the West Indies, Cave Hill, Barbados. She is the author of The Political Economy of Gender in the TwentiethCentury Caribbean. Her most recent publications include Confronting Power Theorizing Gender: Interdisciplinary Perspectives in the Caribbean, and “Women Entrepreneurs and Economic Marginality: Rethinking Caribbean Women’s Economic Relations” in Gendered Realities: Essays in Caribbean Feminist Thought. She is the inaugural Dame Nita Barrow Women in Development Fellow, Ontario Institute for Studies in Education, University of Toronto. Suzanne Bergeron holds a joint appointment in Women’s Studies and Social Sciences at the University of Michigan Dearborn, USA, where she is also the director of the Women in Learning and Leadership program. She is the author of Fragments of Development: Nation, Gender, and the Space of Modernity, as well as numerous articles on gender, development and the World Bank.

xii Contributors Cheryl Doss is the Director of Graduate Studies for the M.A. Program in International Relations and a Lecturer in Economics at Yale University, USA. She has a Ph.D. in Applied Economics from the University of Minnesota. Her research focuses on gender and intrahousehold issues related to rural households in Africa, especially those in Ghana, Kenya and Ethiopia. In particular, she is interested in the gender implications of technology for agricultural households. Currently, she is working to understand risk management strategies among pastoralist households in East Africa, with a focus on the implications for women, and she is involved with a project looking at women and the distribution of wealth. Diane Elson teaches courses on gender, development and human rights in the Department of Sociology and Human Rights Centre at the University of Essex. She has previously held posts in the universities of Manchester, Sussex, York and Oxford. In 1998–2000, she served as Special Advisor to the United Nations Development Fund for Women (UNIFEM); and in 2004 she was the vice-president of the International Association for Feminist Economics. She is a member of the UN Taskforce on the Millennium Development Goals. Her recent publications include Progress of the World’s Women 2002; “International Financial Architecture: A View from the Kitchen”, Femina Politica, 11(1) 2002; and “Human Rights and Corporate Profits: the UN Global Compact-Part of the Solution of Part of the Problem?” in Global Tensions, also published by Routledge. Edith Kuiper is Researcher at the Department of Economics and Econometrics at the Universiteit van Amsterdam. Her research interests are in the history and philosophy of economics. She is co-editor of Out of the Margin: Feminist Perspectives on Economics (Routledge, 1995), author of The Most Valuable of All Capital: A Gender Reading of Economic Texts (2001) and she co-edited (with Drucilla K. Barker) Towards a Feminist Philosophy of Economics (Routledge, 2003). She is a founder and former President of the Dutch Network of Feminist Economics (FENN) and coordinator of the European Chapter of the International Association for Feminist Economics (IAFFE-Europe). Carolyn M. Long, now an independent consultant, was previously VicePresident of InterAction, the US coalition of NGOs, where she oversaw the organization’s initiatives to strengthen the development operations of US NGOs and to promote appropriate policies and programs for sustainable human development at the World Bank and other public donor agencies. She is the author of Participation of the Poor in Development Initiatives: Taking Their Rightful Place, which analyzes progress made in incorporating participation of the poor in projects and policies of large public donor agencies and recipient government agencies over a ten-year period. Focused largely on the World Bank, the book also analyzes strategies used by NGOs aligned with activists inside donor agencies to promote this work. She has worked extensively in sub-Saharan Africa.

Contributors xiii Sakuntala Narasimhan is a national award-winning Indian journalist specializing in gender and development. She taught a course on women and development at Virginia Tech, USA, on a Fulbright assignment (1990) and is the author of ten books including Empowering Women: An Alternative Strategy from Rural India. She has presented papers at the annual conferences of the International Association for Feminist Economics at Oslo (2001), Barbados (2003) and Oxford (2004) and was commissioned to report on the Fourth World conference on women (Beijing 1995), UNGASS (Beijing +5, New York 2000) and the World Summit at Johannesburg (2002). Aida Orgocka received her doctoral degree in Human Development and Family Studies from the University of Illinois at Urbana-Champaign. She helped found one of the ﬁrst women’s non-governmental organizations in Albania. Her research interests address gender, health, international migration and development. Her most recent work on Albanian female high-skilled migration will appear in The New Albanian Diaspora. She is also publishing her work on Muslim immigrant women in the USA in professional journals such as Sex Education. Laura Parisi is Assistant Professor of Women’s Studies at the University of Victoria in Victoria, BC, Canada. Her areas of specialty are gender, feminist theory, international political economy, human rights, and international law and organization. She was a delegate to the UN Beijing +5 women’s conference in New York in 2000, and is the author of several publications on women’s human rights issues, including her most recent publication “Feminist Praxis and Women’s Human Rights” in the Journal of Human Rights (2002, vol. 1). Karin Schoenpﬂug received her Ph.D. in economics from the University of Vienna and works in the economic policy division of the Austrian Ministry of Finance. She has taught at the Universities of Klagenfurt and Graz in Austria and at the University of Southern Maine in the USA. Her cultural and political commitments include working at Rosa Lila Villa, a GLBTQ center in Vienna, attending and organizing the world’s oldest Women’s Camp on Femø, Denmark, recently focusing on the lessons from the Northern European women’s movement of the 1970s. She is a member of International Association for Feminist Economics and her research interests include exploring utopian visions of feminist economics. Stephanie Seguino is Associate Professor and Chair of the Department of Economics at the University of Vermont, USA. A central issue her research explores is the welfare effects of globalization. She has also studied the interrelationship between output, growth, and income distribution, exploring how the distribution of income between genders affects macroeconomic outcomes. Her most recent work has explored the effects of economic growth on gender equity in well-being in Latin America and the Caribbean as well as Asia. A second branch of her work explores the determinants

xiv Contributors of investment and the effect of capital mobility on productivity, wages, and growth. She has also worked on living wage issues. Her worked is published in a wide variety of journals, including the Journal of Development Studies, the Cambridge Journal of Economics, World Development, and the Journal of PostKeynesian Economics. She has also served as consultant to the World Bank, USAID, the United Nations, and UNRISD. Gale Summerfield is Director of the Women and Gender in Global Perspectives Program and Associate Professor of Human and Community Development at the University of Illinois at Urbana-Champaign. She has written extensively on gender aspects of reforms in post-socialist and developing countries. Her articles appear in journals such as International Journal of Politics, Culture, and Society, Feminist Economics, World Development, and Review of Social Economy. She is co-editor (with Nahid Aslanbeigui and Steven Pressman) of Women in the Age of Economic Transformation (Routledge, 1994), and (with Irene Tinker) of Woman’s Rights to House and Land: China, Laos, Vietnam. Her current research interests address gender, human security, and transnational migration. Zafiris Tzannatos is Advisor to the World Bank Institute and at the time of writing the chapter in this volume he was Professor and Chair of the Economics Department at the American University of Beirut, Lebanon. He has previously held senior appointments in UK universities and the World Bank including being Advisor to the Bank’s Managing Director for Human Development, Manager for Social Protection in the Middle East and North Africa Region, and Leader of the Child Labor Program. He has advised many governments and international organizations on issues relating to development and social policy and has numerous publications, including 14 books, in the areas of gender equality, child labor and trade unions and more broadly labor economics. Mariama Williams, Ph.D., is an international economics consultant and an Adjunct Associate at the Center of Concern, Washington, DC. She is also the Research Adviser for the International Gender and Trade Network/ Caribbean Gender and Trade Network and co-research Coordinator, Political Economy of Globalisation (Trade), and Development Alternative with Women for a New Era (DAWN). Williams is also a former member of the board of the Association for Women’s Right and Development (AWID) and a Director of the Institute for Law and Economics (ILE-Jamaica). She is also a member of the International Working Group on Engendering Macroeconomics and International Trade and a consultant adviser for Gender and Trade to the Commonwealth Secretariat. Cynthia A. Wood is an economist and Associate Professor of Interdisciplinary Studies at Appalachian State University, whose work has appeared in Feminist Economics and Nepantla: Views from South. Her research interests include

Contributors xv gender and development, Latin American economics, and postcolonial feminist theory. Brigitte Young is Professor of International and Comparative Political Economy at the Institute of Political Science at the University of Muenster. She also taught at the Institute of Political Science, Free University of Berlin from 1997 to 1999. From 1994 to 1995 she was Research Associate at the Centre for German and European Studies, School of Foreign Service, Georgetown University, Washington, DC, and from 1991 to 1997 she was at Wesleyan University, Connecticut. Her research areas are globalization, financial markets, GATS, international political economy, feminist economics, and multimedia intercultural education. She has published widely in English and German. Her recent publications include: Gender, Globalization, and Democratization (co-edited) and Global Governance (co-edited). She is presently working on two manuscripts, Engendering the WTO (forthcoming) and Finance, Power and Social Reproduction (forthcoming).

Acknowledgments

This book is the result of both our enthusiasm about the leap made by the World Bank Gender Ofﬁce in taking feminist economic interests and research into account in Engendering Development: Through Gender Equality in Rights, Resources, and Voice, and our concerns about its limitations. We acknowledge Andrew Mason, of the Gender Office of the World Bank, for his participation in conference sessions on the report. We also thank Zaﬁris Tzannatos for his support and years of commitment to open discussions on these matters in an international forum of feminist economists. We thank all those who joined the discussions at the conference of the International Association for Feminist Economics (IAFFE) in Los Angeles. Thanks also to IAFFE for providing us with an international forum and network for these discussions and a place to bring together scholars committed to critical discussions on gender and development. We also thank Robert Langham, Routledge economics editor, for his continued support and interest in this project. We thank the organization of the Marie-Jahoda Visiting Professorship for International Women’s Studies Chair of the Ruhr-Universität Bochum (Germany) for providing the environment in which to work on the completion of this book. Thanks also to Ilse Lenz and Brigitte Young for their discussions that helped us to further articulate the conception of this book. Thanks also to Hollins University (USA) for their research support. We acknowledge and thank the Women’s Edge Coalition for permission to adapt “The Advocate’s Guide to Promoting Gender Equality at the World Bank.” The full paper is available at www.womensedge.org. We also acknowledge Routledge for permission to reprint “Adjustment With a Woman’s Face: Gender and Macroeconomic Policy at the World Bank,” by Cynthia Wood, reprinted from Struggles for Social Rights in Latin America, edited by Susan Eckstein and Tim Wickam-Crowley, New York: Routledge, 2003.

Feminist economics and the World Bank An introduction Drucilla K. Barker and Edith Kuiper

In the past decade the World Bank has made a substantial shift both in theory and policy. Previously economists in Washington determined the boundaries and conditions for World Bank programmes and loans. Now macroeconomic models and top-down approaches are augmented by cooperation with local groups and non-governmental organizations. This has been accompanied by a renewed focus on fighting poverty, a decentralization of authority, and an increased involvement of national governments. An increased attention to gender issues has been an important part of this process. During the 1980s only scant attention was paid to gender and women’s issues in the Bank’s publications and policies. This changed in the 1990s as the Bank began to study the effects of women’s education on poverty and fertility rates. The 1995 Fourth World Conference on Women in Beijing provided further impetus for the Bank to pursue research on gender and development. At that conference James Wolfensohn, the President of the World Bank, made a commitment to develop a framework for gender and development in cooperation with NGOs and other international organizations working on these issues. In this process the attention paid to the relationship between gender and development broadened from a relatively narrow economic focus to a more interdisciplinary one that included discussions of identity, culture, and North–South relations. This research, which culminated in the 2001 report Engendering Development: Through Gender Equality in Rights, Resources, and Voice, calls for including women’s economic status as a major economic indicator in the evaluation of World Bank projects (World Bank 2001). Taking the economic framework of the World Bank as a starting point for analysis, Engendering Development makes the case that economic growth and development cannot be effectively addressed when gender inequality is not taken into account. It starts from the premise that poverty increases gender inequalities and that gender inequalities hinder economic development. As James Wolfensohn puts it in his Foreword to the report, The World Bank is committed to a world free from poverty. And it is clear that efforts to achieve this must address gender inequalities . . . ignoring gender disparities comes at great cost – to people’s well being and to

2 Drucilla K. Barker and Edith Kuiper countries’ abilities to grow sustainably, to govern effectively, and thus to reduce poverty. (Wolfensohn 2001: xi) Of course, promoting gender equality is a complex undertaking, as the authors of the report, Elizabeth M. King and Andrew D. Mason, acknowledge in their preface. The research informing the report is multidisciplinary and draws on economics, law, demography, sociology, and other disciplines. The report is organized around the principal pathways through which gender inequalities are generated and persist: the role of social institutions, of economic institutions, of household power relations, resources and decision-making, and of economic change and development policy. Acknowledging that addressing persistent gender inequalities requires more than just economic growth, the Report proposes a three-part strategy for promoting gender equality: • • •

Rights: Reform institutions to establish equal rights and opportunities for women and men. Resources: Foster economic development to strengthen incentives for more equal resources and participation. Voice: Take active measures to redress persistent disparities in command over resources and political voice (World Bank 2001: 1–2).

Given the importance of the topic, and the sustained level of interest in it, Engendering Development can, in many ways, be seen as a major achievement for scholars, policy-makers and activists who have argued for including gender in economic theory and policy. The publication of this report, its explicit claim that gender issues are economic issues, the use made of feminist economic research, and the reception of the report in the World Bank motivated the editors of this volume to organize a set of sessions at the annual meetings of the International Association for Feminist Economics (IAFFE) in 2002 in Los Angeles. The overwhelming interest in these sessions, attended by many of the authors in this anthology, most of whom have been involved in discussions about women, development, and the Bank for many years, as well as Andrew Mason, one of the authors of the report, motivated this project. It brings together a range of viewpoints from feminist economists and other social researchers on the paradigm shift articulated in Engendering Development. The essays in this anthology critically examine the relationships between gender, growth, development, and the World Bank from historical, empirical, and theoretical perspectives, and explore futures directions for research, policy, and activism.

Gender and the World Bank: an institutional history and assessment The first section of this book sets the stage for a feminist examination and critique of the World Bank’s policies on gender and development by providing

Feminist economics and the World Bank 3 a historical overview of the Bank’s gender policies and an examination of the institutional processes that culminated in the Engendering Development report. Zafiris Tzannatos, who has been involved in the research and development of gender policies and strategies at the Bank for many years, provides an insider’s perspective. Carolyn Long, a consultant on the civil society organizations in development, connects the report to advocacy efforts carried out by both Northern and Southern women’s organizations, and Sakuntala Narasimhan brings a journalist’s eye to the issues. Tzannatos provides a historical overview of the World Bank’s genesis, mandate, and approach to development, poverty, and gender. He locates the evolution of the Bank’s position on women and gender against the background of the larger changes in the Bank’s policies from reconstruction, to structural adjustment, to its current focus on poverty alleviation. His discussion of the evolution of gender issues within the Bank in terms of philosophy, internal organization, and activities allows the reader to better understand the institutional character and culture of the institution. The Bank’s preoccupation with macroeconomic issues and structural adjustment policies meant that thematic issues such as gender equality received scant attention until it began to prepare for the 1995 Beijing conference. Since then the Bank came to endorse the position that gender inequalities hamper growth, called for attention to gender equality in development efforts, and outlined how gender issues were to be identiﬁed and incorporated into Bank-supported projects and programs. Tzannatos concludes that although attention to gender issues at the Bank continue, and have become particularly visible since the onset of the new millennium, their impact on development and on women in developing countries and more generally on development is unclear. From her extensive experience in women’s organizations and other international NGOs, Carolyn Long assesses the Bank’s attention to the promotion of gender equality in its policies and projects. Her account of the Bank’s approach to gender in terms of history, organizational structure, and institutional culture is framed in terms of gender advocacy efforts carried out by women’s organizations and other external actors, and complements Tzannatos’ insider’s account. She includes a discussion of the actions that a number of Northern and Southern women’s organizations consider essential in promoting gender equality in Bank-funded initiatives. Long’s account is particularly critical of the lack of real institutional support for gender at the Bank. She considers the Bank’s gender strategies and adoption of Country Gender Assessments (CGAs), which analyze the gender dimensions of development across sectors and identify gender-responsive measures that will contribute to poverty reduction, growth, and human well-being, as an important step forward. The problem, however, is the lack of real institutional support, in terms of either budgets or administrative authority. The integration of gender concerns into projects and policies varies by region and depends on the interest, or lack thereof, of the particular Bank employees and countries in promoting gender equality. External pressure from NGOs and women’s

4 Drucilla K. Barker and Edith Kuiper organizations in both the North and South will be necessary to create the institutional changes necessary to turn the rhetoric of gender equality into real integration of women and women’s issues into Bank policies. This section concludes with a comment by Sakuntala Narasimhan. She asks whether the Bank has learned from its past mistakes and whether the changes now under way will make a positive difference in the ways that development funding affects women. Dam projects on the Narmada river in upper India, a road development project on the slopes of Kilimanjaro in Tanzania, and the building of a coal-fired power plant in Orissa state in India are just three examples that lead her to doubt the learning abilities of the Bank. She concludes that development funding continues to have a negative effect on the well-being of many poor women in the South. Narishimhan agrees with Long that external pressure on the Bank is necessary to effect real reforms, and this in turn requires that the voices of poor, mostly rural, mostly uneducated, women in the South be heeded and the chasm between them and privileged, urban, educated women be closed.

Policy evaluations The second section of the book examines the explicit and implicit policy implications of the report in light of the actual economic conditions in three different regions: Latin America, Africa, and Southeast Asia. As all of these chapters point out, the Bank assumes that economic growth is, for all practical purposes, synonymous with development. In development literature and in development practices, however, the concept has evolved to include a variety of non-economic dimensions including freedom, dignity, capabilities, inclusion, and so forth. Defined in this broader sense, development and gender equity can go hand in hand. In contrast, economic growth alone may not improve the well-being of the vast majority of people in the developing world, nor will it necessarily contribute to gender equity and the empowerment of women. Rose-Marie Avin’s chapter challenges the report’s assertion that women’s status and gender equality actually improved in parts of Latin America during the structural adjustment period of the 1980s and 1990s. Case studies from Brazil and Nicaragua provide the evidence for her challenge. In the Brazilian case, large numbers of women migrated to Rio de Janeiro and São Paulo to find work. However, due to capital-intensive import-substitution many of these women were not able to ﬁnd jobs in the manufacturing sector and found work in the informal sector as street vendors, prostitutes, cooks, maids, and nannies. All women were not, of course, affected equally. Segmented labor markets meant that middle-class, Euro-Brazilian women were able to take advantage of educational and occupational opportunities while millions of poor, workingclass women were relegated to the low-paying, low-status occupations. Thus the Bank’s assertion that economic growth improves the status of women depends on its disregard of the importance of race, ethnicity, and class in determining women’s economic status.

Feminist economics and the World Bank 5 The Nicaraguan case tells an interesting story. During the period between 1979 and 1990, when the Sandinista government was in power, major policies were introduced that enabled women to participate fully in the political, economic, and social life of the country. For the ﬁrst time working-class and peasant women had voice. With the defeat of the Sandinistas in 1990, advancements toward gender equity came to a halt. Structural adjustment policies resulted in widespread female unemployment and women’s burdens in households also increased. Cheryl Doss explores gender and development in an African context. She begins with the premise that improved agricultural technologies are key to reducing poverty and increasing the standard of living in Africa and notes that the report does not directly address issues of agricultural and technology, nor does it directly address issues of power relations. In order to improve the wellbeing of women in Africa it is necessary to develop policies and technologies that will improve agricultural productivity, and for such polices and technologies to be efﬁcacious, it is necessary to investigate power and gender relations in households, communities, and countries. Women will only beneﬁt from new technologies if the accompanying policies increase their options and control over their own lives. Doss considers three dimensions of the effects of technologies on women farmers. First she considers their effects on women’s workloads. Technologies that increase productivity may have a negative effect on well-being if women have to work harder or longer or lose control over the output. Technologies can also impact women’s access to resources, land in particular. Since women often acquire land rights through male relatives or village heads, technologies that increase the value of land may actually make it more difﬁcult for them to acquire such rights. Finally, if women lose control over their activities once they become profitable, then their well-being will be harmed. Since women’s lives and situations are so varied, it is not possible to tell a priori how technologies will affect them. A nuanced sense of power relations in particular contexts is necessary to understand whether technologies will enhance or diminish women’s power and hence their well-being, and the report fails to provide this. Concerns about the effects of economic growth on women’s well-being inform the chapter by Stephanie Seguino. She notes that well-being is improved by paid work at living wages and by gender-wage parity. In developing countries, however, the combination of women’s segregation in export industries and increased capital mobility constrains their ability to improve their wages and working conditions. Credible threats of firm relocation effectively dampen women’s bargaining power relative to that of capital even as the demand for their labor increases. This leads her to question the Bank’s assumption that trade and market liberalization are beneﬁcial to women’s well-being. Using the case of the Asian semi-industrialized countries (SIEs), she demonstrates an inverse relationship between economic growth and gender-wage parity. Evidence shows that the countries with the largest gender-wage gaps were the ones that grew the most rapidly during the period between 1975 and

6 Drucilla K. Barker and Edith Kuiper 1995. Using a neo-Kaleckian model to further illuminate the relationship between female wages and macroeconomic outcomes in an open economy, she shows that the decline in output and female employment resulting from higher female wages is more likely to occur if capital is footloose, if the price elasticity of exports is high, and if the spending propensities of capitalists and workers are similar. A more optimistic scenario in which higher female wages are consistent with economic growth is possible if ﬁrms have an incentive to innovate or if higher wages stimulate productivity. The case of South Korea demonstrates that this is possible, and the precise conditions need to be investigated further. A comment by Karin Schoenpﬂug concludes this section. She uses Michel Foucault’s discourse analysis to reﬂect on the arguments made by Avin, Doss, and Seguino and to deconstruct the rhetoric and policies of the World Bank. Four localities are particularly germane to her analysis: the constitution of subjectivity and issues of voice, processes of legitimization and the power of deﬁnition, the role of hierarchal dualisms, and the material consequences of practices of inclusion and exclusion. From this framework she concludes that Engendering Development is a powerful piece of institutional rhetoric that creates its own logic and legitimizes World Bank policies. These issues are explored further in the next section.

Disciplinary paradigms/development paradigms The third section of this book examines the ideological and methodological commitments of the report from a variety of feminist and interdisciplinary social science perspectives. It explicitly considers the ways in which disciplinary paradigms and boundaries contribute to the sense-making and legitimization of the report. These chapters note that although the Bank has adopted a new, multidisciplinary paradigm that includes attention to problems of poverty and other social issues, the changes are more apparent than real. We begin this section with Suzanne Bergeron’s examination of the Bank’s break with a narrow, economistic approach in favor of an approach that is responsive to poverty and inequality. She argues that although economists are encouraged to take a multidisciplinary approach and account for social and cultural factors in their analyses, they do in ways that preserve and strengthen the disciplinary core of economics. Rationality and market efﬁciency remain privileged concepts, and the critical and the transdisciplinary methods that characterize feminist theorizing remain outside the Bank’s framework. The new approach does account for the signiﬁcance of market failures and externalities, norms, values, and institutions. It draws attention to the importance of health and education for women. However, the disciplinary boundaries of economics remain intact and intersectionality, a central concept in women’s studies, remains absent. Differences among women are obscured by universalistic accounts of women’s behavior and experiences. Integration into formal labor markets is seen as the key to women’s emancipation from traditional, patriarchal cultures. Women’s emancipation is not, however, seen as an end in itself, but

Feminist economics and the World Bank 7 rather as instrumental to the goals of economic growth. In contrast, a feminist transdisciplinary methodology would call on social constructivist and postcolonial insights to consider difference and representations and consider how both market and non-market social processes affect women’s agency and wellbeing. Cynthia Wood agrees with Bergeron that neoclassical economics remains the dominant paradigm at the Bank. She argues that as long as it remains dominant, neither political pressure nor other institutional changes will eliminate gender bias in adjustment policies. Neoclassical economics, with its emphasis on markets, prices, and economic growth, precludes gender analyses and marginalizes any consideration of the effects of structural adjustment on unpaid domestic labor. The Bank’s discussion of gender downplays considerations of structural adjustment and macroeconomics and focuses instead on social issues such as education, population, and health. Wood argues that publication of Engendering Development will do little to change this because, in reality, engendering macroeconomics entails considering unpaid domestic labor, and the current paradigm precludes doing this in any meaningful way. Moreover, eliminating gender bias in economic theory and policy requires not only that non-market activities be recognized and valued, but that differences in intrahousehold relations be considered. Intrahousehold relations are the subject of the next essay by Aida Orgocka and Gale Summerﬁeld. Although the fourth chapter in Engendering Development analyzes intrahousehold bargaining processes, Orgocka and Summerﬁeld argue that it has limitations. It does not include new interdisciplinary developments, and it is quite selective in the work it does include, stressing econometric studies rather than qualitative approaches. Moreover, the Bank’s analysis is focused on adult, heterosexual partners, and considers children mainly as the recipients of the parents’ largesse or as investments for the parents’ future support. Orgocka and Summerﬁeld, in contrast, argue that understanding parent/child dynamics would contribute to a deeper understanding of the distribution of power and resources in households. And a better understanding of mother/daughter relationships is fundamental to constructing policies that promote gender equity since history is replete with examples of mothers passing down hurtful traditions to their daughters because their economic well-being was dependent on adherence to patriarchal gender norms. They use their qualitative research on gender and Muslim immigrants in the Midwestern United States to illustrate how factors overlooked in the Bank’s discussion enrich our understanding of mother/daughter interactions within the household regarding sex education for young women. Sex education is, of course, a key component of policies concerning health and reproduction. Orgocka and Summerﬁeld’s work points to the importance of using qualitative research methods so that policy-makers can better understand different contexts and different expressions of agency in diverse groups of people. V. Eudine Barriteau provides a contextualized critique of the Bank’s report. She argues that it is best understood as a form of gender mainstreaming and

8 Drucilla K. Barker and Edith Kuiper explores the major shortcomings of gender mainstreaming in the context of the Commonwealth Caribbean. Gender mainstreaming views gender as arising from differences in male and female bodies and does not include feminist inquiries into ideology, power relations, and conditions of subordination. The report suffers from the same ﬂaw and so its policies will not disrupt gender hierarchies. Barriteau points out that the report does not differentiate between the Caribbean and rest of Latin America. The experience of the Commonwealth Caribbean, however, provides evidence that women’s access to material resources is insufﬁcient for bringing about positive change. Despite the fact that women have equal access to education and are becoming dominant in traditional male occupations, they remain economically disadvantaged relative to men. Men continue to be politically, economically, and socially dominant because the ideological dimensions of gender hierarchies are not fundamentally challenged. Thus, she argues that the Bank needs to revisit its conceptual analysis of gender. For Barriteau, material and ideological relations of gender are relations of power. The former concerning the relative access to resources by women and men, the latter concerning the social meanings of masculinity and femininity. Dismantling gender inequality requires a consideration of both. This section ends with a comment by Laura Parisi. She concludes that rather than promoting gender equality, the World Bank’s approach reiﬁes masculinist neoclassical/liberal ideology. Observing that the term “engendering” can also mean “facilitating” or “enabling,” she argues that its use in the title of the report serves to divert attention away from the intersections of gender, race/ethnicity, class, and sexuality in the development dialogue. The report focuses on promoting neoclassical economic development rather than gendering development processes. Like women in development frameworks (WID), the World Bank seeks to include women in existing development projects rather than challenging the notion of development itself.

Explorations: future directions of feminist research The chapters in this section of the book brieﬂy explore further directions for feminist economic research, activity, cooperation, and dialog on engendering development. It begins with an essay by Drucilla K. Barker, which explores some of the methodological challenges and dilemmas facing feminist scholars and practitioners who engage with large development institutions. She brieﬂy summarizes some of the history of feminist economic engagement with gender, development, and transnational institutions. Throughout this history, feminist policy-makers and academics have endeavoured to start from the perspective of poor women and other marginalized people. The question is, how can they, located as they are in places of privilege such as academe or transnational organizations, discern what this perspective is. This chapter considers some possible answers to this query and concludes with a discussion of disciplinary and institutional power in feminist economics. Feminist economists are both insiders and outsiders in institutions such as the World Bank, a contradictory position

Feminist economics and the World Bank 9 that ultimately makes intervention and transformation possible. Some possible transformations and interventions are discussed in the next three chapters. Mariama Williams argues for feminist intervention in the coherence framework being constructed by the World Bank, the International Monetary Fund (IMF) and World Trade Organization (WTO). The coherence framework coordinates the activities of these institutions in order to ensure consistency and complementarity in international economic policy-making, and it promotes trade liberalization as a tool for achieving economic development and growth. This should be troubling to feminists because the legally binding trade agreements of the WTO are already superseding national macroeconomic policies, regulatory regimes, and social welfare policies. The coherence framework being pursued will speed up and strengthen this process. Williams argues that the farreaching inﬂuence of the policies prescribed by the World Bank, the IMF, and the WTO, backed up by a coherence framework, will not guarantee sustainable development nor will it foster a gender-balanced approach to economic growth. Feminist economists need to analyze and promote awareness of the impact of the coherence framework on social reproduction and social relations. The chapter by Brigitte Young provides an insight into promises and limitations of integrating gender into a high-level parliamentary report. In response to increasing controversies regarding the political, social, and economic effects of globalization, the German Parliament set up a commission to systematically investigate issues relating to globalization. Its mandate was to study the economic, social, and political impacts of globalization. Gender-mainstreaming was central to this mandate. Young was the only woman on the commission and provides the readers with an insider’s view of the processes through which gender was incorporated into a parliamentary report. Diane Elson’s chapter concludes this section. She critiques the treatment of women’s rights in Engendering Development on both a conceptual and an empirical level and offers an alternative account. The articulation of women’s rights in the human rights declarations and treaties of the United Nations, particularly the Convention on the Elimination of All Forms of Discrimination Against Women (CEDAW), informs the conceptual alternative. In this context, women’s rights are ends in themselves and subordinate to economic policy. Her empirical alternative focuses on outcomes rather than inputs and brings in new data sources to evaluate the progress. The evidence presented belies the report’s rosy picture of progress toward gender equality and suggests instead that there has been regress rather than progress for women’s economic and social rights in many countries. Having no interest in rosy pictures, the authors in this anthology present complex and nuanced analyses of the social relations of gender in a global context. We hope these insights will inform policy-making and motivate further research.

10 Drucilla K. Barker and Edith Kuiper

References Wolfensohn, James D. (2001) “Foreword”, in Engendering Development: Through Gender Equality in Rights, Resources, and Voice, New York: Oxford University Press. World Bank (2001) Engendering Development: Through Gender Equality in Rights, Resources, and Voice, New York: Oxford University Press.

Part I

Gender and the World Bank An institutional history and assessment

The World Bank, development, adjustment and gender equality Zaﬁris Tzannatos

Introduction Though the world is integrating in many respects and location does not matter, capital constraints and gender inequality still matter big time in developing countries which are the focus of operations of the World Bank (from now on, the Bank). And location may also matter for the Bank: being situated between the International Monetary Fund and the White House, the Bank has not infrequently attracted the charge that it is an agent of uncritical liberal policies or an arm of US economic and political interests and more broadly foreign policy (Wade 2000). Thus, the “initial conditions” for an objective or at least a generally acceptable assessment of what the Bank does in the area of gender (or any other area) are not promising: the Bank’s resources are limited,1 critics of mainstream economics are many (though often short of providing a unified alternative), and the efforts by the remaining superpower to inﬂuence international outcomes are on the increase especially post-September 11, 2001. Against this context it is easy to come up with a verdict of failure. This is so because, first, the scale of poverty (hence development) is huge: two billion people live on less than $2/day. Second, any deviation even from the simplistic notion of what unfettered markets can achieve in theory must encounter the difﬁcult issue of what the correct mix of public/private shares should be, when and in what sectors. Third, the charge that the Bank can swing at the wish of any single government needs to take into account the Bank’s collective governance structure in which practically all countries, industrialized and developing ones, are represented.2 Finally, the Bank is a changing institution, and its assessment – friendly or critical – must take into account the time period in which evaluations are undertaken: the recent statement “The Washington Consensus has been dead for years. It’s been replaced by all sorts of other consensuses” may have come late but is no less than a revolution in the Bank’s history (Wolfensohn 2004).3 These methodological, political and philosophical issues have not been resolved in the six decades of the Bank’s existence and are beyond the ability of a single chapter to summarize them or come up with deﬁnitive conclusions. In less ambitious terms, this chapter focuses on the activities and approach of a living

14 Zaﬁris Tzannatos international organization, the Bank, which tries to accomplish its mission of poverty reduction amidst conditions of changing economic realities, improvements in knowledge (including richer economic analyses), political realignments and international ﬂux. The chapter starts with a review of the Bank’s genesis and its changing focus from reconstruction and structural adjustment to poverty reduction and more comprehensive approaches to development. It then focuses on the evolution of gender issues within the Bank in terms of philosophy, internal organization and activities. Given that many criticisms of the Bank evolve around the issue of human rights, the chapter discusses how the Bank’s economic mandate positions it versus a rights-based approach. Finally, the chapter summarizes some assessments of the Bank’s involvement with gender issues.

The Bank Genesis and structure The World Bank was created at the United Nations Monetary and Financial Conference held in Bretton Woods, New Hampshire in 1944, to facilitate the ﬂow of international capital to a war-torn Europe.4 Following the successful growth of Europe, the Bank subsequently expanded its operations in developing countries and the ofﬁcial name became “International Bank for Reconstruction and Development” (IBRD).5 The name “World Bank” came to be used for the International Bank for Reconstruction and Development (IBRD) and the International Development Association (IDA – see below). These two organizations provide low-interest loans, interest-free credit, and grants to developing countries. The World Bank has now evolved into the World Bank Group (WBG) that consists of ﬁve closely associated institutions (the year is when each institution was established): 1 2 3 4 5

The International Bank for Reconstruction and Development (IBRD) (1945); The International Finance Corporation (IFC) (1956);6 The International Development Association (IDA) (1960); The International Centre for Settlement of Investment Disputes (ICSID) (1966); The Multilateral Investment Guarantee Agency (MIGA) (1988).

All five specialize in different aspects of development but all focus on the “overarching goal” of the Bank, which is poverty reduction. The Bank is owned by 184 countries and is run like a cooperative.7 Member countries are represented by a Board of Governors. The Governors (typically ministers of ﬁnance or development) carry ultimate decision-making power in the Bank. They meet annually to decide on key Bank policy issues, admit or

World Bank, development and adjustment 15 suspend country members, decide on changes in the authorized capital stock, determine the distribution of net income, and endorse ﬁnancial statements and budgets. Because the Governors meet only once a year, they delegate speciﬁc duties to their Executive Directors. The Bank currently has 24 Executive Directors who work onsite and constitute a “resident board” based in its headquarters in Washington, DC. The five largest shareholders (the United States, Japan, Germany, the United Kingdom and France) have one Executive Director each as do some others (such as China, Russia and Saudi Arabia). Other member countries are represented in groups (“multi country constituencies”) by the remaining Executive Directors. The Executive Directors and the President of the Bank – who serves as Chairman of the Board – are responsible for the conduct of the Bank’s general operations and perform their duties under powers delegated by the Board of Governors. While the Bank is best known as a ﬁnancier, one of its most important roles is in the provision of analysis and advice. In this respect the role of the Bank as an “intellectual actor” has been significant. In short, one can think of the Bank today not narrowly as a bank but rather a United Nations’ specialized (that is, development) agency. It is a complex institution as (a) its operations cover practically every aspect of development; (b) despite its focus on economics, its cooperative nature implies that its Executive Directors (acting on behalf of the Governors) bring into the Bank’s policies and operations the political views of the countries they represent; and (c) political considerations aside, the Bank is often stranded by conflicting guidance provided by economics – the dismal science.8 Internal dissatisfaction with the effectiveness of its operations as well as external criticisms coupled with changes in the economic paradigm (from Keynesianism and fixed exchange rates to monetarism and structural adjustment and then adjustment with a human face) have led the Bank to evolve over time and also to introduce internal and external oversight and monitoring mechanisms which range from objective assessments to (what amounts to) policing. More speciﬁcally, the Bank has a number of internal and external monitoring systems which include: 1 2

3 4

The Operations Evaluations Department (OED) which reports directly to member countries on the Bank’s performance;9 An independent inspection panel which responds to concerns of people affected by Bank projects and ensures that safeguard policies are being enforced; An Internal Quality Assurance Group (QAG) which monitors projects during implementation when adjustments can still be made; An Internal Auditing Department which oversees risk management and internal controls in accordance with the Standards for the Professional Practice of Internal Auditing of the Institute of Internal Auditors; A corruption and fraud investigation unit (called the Department of

16 Zaﬁris Tzannatos

6 7

Institutional Integrity, INT) that investigates allegations of fraud and corruption within the Bank Group or in connection with Bank-ﬁnanced contracts, and any allegations of unethical behavior by Bank staff; The Ethics Office which reports directly to the Bank’s President and is responsible for outreach and communications on ethics matters; and A Quality Assurance and Compliance Unit which gives advice on safeguard issues in projects and ensures the consistent implementation of safeguard policies, such as the policy on indigenous people or the policy on involuntary resettlement.

In addition the Bank undertakes periodic assessments of its overall operations as well as of more general development assistance. For example, in 2002 the Bank undertook a study of the effectiveness of development assistance over the past half-century (World Bank 2002a).10 Through its own learning based on the results of these monitoring and evaluation mechanisms, as well as changes in economic realities and thinking, the Bank started moving away from supporting narrow infrastructure projects, and its mission evolved into poverty reduction incorporating more explicitly the gender dimension, as the next sections indicate. But the process has been neither linear nor smooth. The Bank went through a phase in the 1980s when its policies were derived from what came to be known as the “Washington Consensus”11 while its operations focused on structural adjustment without explicitly incorporating social concerns (in a belief that the benefits of macroeconomic adjustment will quickly “trickle down” to the social sectors).12 Then, by 2004, the Washington Consensus was declared dead. This long journey is discussed in the next sections. Changing focus: from reconstruction to adjustment and poverty reduction Postwar international economic order was very much shaped by the International Monetary Fund (IMF), the General Agreement on Tariffs and Trade (GATT) and the Bank. The creation of the former two organizations was based on rather clear justiﬁcations. The monetary disorder of the interwar period had taught policymakers the value of international monetary cooperation. Similarly, there was a need for a multilateral mechanism to induce “closed economy/inward looking” governments to commit themselves to nondiscriminatory trade policies. Thus the debate surrounding the creation of the IMF and GATT was based primarily on how they should operate, not over what they should do. In contrast, the writing of the Articles of Association (constitution) of the Bank required a lot of elbow grease: why is there a need for a multilateral institution to lend to governments when capital had been traditionally channeled through private capital markets or loans and grants provided on a bilateral basis? Thus the debate surrounding the creation of the Bank was (and still remains) very much about what it should do, not just how it should do it. The four add-on

World Bank, development and adjustment 17 institutions to what was originally one World Bank and the switching of focus from reconstruction to development to adjustment and later on to poverty reduction and a more comprehensive approach to development attest to the struggle of an organization that tries to deﬁne what it does. A critical junction in the Bank’s history was the appointment as its president of Robert McNamara, formerly US Secretary of Defense during the Vietnam era. During his tenure at the Bank (1968–81), the Bank shifted from an emphasis on infrastructure to agriculture and rural development in an attempt to address people’s basic needs, particularly the rural poor. Poverty was adopted “big time” into the Bank’s thinking in 1972 backed by the admission that 40 percent of the world’s population lived in absolute poverty. During the rest of the decade the Bank’s work on poverty was very much linked to human development which culminated with the 1980 World Development Report13 on Poverty. During the McNamara era, the Bank promoted the view that directing investment (in basic infrastructure and human resources development) toward the poor through redistribution would contribute to economic growth as well as to poverty reduction. It should be noted that it was during McNamara’s era that the Bank ﬁrst introduced structural and sectoral adjustment lending (in 1979)14 though the Bank was not then part of the general movement towards supply-side economics that was already taking place in the academic profession since the 1970s. The adoption of supply-side economics came to be a prominent feature of the Bank in the 1980s. An appropriate instrument (adjustment lending) was already in place when the steady advance of poverty reduction was halted by the global economic crisis of the early 1980s, soon after McNamara left the Bank. In fact adjustment (or policy-based) lending as a broad instrument of aid was widely used by the Bank during the early years after World War II, but thereafter the focus shifted to investment lending. In the years after McNamara left the Bank, which coincided with the Ronald Reagan presidency in the United States, program lending was resurrected and poverty was moved to the back burner (Stern and Ferreira 1997: 549). Under a new president (A. W. Clausen, 1981–86) the Bank changed its poverty focus and gave less prominence to measures hinging on ideological stances (such as population control). Attention switched mainly on to achieving growth through structural adjustment. Adjustment, in terms of its theoretical justiﬁcation or results on the ground, became a thorny issue for the Bank in the 1980s. The charge here is that the Bank directed reforms in a specific direction (the Washington Consensus) through its lending program, policy dialogue, and selective research (Edwards 1997). The developing countries’ massive move toward trade liberalization and structural adjustment in the late 1980s and early 1990s has been associated by many analysts with the role of the Bank (and the IMF). In addition, there were “less impressive performances” of the Bank throughout the 1980s associated with the mounting levels of debt in developing countries (where more stringent monitoring by the ﬁnancial institutions was called for), the reluctance of client countries to adopt the Bank’s policies (bypassed through “conditionality”),

18 Zaﬁris Tzannatos pessimism that the resumption of growth will be quick (slow “trickle down”) and a neglect of environmental issues. Somewhat paradoxically, in the 1980s there was little emphasis on cost-beneﬁt analysis and even that was in decline (Little and Mirrlees 1991) notwithstanding the fact that the Bank’s focus was to “get the prices right.” The worse-than-anticipated effects of adjustment policies, if not noticed internally, were noticed by others. For example, UNICEF’s Adjustment with a Human Face (Cornia et al. 1987) and UNDP’s Human Development Report (1990) called for a richer approach to development than just focusing on structural adjustment programs in a hopeful pursue of economic growth. In fact, during the tenure of the next president (Barber B. Conable, 1986–91), the Bank continued seeing economic growth brought about by conventional policies as a precondition for reducing poverty but also acknowledged that supplementary actions were necessary. This was reflected in the 1990 World Development Report as a two-part strategy, namely, to increase the income of the poor through broad-based growth and ensure that the poor had access to basic social services. This World Development Report also noted transfers and safety nets as important complements to the basic two-part strategy.15 The fiftieth anniversary of the genesis of the Bank (1994) provided an additional impetus for renewing demands for a radical transformation in the Bank’s approach and practices. In the meantime (1992) “a damning document came to be written” (Kapur et al. 1997). The document was commissioned by yet another president of the Bank (Lewis Preston, 1991–95). The so-called Wapenhans Report (Effective Implementation: Key to Development Impact, 1992) was an in-house assessment of the Bank’s entire loan portfolio (World Bank 1992b). The gist of the Wapenhans Report was that a large and increasing number of Bank projects were failing, even when assessed only on narrow economic criteria. The “50 Years is Enough” campaign and the appointment of a new president (James D. Wolfensohn, 1995–2005) resulted in greater attention being given to openness, accountability, debt reduction, participatory approaches, institutions and an end to environmentally destructive lending and structural adjustment. The experience with adjustment lending and its mixed outcomes during this period lent impetus to the Bank’s internal learning process. Several OED and Bank reviews analyzed the economic and social impact of adjustment and fed back the lessons learnt into the design of adjustment operations. Today, internal evaluations no longer shy away from admitting that the performance of adjustment operations has been mixed, especially during the 1980s.16 They also indicate that heavy reliance on conditionality was ineffective for several reasons: it can be difﬁcult to monitor whether a government has in fact fulﬁlled the conditions, particularly when external shocks muddy the picture; governments may revert to old practices as soon as the money has been disbursed; and when assessments are subjective, donors may have an incentive to emphasize progress in order to keep programs moving. They ﬁnally notice the difﬁculty in securing country commitment to reform: naturally, governments change over time and do not

World Bank, development and adjustment 19 Percent 100

0 1974

1977

1980

1983

1986

1989

1992

1995

1998

2001

Year

Figure 2.1 Projects rated as achieving satisfactory outcomes, 1974–2001. Source: World Bank, internal evaluations.

feel obliged to fulﬁll their predecessors’ policies. It is thus not surprising that only 60 percent of adjustment operations achieved (per OED evaluations) satisfactory or better outcomes in the 1980s, a ﬁgure that rose to above 80 percent in the 1990s following design improvements and increased country selectivity based on lessons of experience (Figure 2.1). Similarly, the issue of insufﬁcient attention to the social sectors, which was a particular problem of early adjustment operations, is now given more attention. Though the Bank is still thoroughly committed to the statement that “on balance poor people beneﬁt from adjustment and reform as much as other groups,” it does now recognize that “even successful, market-friendly reforms have beneﬁts and costs that are distributed unevenly, especially in the near term, and the design and implementation of adjustment operations need to take this factor into account.” Irrespective of whether the Bank could have done better in the past or can do better today, it is very much a different organization than it was in the 1980s. In the 1980s the Bank’s support for education, health, nutrition and other social services accounted for less than 6 percent of its lending but exceeded 18 percent in the 1990s (World Bank 2001a: 5). Today the Bank is the world’s largest external funder of education and of the fight against AIDS, while it remains among the most important ones of health programs. Similarly, since 1988, the Bank has become one of the largest international funders of biodiversity projects and its environment strategy covers from poor people in rural areas in developing countries to broader issues of climate change, forests and water resources.

20 Zaﬁris Tzannatos In 1996, the Bank and the International Monetary Fund (IMF) launched the Heavily Indebted Poor Countries (HIPC) Initiative to reduce the external debt of the world’s poorest, most indebted countries, thus enabling them to reorient their budgetary priorities toward key social and human development sectors.17 The way the Bank is doing business also changed over time with increasing reliance on stakeholder participation and partnerships, both local and international.18 Unlike in earlier periods when most projects were prepared and supervised from Washington, the Bank has now ofﬁces in 70 countries around the world that reach out to and collaborate with nongovernmental organizations (NGOs) in a variety of areas ranging from education to the environment and AIDS. In recent years, more than two-thirds of development projects approved by the Bank involved the active participation of NGOs in their implementation, and most of its country strategies beneﬁted from consultations with civil society. The Bank has also increased its activities in the areas of corruption and conﬂict-affected economies.19 More recently (1999) the Bank introduced the Poverty Reduction Strategy Papers (PRSP) approach which was approved by the Boards of both the IMF and the Bank. This approach is based on poverty reduction strategies prepared by the countries themselves and serve as a framework for development assistance. The underlying goals of the PRSP are to ensure broad-based country ownership of poverty reduction strategies; develop strategies that take a comprehensive, long-term perspective; focus on results that matter for the poor; and build stronger partnerships between low-income countries and the international donor community. An internal Bank review of the PRSP process claimed that there seems to be widespread agreement (among national PRSP teams, donors and civil society organizations) that in low-income countries with active PRSP programs there is an increased emphasis on “pro-poor spending.” While these results are encouraging, this is obviously a narrow measure of success. It is important to bear in mind that these budgetary reallocations will translate into improved poverty outcomes only if there are corresponding improvements in budget execution and service delivery. All this does not mean that the situation is rosy. A more interesting issue relates to whether the much-heralded international debt relief policy in the 1990s had an impact on economic growth. A recent independent study found that international debt relief in the 1990s had only limited efﬁciency, effectiveness and relevance (Ministry of Foreign Affairs (The Netherlands), 2003). The amounts of debt relief were too small to start with, were given in the wrong modalities and were accompanied by too many new loans. Only in a minority of countries has debt relief led to visible stock or ﬂow effects and in none has the debt burden become sustainable. Debt relief has rarely contributed to economic growth, particularly in the poorest and most highly indebted countries. This may suggest that the inﬂow of new loans from multilateral institutions to HIPC threatens their long-term debt sustainability. There is an element of déjà vu in all this as it resembles the case of structural adjustment: sound framework and intentions but perhaps too good to be true given the inherent conﬂict of

World Bank, development and adjustment 21 interest when the IFIs (International Financial Institutions) act as both gatekeepers and creditors, a conﬂict that entails the risk that countries receive aid precisely because they have already a high debt burden.

The Bank and gender equality The Bank’s thinking and activities on gender (or other thematic) areas20 can be only as good as its overall philosophy and quality of operations. In situations like those outlined in the previous section, when the Bank shifted to broad macroeconomy issues accompanied by an ideological belief that all that is needed is to just get some liberal economic fundamentals right, gender issues were sidestepped. Indeed, the Bank has been relatively slow in addressing the role of women and broader gender issues in the economic development process. Academic attention on gender issues may be said to have started with Esther Boserup’s seminal publication on Women’s Role in Economic Development (1970). Around that time, gender anti-discrimination policies started being introduced (at least in the majority of OECD economies). The Bank created a position of an Advisor for Women in Development only in 1977 (Table 2.1). It then gradually started paying some attention to gender issues and issued the Operational Manual Statement 2.20 21 (OMS 2.20) in 1984 (World Bank 1984). This statement presented an overview of the general objectives and key features of the appraisal of investment projects. Within this broad framework, the statement discussed how to investigate and address the impacts of investment (not adjustment) projects on women.22 The Bank’s regularly published statistical compendium (World Development Indicators) included information on some female indicators (such as maternal mortality, education and health) for the ﬁrst time in 1988 at a time when only one-quarter of Bank reports included references to women in development. Then things starting changing relatively fast, for example, two years later substantial discussion or recommendations on women in development were included in 62 percent of all economic and sector reports (ESW) produced by the Bank. This was partly because of external pressure either directly (for example, the “50 Years Is Enough” Campaign) or indirectly (for example, the publication of critical research on gender, adjustment and macroeconomics). Viewed from heterodox macroeconomics and feminist analyses, the Bank’s operations did not score highly (Çagatay et al. 1995). A special issue of the journal World Development 23 contained papers arguing that, while progress has been made in the inclusion of reproductive work into national income accounting and gender-disaggregated analysis of structural adjustment policies and programs (Benería 1995), there were major shortcomings in the way gender issues were conceptualized (or not) in the macroeconomic approaches and programs of the World Bank (and IMF). Emphasis on gender issues started increasing relatively fast in the ﬁrst half of the 1990s – partly in preparation for the Bank’s participation in the Beijing

22 Zaﬁris Tzannatos Table 2.1 Evolution of the World Bank’s attention to gender and development issues Year

Event

1977

First Women in Development Advisor is appointed.

1984

Operational Manual Statement 2.20 addresses the impact of Bank assistance on women as part of project appraisal for certain types of projects.

1986

Women in Development (WID) Unit is created.

1990

Operational Directive (OD) on poverty reduction recommends that women’s issues be considered in designing poverty reduction programs.

1994

In the Policy Paper Enhancing Women’s Participation in Economic Development the shift to gender analysis becomes visible. Operational Policy (OP) 4.20 is issued establishing the goals of reducing gender disparities and enhancing women’s participation in economic development.

1995

The publication Toward Gender Equality: The Role of Public Policy explicitly recognizes that Governments can play a signiﬁcant role in the reduction of gender inequalities.

1996

External Gender Consultative Group is established.

1997

Gender and Development Board is established.

1998

Position of head of the Gender and Development Board is raised from a chief/manager to a director level.

2001

The publication of Engendering Development: Through Gender Equality in Rights, Resources, and Voice provides an analytical framework for gender inequalities and derives implications for relevant policies. Publication of Integration of Gender Into the World Bank’s Work: A Strategy for Action.

2003

Operational Policy 4.20 revised. Bank Procedure 4.20 statement issued.

Conference (1995).24 By 1994, the Bank’s Board endorsed for the ﬁrst time a Policy Paper in the area of gender (Enhancing Women’s Participation in Economic Development), and the Bank issued an Operational Policy statement on the gender dimension of development (Operational Policy 4.20). The Bank’s policy was to focus operationally on five sectors (agriculture, education, health, financial services and labor) but did not – and still does not – require that all investment programs address gender issues. Structural adjustment programs – representing at least one-third of total Bank commitments – were not included under the Operational Policy.

World Bank, development and adjustment 23 In 1995 the Bank prepared a report Toward Gender Equality: The Role of Public Policy which followed a simple but powerful line.25 It ﬁrst observed that gender inequalities persist or, alternatively (though not so bluntly) that “trickle down” takes time. It then discussed how inequalities hamper growth and concluded that public policies can have a signiﬁcant effect on development. More importantly, the report argued that such policies were not to be restricted only to economic ones but to be extended to other areas (such as the legal framework). The report was part of the preparation for the Bank’s participation at the Fourth World Conference on Women held in Beijing in 1995 and was followed by another report (Advancing Gender Equality: From Concept to Action) also published in the same year. The latter report provided an overview of the Bank’s support to the efforts of member countries to integrate gender equality into the development efforts and outlined how gender issues are identiﬁed and incorporated into Bank-supported projects and programs. In its “Beijing+5” report (World Bank 2000) the Bank noted three main points: disparities persist; they are costly; and rights and institutions are important complements to growth policies. The report recommended establishing legal, economic and social institutions, including equality of rights, promoting growth and undertaking active measures to promote equality in the command of resources and political voice. Though these recommendations were not derived from a human rights perspective, they did nevertheless acknowledge the link between economics and rights – an issue discussed below. The link between gender equality and poverty reduction was forcefully made in another publication (the 2000/2001 World Development Report, Attacking Poverty) that argued that gender relations affect all aspects of poverty, including income, opportunity, security and empowerment (World Bank 2001b). Examples cited were that girls in poor families may receive lower quality nutrition, less health care and inferior education than their brothers. Likewise, female household members often have less access to and control over the household’s productive resources and income than do male family members. Overall, the report questioned at times the relevance of the standard neoclassical economic theory (whereby households are usually regarded as sharing a single utility function and an equitable distribution of resources and well-being) while results of studies based on less conventional paradigms were openly acknowledged. By far the most important analytical basis for the Bank’s involvement in gender issues was included in the report Engendering Development: Through Gender Equality in Rights, Resources, and Voice (World Bank 2001c).26 The subtitle is in itself indicative of the Bank’s broadening its approach to the study of gender issues. The report focused on gender inequalities and their economic and social implications in developing and transition economies. It examined the conceptual and empirical links between gender equality, public policy and development outcomes, and made a strong case for applying a gender perspective to the design of development policies. It argued that societies that discriminate by gender pay a high price in terms of their ability to develop and reduce poverty. The report recognized that

24 Zaﬁris Tzannatos there is growing evidence that several aspects of gender relations – the gender-based division of labor, disparities between males and females in power and resources, and gender biases in rights and entitlements – act to undermine economic growth and reduce the well-being of men, women, and children. Gender-based divisions of labor and gender inequalities also contribute to poverty. The report spent some time in identifying (or restating ﬁndings from previous research) that the primary channels through which gender systems affect growth are the productivity of labor and the allocative efficiency of the economy (speciﬁcally through investments in human capital, especially girls’ and women’s education27 and health), investments in physical capital (especially women’s access to capital) and the functioning of markets and institutions.28 The report also emphasized the beneﬁcial effects on improving women’s rights, resources and voice (World Bank 2001c: Chapter 2) as well as the role of infrastructure, equal access to productive assets and resources, and quality of governance. As far as the implications for countries are concerned, the report proposed a threepart approach that emphasizes institutional reforms based on a foundation of equal rights for women and men, policies for sustained economic development, and active measures to redress persistent gender disparities. The Bank’s gender strategy By the start of the millennium, the World Bank was in a position to recognize from its own learning that its effectiveness in helping member countries can increase by paying more attention to gender issues. The time was ripe to come up with a more coherent and holistic gender approach. This materialized with the discussion of the Bank-wide gender mainstreaming strategy at the Board of Directors in 2001 (World Bank 2001d). The Gender Strategy called for a selective integration of gender issues into the Bank’s work based on the readiness of countries. It thus adopted a soft-pedal approach29 towards the clients (developing countries). It proposed to work with governments and civil society in client countries, and with other donors, to diagnose the gender-related barriers to and opportunities for poverty reduction and sustainable development; and will then identify and support appropriate actions that will reduce these barriers and capitalize on the opportunities. The Gender Strategy thus envisages that the Bank should play a supportive vis-à-vis member countries creating an enabling environment. More speciﬁcally, the basic process for the implementation of the Gender Strategy has three steps: preparation of periodic, multi-sectoral Country Gender Assessments (CGAs – see below); development and implementation of relevant policy and operational interventions; and monitoring the implementation and impacts of these policies

World Bank, development and adjustment 25 and interventions. The Bank thus would play an advisory role towards creating an enabling environment and, where governments choose particular genderresponsive actions involving Bank assistance, the Bank would provide additional support for changing the gender patterns that are deemed to be costly to growth, poverty reduction and human well-being. The process of mainstreaming gender issues within the Bank is to be implemented through four actions: integration of the gender dimension into analytical work; support of the integration of gender issues into Bank operations; alignment of resources with the new Gender Strategy; and development and implementation of systems to monitor and evaluate processes and impacts.30 The implementation of the Gender Strategy thus involves a potentially wide range of activities that range from analytical work to policy advice and lending. The Country Gender Assessments A key component of the Gender Strategy is the Country Gender Assessment (CGA), that is, a study analyzing country-level gender critical areas in which gender-responsive actions are likely to enhance growth and reduce poverty. A CGA would normally include: •

•

A proﬁle of: (i) the different socio-economic roles of males and females, including their participation in both the market and household economies; (ii) gender disparities in access to, control over and use of assets and productive resources; (iii) gender disparities in human development indicators; (iv) inequalities between males and females in the ability to participate in development decision-making at the local and national levels; and (v) laws, institutional frameworks, norms and other societal practices that lead (implicitly or explicitly) to gender discrimination and/or gender inequality; the country context, including the country’s policies, priorities, legal and regulatory framework and institutional arrangements for implementing its gender and development goals; and a set of gender-responsive priority policy and operational interventions that are important for poverty reduction and development effectiveness.

The CGA may be a stand-alone document or a section of another report. It may contain original analytical work or may simply refer to such work, produced by the Bank or by other agencies (government, international, academic). Operational Policy/Bank Procedure on Gender The developments since the mid-1990s culminated with the issuing of a new version of the existing Operational Policy (OP) as well as the introduction of a Bank Procedure (BP) on Gender and Development (OP/BP 4.20). The foundation for both the OP and the BP was indeed the Gender Strategy, especially

26 Zaﬁris Tzannatos the basic process of creating an enabling environment for country-led, countryspeciﬁc strategies with the Country Gender Assessments as the basic instrument. OPs are Bank policies that follow from the Bank’s Articles of Agreement and the general conditions and polices approved by the Board of Executive Directors. They establish parameters for the conduct of operations and are focused statements of policies. The OP on Gender and Development (OP 4.20) includes a statement of philosophy, clearly linking the OP to the Bank’s poverty mandate and emphasizing the need for sustainability. The OP also includes a statement of policy for all Bank assistance, detailing how the Bank will work with its clients, and a commitment to monitoring and evaluation. The new OP ﬁnally stresses the need to cross reference other relevant Bank policies, such as those on poverty reduction, investment and adjustment lending, and project supervision. The objective of Bank Procedures ( BPs) is to explain how Bank staff are to carry out the policies set out in the OPs. Hence, BPs spell out the procedures and documentation required to ensure Bank-wide consistency and quality. The BP 4.20 focuses on the procedural steps for the preparation of CGAs and lays out the minimum content of a CGA. The BP also prescribes that consultation with country stakeholders is a central feature of the CGA process. Included in the BP are also steps for integrating the CGA ﬁndings into the Bank’s assistance program for a country and for monitoring and evaluation. In conclusion, the Strategy, the Country Gender Assessments, the Operational Policy and the Bank Procedure provide a clear framework and instruments for the Bank to deal with gender issues in the context of development and poverty reduction. The internal arrangements for dealing with gender issues are examined separately below. Gender and family issues within the World Bank The internal treatment of gender equality was naturally perceived by the Bank as key to its credibility when discussing gender issues with developing countries in the sense of aligning preaching to practices. In 1992 the Bank commissioned an internal report (the Stern Report31) that provided its first action plan for recruiting, retaining, and promoting women at decision-making levels. It had no less than 33 recommendations and introduced numerical targets. For example, the ratio of women in middle level management jobs was to increase to 27 percent by 1997 (a target reached before that date) while the target to increase women in top jobs was to increase from 7 percent to 15 percent (this was met by 1997). The report also recommended more attention to work-life issues. Many of the ideas of those days have subsequently become reality as AWS (Alternative Work Schedule),32 part-time work options, telecommuting, and supportive services such as advice on child care, elder care, housing, education.33 Additional recommendations related to sexual harassment. Since 1992, the Bank has adopted strong sexual harassment policies and created a network of Anti-Harassment Advisors as well as an integrated Conflict Resolution System; and has provided guidelines on Working with Respect in the Bank Group

World Bank, development and adjustment 27 to every staff member. Many of the gender issues are now rightly addressed in the broader context of diversity as many of the strategies for improving women’s representation are applicable to other subgroups. There was a follow-up to the Stern Report in 1998 (World Bank 1998b). The new report reviewed pay, promotion and performance with a view to identifying and redressing remaining gender differences among staff. It was based on econometric analysis and also a qualitative assessment undertaken by external (consulting) ﬁrms. The report noted that while differences in pay still exist, there was a marked decline in the unexplained part of salary gender differentials from about 17–25 percent in 1992 down to 6 percent in 1998. It identiﬁed that the main source of current differences is not discrimination in promotion but at the recruitment stage (grade assignment at entry and resulting salary offers).

Comparisons and evaluations The International Finance Corporation (IFC) Recall, the World Bank refers conventionally to two organizations: IBRD and IDA. The discussion so far referred to these two units but the role of the private arm of the Bank (IFC) is potentially increasing in the era of globalization and the greater role of the private sector. Though the Bank and IFC share the same Board of Directors, their policies are not always identical. For example, IFC does not have an operational policy on gender issues. Still, all IFC projects must comply with applicable environmental, social and disclosure policies designed to ensure that the business areas in which it invests are implemented in an environmentally, socially and ﬁnancially responsible manner. The IFC’s manual on public consultation (World Bank 1998a)34 explicitly states that consulting women and being gender sensitive are key components of a successful consultation program. In terms of monitoring results, IFC focuses on areas such as job creation for females working in industry and small-to-medium enterprises; access to credit for poor and low-income women; increased access to affordable education for poor and middle-income females; and increased opportunities, advice and ﬁnancing for female entrepreneurs. Thus, technically, the Bank’s Gender Strategy is of limited relevance to the way IFC conducts its business. For example, IFC does not have to conduct the same kind of countrylevel analytical studies that the Bank (IBRD and IDA) conduct. Still, there may be areas where the Bank and IFC can collaborate with IFC, for example on issues of common interest such as female entrepreneurship or women-owned businesses. Comparison with other multilateral and bilateral development agencies The Bank’s current approach to gender equality and its Gender Strategy compare favorably to what other multilateral and bilateral organizations do (see Table 2.2).

28 Zaﬁris Tzannatos Table 2.2 Gender policies of multilateral donor agencies Institution

Gender-related objective

AfDB

Increased support for functional literacy programs; elimination of gender disparity in primary and secondary education; infant and maternal mortality; one of two cross-cutting issues in four areas of focus. The focus is on women.

ADB

Improved status of women and elimination of discrimination is one of ﬁve strategic objectives. From 1998, it adopted gender mainstreaming as the key strategy to promote “ gender equity,” but the focus remains the improvement of the status of women with concurrence and cooperation of men.

EBRD

No gender considerations explicitly indicated.

IDB

Fuller integration of women into all stages of the development process and improvement in their socioeconomic situation through a gendered analysis.

UNDP

Gender equality and advancement of women. Focus on equality of capacities and opportunities to ensure that women and men participate equally in the development process as agents and beneﬁciaries.

Reduced gender disparity and increased women’s participation in economic development as part of overall poverty reduction strategy.

Source: World Bank (OED), 2001a: 7.

In fact, the Bank was the first multilateral donor organization to mandate consideration of impacts on women as part of project appraisal. However, other multilaterals, like the Asian Development Bank (ADB), African Development Bank (AfDB), and Inter-American Development Bank (IDB) have adopted policies and process-related goals focused on increasing economic opportunities and reducing barriers to the economic and social well-being of women. Their approaches are justiﬁed, as in the case of the Bank, on the grounds of efﬁciency, equity and poverty reduction. Evaluations The Operations Evaluation Department (OED) undertook two recent assessments in the area of gender equality: Integrating Gender in World Bank Assistance (World Bank 2001a) and The Gender Dimension of Bank Assistance: An Evaluation of Results (World Bank 2002b). Both noted progress and identiﬁed areas of improvement. The ﬁrst report examined how the Bank has performed at country level. It noted that, on the positive side, gender issues have been better integrated in the Bank’s work since the 1994 policy was introduced. At regional level, some regions (such as Africa and South Asia) achieved better integration in their

World Bank, development and adjustment 29 Table 2.3 Gender analysis in Bank projects by sector 1988–1999 (%) Sector group

Investment only % of projects approved with gender analysis

Investment and adjustment % of projects approved with gender analysis

Population, health and nutrition Agriculture Education Water supply and sanitation Social protection Electric power and energy Environment Finance Industry Mining Multisector Public sector management Transportation Urban development Oil and gas Telecommunications

89 56 68 34 73 7 34 16 14 12 13 11 7 30 4 0

88 53 66 34 74 7 34 12 13 17 24 11 7 29 4 0

Weighted average

Source: World Bank (2001a).

country programs. The report also noted that the Bank’s analytical work in these regions is becoming clearly more gender aware. Improvements in the other regions (such as East Europe and Central Asia as well as Latin America and the Caribbean) were either more recent or lacking. In the health and education sectors, two of the ﬁve sectors of focus identiﬁed in the 1994 strategy paper, the Bank has progressed well. In another sector, agriculture, the Bank has made progress, but there still remains a need for uniform treatment of critical issues such as access to and control over productive assets and services. The report concluded that there remains a long way to go in terms of implementation especially in the so-called hard sectors (see Table 2.3). The second report examined the results of the Bank’s activities during the previous decade. It addressed the extent to which the Bank helped to (a) reduce gender disparities in the education and health sectors; ( b) increase the participation of women in economic activities; and (c) influence institutional changes that support the advancement of women. The evaluation was based on 12 countries and, like the 2001 report mentioned above, found satisfactory results in the education and health sector. Though the report acknowledges that causality is difﬁcult to establish, it found weak or mixed results with respect to the other two areas (economic participation and institutional aspects). The report noted that projects tended to have better on-the-ground impacts when gender issues had been analyzed at the country and project levels, and genderdifferentiated needs or impacts were recognized in project and program design.

30 Zaﬁris Tzannatos More recently (2003) a monitoring report on the implementation of the Gender Strategy during its ﬁrst year (World Bank 2003a) noted that ﬁve CGAs were prepared (in addition to the 7 CGAs already completed) and another 26 were planned. This (38 CGAs) compares to about 100 countries that are active borrowers of the Bank. While this is encouraging, the report noted that of 39 core diagnostic reports prepared since the Gender Strategy was issued only seven had integrated the gender dimension into their analyses. The situation was better (in terms of increased attention to gender issues) in terms of lending, especially in selected sectors such as education, health and social protection. The report identiﬁed as challenges the need to increase the attention to gender issues in analytical and lending activities, ability to work with partners, and the evaluations not just of inputs but of outcomes. Needs versus rights The objective of the Bank’s gender and development policy is to assist member countries to reduce poverty and enhance economic growth, human well-being, and development effectiveness by addressing the gender disparities and inequalities that are barriers to development, and by assisting member countries in formulating and implementing their gender and development goals. In other words, in the economic context within which the Bank operates it assists member countries to design gender-sensitive policies and programs to ensure that overall development efforts are directed to attain impacts that are equitable and beneficial for both men and women. Still, as an international organization, the Bank may need to go further than this as it cannot ignore key universal declarations (such as, for example, the Universal Declaration of Human Rights, 1948). In addressing human rights issues the Bank has traditionally exercised caution.35 In principle, the Bank’s Articles of Association prohibit interference in the political affairs of its members. Also, the Bank’s clients are sovereign. Finally, the Bank needs to exercise restraint in its use of conditionality. Despite its economic focus, the Bank is increasingly becoming more vocal (though more often through its officers than in official forms) about the links between the economic participation of poor people and equality in dignity and rights (Stern 2003).36 A series of Bank publications (such as World Development Report 2000/01) recognize the importance of fair and inclusive institutions, empowerment, accountability, transparency and participation. The Bank supports the Millennium Declaration and the Monterrey Consensus which endorse respect for human rights as part of a multifaceted response to the development challenge. The Bank would argue that the objectives of sustainable poverty reduction, human wellbeing and development effectiveness underlie its operational policies, including speciﬁcally those aimed at promoting gender equity and protection of vulnerable

World Bank, development and adjustment 31 groups, in particular, indigenous peoples and those subject to involuntary resettlement. Two issues that need to be taken into account in assessing the Bank’s approach to human rights are, first, the Bank has no means to legislate human rights obligations for its members, nor is it nor can be the world’s policeman that enforces its members’ human rights obligations: the Bank has neither the mandate nor the instruments to do so. Second, rights-based approaches as stated by the United Nations, “with a rights-based approach, effective action for development moves from the optional realm of charity into the mandatory realm of law with identiﬁable rights, obligations, claim-holders and duty-bearers” (UN 2000) are rather categorical for an organization that seeks to develop effective and efﬁcient policies under economic constraints and budget choices. The principle of indivisibility of human rights results in the absence of guidance on how to prioritize in a resource-constrained environment which characterizes an economic organization. An unsatisﬁed need as a violation or abuse of human rights provides little guidance as to how to resolve a policy or resource constraint or to rectify what all may agree is an unsatisfactory situation. The generality and ambiguity of many human rights provisions present difﬁculties of interpretation and implementation. And many will recognize that enactment of laws alone does not ensure fair and equitable treatment of all citizens. A more relevant issue here is whether the Bank’s contributions are consistent with efforts to promote achievement of human rights. It can be argued that many of the Bank’s practices are or can be supportive, albeit indirectly, of practices which are fundamental to human rights. Such practices include, ﬁrst, the identiﬁcation of the poor (especially with respect to particular subgroups, such as women and men, children, youth and the elderly, or particular ethnic or regional groups), second, participatory approaches to project design and implementation and third, accountability and transparency of information systems.37 The Bank’s move (if only in rhetoric terms – as critics would put it) into the area of human rights was preceded by its in-house research. For example, it is now acknowledged that gender equality – not only in health and education, but also in voice and rights – is an important element in development (Goldin et al. 2003). Aside from the obvious direct beneﬁts for women, equality in these dimensions is seen having instrumental beneﬁts in terms of growth and poverty reduction. The Bank’s cross-country research conﬁrms that low investment in female education has been a barrier to growth in South Asia, Sub-Saharan Africa, and the Middle East and North Africa, compared with East Asia, which has closed the gender gap in education more rapidly. Since the 1990s, the Bank has increasingly emphasized the importance of establishing the rule of law and of the state playing a role of facilitator and enabler of development. It acknowledges that increasing the efficiency of legal and judicial systems, providing greater access to justice, and reforming the public sector at large to enhance accountability and transparency can significantly contribute to the achievement of human rights. The Bank now supports judicial reform projects that strengthen good government and good governance. Such

32 Zaﬁris Tzannatos projects are not justiﬁed on the grounds of human rights per se but because they can enhance the effectiveness of Bank support as it is the absence of information on people’s rights and means to pursue them which often prevents policies and projects from realizing their intended beneﬁts.

Summary and conclusions For more than half of its history the Bank has addressed women’s and gender issues both in development and also in its internal organization and, since the 1990s, it started making accelerating progress in integrating gender issues into its activities. For example, between 1995 and 2000 the Bank lent more than $3.4 billion for girls’ education programs, and was also the single largest lender in the world for health, nutrition and population projects, three-quarters of which contained gender-responsive actions. Attention to gender issues in various key Bank documents38 also increased during the 1990s. Several organizational changes designed to facilitate greater attention to gender and development issues were also institutionalized, including the issuing of an Operational Policy on the gender dimension of development in 1994 and its subsequent revision (2003), the creation of a Gender and Development Board (1997), the adoption of a Gender Strategy in 2001 and the issuing of the Bank Procedure in 2003. The Bank has been contributing to capacity-building and has been organizing conferences on diverse areas relating to gender issues such as laws and more broadly institutions. These typically cover issues such as gender equality and citizenship, women’s relationship to the State, gender-based violence, how laws can provide equality of rights and access, and how institutions can create the enabling environment to enforce rights. The Bank’s lending continued to be on a wide range of areas affecting gender equality such as support for artisan enterprise institutes, “women and justice” programs, community empowerment projects, water and sanitation, reducing female genital cutting, and helping poor women and girls obtain identity cards and birth certificates to enable them to “become citizens” (and, thus obtain, for example, microcredit and access to basic services). The Bank is also active in promoting gender equality through nonlending activities, including work on gender and human immunodeﬁciency virus/acquired immune deﬁciency syndrome and on emerging issues such as gender and the digital divide. A new initiative on gender and macroeconomic policy is under way that aims to identify effective actions for integrating gender into country level analysis and policy dialogue. Though the Bank was slow to start with, it moved relatively fast in the 1990s and found itself well positioned in the new millennium. Engendering development is at least being discussed and (to an extent still to be assessed) mainstreamed in the Bank’s approach and operations. Internally, there have been signiﬁcant improvements in the employment conditions and prospects of female staff at all levels as well as their partners and families. And more recently the Bank became fully committed to the attainment of the Millennium Development Goals (World Bank 2003b).39

World Bank, development and adjustment 33 The Bank has now an analytical understanding that authorizes it to get involved with gender issues through economic justifications (instead of human rights ones). It has a clear Gender Strategy and also the operational tools to put theory into practice. As a broad development institution (rather than a specialized one – such as UNIFEM) the Bank would never be able to focus 100 percent on a speciﬁc issue. This means that it will always be debatable whether dedicating 30 percent or 60 percent of its activities to gender issues would constitute little or enough attention.40 At the broadest level, the Bank’s evolution and current state is effectively the result of an organization that has a mission related to a difﬁcult (if not an unsolvable, in some specific time horizon) problem, that is, development and poverty reduction. In short, the Bank does not have a narrow technical mission. Approaches to development and poverty are characterized by analytical uncertainty and easily dominated by ideological or interest-based disagreements about goals, priorities and causation. Although in the 1980s the Bank adopted with little reservation the recipes of supply-side economics, this was not the case before and surely things are different today. Less easy to agree is whether the change has been substantial and whether progress has been too little too late. Some or many may argue that the 1980s were a lost decade for the Bank. Its belief that markets in developing countries would respond fast and efﬁciently to structural adjustment did not materialize. The collapse of the planned economies did not help either as it lead towards the belief that since one economic system was proven wrong, its theoretical alternative (that of unfettered markets) must be right. Whether it was due to the “50 Years Is Enough” campaign or the Bank’s new president in 1995 or the ﬁnancial crises of the late-1990s, the Bank today is a different institution than it was during the post-McNamara era till sometime in the 1990s. Different does not necessarily mean better and the point here is that, by focusing on the aggregate macroeconomic picture, women and development got the short stick from the Bank in the 1980s and this continued for sometime into the 1990s. The more recent treatment of gender issues at the Bank does not, however, support the view that the Bank is now gender blind. This can improve further somewhat paradoxically because of the hegemonic tendencies of some of the Bank’s stakeholders (Wade 2001). For example, the US Executive Director at the Bank in the 1990s did push hard for augmenting the Bank’s gender agenda. The ﬁrst year’s implementation report of the Gender Strategy concluded despite the progress to date in gender mainstreaming, the Bank’s effectiveness can be improved by paying more systematic and widespread attention to gender issues in the context of our poverty reduction mandate. The opportunities for improving the development impact of the Bank’s work through gender mainstreaming include making Bank interventions more responsive to country gender conditions and commitments, making these

34 Zaﬁris Tzannatos interventions more strategic, and improving the alignment of Bank policies, processes, and resources to support such interventions. In conclusion, if the economist’s answer is “the glass is twice as big as it ought to be,” then perhaps the Bank is too small to address all developmental issues. But despite past failures (partly due to reasons of biased ideology or hegemonic interferences), the Bank seems to have come a long way on gender issues. Again from an economist’s perspective “on the other hand, the Bank has some way to go.”

Acknowledgments The author would like to thank Drucilla Barker, Edith Kuiper, Karen Mason, Susan Razzaz and Haneen Sayed for useful comments on an earlier draft. Remaining omissions and errors are of the author.

Notes 1 The Bank provides approximately $20 billion/year in development assistance against another $40–50 billion provided by other multilaterals and bilaterals. This total of $60–70 billion of ofﬁcial aid is only a fraction of private ﬂows which can be more than ﬁve times the amount transferred through ofﬁcial aid in a typical year. 2 The Bank has 184 members. 3 Jim Wolfensohn was then the Bank’s president. In July 2005 he was succeeded by Paul Wolfowitz, former US deputy defense secretary in the George Bush administration. 4 The Bank and the International Monetary Fund (established at the same time) constitute the Bretton Woods institutions and can be viewed as the crowning achievement of the English economist John Maynard Keynes. Another, lesser known protagonist was the American Harry Dexter White who later became a victim of Senator Joseph McCarthy’s anti-communist campaign. 5 Though the initial emphasis of the Bank was on Europe, countries in other regions were also getting signiﬁcant support, for example, Japan and India. By 1960 the Bank had lent $1.4 billion to eight Asian member countries out of 13 total members. The early years reﬂected the Bank’s worldwide emphasis on infrastructure projects. 6 IFC is the private arm of the Bank and coﬁnances private undertakings and provides services to the private sector in developing countries. 7 To become a member of the Bank a country must first join the International Monetary Fund (IMF). Membership in IDA, IFC and MIGA are conditional on membership in IBRD. 8 It is a little known fact that the Bank made a loan equivalent to $40 million to Italy to build an atomic power plant in 1959. There were no protests at the time. 9 OED is the evaluation unit at the Bank, independent of management and reports directly to the Board of Executive Directors. It evaluates outcomes by considering (a) the relevance of the intervention’s objectives in relation to country needs and institutional priorities; (b) the efﬁcacy of the intervention – that is, the extent to which the developmental objectives have been or are expected to be achieved; (c) the efﬁciency of the intervention – that is, the extent to which the objectives have been or are expected to be achieved, using the minimum resources. In addition, the benchmark for a satisfactory investment project is an economic rate of return of

World Bank, development and adjustment 35

13 14 15 16 17

at least 10 percent. Additional consideration is given to sustainability of benefits over time and the institutional development impact on the ability of a country to make more efﬁcient, equitable and sustainable use of its human, ﬁnancial and natural resources. The study noted the progress overall during this period which included global gains of life expectancy at birth in developing countries by 20 years, the drop in adult illiteracy by nearly half (from 47 percent to 25 percent), and the decline of the number of people living in abject poverty (deﬁned as living on less than $1 a day) after rising through most of the nineteenth and twentieth centuries. The study noted, however, that development assistance has not been always fully effective and the development process has been uneven. For example, Africa as a region saw no increase in its overall per capita income between 1965 and 1999, AIDS reversed gains in social indicators, and many of the transition economies of Eastern Europe and Central Asia experienced sharp rises in poverty in the 1990s. This refers to a package of policy prescriptions attributed to the World Bank and IMF and includes ﬁscal discipline, reorientation of public expenditures, tax reforms, interest rates liberalization, competitive exchange rates, trade liberalization, openness to foreign direct investment, privatization, deregulation and securing property rights. “Trickle down” (or Trickle Down Economics – TDE) refers to the belief that once free market policies are introduced at the macro level, there will be signiﬁcant and quick beneﬁcial effects on the whole economy (especially in terms of employment creation and improvements in the social sectors). World Development Reports are annual and constitute the ﬂagship publication of the Bank. They vary in the subjects they cover (for example, poverty, the role of the state, labor or infrastructure). McNamara left the Bank in 1981 having maintained the position that the Bank should remain independent of the US Government. Subsequent World Development Reports started presenting a more balanced view on the appropriate role for the state in human development, infrastructure and began undertaking some excursions into the areas of political economy. They go on to identify the fact that this is more the result of insufﬁcient country ownership and insufﬁcient attention to social dimensions. As of 2003, 26 countries are receiving debt relief projected to amount to US$40 billion over time. With other forms of debt relief, the HIPC Initiative will cut by two-thirds the external debt in these countries, lowering their indebtedness to levels well below the average for developing countries overall. Examples of such partnerships are: with the World Wildlife Fund to protect forests; with the Food and Agriculture Organization (FAO) and the United Nations Development Programme (UNDP) to sponsor the renowned Consultative Group on International Agricultural Research; with the Consultative Group to Assist the Poorest (CGAP) which aims to improve the capacity of microﬁnance institutions to deliver ﬁnancial services to the poorest people; with UNICEF and ILO in the area of child labor; with WHO, UNAIDS and the Global Fund for AIDS, Tuberculosis and Malaria (GFATM); or with partnerships which are formed around specific initiatives such as the one to fight river blindness which has successfully prevented more than 600,000 cases of blindness, has opened 25 million hectares to cultivation, and annually treats more than 22 million people for the disease. The Bank has launched (since 1996) more than 600 anticorruption programs and governance initiatives in nearly 100 client countries and currently works in 40 conflict-affected countries supporting international efforts to assist war-torn populations, resume development and prevent relapse into violence. The Bank deﬁnes “gender” as a “theme,” not a sector. A theme is cross-sectoral and applies to all sectors, and this is obviously the right way of approaching the issues concerned.

36 Zaﬁris Tzannatos 21 The Bank subsequently started a process of converting Operational Manual Statements to Operational Policy and Bank Procedures statements. This process resulted in richer language on the economic, ﬁnancial, social and institutional analysis of investment projects. 22 However, OMS 2.20 was never clear in its intent as (a) some argued that gender analysis is required in projects where it is generally acknowledged that women are a particularly important group of project participants or beneﬁciaries and (b) others have argued that an analysis is required in all projects to determine when women are important participants or beneﬁciaries. Still, in the mid-1990s (and till the onset of the new millennium) the Bank’s gender and development policy was embodied in the aforesaid document plus a policy paper, Enhancing Women’s Participation in Economic Development (World Bank 1994a) and Operational Policy 4.20 (The Gender Dimension of Development). 23 World Development, Vol. 23, No. 11, November (1995). 24 It would be wrong, though, to assume that the Bank was idle on gender analysis before the mid-1990s. For example, Elson (1995) rightly criticized the Bank for applying gender-blind structural adjustment programs in the 1980s but did not fail to note the signiﬁcant analytical work the Bank or researchers sponsored by the Bank were carrying out in key areas for gender and development (such as agriculture: see Collier (1990, 1994) or Moock (1976)) or the “pioneering study by Tzannatos (1992) [that] demonstrated that if gender discrimination patterns of occupation and pay were eliminated, total output, as well as women’s income, could increase signiﬁcantly)” (Elson 1995: 1857). 25 Some of the theoretical underpinnings of the report can be found in Tzannatos (1999). 26 This research report was supported in part by the governments of Norway and the Netherlands. 27 It should be recognized that though education does not guarantee rights or power, it is a critical condition for empowering women. 28 Greater gender equality and a less rigid or extreme gender-based division of labor is expected to promote growth by raising the total level of productive capital in the society, and specifically by increasing female productive capital – capital that has important pro-growth effects. 29 There is a lively debate whether policies (including legislation) should reinforce/ legitimize existing trends or lead to new directions. 30 The report went further to specify accountabilities, funding, stafﬁng, tools, capacitybuilding and partnerships for the successful implementation of the strategy. 31 Named after Ernie Stern, a Managing Director at the Bank. 32 Flexibility over weekly working hours. 33 Since 2002, the Bank has extended conventional family beneﬁts to same sex partners. Ironically, despite its mission to support the poor in developing countries, the equal treatment of same sex partners for beneﬁt purposes has the practical implication that the additional benefits the Bank will have to pay will most likely accrue almost exclusively to US nationals as foreigners have to be heterosexual spouses to get a visa from the US authorities. 34 See also http://www.ifc.org/enviro/Publications/Practice/practice.htm. 35 The Bank does not use the human rights record of its client countries to determine its assistance to them, although the quality of policies and institutions is used to determine the level of IDA assistance the poorest countries receive. 36 Nicholas Stern was the Chief Economist of the World Bank from 2000 to 2003. 37 For example, in the area of core labor standards where many Bank members have signed declarations and research can show that there can be positive economic effects (Aidt and Tzannatos 2002), the Bank encourages the promotion of good practice associated with human rights such as freedom of association.

World Bank, development and adjustment 37 38 Such as the Country Assistance Strategies (CASs) which outline the Bank’s program in member countries. 39 This document reiterates the strong linkages between gender equality and all the MDGs and argues that working for gender equality offers a compelling, win-win approach for policy-makers and planners towards attaining and implementing the goals in areas such as poverty, education, health, nutrition and the environment. It also provides examples of how gender equality can be integrated into MDG policies and interventions. It concludes with a call to integrate gender into MDG policies and interventions especially in the areas of poverty, education, health and nutrition, and the environment. 40 Of course, if gender is deﬁned not as a sector but as a (cross-cutting) theme, then nothing less than 100 percent can be satisfactory, and the issue then becomes one of both substance and measurement.

References Aidt, T. and Z. Tzannatos (2002) Unions and Collective Bargaining: Economic Effects in a Global Environment, Washington, DC: World Bank. Benería, L. (1995) “Toward a Greater Integration of Gender in Economics,” World Development, 23(11): 1839–50. Boserup, E. (1970) Women’s Role in Economic Development, London: Gower. Çagatay, N., D. Elson and C. Grown (1995) “Introduction” (to special issue on Gender, Adjustment and Macroeconomics), World Development 23(11): 1827–38. Collier, P. (1990) “The Impact of Adjustment on Women,” in World Bank Analysis Plan for Understanding the Social Dimensions of Adjustment, World Bank, SDA Unit, Africa Region. —— (1994) “Gender Aspects of Labor Allocation During Structural Adjustment – A Theoretical Framework and the Africa Experience,” in S. Horton, R. Kanbur and D. Mazumdar (eds), Labor Markets in an Era of Adjustment, Vol. 1, Washington, DC: World Bank. Cornia, G., R. Jolly and F. Stewart (eds) (1987) Adjustment with a Human Face, Oxford and New York: Clarendon Press. Edwards, S. (1997) “Trade Liberalization Reforms and the World Bank,” American Economic Review, 87(2), Papers and Proceedings of the Hundred and Fourth Annual Meeting of the American Economic Association, May: 43–8. Elson, D. (1995) “Gender Awareness in Modelling Structural Adjustment,” World Development, 23(11): 1851–68. Goldin, I., H.R. Rogers and Nicholas Stern (2003) The Role and Effectiveness of Development Assistance Lessons from World Bank Experience, Washington, DC: The World Bank (A Research Paper from the Development Economics Vice Presidency of the World Bank). Kapur, D., J. Lewis and R. Webb (eds) (1997) The World Bank: Its First Half Century, Washington, DC: Brookings Institution Press. Little, I.M.D. and J.A. Mirrless (1991) “Project Appraisal and Planning Twenty Years On,” in Proceedings of the World Bank Annual Conference on Development Economics, Washington, DC: World Bank: 351–82. Ministry of Foreign Affairs (Netherlands) (2003) Results of International Debt Relief 1990–1999, Policy and Operations Evaluation Department, May. Moock, P. (1976) “The Efficiency of Women as Farm Managers: Kenya,” American Journal of Agricultural Economics, 58: 831–5.

38 Zaﬁris Tzannatos Stern, N. and F. Ferreira (1997) “The World Bank as an Intellectual Actor,” in D. Kapur, J. Lewis and R. Webb (eds), The World Bank: Its First Half Century, Washington, DC: Brookings Institution Press: 523–609. Stern, Nicholas (2003) “Development and Human Rights,” Address to Panel at the LSE Workshop, Washington, DC, March 5. Tzannatos, Z. (1992), “Potential Gains from the Elimination of Labor Market Differentials,” in Women’s Employment and Pay In Latin America, Part I: Overview and Methodology, Regional Studies Program Report No. 10, Washington, DC: World Bank. —— (1999) “Women and Labor Market Changes in the Global Economy: Growth Helps, Inequalities Hurt and Public Policy Matters,” World Development, 27(3): 551–69. UN (2000) The United Nations System and Human Rights: Guidelines and Information for the Resident Coordinator System, New York: United Nations Administrative Committee on Coordination (ACC), March. UNDP (1990) Human Development Report, New York. Wade, R. (2000) “US Hegemony and the World Bank: Stiglitz’s Firing and Kanbur’s Resignation,” mimeo. —— (2001) “The US Role in the Malaise at the World Bank: Get Up Gulliver!” paper presented at a panel on International Institutions and North/South Conﬂict, American Political Science Association Annual Meetings, San Francisco, August 28–30. Wolfensohn, J.D. (2004) Address at the Shanghai Conference on Scaling Up Poverty Reduction. Shangai: People’s Republic of China, May 26. World Bank (1980) World Development Report on Poverty, New York: Oxford University Press. —— (1984) Operational Manual Statement 2.20, Washington, DC: World Bank. —— (1990) World Development Report on Poverty, New York: Oxford University Press. —— (1992a) The Stern Report (Excellence Through Equality: An Increased Role for Women in the World Bank. A Report of the Advisory Group on Higher-Level Women’s Issues, April, Washington, DC: World Bank. —— (1992b) The Wapentians Report (Effective Implementation: Key to Development Impact), Washington, DC: World Bank. —— (1994a) Enhancing Women’s Participation in Economic Development: A World Bank Policy Paper, Washington, DC: World Bank. —— (1994b) Operational Policy 4.20, Washington, DC: World Bank. —— (1995a) Toward Gender Equality: The Role of Public Policy, Washington, DC: World Bank. —— (1995b) Advancing Gender Equality: From Concept to Action, Washington, DC: World Bank. —— (1998a) Doing Better Business Through Effective Public Consultation and Disclosure: A Good Practice Manual, Washington, DC: International Finance Corporation. —— (1998b) Assessing Bias in Pay and Grade at the World Bank, Development Research Group, Washington, DC: World Bank. —— (2000) Advancing Gender Equality: World Bank Action Since Beijing, Washington, DC: World Bank. —— (2001a) Integrating Gender in World Bank’s Assistance. Operations Evaluation Department, Washington, DC: World Bank. —— (2001b) 2000/2001 World Development Report: Attacking Poverty, New York: Oxford University Press.

World Bank, development and adjustment 39 —— (2001c) Engendering Development: Through Equality in Rights, Resources, and Voice, Washington, DC: World Bank. —— (2001d) Integrating Gender into the World Bank’s Work: A Strategy for Action, Washington, DC: World Bank. —— (2002a) The Role and Effectiveness of Development Assistance: Lessons from World Bank Experience, Washington, DC: World Bank (OED). —— (2002b) The Gender Dimension of Bank Assistance: An Evaluation of Results, Washington DC: The World Bank (OED). —— (2003a) Implementation of the Gender Mainstreaming Strategy: First Annual Monitoring Report, Washington, DC: World Bank. —— (2003b) Gender Equality and the Millennium Development Goals, Washington, DC: World Bank. —— (2003c) Operational Policy 4.20 (revised), Washington, DC: World Bank. —— (2003d) Bank Procedure 4.20, Washington, DC: World Bank.

An assessment of efforts to promote gender equality at the World Bank Carolyn M. Long

Introduction During the past 25 years, donor agencies and other development organizations have worked to promote gender equality in their programs with varying degrees of commitment and success. As time has gone by, research and actual experience have shown that gender equality is essential to development effectiveness. In order for countries to grow and prosper, all human resources must be maximized. If left unaddressed, gender disparities undermine otherwise wellconceived policies and programs. One of the most important public donor agencies is the World Bank. Its record thus far in promoting and institutionalizing gender equality in its policy and operational initiatives has been quite unsatisfactory. However, by the Bank’s own admission, the Fourth World Conference on Women held in Beijing in 1995, and, in particular, NGO advocacy efforts which were launched there, motivated the Bank to increase its efforts to promote gender equality within its initiatives (World Bank 2000: 2). Since 2001, the World Bank has taken three important steps to increase its commitment to gender equality. These are: 1

Public release in 2001 of a critical evaluation of the Bank’s efforts to promote gender equality during the previous four years. The evaluation was carried out by the Operations Evaluation Department (OED), an internal but autonomous part of the Bank. Publication of a major policy research report which documents the pervasiveness of gender discrimination and deﬁnes a three-part strategy to promote gender equality. Engendering Development: Through Gender Equality in Rights, Resources, and Voice (2001a) proposes a strategy to reform institutions to establish equal rights and opportunities for women and men; to foster economic development to strengthen incentives for more equal resources and participation; and to take active measures to redress persistent disparities in command over resources and political voice. Adoption by Bank management and endorsement by its board of Executive Directors of a new gender mainstreaming strategy, a central component of which is a Country Gender Assessment (CGA) to be prepared for all active

Efforts to promote equality 41 borrowing countries. The results of these assessments will be used in dialogue with borrower governments to identify priority gender-responsive policies and interventions in high-impact sectors important for poverty reduction and economic growth. Whether these steps will result in signiﬁcant progress by the Bank in promoting gender equality remains to be seen. To fully understand these recent steps, it is important to place them within the historical context of the Bank’s efforts to incorporate a focus on gender equality in its operations. This chapter provides a brief history of steps taken by the Bank to promote attention to gender equality in policies and projects, an assessment of how well the organization has done in these activities thus far, current thinking at the organization and a description of advocacy efforts carried out by women’s organizations and other external actors. The chapter also includes views of a number of Southern and Northern women’s organizations on gender advocacy at the Bank and what actions they think are essential in order to promote gender equality in Bank-funded initiatives.

Where gender staff is situated in the World Bank In 1997, the most recent reorganization of the Bank created a matrix organization. As of November, 2003, there were four managing directors, six regions and six networks, each headed by a vice president. These networks are Poverty Reduction and Economic Management (PREM), Environmentally and Socially Sustainable Development (ESSD), Human Development (HD), Financial Sector (FSI), Operations Policy and Country Services (OPCS), and Private Sector Development and Infrastructure (PSI). The role of the networks is to provide corporate and human resources functions, knowledge management, quality control and technical assistance to the regions. The PREM network houses the Gender and Development Unit (also called the “anchor”) which has 14 staff members. The PREM network has a Gender and Development Board (a “sector” board) chaired by the Unit’s director. Members of the board currently include representatives from each region as well as several sectors and networks across the Bank. Examples of sectors and networks represented in the Gender and Development Board include the World Bank Institute, the legal department, transport, human development, social development and human resources. The sector board functions as a board of directors for the gender anchor, and as such, determines the overall approach to be taken to promote gender concerns (subject to approval by management and the Executive Directors (EDs)). Unlike most sector boards which are made up of managers with budgets, the Gender Sector Board’s members are primarily gender specialists who are not managers and, in most instances, have no budgets. Each region has a gender coordinator. The effectiveness of these specialists varies considerably, depending on their particular approaches to their task and their ability to inﬂuence their regions or to raise money from donor trust funds.

42 Carolyn M. Long

Philosophical approach to gender issues at the World Bank First and foremost, gender issues at the World Bank are analyzed within the organization’s prevailing paradigm of economic reform and globalization. Although in recent years, social inclusion, empowerment and good governance have come to be seen by the Bank as essential elements of development, the core of the paradigm continues to be economic reform and globalization. Bank staff members striving to promote gender equality do not question whether this emphasis, itself, contributes to the continuing marginalization and impoverishment of women in their societies. Within the Bank’s economic paradigm, gender equality is promoted primarily through an “economic efﬁciency” argument, i.e. that attention to gender concerns is important so as to develop better projects and policies in order to achieve economic growth and reduce poverty. A different philosophical approach – which feminists around the world espouse – is that women have the right to equality, and they should be assisted in standing up for their rights vis-à-vis their husbands and families, communities, local government and the state. This second argument is not voiced by many in the Bank because it implies that women’s rights are human rights. The Articles of Agreement that established the Bank specify that the organization is not to engage in politics. Therefore, historically, the Bank has not taken stands on human rights issues but, under pressure from civil society groups and UN organizations, it is beginning to consider a human rights approach. The Bank’s recently completed policy research report (World Bank 2001a) should help advance the debate in the direction of gender equality as a woman’s right. For now, gender specialists in both the anchor and the regions still primarily use the efficiency argument and see their task being to win the “hearts and minds” of Bank staff regarding the importance of gender concerns. Many, perhaps even most, staffers interpret the Bank’s gender policy as representing a Women in Development (WID) rather than a gender-analysis approach. A WID strategy involves efforts to incorporate or include women and their concerns in development initiatives. A gender-analysis development approach (GAD) examines both women’s and men’s roles to determine how they contribute to development, the constraints each faces in striving to be productive members of society, and appropriate interventions to enable each to achieve a higher level of well-being. At the Bank, the WID approach remained dominant. This may explain partly why girls’ education is an area where the Bank has vastly increased its investments over the past several years. Clearly, the Bank increased its investments as a response to research evidence showing the great beneﬁts of closing the gender gap in school. But as a staff person noted, “The Bank seems to have done very well in girls’ education because it didn’t threaten anyone.” Bank staffers may also be more comfortable with a WID approach in terms of their interactions with borrower governments since the Bank maintains that it cannot impose actions on governments. However, as signatories to the 1995

Efforts to promote equality 43 Beijing Platform for Action from the Fourth World Conference on Women, governments have already espoused gender equality. By promoting attention to concerns of gender equality, Bank staff would simply be encouraging governments to implement these commitments.

Brief history of promotion of gender equality at the World Bank In 1977, the Bank appointed a Women in Development Advisor and in 1986, established a WID Unit. According to a 1996 joint analysis of Bank efforts toward gender equality done by the International Center for Research on Women and the Overseas Development Council, “Serious momentum in the Bank began in the mid-1980s.” However, the report also states The Bank has done signiﬁcantly more on behalf of women as mothers than as workers . . . There is in the Bank an intellectual consensus surrounding the importance of investing in women in the social sectors that does not exist with regard to women’s roles in economic development. (Buvinic et al. 1996: 2) In 1994, four years after the Bank adopted poverty reduction as its overall mission, the Bank’s board endorsed the ﬁrst gender policy paper from which the Operational Policy statement on the gender dimension of development (Operational Policy 4.20) was derived. The Bank’s policy was focused operationally on five sectors (agriculture, education, health, financial services and labor) and did not – and still does not – require that all investment and structural adjustment programs address gender issues. It is important to note that structural adjustment programs – representing at least one-third of total Bank commitments – were not included under Operational Policy 4.20. After the arrival of James Wolfensohn as Bank president in 1995, and his statements of support for gender integration into Bank work – including his attendance at the Fourth World Conference on Women in 1995 – the Bank began to increase its emphasis on the promotion of gender equality. Persuaded by the importance of girls’ education for advances in family health and well-being, the Bank has lent approximately US$5.3 billion for girls’ education since the Beijing conference (Mason 2002). In the areas of health, population and nutrition, the Bank is now the largest provider of external funds for such programs in low and middle-income countries. Since Beijing, two-thirds of these loans have included gender-related goals. In terms of the future, a report prepared by the Bank in early 2000 for the Beijing Plus Five meeting reinforces the Bank’s overall commitment to gender equality, at least rhetorically, when it says Recently the Bank has renewed its focus on poverty reduction as its primary mission and adopted a broad deﬁnition of poverty that includes empowerment, opportunity, and security as well as income as necessary to the ﬁght

44 Carolyn M. Long to end poverty. These changes have set the stage for the Bank to sharpen its focus on gender equality. (World Bank 2000: vi)

Assessment of progress on gender equality at the World Bank through 2000 Mainstreaming a focus on gender issues According to the Evaluation of the Gender Dimensions of World Bank Assistance’s Phase I Report (World Bank 2001b), prepared by the OED and reviewed by the Bank’s Executive Directors in October, 2001, three basic steps are necessary to successfully mainstream a focus on gender issues in any institution. These include clear instructions to staff, measures to institutionalize and operationalize the policy, and monitoring and evaluating the policy and its implementation (World Bank 2001b: 29). Measured against these criteria, the Bank still has a great deal to do to systematically integrate gender concerns into its initiatives. In the OED evaluation, the following reasons are given for such limited progress: • • • • • • •

The 1994 gender policy is seen as poorly understood and open to wide interpretation by Bank staff, even among staff who work on gender issues; There have been no clear plans set in place to implement the gender mainstreaming strategy; There are no effective systems in place to institutionalize the policy; Resources for gender mainstreaming have been inadequate; No time-bound benchmarks were set to measure the degree of mainstreaming nor has implementation of the strategy been rigorously monitored; There aren’t incentives for staff to integrate gender issues, nor are there mechanisms in place to determine accountability; The Bank hasn’t built the necessary capacity among its staff to mainstream gender issues (World Bank 2001b: 5).

These very same issues were noted in OED’s evaluation of gender dimensions of Bank assistance done in 1997, an indication of little, if any progress on mainstreaming in recent years!2 The Bank’s gender policy As an operational policy, the gender policy is required to be followed by staff. However, because the 1994 policy was vague, it was open to wide interpretation, and could be safely ignored by staff without consequence. The gender policy could be contrasted with “safeguard” policies, a group of ten individual polices which cover such issues as environmental assessment and protection, involuntary resettlement, indigenous peoples and dam safety. Bank Procedures (which specify

Efforts to promote equality 45 procedures and documentation required for Bank-wide consistency and quality) and Good Practices (statements containing advice and guidance on policy implementation based on experience) are both prepared for safeguard policies but were not for the 1994 gender policy. The absence of these further undermined the policy’s effectiveness since the Bank Procedures provide the framework and detailed requirements of a policy which then can be monitored for compliance. Historically, even in cases where staff has genuinely wanted to pay attention to gender equality in projects and policy formulation, such efforts have been dependent on approvals by country directors or project task team leaders for funding. Since the reorganization in 1996, country directors have controlled the budgets for their country programs. It has been only when these directors agreed to speciﬁc requests that gender-related activities have been funded. This has been a major impediment to gender mainstreaming. Beyond one’s own interest, there have been no incentives for including gender-related activities or sanctions for not doing so. Indeed, trust funds provided by donors (e.g. the Norwegians and the Dutch) have ﬁnanced at least a portion of gender-related work done by the Bank over many years. Gender issues in Bank-funded projects Given the Bank’s limited monitoring of its own work to incorporate gender issues into its initiatives, it’s difﬁcult to give an accurate portrayal of the extent of mainstreaming. The data that exists indicates the following. The Bank’s published reports include these results: • • • •

From 1965 to 1985: less than 10 percent of Bank-funded projects included speciﬁc gender-related actions. From 1988 to 1994: the percentage increased to 30 percent, with the strongest efforts made in improvements to women’s health and education, and to increase their options in agriculture. As of 2000: the percentage of projects including some consideration of gender issues in their design had almost doubled since 1995, climbing to more than 40 percent of all Bank projects. The percentage of projects that ﬁnance gender-related activities or support policy changes designed to reduce gender inequalities averaged 26 percent in 1995–99 (World Bank 2001b: 9).

Although these statistics show a slow, steady progression, internal Bank evaluations reveal a somewhat different story. The most recent evaluation by OED, completed in late 2000, indicates: • •

Between 1988 and 1999, only 38 percent of the Bank’s investment projects have any meaningful reference to gender issues; Less than 28 percent of these projects included gender action plans or components;

46 Carolyn M. Long • •

At least 60 percent of all projects didn’t explicitly consider gender issues in a meaningful way during preparation; Less than 15 percent of projects used gender disaggregated data in preparation (World Bank 2001b: 17).

Women’s participation in policy formulation Starting in 1997, NGOs around the world began an effort to promote participation of civil society organizations in the formulation of the Bank’s Country Assistance Strategy (CAS) in borrower countries. As part of this larger effort, Southern women’s organizations began to be invited to meetings held by the Bank. The OED evaluation found that all CASs reviewed which were prepared after 1997 were rated satisfactory as regards gender awareness. However, the evaluation states “even in CASs with good gender analysis, the link with recommendations is weak” (World Bank 2001b: 24). NGOs have done extensive work to promote systematic and genuine participation of civil society in formulation of the CAS, and the Bank has developed guidelines for staff members to promote such involvement. Nevertheless, NGOs are still not involved systematically in the formulation of the CAS. The more common experience is that NGOs are consulted, and their views may or may not be taken into consideration when the CAS is ﬁnalized. Bank systems and procedures that ensure that real participation happens should be enforced more strongly and need to guarantee that women’s organizations are always included. The Poverty Reduction Strategy process, launched in 1999 by the World Bank and the International Monetary Fund (IMF), requires governments to carry out a participatory process with civil society and other stakeholders involved in its formulation. However, according to available reports, gender concerns have been largely overlooked in the ﬁrst 33 interim Poverty Reduction Strategy Papers (PRSPs) and while certain gender issues were addressed for the ﬁrst time in some of the ﬁrst nine ﬁnal PRSPs completed in November, 2001, they were not treated in the depth necessary (World Bank 2002). One exception is Rwanda whose PRSP does mainstream gender issues with very few exceptions and treats gender concerns in some depth (Zuckerman 2002: 10). The promotion of gender equality in Bank project and policy initiatives has been largely dependent on individual initiative by gender specialists working in the Bank’s regional ofﬁces and in the gender “anchor.” Therefore, success in integration of gender concerns into projects and policy formulation varies considerably by region, depending on the particular gender specialist, the regional vice president, the regional chief economist and the individual country directors and their willingness, or lack thereof, to promote gender equality. The Latin America and Caribbean region is considered by many to have been successful in its approach to gender mainstreaming in recent years, largely due to the efforts of the regional gender specialist there until 2002.3 She deﬁned her work as providing services to task managers based on an efﬁciency model. She believed that the majority of task team leaders were willing to incorporate

Efforts to promote equality 47 gender concerns in their projects if they were given the proper help. She raised money from donor trust funds to underwrite her work and had good results in having gender issues addressed in project formulation and design. She also identiﬁed implementation constraints with clients. Because very little technical assistance is provided to borrowers during the implementation phase of projects, she piloted a technical assistance (TA) facility in Ecuador with a gender specialist available to coordinate assistance to projects in several countries. The coordinator identiﬁed providers of TA (either consultants or NGOs) in the countries themselves. When the TA facility was ﬁrst set up, the regional gender specialist sent an e-mail to 11 Bank task team leaders to ascertain their interest in using it. Nine responded positively within a week (Correia 2001). According to some Bank staff members, successful work in gender mainstreaming is also being done in other regions. However, in Europe and Central Asia, the gender coordinator post remained vacant for well over a year before being filled in 2002, and task team leaders in Africa express little interest in promoting gender equality.

Current thinking and action on promotion of gender equality at the World Bank Speciﬁc work being carried out in the Bank at present is encouraging. The policy research report noted on page 40, this volume, which was completed in 2001, makes the case as to why gender equality must be viewed as an integral part of development initiatives that aim to reduce poverty and encourage economic growth. Other research done earlier by the Bank within the context of the Special Program of Assistance to Africa shows the linkages between gender inequality, growth and poverty and concludes that reducing gender-based asset inequality increases growth, efﬁciency and welfare (Blackden and Bhanu 1998: 2). Undoubtedly the most important decision taken in recent years regarding gender equality was the Bank Board’s endorsement on September 18, 2001 of a new gender mainstreaming strategy. Under development for much of the previous year, the new strategy has as its centerpiece a Country Gender Assessment (CGA) to be completed in all active borrowing countries. The CGAs are based on analysis and inclusive consultations at the country level and identify gender-responsive policy and operational interventions important for poverty reduction, economic growth and sustainable development. The Bank’s Gender and Development Board developed the new strategy which was written by a team headed by the Director of Gender and Development. Consultations held during the strategy’s development in the six regions where the Bank works included civil society groups and government officials, and informal discussions were held with bilateral and multilateral donors. The External Gender Consultative Group (see page 49, this volume) commented on several drafts. The gender mainstreaming strategy is being made operational through a basic three-step process:

48 Carolyn M. Long 1

2 3

Prepare, for each country in which the Bank has an active lending program, a periodic Country Gender Assessment analyzing the gender dimensions of development across sectors and identifying the gender-responsive actions that are important for poverty reduction, economic growth, human wellbeing and development effectiveness, and use it to inform the Bank’s country assistance program; Develop and implement, as part of the Country Assistance Program, priority policy and operational interventions (if any) that respond to the CGA; and Monitor the implementation and results of these policy and operational interventions (World Bank 2002: 18).

Expectations are that, based on the diagnosis carried out through the CGA, a gender dimension will be integrated into relevant analytical work and lending instruments. For example, this would include analytical work in high-priority sectors and social impact analysis associated with adjustment lending, including Poverty Reduction Support Credits, among others. An important part of the strategy has been the revision of the Bank’s gender policy, Operational Policy 4.20 (see page 43, this volume). In March, 2003, the Bank’s board approved the new “Gender and Development” Operational Policy (OP 4.20). In terms of accountability, regional vice presidents are expected to submit annual gender mainstreaming plans and to prepare year-end reports outlining fulﬁllment of these plans. In October, 2002 the Gender and Development anchor completed preparation of a new monitoring and evaluation system to track and evaluate implementation of the strategy in Bank work. The new gender mainstreaming strategy, with its emphasis on Country Gender Assessments, represents a new level of seriousness in how the Bank is to examine gender status and gender disparities in speciﬁc countries. Whether the assessments will result in any significant increase in incorporation of gender-responsive actions into policy and operational interventions remains to be seen. The strategy states that while the process should be led by the speciﬁc countries, the Bank will play a supportive and proactive role. The mainstreaming strategy has been adopted at a time when traditional pressures on staff remain, i.e. to formulate, prepare and bring loan proposals to the EDs for approval and to get money disbursed to borrowers. While personnel evaluations have been changed somewhat in the recent past to encourage teamwork and a client orientation, staff are still rewarded primarily for getting loans approved. There are no speciﬁc incentives offered to encourage Bank staff to embrace this strategy other than the research evidence stated earlier in this report. As noted, accountability for implementation of the strategy is to be through the annual plans and reports of the regional vice presidencies. However, adherence in the past to accountability mechanisms for gender, participation and other social issues has been quite unsatisfactory. There is even some difﬁculty achieving compliance with the safeguard policies which are considered absolutely mandatory. Therefore, successful mainstreaming will depend in large measure

Efforts to promote equality 49 on the ability of gender advocates within the Bank to persuade their colleagues to embrace this plan and on the commitment of the country directors and regional vice presidents to ensure implementation. External monitoring of the strategy’s implementation by gender advocates is essential, both at headquarters and in borrower countries. As regards the board of the Bank, some progress has been made regarding attention to gender. At the present time, those countries whose Executive Directors are most attentive to gender concerns are the Netherlands, the Nordic countries, Germany, the United Kingdom, France and Canada. While in the past, there was a deﬁnite split between North and South regarding support for gender equality, the situation among the Southern EDs is becoming more encouraging, with support having been expressed by at least a few of the EDs from borrower countries.

Bank mechanisms for expanded consultation on gender issues Since the 1995 Fourth World Conference on Women in Beijing, the Bank has created two important mechanisms to increase consultation with gender specialists and women’s organizations around the world. External Gender Consultative Group In 1996, President Wolfensohn created the External Gender Consultative Group, composed of 14 gender specialists from around the world. Its purpose is to assist the World Bank in the design and implementation of its gender policies, and to help strengthen dialogue on gender-related issues between the Bank, its partners and interested sectors of civil society. The members of the EGCG were chosen by the Bank based on their expertise, experience and geographical representation. The EGCG has succeeded in raising important issues for consideration by the Bank such as the need to approach gender from a rights perspective rather than simply for efficiency, and to address the harmful effects of structural adjustment and globalization on women. They also provided important input to the policy research report and to the gender mainstreaming strategy. However, the group is a set of individuals who do not represent their organizations in their capacity as members of the EGCG. The group has no formal advocacy program, extremely limited funds, and a once-a-year encounter with the Bank at headquarters. Therefore, their inﬂuence is limited. In 2000, the Bank reduced the EGCG membership from 14 to 9 people, most likely as a cost-saving measure. The EGCG is an important reference group for the Gender and Development anchor and provides a sounding board for internal gender advocates as they consider how to increase attention to gender at the Bank. However, some gender-focused Bank staffers perceive the orientation of the Bank towards the EGCG as a sort of “ﬁreﬁghting” exercise, i.e. Bank

50 Carolyn M. Long management listens to the EGCG in order to keep a positive public relations proﬁle on the topic but the group does not signiﬁcantly affect Bank behavior toward gender issues. Consultative Council on Gender for the Middle East and North Africa This group was created by the Bank in January, 1999 largely as a result of the commitment and dynamism of the MENA gender coordinator to enable it to consult with members of civil society in countries of the region.

NGO advocacy efforts on gender equality at the World Bank Shortly before she left the Bank in 2001, Jan Piercy, the former US Executive Director, said “Having an external constituency on gender is indispensable” (Piercy 2001). She noted how struck she had been by how priorities of the Bank shift depending on advocacy by external groups. In the months preceding the adoption of the new gender mainstreaming strategy, comments expressed by several Bank staff regarding the status of gender promotion in the Bank revealed the current dearth of external advocacy as well as the work needed to be done inside the Bank. “Gender is nowhere here,” said one high-placed female manager who doesn’t work on gender at the Bank but who did in another development organization. “Gender is not even on the radar screen,” said another Bank staffer familiar with social development issues in the Bank. “No one cares about gender here,” said another. Evidence shows that advocacy by women’s organizations around the world has helped to promote the gender advances that have occurred so far in the Bank. While some advocacy work had been carried out subsequent to the imposition of structural adjustment policies in the early 1980s and in the decade following the establishment of the Bank’s WID Unit in 1986, efforts began to intensify at the 1995 Fourth World Conference on Women where Bank President James Wolfensohn made a speech indicating the Bank’s commitment to promote gender equality. He was the ﬁrst Bank president to attend any of the United Nations conferences on women. Women’s eyes on the Bank At the Beijing conference, representatives of women’s organizations presented Wolfensohn with a letter signed by over 800 people urging the Bank to take the following four steps: • •

To increase the participation of grassroots women in the Bank’s projects and economic policy making; To institutionalize a gender perspective in all Bank policies and programs;

Efforts to promote equality 51 • •

To increase Bank investments in women’s health services, education, agriculture, land ownership, employment and ﬁnancial services; To increase the number and diversity of women in senior management positions within the Bank (Williams 1997: 33).

This action launched the Women’s Eyes on the Bank Campaign (WEOB) which took root in the US and Latin America and the Caribbean (LAC). Following the Beijing conference, 8 to 12 organizations based mainly in Washington created the campaign’s US chapter. The only other region where an active campaign ensued was in Latin America and the Caribbean which has continued to have a group of women involved in advocacy in 10 countries (Frade Rubio 2000: 1). In 1997, the US campaign produced an excellent assessment of the Bank’s work in the four areas noted above. Three or four individuals in Washington-based NGOs became the backbone of the US campaign, with most working on WEOB in addition to their regular professional responsibilities. When they left their respective NGO jobs in 1999, work on the WEOB essentially stopped. Women’s EDGE In 2000, Oxfam America asked Women’s EDGE to consider taking up advocacy on gender equality at the Bank after the US coordinator of the Women’s Eyes on the Bank Campaign left Oxfam for a job elsewhere. Through support provided by Oxfam America, Women’s EDGE researched this topic and began a dialogue with Southern gender activists about advocacy topics of importance to them. The results of that work were published in April, 2002, in a report entitled “The advocate’s guide to promoting gender equality at the World Bank.” The focus of the Women’s EDGE coalition is women and macro-economic issues, and it works to ensure that complex international trade agreements do not harm people living in developing countries. Its constituency includes more than 65 member organizations and more than 7000 individuals on campuses and other key places around the country. The coalition’s diverse membership – from the Christian Children’s Fund to the Feminist Majority – have come together around a common mission to inject women’s economic, social and political rights into US interactions with developing and transitional countries. The organization deﬁnes and targets US action broadly; EDGE advocates for women’s rights in global trade, ofﬁcial development assistance and US participation in multilateral organizations. It has established relationships with Southern women’s organizations interested in and active on macro-economic issues. Unfortunately, due to a lack of funding, it has been unable to continue its work to promote gender equality at the World Bank.

52 Carolyn M. Long Gender Action Gender Action is a nonproﬁt organization established in 2002 to ensure multilateral organizations promote gender equality and women’s rights in all their investments worldwide. Gender Action’s strategies include partnering with Southern civil society groups to convince their governments to mainstream gender concerns into MDB investments in their countries; persuading donor governments to hold the MDBs accountable on gender issues; monitoring and reporting on multilateral progress in mainstreaming gender issues; and conducting gender advocacy directly on the multilaterals. During its initial months, Gender Action has concentrated on engendering PRSPs because of the critical role PRSPs play in country eligibility for multilateral investments and in national planning and budgeting. Gender Action has undertaken PRSP fieldwork collaboratively with Southern partners and direct advocacy with bilateral agencies and other nonproﬁts. Gender Action has actively participated in the Coalition for World Bank Reform’s advocacy campaign by engendering proposed US World Bank (IDA-13) funding replenishment legislation and conditions. Gender Action also contributed to the Coalition report, “Responsible Reform of the World Bank.”

Integrating gender into PRSPs The advent of the Poverty Reduction Strategy process has meant that women’s organizations have begun to be involved in these efforts although, as noted on page 46, participation by women’s groups and attention to gender issues have been inadequate thus far. According to a Bank staffer, the Bank, itself, is precluded from helping civil society groups become involved in the PRSP process. The UN Development Fund for Women (UNIFEM) analyzed the treatment of gender concerns in the PRSP as part of a comprehensive review of this process undertaken by the Bank and IMF in late 2001. UNIFEM’s comments were drawn from assessments done by NGOs, consultants and national women’s machineries in particular countries, and noted these six points: •

•

• •

One of the key areas where there is a singular lack of gender dimension in the PRSPs is that of data collection to inform poverty diagnoses, policy development and monitoring and evaluation. Gender is not addressed as a cross-cutting issue; instead it is often limited to a few speciﬁc sectors such as health and education where gender issues are traditionally addressed. Gender is rarely addressed in relation to macro-economic and structural policies. These are assumed to be gender-neutral when in fact they are not. PRSPs emphasize the market economy at the expense of the household economy, which is mostly female-dominated, based on unpaid labor and has direct contributions to the market economy, especially during business cycle downturns.

Efforts to promote equality 53 •

•

Participation of civil society in the development of PRSPs is often limited, without institutionalized mechanisms that can ensure the views of all groups are actually reﬂected in the various stages of a PRSP. One important omission in PRSPs has been the very limited discussion on violence against women (World Bank 2001d: 1).

Other research undertaken has uncovered additional reasons why gender concerns have not been included in PRSPs thus far: •

•

Although some participatory processes undertaken to ensure civil society input into preparation of PRSPs have considered gender concerns, the ﬁnal reports or workshops obscured gender differences by aggregating previously sex-disaggregated data. Participants in PRSP writing teams most often are staff from government ﬁnance and economics ministries and external consultants, many of whom are insensitive to gender issues. Many PRSP stakeholders lack gender-mainstreaming skills; they see the process of engendering the PRSP as mysterious; and they find gender discussions largely conceptual rather than practical. The PRSP sourcebook, prepared by the World Bank, is seen as impractical to use. The gender chapter, while very useful in terms of diagnostic and monitoring indicators and tools, does not deal with gender advocacy and gives little attention to the need for gender analysis of macro-economic issues including the national budget which is central to the PRSP (Zuckerman 2002: 10).

Views of Southern NGO gender specialists regarding gender advocacy at the World Bank The most inﬂuential voices regarding gender issues at the World Bank must be those of Southern women themselves. However, historically, women in borrower countries have not been given the opportunity to participate adequately or appropriately in Bank policy and project formulation, design, implementation and evaluation. The recent positive steps taken by the World Bank may represent an opportunity to redress this situation. Although at present there is no way to systematically catalogue or analyze advocacy by women’s organizations at the national level, Southern women’s groups have been promoting gender concerns to the Bank for many years. Southern women’s groups have long been among those civil society organizations working to convince the Bank to end or signiﬁcantly alter structural adjustment policies imposed on their countries. The LAC-WEOB campaign has been the most active World Bank-focused effort in a Southern region in recent years. Southern women have also been in the forefront of groups urging an end to user fees in health, water and education which governments have imposed as conditions of certain World Bank lending. ISAAF International in India and

54 Carolyn M. Long the Tanzania Gender Networking Program are two of many women’s organizations advocating these changes. In many instances, women’s groups have been included in the consultative processes with civil society conducted for the formulation of the CAS in borrower countries. Interviews in recent years with women from Asia, Africa, Eastern Europe, and Latin America and the Caribbean reveal the following issues as those they perceive as most urgent regarding the promotion of gender at the Bank: • • • • •

The Bank’s promotion of gender equality within the prevailing paradigm of economic reform and globalization; The technical compartmentalization of gender efforts rather than viewing them within the political context of structural inequalities and patriarchy; The Bank’s focus on economic efﬁciency rather than empowerment and rights; The emphasis on women as mothers, not as workers; The role of the US in maintaining the current World Bank approach to economic reform and globalization.

Southern gender specialists think advocacy should start by questioning the assumptions and values that form the basis of the Bank’s policies, as well as its system for delivering loans to Southern countries. This includes economic conditionality and the process of negotiation of macro-economic policies. In the view of Southern gender specialists, the PRSP is an example of a Bank-led initiative to continue imposition of liberal economic reforms, with ofﬁcial claims for poverty reduction goals and broad-based public participation, coupled with conditionality likely to conﬂict with national development imperatives and longterm poverty reduction. The negotiation process is conducted for the most part without the full knowledge and participation of affected citizens, including women. Many Southern gender specialists think that the Bank is promoting gendermainstreaming while it supports macro-economic policies that further the interests of transnational corporations in search of new markets and new sources of cheap labor, thus reinforcing the exclusion of disadvantaged social groups, especially women. Southern women also think that these policies have led to the removal and/or de-legitimization of public policy instruments for the implementation of social development policies and afﬁrmative action policies, including the Beijing Platform for Action.

Conclusion As detailed in these pages, during the past 25 years the record of the World Bank in promoting gender equality in its policy and operational initiatives has been quite unsatisfactory. However, the recent steps taken by the Bank which have been described in this chapter are encouraging. Gender activists and feminist economists around the world need to seize this opportunity to monitor the

Efforts to promote equality 55 implementation of the new gender-mainstreaming strategy, both at the country level and globally, to analyze gender impacts of Bank-funded initiatives and to promote as vigorously as possible gender equality and equitable access to resources and power.

Notes 1 2

This chapter is adapted from “The advocate’s guide to promoting gender equality at the World Bank,” published by the Women’s EDGE Coalition in April, 2003. The full paper is available at www.womensedge.org Although the current director of the Bank’s Gender and Development Unit indicated in July, 2002 that the 1994 gender policy paper was not a mainstreaming strategy, “mainstreaming strategy” was the term used in the OED evaluation cited in this paper. For another view on this matter see Chapter 5 by Avin and Chapter 12 by Barriteau in this book.

Acronyms and abbreviations CAS CGA ED EGCG IBRD ICSID IDA IFC IMF LAC LAC-WEOB MENA MIGA NGO OED PREM PRS PRSP TA UNIFEM WEOB WID

Country Assistance Strategy Country Gender Assessment Executive Director External Gender Consultative Group International Bank for Reconstruction and Development International Centre for Settlement of Investment Disputes International Development Association International Finance Corporation International Monetary Fund Latin America and the Caribbean Latin America and the Caribbean Women’s Eyes on the Bank Middle East and North Africa Multilateral Investment Guarantee Agency Non-governmental organization Operations Evaluation Department Poverty Reduction and Economic Management Network Poverty Reduction Strategy Poverty Reduction Strategy Paper Technical assistance United Nations Development Fund for Women Women’s Eyes on the Bank Women in Development

References Blackden, Mark and Chitra Bhanu (1998) 1998 Poverty Status Report, Special Partnership with Africa, Washington, DC: World Bank. Buvinic, Mayra, Catherine Gwin and Lisa M. Bates (1996) Investing in Women: Progress and Prospects for the World Bank, Washington, DC: Overseas Development Council. Correia, Maria, interview, January 24, 2001.

56 Carolyn M. Long Frade Rubio, Laura (ed.) (2000) Equity, Participation and Consistency? The World Bank at the Beijing + 5, Women’s Eyes on the World Bank Campaign, Latin American Region, Mexico: Alcadeco, AC. Long, Carolyn M. (2003) “The advocate’s guide to promoting gender equality at the World Bank,” published by Women’s EDGE Coalition. Online. Available http:// www.womensedge.org Mason, Karen, notes to author, July 18, 2002. Piercy, Jan, interview, January 25, 2001. Williams, Lydia (ed.) (1997) Gender Equity and the World Bank Group: A Post-Beijing Assessment, Women’s Eyes on the World Bank – US in collaboration with Women’s Eyes on the World Bank – Latin America, 50 Years is Enough, and US Network for Global Economic Justice, Washington, DC. World Bank (2000) Advancing Gender Equality. World Bank Action Since Beijing, Washington, DC: World Bank. —— (2001a) Engendering Development: Through Equality in Rights, Resources, and Voice. Washington, DC: The World Bank. —— (2001b) Evaluation of the Gender Dimensions of World Bank Assistance, Operations Evaluation Department, Fourth Draft, Washington, DC: World Bank. —— (2001c) “Development Topics: Participation, Participation in the Poverty Reduction Strategy Formulation, External Reviews,” Online. Available http://www. worldbank.org —— (2001d) “United Nations Development Fund for Women (UNIFEM) Contribution to the World Bank and IMF PRSP Review,” Online. Available http:// www.worldbank.org —— (2002) Integrating Gender into the World Bank’s Work: A Strategy for Action, Washington, DC: World Bank. Zuckerman, Elaine (2002) A Primer on Poverty Reduction Strategy Papers and Gender, Washington, DC: Gender Action.

Rhetoric and realities A comment Sakuntala Narasimhan

The two preceding chapters, by Zaﬁris Tzannatos and Carolyn M. Long respectively, seem to be broadly in agreement over the assessment that the priorities of the ﬁrst four decades of the Bank’s operations were not merely gender-blind but actually caused the feminization of poverty (increasing the percentage of women among those living below the poverty line), leaving women worse off and further marginalized. Both chapters also concede that in recent years (and speciﬁcally, after the Beijing conference of women in 1995) the organization has declared its commitment to a more gender-sensitive perspective, and that a number of initiatives, ranging from policy guidelines to setting up gender consultative groups, have been taken. Where once the Structural Adjustment Policies (SAPs) urged and imposed by the Bank on developing countries desirous of borrowing from the Bank eroded the quality of life for women, while exports rose to pay for debt-servicing, the new policies and initiatives were meant (at least, on paper) to ensure that gender equity did not get pushed off the radar in future funding operations. But have these initiatives been matched by performance? Long’s assessment is that the “corrective measures” that the Bank has undertaken are worth taking note of, but are “not enough.” In other words, the scorecard is a C minus or thereabouts, from a feminist point of view. Not satisfactory. Tzannatos likewise, concludes that the Bank has a long way to go in terms of ensuring that gender equity is not compromised by its funding policies. In reflecting on the thrust of the arguments in the two previous chapters, I ﬁnd myself raising two basic questions: •

•

Has the Bank really learnt from past mistakes, or is there still a gap between rhetoric (about becoming gender sensitive) and the reality of the organization’s current policies and patterns of functioning? and Do the changes initiated within the Bank, as policy shifts and evaluation mechanisms, promise (at least in the long run) to make an abiding difference to the way developmental funding in the borrowing countries affects women?

The ﬁrst question is best answered in terms of a few examples from around the world. After committing a whopping $450 million for the massive Sardar Sarovar Project on the Narmada river in upper India, the Bank parted ways with the

58 Sakuntala Narasimhan Indian government in the wake of widespread protests by activists. The project displaced tens of thousands of people, inundated hundreds of villages, and caused massive dispossession among poor tribals. The Narmada project in fact became the world’s “test case” of the Bank’s willingness and ability to adhere to its own environmental and social guidelines (Mehta 1993). Nonetheless, a decade later, the Bank is reported to have resumed lending for big dams in India despite years of trenchant criticism of its lending policies. In the lush green northeastern states alone (which have a large population of tribals), dams and river diversion projects are being planned that will cause “ecological disasters,” as activists put it. That does not quite look like “learning from past mistakes.” Forced migration and displacement put a disproportionately greater burden on females than on males. As recently as in February 2004, activists from Tanzania drew international attention to the fact that the World Bank was about to fund a major road project around the slopes of Kilimanjaro. “This will spell disaster for the people and pose a serious threat to a world heritage site,” the activists point out. This case too, ﬂies in the face of the claims that the Bank makes, of having become “more ecology and environment conscious in its assessment of projects.” In Uganda, a curious resurgence of the practice known as “mikayi” has been noted in recent research studies. Mikayi also refers to the eldest wife (who traditionally enjoyed seniority and privileges among wives in an African polygamous marriage). Young girls enrolled at the university (even in “modern” courses like gender studies and information technology), it was found, become the mikayi of male classmates and provide housekeeping services (cooking, cleaning) in return for money that will go towards paying college fees. Thanks to the structural adjustment policies dictated by the Bank, state allocations for education got cut, and girls who found there was no money forthcoming for their education have ended up reviving, as a coping strategy, a practice that seems anomalous and socially regressive in the context of gender equity. These girls actually end up defaulting on their educational goals because housekeeping chores leave them no time for studying or attending classes. Many even end up victims of violence inﬂicted by their boy friends (Abuya 2002). This links up with the main argument, because it is the World Bank’s policies (of cutting down on social sector expenditure) that drive girls to adopt strategies that could help overcome the reduction in educational funds available. It is a classic example of the feminization of poverty caused by structural adjustment policies. A more recent report of June 2003 describes how young women trade unsafe sex for survival in Lesotho, because SAP has contributed to rising shortages of food. The World Food Program says increasing prostitution has become a problem in at least six southern African countries (Itano 2003). Similar examples of women’s immiseration can be found across the continents, from Thailand to Kenya and Jamaica to Brazil. Despite professions of greater gender sensitivity, the Bank continues to let “economic proﬁtability” dictate its lending and funding decisions in developing countries across the world. If that is one aspect of “past mistakes,” the other is that the damage inﬂicted,

Rhetoric and realities 59 particularly on women’s lives, through blinkered policies thrust upon developing economies in the earlier years, cannot be undone – thousands of indigent women in Asian countries (India, Bangladesh, Vietnam) who are caught in exploitative industrial units (garment exports, ﬂoriculture and food processing, for instance) ﬁnd themselves in a no-win situation, with their traditional communities and support systems destroyed and no alternatives to turn to. The People’s Charter for Health drawn up by the People’s Health Movement (a global coalition of health activists and medical personnel) took note in May 2004 of the “disastrous effects on women’s health, of the policies dictated by the World Bank.”1 Thanks to the emphasis on privatization of medical services, maternal mortality and morbidity has actually risen in some regions, as several studies have shown. Gender and environmental concerns go hand in hand with overlapping agendas, because whatever degrades the environment puts a greater burden on women, especially in the lower economic strata (having to walk further to fetch water for the family’s needs, fodder for the cattle and ﬁrewood for cooking, more time spent on caring for infants developing respiratory ailments caused by pollution). Regardless, the Bank has, even after 1992 (when the international convention on climate change and environment, also known as the Kyoto Protocol, was signed), ﬁnanced projects that were actually climate-changing to the tune of $2–3 billion per year, which is 20–30 times the Bank’s ﬁnancing of climate-friendly activities, in terms of renewable energy options. The new coal-ﬁred power plant at Talchar in Orissa state in eastern India (described as “the pride of the Bank”) has resulted in the river water becoming coated with thick ash deposits, so women of the region are reduced to digging in the sand with their hands to scoop up water for the family’s daily needs. The plant uses millions of gallons of water for its cooling towers, robbing the villages of a once clean and abundant water supply (Wysham and Valette 1997). That, again, is neither “gender-sensitive” nor “learning from experience.” In Bangladesh, the Bank’s advice urging a greater focus on export promotion (for earning foreign exchange to pay for debt-servicing) led to rice ﬁelds getting converted to shrimp farms. Where once the rice crop fed the farmer and his family, the shrimp cultivated instead of paddy got sent overseas as export. The salination of rice lands caused by the ponds built to raise shrimps in also led to what activists describe as an “ecological disaster” (Ahmad 1992). Agricultural production has fallen by 30 percent, and to the 25 million people who are suffering hunger, the increase in export earnings is hardly a measure of “development” or cause for rejoicing. This brings me to the second question about the developmental paradigm that the Bank is sworn to, and promotes in the borrowing countries. As Long points out, the original brief was, purely and strictly, economic – increasing money incomes and returns from investments, without reference to dimensions of “equity” or “rights” (in terms of women’s right to having their voices heard during processes of decision-making, for example), because these were considered “political” dimensions (to be strictly avoided). In recent years, however, the Bank’s policy is supposed to have shifted towards a rights-and-resource

60 Sakuntala Narasimhan entitlements approach, taking into consideration (at least in theory) the effect of policy decisions on gender equity. Nonetheless, the basic deﬁnition of development remains unchanged – the indices are power generation, highways built, infrastructure base per capita, foreign exchange earnings. If a project promises high returns for investment (through power generation that will bring in proﬁts, for instance) it is still OK under the Bank’s criteria. (The benchmark for a satisfactory investment project is an economic rate of return of at least 10 percent, as Tzannatos observes in a note.) If it is found that this will create gender imbalances and social problems, the approach is to “examine” how these effects could be mitigated rather than questioning the basic deﬁnition of “development” and “progress” itself. As the Bangladeshi activist Nilufar Ahmad asks, what kind of progress is it, if shrimp exports bring in extra foreign exchange earnings, when at the same time 25 million people are pushed into starvation because of the damage to rice ﬁelds? The global consensus today, as Tzannatos points out, is that economics alone is quite inadequate as a matrix for development planning, and that a multi-disciplinary (economic-social-cultural-human rights) approach is necessary. In this, the Bank has fallen far short of its avowed, “updated” intentions that are supposed to be based on “lessons learnt from the past.” The Meltzer Commission report2 described the Bank’s record as “miserable”; Walden Bello has called for the “decommissioning and neutering of the Bank”; and even former senior Bank officials like Nicholas Stern have attacked the Bank’s policy of pampering the rich and penalizing the poor. Whether any of this has altered the perspectives of the Bank is a moot point. Judging from the comments from Long and Tzannatos, the answer seems to be, “no” or at best, “not much.” As part of its “new” and “more sensitive” agenda, the Bank has also mentioned “a greater participatory role for civic society” – and yet, how many displaced tribal hordes, or women impoverished by Structural Adjustment Programs, have access to inspection panels or evaluation units of the Bank ? In a way the Bank is able to conveniently avoid responsibility in this area since it puts the onus on the country representatives. Tzannatos points out, however, that the country representatives are themselves either the ﬁnance minister or similar high-ranking official (who might not necessarily be sensitive to issues of gender equity). Moreover, those authorized to voice opinions at the Bank’s deliberations are not all democratically equal in effect – the top donor countries (US, Japan, UK, Germany, France – none of them belonging to the category of “developing countries”) have one executive director each, while 176 developing countries out of the total membership of 184 nations have to make do with just 16 executive directors representing them. Country representation is one thing, having its voice heard and heeded is another. He who pays the piper, calls the tune – and it comes as no surprise that global criticism of the Bank’s functioning includes the charge that the Bank’s decisions prioritize US multinational companies’ needs (in terms of market access or proﬁts) rather than the needs of the beneficiary country (or for that matter, the women’s component of the populations therein).

Rhetoric and realities 61 The organizational descriptions provided in the earlier portions of the foregoing chapters reveal other details that strengthen the image of the Bank – even after its recent policy “ﬁne-tuning” exercises – as “insensitive” even to the basic concept of poverty. When the Bank evaluates a proposal or development project, it gets consultants to assess the feasibility and proﬁtability dimensions before the project is sanctioned. These consultants, as the Bangladeshi activist Nilufar Ahmad points out, are invariably from the West (American ﬁrms, mostly – one of the conditions the Bank imposes is that foreign consultants have to be engaged) who ﬂy in, and give their expert opinion (regardless of their familiarity with the ground realities of the socio-cultural matrices in the country). Long’s chapter takes note of the fact that, in terms of positive measures, the Bank has recently “decentralized” its working and “moved staff to the ﬁeld.” Out of a total of 10,000 employees, over 3,000 are reported to be based in borrower countries. That’s a half-filled glass (less than, in fact – 30 percent is still way short of “equity”). Imposing assessments from outside and not heeding local voices on priority needs causes wasteful (and at times even harmful) expenditure. Criticism of the Bank’s policies have included the viewpoint that the Bank’s funding has been like pouring water into a leaking tub, because part of the problem in developing countries has been poor implementation – for instance, in power generation, reducing transmission losses and plugging wastage would be a better and far more effective strategy than funding new dam projects that displace communities and cause social unrest. Even after the introduction of “gender-sensitive mechanisms” for ﬁne-tuning project evaluations, as Long’s chapter points out, there is neither an incentive for prioritizing a strong gender component, nor accountability (or penalties for ignoring women’s needs). The organizational initiatives, in other words, have served as no more than cosmetic moves. “The [External Gender Consultative] group has no formal advocacy program, extremely limited funds, and a oncea-year encounter with the Bank at headquarters. Therefore, their influence is limited.” Also, “some gender-focused Bank staffers perceive the orientation of the Bank towards the EGCG as a sort of ‘firefighting’ exercise, i.e. Bank management listens to the EGCG in order to keep a positive public relations proﬁle on the topic but the group does not signiﬁcantly affect Bank behavior toward gender issues” (Long, this volume, page 49). More eloquently, the Bank has “reduced the EGCG membership from 14 to 9 . . . most likely as a costsaving measure” (Long, page 49). Cost-saving does not, presumably, get in the way while ﬂying out, at enormous expense, foreign consultants into developing regions where the average citizen earns in a whole year a fraction of what the consultants spend in a day (and women earn even less). Mindsets, rather than minor adjustments to perspectives and priorities, are the problem. Shifting the perspective and emphasis towards “poverty reduction” makes little sense in the face of insensitivity to, and poor understanding of, the nature and ramiﬁcations of poverty. Even where Bank-funded projects have raised incomes, they have resulted in widening the chasm between the haves and the have-nots, with the rich becoming richer and the poor becoming further impoverished (despite

62 Sakuntala Narasimhan the Bank’s mission statement of 2003 claiming that it fights poverty with “passion”). Globally, money incomes rose at the rate of 2.5 per cent annually, while the number of people in poverty increased by 100 million, during the decade of the 1990s (Stiglitz 2002). A stark example of the Bank’s insensitivity to the human dimension of development is its disregard of the recommendations of the World Commission on Dams set up by the Bank itself. Two more indicators of the lack of political will to mainstream gender in the Bank’s operations are the woefully inadequate use of gender-disaggregated data (less than 15 percent, by the Bank’s own admission of 2001, as Long points out) and the failure to ﬁll posts of gender coordinator lying vacant for “well over a year” (Long, page 47). The assessment given by Long contains the crucial observation: “Evidence shows that advocacy by women’s organizations around the world has helped to promote the gender advances that have occurred so far in the Bank” (Long, page 50). Women’s groups, in other words, need to become not only vocal but also network globally, knitting the developed and developing regions together, in order to articulate cross-cutting gender concerns and make a dent in the Bank’s operations for promoting gender equity. More importantly, the voices of women from the grassroots in the developing countries need to be heeded because they are the ones who are hit hardest by the inequitable and inappropriate policies of the Bank.

Notes 1 2

Comment from the People’s Health Movement (PHM) in a media report on the WHO director general’s speech, May 18, 2004. Walden Bello, executive director of Focus on Global South, in the March issue of Focus on Trade (2001a).The Meltzer Committee report argued that the Bank is “irrelevant to the question of solving poverty and should be radically reduced in terms of scale and functions.” See also Bello (2001b).

References Abuya, Pamela (2002) “The Institution of Mikayi: Gains or Challenge?”, paper presented at the Eighth Women’s Worlds 2002 International Conference, at Kampala, Uganda, July 2002. Ahmad, Nilufar (1992) “Battling the World Bank: an interview with Nilufar Ahmad,” Multinational Monitor, October 10, 1992. Available online at http://www.multi nationalmonitor.org/hyper/issues/1992/10/mm1092_09.html (Accessed September 16, 2005). Bello, Walden (2001a) “The Global Conjecture,” Focus on Trade, 60, March. —— (2001b) “Genoa and the multiple crises of globalization,” Focus on Trade, 64, July. Itano, Nicole (2003) “Women trade unsafe sex for survival in Lesotho,” Women’s eNews, June 1, 2003. Available online at http://www.womensenews.org/article.cfm/dyn/ aid/1350/context/archive (Accessed September 16, 2005). Mehta, Pradeep (1993) “Fury over a River,” Multinational Monitor, December. Stiglitz, Joseph (2002) Globalization and its Discontents, London: Penguin Books. Wysham, Daphne and Vallette, Jim (1997) “Changing the Earth’s Climate for Business – the World Bank and the Greenhouse Effect,” Multinational Monitor, October.

Part II

Policy evaluations

Engendering Development A critique Rose-Marie Avin

Introduction The World Bank Policy Research Report entitled Engendering Development: Through Gender Equality in Rights, Resources, and Voice explores the connections between economic development, existing institutions, and gender inequality in the Third World. The Report also discusses in great detail the role of public policy in promoting gender equality in the world. Given that the Report emphasizes the role of economic development as a key element in improving gender equality in the Third World, this chapter addresses four concerns dealing with the concept of economic development and its connections to gender equality. First, the Report does not explain clearly the objectives of economic development and does not make explicit the distinction between economic growth and economic development, sometimes using the two terms interchangeably (World Bank 2001: 1,182). Second, the Report fails to take into consideration the issues of class, race, and ethnicity, a very important omission in the case of Latin American countries where huge income disparities exist. The dynamic process of economic development and its impact on gender equality will play out differently for working-class, middleclass and upper-class women. It is important to acknowledge differences among women in terms of race and ethnicity. Third, I disagree with the Report’s conclusion that “on balance the evidence suggests that females’ absolute status and gender equality improved, not deteriorated” (ibid.: 215) during the structural adjustment period of the 1980s and 1990s in Latin America. My contention is that the neoliberal economic model that resulted from Structural Adjustment Policies (SAPs) has worsened the economic conditions of the majority of women in Latin America. Women have less access to resources and are still bearing the economic, emotional, and physical costs of structural adjustment. Fourth, the Report needs to acknowledge that Third World women, and especially women’s organizations, have been and are agents of change in transforming their own lives and societies. Not only should women become the beneficiaries of economic development but also the agents of economic development. All four concerns will be discussed in the context of Latin America, more speciﬁcally in the realities of Brazil and Nicaragua.

66 Rose-Marie Avin

Economic development, growth, and gender equality Why is it important to start the Report with a deﬁnition of economic development? Throughout the Report, the authors use the terms “economic growth” and “economic development” interchangeably, which makes it difﬁcult to follow their argument considering that the two terms convey different meanings. It is quite possible that the authors believe that economic growth leads automatically to economic development, a link that has been widely rejected in the development literature. In this case, the authors should reconceptualize their meaning of economic development to include the new scholarship that is widely accepted in the development literature. Incorporating the new meaning of economic development in the analysis is important for three reasons: 1 2

To present the multidimensional aspect of the concept; To develop a more coherent argument establishing effectively the interconnections between the different dimensions of the development process and the various ways they contribute to gender equality; To help policymakers at the World Bank develop more effective policies to deal with problems in the Third World.

Since World War II, the meaning of development has moved away from the narrower view associated with economic growth, income growth, industrialization, and modernization and toward a broader definition that includes both economic and non-economic dimensions. According to Michael Todaro, there are three “core values of development” (Todaro 2000: 16): sustenance – the ability to meet basic needs; self-esteem – the ability to develop a sense of self-worth and self-respect; and freedom from servitude – the ability to choose and be emancipated from alienating social, material, and political conditions. This deﬁnition involves major social, political, and economic changes that can radically alter institutions in Third World societies. Todaro’s deﬁnition is also consistent with Amartya Sen’s deﬁnition of development as “a process of expanding the real freedoms that people enjoy” (Sen 1999: 3). According to Sen, there are ﬁve distinct types of freedom: political, economic facilities, social opportunities, transparency guarantees, and protective security. All ﬁve elements of Sen’s deﬁnition are interrelated and are the objectives of development. Based on this deﬁnition, a successful development strategy can lead to the improvement of the well-being of all individuals in society. In addition, Sen argues that “freedoms are not only the primary ends of development, they are also among its principal means” (ibid.: 10). His meaning of development is also an agent-oriented view where individuals can shape their own destiny. Sen’s deﬁnition is important because it captures well the process of social, economic, and political transformation that needs to happen in order to achieve an economic development that promotes gender equality. Sen’s analytical framework will be analyzed again at the end of this paper to explain the process of gender empowerment that has been happening in Latin America.

Engendering Development 67 The deﬁnitions of economic development given above make it obvious that economic development is a multidimensional process that can contribute to the empowerment of women and reduce gender inequality. The same argument can not be made for economic growth in the way that the World Bank Report implies. Empirical studies have shown that economic growth alone may not improve the well-being of the vast majority of people in the Third World. “Growth of GNP or of individual incomes can, of course, be very important as ‘means’ to expanding the freedoms enjoyed by the members of the society” (ibid.: 3). Some empirical studies (Adelman 1975), however, have shown that the poorest members of society may become worse off with economic growth. An extensive empirical study of 43 developing countries by Irma Adelman showed that the primary impact of economic growth has been to decrease both the absolute and the relative incomes of the poor. So, there is no trickle down with economic growth as this Report seems to imply. According to Adelman, “the poorest segments of the population typically beneﬁt from economic growth only when the government plays an important role and when widespread efforts are made to improve the human resource base” (ibid.: 302–3). During the 1960s, many developing countries such as Brazil and Mexico that experienced high rates of economic growth realized that the growth did not bring significant beneﬁts to the poor (Todaro 2000: 152). In Latin America, standards of living of the poor either stagnated or declined in real terms (ibid.: 152). At the same time, gender inequality did not seem to diminish for the vast majority of poor, working-class women. Considering the wealth of evidence, it is imperative that policymakers at the World Bank reconceptualize the meaning of economic development to show more effectively the relationship between economic development and gender equality and, more importantly, develop more successful policies. The case of Brazil discussed below shows that economic growth is necessary but not sufﬁcient to improve the lives of the vast majority of people and to improve gender equality.

Economic transformation and women in Brazil: 1950–2000 What has been the impact of economic growth on the lives of women in Brazil? Did it lead to an improvement in gender equality as stipulated by the World Bank Report? According to Silvia Berger, women in South America “experienced considerable changes in their economic status as a result of deep transformations in the region’s societies” (Berger 1999: 246). The region’s emphasis on industrial growth after the 1930s, urbanization, and social movements have all contributed to the transformation of women’s lives. The emphasis on industrialization combined with the existence of a highly unequal land distribution in the rural areas led to rapid migration into the major cities of South America, with women forming the largest component of the migrant population (ibid.: 247).

68 Rose-Marie Avin Economic growth and gender equality Brazilian women migrated to the urban areas in large numbers, with the vast majority going to Rio de Janeiro and São Paulo. As a consequence, the percentage of women working in agriculture fell from 31 percent in 1950 to 15.3 percent in 1980 (Inter-American Development Bank 1987: 104–5). At the same time, their share of services increased from 43.6 percent to 65.7 percent during the same period (ibid.: 104–5). Their labor force participation rate in the formal sector also increased from 16.8 percent in 1960 to 26.6 percent in 1985 (Arriagada 1995: 334) to 30.3 percent in 1990 (Delgado et al. 2000: 112). These official data do not capture the full contribution of women to the economy. They tend to underestimate women’s participation in the labor market because their work in the household is not measured and also because many women work in the informal sector. The emphasis on import-substitution industrialization (ISI) based on capitalintensive manufacturing absorbed only a fraction of the labor force. Consequently, the vast majority of female migrants were not able to ﬁnd jobs in the growing manufacturing sector. Some women found jobs in the textile and garment industries. However, the vast majority of jobs for women were in the services sector. Many women were able to ﬁnd jobs as shop clerks, secretaries, teachers, and nurses as a result of expanding state services in public education, health, and social security (Berger 1999: 246–57). A few well-educated women found positions in academia and law. For the women with little education who migrated from the rural areas, it was not easy to ﬁnd a job in the formal sector. They ended up in what is known as the informal economy, which comprises the self-employed and small-scale family-owned enterprises. They engaged in an array of activities that include street vending and prostitution. Others found jobs as domestic servants. During the 1980s, it was estimated that one-third of the employed women in Brazil were in domestic service, acting as nannies, cooks, and maids (Library of Nations 1987: 65). According to Todaro, workers in the informal sector do not have job security, decent working conditions, or old-age pensions. Most live in “favelas” (shanty towns) and lack access to public services such as transportation, and educational and health services (Todaro 2000: 297). Empirical studies also show substantial segmentation of the labor market by gender, that is women in the labor market are concentrated in a few low-level occupations. In 1980, the vast majority of the personal service workers were women (92.4 percent), and of the maids and washerwomen (92.4 percent) (Arriagada 1995: 33). In spite of the gains realized by educated Brazilian women in terms of better access to professional jobs, income received by women is lower than income received by men at all levels of education and for all professions. In 1985, women received 52.8 percent of the average male income in São Paulo. Empirical studies have shown that the higher the level of education, the larger the income gap is. Furthermore, male income is always higher for the same occupations. One

Engendering Development 69 interesting fact is that, in Latin America, the female working population has higher levels of education than the female population that does not participate in the labor market, and also more education than both the working and nonworking male population (ibid.: 335). In Brazil, there does not seem to exist differences in educational attainment between men and women but there are differences between women at different class levels. In 1998, the female adult literacy rate as a percentage of the male rate was 100 percent (Human Development Report 2000: 256). In 1997, the ratio of the female primary age group enrollment as a percentage of the male rate was 94 percent (ibid.: 256). In 1997, the ratio of the female secondary age group enrollment as a percentage of the male ratio was 103 percent. During the period 1994–7, the percentage of female tertiary students as a percentage of males was 116 percent (ibid.: 256). However, huge inequalities based on race and class remain in the educational system. According to the Brazilian Ministry of Education, “although 44 percent of all Brazilians are blacks and mestizos, they constitute about 70 percent of all poor and indigent” (Ministry of Education 2000: 4). Furthermore, while the illiteracy rate is 9 percent among Caucasians, it is 22 percent among Afro-Brazilians. As for schooling, the Caucasian population has an average schooling rate of 6.2 years compared to 4.2 years for blacks and mestizos (ibid.: 4). Also, it was estimated that 93 percent of domestic workers are women of color (Castro 1995: 28). Significant demographic changes also took place during the time of rapid economic transformation. Brazil experienced decreases in the mortality rate, increases in life expectancy, and a fall in fertility rates. The fertility rate declined from 5.6 children in 1965 to 3.9 in 1980 and to 2.3 in 1997 (World Development Report 1999/2000, 2000). During the period 1990–8, 77 percent of Brazilian women had access to contraception (ibid.: 242). This signiﬁes that women have more control over their lives, and are able to increase their participation in the labor market because of fewer responsibilities at home. It is obvious that modernization and industrialization in Brazil had different effects on the roles of women of various races and social classes. On the one hand, economic growth and the expansion of higher education created new jobs and educational opportunities for Euro-Brazilian, middle-class women (Alvarez 1990: 9). Gender discrimination continues, however, in the labor market in terms of occupational segregation and earning gaps. Not only does there exist a large income gap between men and women in Brazil, it seems that the higher the educational level, the higher the income gap is. On the other hand, millions of poor and working-class women were pushed into low-paying, lowstatus jobs in the most exploitative sectors of the economy. As Brazil underwent many economic crises during the late 1970s as a result of the oil crisis, hundreds of thousands of poor, working-class women sought solutions to their families’ economic problems by participating in the community self-help organizations and grassroots social movements that sprang up throughout Brazil in the 1960s and 1970s (ibid.). At the same time, the rapid entry of Euro-Brazilian,

70 Rose-Marie Avin middle-class women into academia and other professions started a debate about women’s equality among Brazilian intellectuals. What role did the women’s movement play in the improvement of women’s status in Brazil? Social movements The Brazilian women’s movement played a key role in the promotion of gender equality in Brazil. According to Sonia Alvarez: In the 1970s and 1980s Brazilians witnessed the emergence and development of what is arguably the largest, most diverse, most radical, and most successful women’s movement in contemporary Latin America. By the mid-1980s, tens of thousands of women had been politicized by the women’s movement and core items of the feminist agenda had made their way into the platforms and programs of all major political parties and into the public policies of the New Brazilian Republic. (Alvarez 1994: 3) According to the Feminist Center for Studies and Advisory Services (CFEMEA), the decade of the 1970s represented a milestone for the women’s movement in Brazil. That time period saw the rise of feminist ideology and of women’s groups struggling for the democratization of Brazilian social and political relations. At the same time, the International Women’s Year was celebrated in 1975 around the globe. The Decade of Women was also launched and, in Brazil, hundreds of women’s groups emerged in the periphery of major cities. Over 400 self-professed feminist organizations were formed during the 1970s and 1980s (CFEMEA 1999: 10). At the end of the 1970s and during the 1980s, the women’s movement broadened and diversified, extending itself into political parties, unions, and community associations. As a consequence, changes were made in the Federal Constitution and in public policies in order to ﬁght the oppression of women in Brazilian society. The Constitution guaranteed the right to family planning, to protection against family violence, and to childcare and pre-school centers. Furthermore, common-law marriage was recognized as a family nucleus and discrimination against children born out of wedlock was prohibited. At the same time, many institutions and programs were created to facilitate women’s equality before the law: Women’s Rights Councils, special police stations for victims of domestic violence, speciﬁc integral health programs, and programs for the prevention of and the care for women victims of sexual and domestic violence (ibid.). During the 1990s, the women’s movement took on a broader scope and a large number of Non-Governmental Organizations (NGOs) arose. This led to a diversity of projects and strategies and new structural and mobilization forms were created. For example, the following organizations were created: the Concerted Action of Brazilian Women (AMB), the National Feminist Network

Engendering Development 71 on Health and Reproductive Rights (RedeSaude), networks of rural and urban women workers, women researchers, religious women, Black women, lesbians, and others. Campaigns were launched, such as the For Women’s Life campaign to maintain the right to abortion as codiﬁed in the Brazilian Penal Code when a pregnant woman’s life is at risk or when the pregnancy results from rape (ibid.). It can be concluded that gender equality improved in Brazil for a small percentage of women as the state sector expanded during the period of the 1970s known as the “Economic Miracle,” a period in which many Euro-Brazilian and middle-class women gained access to education and economic opportunities. Furthermore, women’s participation in the democratization process during the 1980s created a political space that translated into increased rights as women and as citizens of Brazil. Brazil’s economic achievements, however, did not improve the lives of the vast majority of Brazilians, including Brazilian women. For Brazilian women who are poor and working-class, the socioeconomic reality was different. As they poured out of the rural areas, they ended up struggling for economic survival in the informal sector of Brazil’s major cities. Women of color, in particular, seem to be at the bottom of the economic ladder. The results discussed above refute the contention of the World Bank Report that economic growth automatically leads to economic development and to improvements in gender equality. By equating economic development with economic growth and by ignoring the role of class, race, and ethnicity in society the Report failed to correctly identify the various mechanisms through which economic development can lead to increased gender equality. In the next section, I will discuss how Nicaraguan women made major strides toward gender equality during the 1980s but lost ground as a result of the structural adjustment policies of the 1990s advocated by the World Bank. This case again shows the need for policymakers at the World Bank to redefine economic development.

Empowerment and women in Nicaragua: 1980–2000 In this section, I discuss why I reject the World Bank Policy Research Report’s conclusion that “on balance the evidence suggests that females’ absolute status and gender equality improved, not deteriorated” (World Bank 2001: 215) during the structural adjustment period in Latin America. My contention is that the neoliberal economic model that resulted from structural adjustment has worsened the economic conditions of the vast majority of women in Nicaragua. Nicaraguan women made far-reaching gains during the socialist Sandinista government that was in power from 1979 to 1990. A society with extreme concentration of income, wealth and political power was dismantled and replaced by a more egalitarian society. Galvanized by their participation in the revolutionary struggle that overthrew the dictator Anastasio Somoza, women demanded that their agenda make its way into the programs and public policies of the new socialist Sandinista government. For the ﬁrst time in Nicaraguan society, women had a government that was somewhat sympathetic to their concerns.

72 Rose-Marie Avin During the 1980s, conditions were created that had the potential to radically alter the private and public lives of Nicaraguan women. Major policies were instituted to remove legal, social, and economic obstacles that prevented women from participating as equal citizens in society. In fact, Nicaragua put into effect many of the same policies advocated in the World Bank Report in its threepart strategy to promote gender equality. According to the Report, countries should: 1) reform institutions to establish equal rights and opportunities for women and men; 2) foster economic development to strengthen incentives for more equal resources and participation; 3) take active measures to redress persistent disparities in command over resources and political voice (ibid.: 15–26). Many laws were instituted to improve women’s productive and earning capacity, and to change sex roles in society. In 1981, the Law of Cooperatives was established, which gave women the right to own land. Other laws required that women be provided with maternity leave and receive equal pay for equal work. Furthermore, there were large investments in education. In 1975, onethird of University students were women. By 1985, the number was 56.3 percent, partly due to the existence of the war that began in 1981 between the Sandinista army and a group of Contra Revolutionaries (Contras, for short), created and sponsored by the United States government. Still, in 1992, two years after the war ended, women constituted 51.7 percent of all university students (Hutauruk 1998). There were also increased expenditures on health care, nutrition, and greater access to credit and job training. In addition, women obtained the right to divorce and to have custody of their children. Moreover, a number of institutions were created to increase women’s awareness of the new laws and to help in their implementation. In 1979, a women’s legal ofﬁce (Oﬁcina Legal de la Mujer) was created to deal with sex discrimination, non-payment of child support, domestic violence, and other legal problems affecting women (Chamorro 1989; Avin 1993; Chinchilla 1995; Babb 2001). Women also played a significant role in the political arena. Their political participation was massive, although it varied among women according to their social status. While many poor women entered the political arena at the grassroots level by joining popular organizations, middle- and upper-class women achieved high positions such as ministers or regional party coordinators. One woman, Dora Maria Tellez, became Minister of Health while six other women were vice ministers. In the National Assembly, 18.5 percent were women (Hutauruk 1998). Furthermore, many women occupied high-level positions in various government ministries (ibid.). An incredible phenomenon took place in Nicaraguan society during the 1980s: working class and peasant women increased their rights, their access to resources, and for the first time had a voice in their community. They were working at the grassroots levels, in both the rural and urban sectors, to increase the level of feminist consciousness among women and in society at large. These women, strong and powerful leaders at the community, national, and international level, were able to articulate the problems of women in Nicaragua and use creative strategies to alleviate them and thus empower women in the process.

Engendering Development 73 At the same time, a number of women’s centers providing health services, family planning services, consciousness-raising and sexuality workshops were formed throughout the country. The electoral defeat of the Sandinistas in 1990 brought all the advancements toward gender equality to a stop. The socialist government was replaced by a coalition government, headed by a woman, Violeta Barrios de Chamorro. Programs of economic stabilization and structural adjustment were put into effect to deal with the debt crisis, hyperinﬂation, and economic stagnation. With the economic restructuring, gone were the subsidies for utilities, public transport, education, health care, and nutritional programs. Gone were the training programs, free lunch meals, paid pregnancy leaves, and childcare programs. In 1991, the economy deteriorated and women struggled to make ends meet. The small degree of control they had gained over their lives during the 1980s seemed to have evaporated overnight. This is not surprising, considering that women were bearing a disproportionate share of the costs of the structural adjustment policies put into effect by the Chamorro government (Chamorro 1989; Avin 1993; Chinchilla 1995; Babb 2001). Women were the ﬁrst ones to be laid off, especially in the state sector where they constituted the majority of state workers. Consequently, many joined the informal economy, becoming street vendors, domestic workers, and prostitutes. Furthermore, it is reported that many small industries “had been particularly hard hit and that women, 54 percent of its membership, were the most affected” (Babb 2001: 166). In addition, most female workers in the garment and food industries lost their jobs because of privatization and lower tariffs on imported goods (ibid.). Women’s burden also increased in the households. As primary caretakers, women spent more time caring for children and other members of the household since public health services have not only declined but are also charging fees for visits and medication. Lower incomes and higher prices have meant that many women were unable to provide for their families. Hence, women were spending more time in the labor force and more time on meeting the basic needs of their families (Chamorro 1989; Avin 1993; Chinchilla 1995; Babb 2001). The inherent gender bias in Nicaraguan society, the restructuring in agriculture and in industry toward large-scale production, the demise of small-scale agriculture and industry, the ﬂood of cheap imports, the collapse of the informal economy where more than 75 percent of women were employed, the elimination of all subsidies have all meant that Nicaraguan women have become more oppressed economically. Currently, 58.8 percent of working-age women are unemployed and underemployed, and of those who are employed 75.5 percent work in the informal sector – the employer of last resort (Agurto Vilchez and Gido Cajina 2001). How are Nicaraguan women surviving today in the new free market economy that emphasizes the manufacturing of export goods in industrial parks?

74 Rose-Marie Avin Growth of Free Trade Zones (FTZs) In Nicaragua, factories operating under Free Trade Zone (FTZ) policies increased considerably during the period 1991–2001. In 1976, Nicaragua had one industrial park, “Las Mercedes,” with 11 factories producing mostly clothes for the United States market (Comision Nacional of Zonas Francas 2001: 12). However, between 1979 and 1980, those factories closed their doors and moved their operations elsewhere in Central America. In 1991, with funds received from the World Bank and the Inter-American Development Bank, new industrial parks were created and the foreign investors came back (ibid.). The sector exploded. Today, the FTZ system consists of 12 industrial parks. The number of ﬁrms increased from 8 in 1992 to 45 in 2001 (ibid.: 13). The sector has also been the most dynamic in terms of job creation: the number of jobs increased from 1003 in 1992 to 38,792 in 2001 (ibid.: 17). At the same time, the value of exports grew from 2.9 million to 230 million dollars during the same period (ibid.: 18). While the sector contributed 7.5 percent to total exports in 1994, its contribution reached 32 percent in 2000 (ibid.). Of the 45 ﬁrms operating in the FTZs, 14 are from Taiwan and 13 from the United States (ibid.: 13). In terms of goods produced, the vast majority of ﬁrms (73 percent) are engaged in the production of clothes, while 16 percent are in tobacco (ibid.). The ﬁrms produce mainly for the United States market. What has been the impact of such growth on women in Nicaragua? There is no doubt that the FTZs sector is an important source of employment for poor Nicaraguan women. It is estimated that 70 percent of the workers are women, with about 50 percent being heads of households. For Nicaraguan women, having a job in the FTZs means not having to sell food and trinkets on the streets and also not having to leave the country to become maids in Costa Rica. The reality is that there are no jobs in other sectors in Nicaragua (Interview with Sonia Agurto Vilchez, 17 January 2002). While the FTZs have been a major source of employment for women, they have not created jobs that can bring women out of poverty and contribute to their empowerment. On the contrary, management in these factories pays low wages, provides difﬁcult working conditions and no health care, and tries to suppress the women’s labor rights. According to Sonia Agurto of FIDEG, a well-known research institution in Nicaragua, the average monthly salary in the FTZs is 850 cordobas (about US$65) higher than the minimum wage of 400 cordobas (US$29). However, the basic market basket for a family of ﬁve cost 2550 cordobas (US$185). It takes three persons working in a FTZ to buy the basic market basket. To make ends meet, many women work extremely long hours (ibid.). Working conditions are also difficult. There have been reported cases of physical mistreatment, as was illustrated by the revolt at Chentex during the mid1990s. At Chentex, workers rose up because of poor working conditions and the fact that they were not treated with respect by management. They were

Engendering Development 75 treated like “cattle” as one former worker said. As for the suppression of labor rights, the various efforts of Nicaraguan workers trying to form unions have been well recorded in the press. According to Gladis Manzanares, former general secretary of one of the labor unions in the FTZs, one of the major problems is the difﬁculty of forming labor unions. Workers suspected of forming labor unions get ﬁred and that can be a devastating blow for a female head of household. Furthermore, many women do not know the law and their rights (Interview with Gladyz Manzanares, 17 January 2002). It is obvious that this neoliberal development model of FTZs has not led to the transformation of women’s lives through economic empowerment, but rather reinforces the conditions of marginalization of women in Nicaraguan society. Job segregation from the outset gives women unskilled production jobs that are the poorest paid. Lacking access to on-the-job training, women are not moving up the job ladder and are not acquiring economic, social, and political power. One organization working to empower women is the “Maria Elena Cuadra” (MEC), a women’s organization working to improve the human rights of women, especially of women in the FTZs. This organization provides scholarships for nontraditional jobs, has a small credit program for unemployed women, teaches women about their labor rights, and provides training for negotiation techniques. According to Sandra Ramos, the head of MEC, women need to have a consciousness-raising that will empower them to take action in the future, but it takes time. The goal of MEC, then, is to create a space for reﬂection for the women of the FTZs so that they can achieve personal empowerment (Interview with Sandra Ramos, 17 January 2002). It can be concluded that structural adjustment and neoliberal policies have not improved gender equality in Nicaragua as stipulated in the World Bank Policy Research Report. On the contrary, the vast majority of women in Nicaragua have lost signiﬁcant economic grounds during the 1990s as discussed above while at the same time ﬁghting to empower themselves. The following section discusses how women in Latin America have contributed to their own empowerment.

Women’s movements and women’s agency In this section, I discuss the need for policymakers at the World Bank to recognize that Third World women have been agents of change in transforming their own lives and societies. As the examples of Brazil and Nicaragua illustrate, women have made a signiﬁcant contribution to their own empowerment. This implies that policymakers need to make women agents of change in their societies, an idea developed by Amartya Sen in his book Development as Freedom. According to Sen, “any practical attempt at enhancing the well-being of women cannot but draw on the agency of women themselves in bringing about such a change” (Sen 2000: 190). Therefore policymakers need to focus not only on the wellbeing aspect but also on the agency aspect of economic development. Women need to become active agents of change in the transformation of their societies, as women in Brazil and Nicaragua seem to do.

76 Rose-Marie Avin In spite of their economic losses during the 1990s, the women’s movement in Nicaragua emerged in 1992 with greater creativity, dynamism, visibility, and diversity than before (Chinchilla 1995: 243). Various sectors of the women’s movement broke with the National Organization of Nicaraguan Women, named the Luisa Amanda Espinosa Association of Nicaraguan Women (AMNLAE), after the first female member of the Sandinista army killed in 1970 during the revolutionary struggle. These break-away groups believe that AMNLAE, closely tied to the Sandinista party, had been more responsive to the party’s needs than to women’s needs. They argue that focus on women’s socioeconomic problems and resistance to male domination must be at the center of their work and not the survival of a political party that had become less and less responsive to women’s needs. With the formation of new women’s centers, groups, and institutes, the women’s movement has become more diverse and dynamic. Currently, there are about 150 social-change groups, centers, institutes, organizations, including a radio station (Montenegro 1997: 375). At the First National Women’s Meeting in January 1992, women decided to set their own agendas and to collectivize their survival problems in order to meet the needs of Nicaraguan women. The meeting was organized according to speciﬁc themes relevant to women’s lives: the economy, violence, and sexuality. At that meeting, national networks were formed to deal with each of them: the Network of Women Against Domestic Violence, the Network of Women for Reproductive Rights, the Network of Women for Health (ibid.: 382). The National Coalition of Women in Politics is an organization formed in 1995 by women politicians from across the political spectrum dedicated to ﬁnd common solutions to the increasing socioeconomic marginalization of Nicaraguan women. By forming these social-change groups, women in the movement have tried to become less polarized by working on issues and problems common to the vast majority of women in Nicaragua: domestic violence, economic insecurity, and serious problems in reproductive health. After the defeat of the Sandinista party in 1990, the women’s movement became autonomous and has tried to become a political force to be reckoned with. Given that the vast majority of women in Nicaragua are poor, and given that these women are at the base of the women’s movement, issues and problems affecting these marginalized women have been at the center of the movement’s campaigns to change the reality of women. Their campaign against domestic violence has been particularly successful: there are now various laws protecting women against domestic violence and two women police stations in Managua. The movement has grown in maturity and diversity to become the most dynamic and creative women’s movement in Central America.

Conclusion The World Bank Policy Research Report emphasizes the role of economic growth/development as a key element in improving gender equality in the Third World. In this chapter, I argue the need for the authors to reconceptualize the

Engendering Development 77 concept of economic development and to incorporate the issues of class, race, and ethnicity in their analysis in order to develop a more coherent argument and effective policies. I also suggest that the three-part strategy proposed by the Report is inconsistent with the neoliberal agenda of international lending organizations, including the World Bank. I argue that the new development model of free market policies has made it difﬁcult for the three-part strategy to be put into effect in Latin America, especially Nicaragua. The neoliberal policies of the 1990s have not led to greater economic growth and gender equality. In fact, the emphasis on Free Trade Zones in Nicaragua limits women’s economic opportunities, reduce their sense of empowerment gained during the 1980s, and exposes them to widespread economic exploitation by multinational corporations.

References Adelman, Irma (1975) “Development Economics: A Reassessment of Goals,” American Economic Review, 65(2): 302–9. Agurto Vilchez, Sonia and Alejandra Guido Cajina (2001) Mujeres: Pilares Fundamentales de la Economia Nicaragüense, Nicaragua: Fundacion Internacional Para el Desafio Economico Global (FIDEG). Alvarez, Sonia E. (1994) “The (trans)formation of feminism(s) and gender politics in democratizing Brazil,” in J.S. Jaquette (ed.) The Women’s Movement in Latin America, second edition, Boulder, CO, San Francisco, CA, and Oxford: Westview Press: 13–63. —— (1990) Movements in Engendering Democracy in Brazil: Women’s Transition Politics, Princeton, NJ: Princeton University Press. —— (1988) Democratizando o Brasil, Rio de Janeiro: Paz e Terra. Arriagada, I. (1995) “Unequal Participation by Women in the Work Force,” in James L. Dietz (ed.), Latin America’s Economic Development: Confronting Crisis, second edition, Boulder, CO: Lynne Rienner Publishers, Inc.: 333–49. Avin, Rose-Marie (1993) “The Status of Women in Changing Economies: The Case of Nicaragua,” paper presented at the annual meeting of the Midwest Economics Association in Indianapolis, Indiana, April 1–3, unpublished. Babb, Florence E. (2001) After Revolution: Mapping Gender and Cultural Politics in Neoliberal Nicaragua, Austin, TX: University of Texas Press. Berger, Silvia (1999) “Economic History, South America,” in Janice Peterson and Margaret Lewis (eds), The Elgar Companion to Feminist Economics, Cheltenham, UK and Northampton, MA: Edward Elgar: 246–57. Castro, Mary Garcia (November 1995) “Mulher negra, resistencia e cidadania; e o lugar da mulata?”, Presenca da Mulher: 26–33. CFEMEA (Feminist Center for Studies and Advisory Services) (1999) Women’s Rights and the Legislative in Brazil, Brasilia, DF. Chamorro, Amalia (1989) “La Mujer: Logros y Limites en 10 Anos de Revolucion,” Cuadernos de Sociologia (January–February): 117–43. Chinchilla, Norma S. (1995) “Revolutionary Popular Feminism in Nicaragua: Ideologies, Political Transitions, and the Struggle for Autonomy,” Women in the Latin American Development Process, Philadelphia, PA: Temple University Press: 242–70.

78 Rose-Marie Avin Comision Nacional de Zonas Francas (2001) Memoria 1997–2001, Managua, Nicaragua. Delgado, Didice G., Paola Cappellin and Vera Soares (eds) (2000) Mulher e Trabalho: Experiencias de Acao Aﬁrmativa, São Paulo: Boitempo Editorial. Human Development Report 2000 (2000) New York, Oxford: Oxford University Press. Hutauruk, Deborah (1998) “Women in Nicaragua: You’ve Come a Long Way,” Nica News, 19 (December). Inter-American Development Bank (1987) Economic and Social Progress in Latin America 1987. Library of Nations (1987) Brazil, Alexandria, VA: Time-Life Books, Inc. Ministry of Education (2000) Education in Brazil: 1995–2000. Montenegro, Soﬁa (1997) “Un Movimiento de Mujeres en Auge,” in Movimiento de Mujeres en Centroamerica, Managua, Nicaragua: Programa Regional La Corriente. Sen, Amarya (1999), Development as Freedom, New York: Anchor Books. Todaro, Michael P. (2000) Economic Development, seventh edition, New York: AddisonWesley. World Bank (2001) Engendering Development: Through Gender Equality in Rights, Resources, and Voice, New York: Oxford University Press. World Bank Development Report 1999/2000 (2000), New York: Oxford University Press.

Engendering agricultural technology for Africa’s farmers Promises and pitfalls Cheryl Doss

Improved agricultural technologies are key to reducing poverty and increasing the standard of living in Africa. The logic behind this statement is straightforward. A majority of Africans live in rural areas and in many countries, as many as 70 percent of people earn their primary income from agriculture. Moreover, African farmers are among the poorest people in the world. One reason for the continuing poverty is that crop yields in sub-Saharan Africa are extremely low. For maize, one of the staple crops, yields in Africa are only 60 percent of those in Latin America and the Caribbean. Similarly, cassava yields are only 75 percent those in Asia and only 40 percent of those in India. Productivity increases in the rest of the world have meant that world prices for these crops have fallen over the past four decades. But these productivity gains have not been matched in Africa. As a result, African farmers face falling world prices but have not had a comparable reduction in the costs of production. Improved technologies are needed in Africa to increase productivity, reduce poverty, and reduce the pressure for unsustainable land use. In particular, technologies are needed to improve the lives of women farmers in Africa, who work long hours in difﬁcult conditions for low levels of output. A wide variety of agricultural technologies have been introduced to improve agricultural productivity. Some of them have been successful in beneﬁting women farmers, while others have not. The failures have generated a sense among many that improved agricultural technologies do not beneﬁt women. The literature critical of improved technologies,1 as well as that critical of policies to promote economic growth, often seems to suggest that women would be better off without improved technologies and economic growth. Yet clearly this cannot be the case. Instead, we should conclude that technologies to improve agricultural productivity are a necessary, but not sufﬁcient, condition for African women farmers to improve their welfare. Similarly, economic growth is necessary but certainly not sufﬁcient for improving women’s welfare, especially in the poorer countries. Thus, we need ask how we can develop agricultural technologies, and more broadly, policies for economic growth, that beneﬁt a broad spectrum of society, including women, and are sustainable. The World Bank Policy Research Report, Engendering Development (World Bank 2001), draws some general lessons on how to ensure that women beneﬁt

80 Cheryl Doss from development. Other chapters in this book provide a detailed critique of this research report. Many of these chapters emphasize the macroeconomic environment and discuss how the World Bank has not incorporated gender issues into its macroeconomic policies. In this chapter, I focus much more on microeconomic issues. I examine the extent to which the lessons derived from the report can be applied to policies to improve the well-being of African women farmers and discuss additional factors that are not considered or emphasized in the report. Although macro policies affect agriculture, this chapter will not spend much time looking at macro issues. Among the other works that have considered the effects of macro policies, especially structural adjustment programs, on women farmers in Africa are Gladwin (1991) and Mehra (1991). The World Bank report does not directly address the issues of agriculture and agricultural technology. In addition, it does not directly address the issues of power relations within households, communities, and countries. These power relations affect how roles and responsibilities are renegotiated when new opportunities, such as new technologies or new market opportunities, are created. Instead, the report emphasizes the importance of economic growth and discusses some of the conditions necessary for women to benefit economically from economic growth. It is critical to take this report an additional step further and ask how we can develop agricultural technologies and policies that will improve the agricultural productivity and well-being of women farmers in Africa. One way in which it is critical to move beyond the lessons in the report is to explicitly recognize the power relationships within households, communities, and countries. Many of the reasons that African women farmers have not beneﬁted from new technologies – even when the technologies were speciﬁcally designed to beneﬁt them – is that the women did not have the power to negotiate the details to their advantage. This concern also means that it is difﬁcult to generalize about policies that will beneﬁt all women farmers in Africa. There are many differences among these women farmers. Some of these differences can be described by statistical variables – wealth, income, education levels, and status within the household (as household head or spouse). Other economic variables also may be important. Women’s access to markets varies considerably: Those in peri-urban areas may have access to urban markets while for others, limited infrastructure and high transportation costs mean that it is difﬁcult to market their products. Ethnicity and kinship networks shape women’s power within their households and communities. National government policies, both economic and non-economic, will affect their abilities to take advantage of new opportunities. Many women still work on subsistence farms, but in some areas, especially in Southern Africa, women work as wage laborers in the commercial agricultural sector. This issue, namely which women beneﬁt and which women lose, is ignored in the World Bank report, but must be taken into consideration.

Engendering technology for Africa’s farmers 81

Women’s roles in agriculture production in Africa It is widely quoted that women in Africa produce 70–80 percent of the food and provide more than 60 percent of the agricultural labor force.2 Formal labor force statistics indicate that 42 percent of the economically active population in agriculture is female (UN 2003), but these estimates certainly underestimate women’s role, since women in farm households are often registered as housewives rather than farmers. Many studies have attempted to quantify women’s contributions to agriculture. Bryson (1981) categorized farming systems as “male predominance,” “female predominance” and “equal participation” and found that 52 percent of a random sample of sub-Saharan African societies were “female predominance” systems, compared with only 34 percent worldwide. Only 19 percent of those in subSaharan Africa were “male predominance” systems. Saito (1994) found that women work more hours than men in agriculture in Burkina Faso, Kenya, Zambia, and Nigeria. Kumar (1991) found similar results for maize-growing households in Zambia, claiming that the average number of hours spent cropping is higher for women than men. As men increasingly become involved in wage labor, both in rural areas and by migrating to cities, women frequently take over many of the agricultural tasks. Thus, women’s contributions are not static and they do vary across locations within Africa, but clearly any consideration of gender issues in development in Africa must consider women farmers. An extensive literature emerged in the 1980s critiquing the lack of attention to women in agriculture (e.g. Staudt 1987; Cloud 1985; Poats et al. 1988). Unfortunately, when this literature was translated into policy circles, it was often simpliﬁed to suggest that what was needed was a kind of sensitivity training for policy-makers and researchers. If only policy-makers and senior scientists understood that their clients were women, then technologies to beneﬁt women farmers would be developed, agricultural productivity by women farmers would increase, and poverty would be reduced. Policy failures were usually attributed to policymakers – and speciﬁcally to their lack of knowledge or commitment concerning the problems of women farmers. In hindsight, this view was quaintly naïve. Many policy-makers did in fact need sensitivity training (and some still do) and it is an important step to realize that women are key actors in agricultural production. But good intentions, even if they are genuine, are not enough. Research institutions, including the international scientiﬁc institutions and the World Bank, have been fairly responsive to the idea that women are farmers in Africa. Pushed by grassroots development workers and feminist researchers, they have tried to target technologies and policies toward women. Yet the ways in which they typically do so are not sufficient to ensure that women will beneﬁt from these innovations. Typically, when looking at gender issues in the agricultural sector, women farmers have been marginalized. In an effort to be responsive to the needs of women, these organizations have tried to prioritize “women’s crops” and “women’s tasks.” Both of these approaches compartmentalize women’s contributions in agriculture and thus limit the understanding of women’s work and women’s needs.

82 Cheryl Doss For example, many rural Africans can tell you which crops are grown by men and which are grown by women in their area. One crop that is considered a “man’s crop” in Ghana is yams. While visiting agricultural households in Ghana, I spoke with the head of a local government ofﬁce for women and development. She spoke at length to me about the distinctions between men’s and women’s crops, emphasizing that yams were a man’s crop. She then invited me to her yam farm without any sense of irony. This example emphasizes the difference between accepted cultural norms and actual practices. It is rare that these two completely overlap. Both may be important when considering how the rights and responsibilities from new technologies are allocated. A generalization is often made that men grow cash crops and women grow subsistence crops. Yet the patterns do not hold very tightly under close scrutiny. Using data from the Ghana Living Standards Survey – a nationally representative household survey – I find that there are no major crops grown exclusively, or even primarily, by women or by men (Doss 2002). For example, cocoa is one of the main cash crops in Ghana, yet 23 percent of the farmers growing cocoa on their plots are women. In addition, of the women holding land in Ghana, 12 percent grow cocoa. The comparable number for men is 18 percent. The cropping patterns are gendered – some crops are more likely to be grown by men and others are more likely to be grown by women but there are many exceptions to the rule.3 Many staple crops, including those thought of traditionally as men’s crops, are grown by women, especially women in female-headed households. Thus, programs that target these crops may have important implications for women. The generalizations have some truth, but the reality of who grows which crops and who performs which tasks is quite complex and varies considerably. Simply accepting the generalizations will mean that much of women’s contributions are ignored. Women provide much of the labor for food production in Africa. Yet the nature of women’s contributions varies tremendously, even within countries. In particular, there are large differences in the extent to which women make the decisions and the extent to which women control the output. Women may own and control their own plots of land and the output from these plots. Women may work on household plots, where they have some rights to the output, especially for the output destined for household consumption. Women may also provide labor on men’s plots and may or not be directly compensated. Individual women may do more than one of these – they may own their own plots and also work on the household and their husband’s plot. This makes it difficult to disentangle women’s contributions from overall household production. Yet, understanding women’s roles within particular contexts is important in order to understand how they will be affected by changes in technologies. Discussions of men’s and women’s tasks usually note that clearing ﬁelds and plowing are men’s tasks while weeding and post-harvest processing are women’s tasks. These distinctions may be clearly deﬁned culturally, yet they may not reﬂect actual practices. Consider the parallels in the USA, where many might claim

Engendering technology for Africa’s farmers 83 that childcare and cooking are women’s responsibilities while maintenance of the lawn and car are men’s responsibilities. These distinctions suggest the cultural norms, but there are many exceptions within any community. Again, this simpliﬁcation may ignore many of the contributions that women make and many of the responsibilities that they face. While there is extensive variation in the gendered patterns of agricultural production across Africa, one of the things that is true throughout the continent is that these patterns are changing. As new economic opportunities develop and as new technologies are available, the traditional patterns change. For example, as men migrate to the cities or take up wage employment in rural areas, women increasingly take over the traditionally male activities of clearing and plowing. It is less common for men to take over women’s tasks. Yet when the returns to a traditionally woman’s activity are increasing, either due to new technologies or new market opportunities, we may see men taking over women’s activities. Men may become involved in producing crops culturally deﬁned as women’s crops when they become profitable. Around Techiman in Ghana, women traditionally have grown tomatoes. When the market forces expanded the demand for tomatoes, the proﬁtability of growing tomatoes increased and men have recently become involved in their production. Similar issues have been documented in Nigeria, where men began to grow crops that were traditionally women’s crops, once the market developed (Ezumah and Domenico 1995; Burﬁsher and Horenstein 1993). Targeting agricultural innovations towards women based on the tasks that women do or the crops that they grow assumes the environment is static. It does not take into consideration how these activities will change after the new technology is introduced. In particular, it does not address any of the underlying ways in which power relations within households and communities will affect who gains and who loses from the new technologies. The outcomes from new technologies are often contested as is discussed in detail below.

Improving agricultural technologies Although agricultural production in Africa increased by an average of 3.2 percent per year during the period from 1981 to 2000, much of that growth was due to increases in the area planted, which grew at an average rate of 2.8 percent per year. Yields only increased by 0.36 percent per year, much lower than the average yield increases of 1.8 percent for all developing countries. Modern varieties actually generated growth in production of 0.47 percent but an accompanying decline in other inputs accounted for a 1 percent annual decline in yields (Evenson and Gollin 2003). Thus, new agricultural technologies have had less of an impact in Africa than elsewhere. Many of the new technologies developed for African farmers have been targeted at smallholder farmers, and hence at women (although that is not always explicitly recognized).

84 Cheryl Doss Improved seeds, especially maize and wheat seed, designed especially for smallholders are being developed through the International Agricultural Research Centers (CGIAR), working in conjunction with National Agricultural Research Centers. These seeds have been developed to overcome some of the problems that smallholder farmers face, including low yields, susceptibility to drought, and susceptibility to weed and pest infestations. A recent series of studies compiled in East Africa by the International Center for the Improvement of Wheat and Maize (CIMMYT) in collaboration with National Agricultural Research Systems indicate that the adoption levels of improved seeds are fairly high in some parts of East Africa. Many farmers, however, are recycling seed rather than purchasing new seed, thereby gaining some, but not all, of the potential beneﬁts.4 One of the other problems facing agricultural productivity of farmers in Africa is soil fertility. In much of East Africa, the proportion of farmers using either organic or chemical fertilizer is relatively small. Although extension services promote the use of chemical fertilizer, farmers complain that it is expensive and often unavailable at the right time or in the right formulations. Given the current infrastructure and prices in many rural areas, it may not be proﬁtable for farmers to use these technologies. Other technologies to improve soil fertility include management practices. These often require additional labor as an input, rather than purchased fertilizer. Many changes are taking place for agricultural households in Africa. Farmers – both male and female – respond to new economic opportunities and to new technologies when it makes sense for them to do so. Yet some households and some individuals are better placed to adopt the new technologies and to beneﬁt from them; women are less likely to be among these farmers.

Women’s use of improved agricultural technologies If one group of farmers, such as women farmers or farmers in female-headed households, does not have access to a new technology, then it is unlikely to be of much beneﬁt to them. They might beneﬁt indirectly, if increased agricultural productivity increased nonagricultural opportunities for them or lowered the prices for the foods that they purchase. These indirect effects are likely to be small without significant increases in productivity. The indirect effects may be overshadowed by the direct effects of providing technology to only some in that the disparities between groups will widen over time, as those using the new technologies will increase their power and ability to gain continued access to these resources and technologies. Kumar (1987) noted that there were three problems with failure to incorporate women’s roles when developing new technologies and designing projects. First, she notes the loss of adaptive efﬁciency from not taking women’s operational knowledge into consideration. Second, women’s bargaining power within the household may decrease and their workload may increase. Finally, not taking women’s roles into account results in lower adoption rates of the new technology

Engendering technology for Africa’s farmers 85 by women. The ﬁrst and third of these are efﬁciency arguments – making the claim that, in the short run, by not including women some of the potential beneﬁts of new technologies are not captured. The second one provides one way to look at some of these broader issues about how technologies may affect women’s well-being. But the real challenge is to know, a priori, about the power relations within households and communities so that we can sense whether women will be able to take advantage of any of the beneﬁts. Generally we ﬁnd that fewer female farmers than male farmers use improved varieties of seeds and fertilizer. In the twenty-one recent CIMMYT studies5 that estimated the probability of adoption of improved varieties of wheat and maize, only three included gender as an explanatory variable. In the fourteen studies that included an estimation of the probability of adopting fertilizer, gender was included in only two of them. The measure included in all of these cases was the gender of the household head, not the gender of the farmer. In all of these studies, female-headed households were less likely to use the improved technologies than male-headed households. In a detailed study examining the adoption of maize technologies in Ghana, we found that only 39 percent of female farmers planted improved varieties of maize in the 1997 cropping year, while 59 percent of male farmers did so (Doss and Morris 2001). Only 16 percent of female farmers used fertilizer while 23 percent of male farmers did so. We were interested in whether these differences could be explained by differential access to the resources needed for farming or whether men and women appeared to making different decisions about technologies, given a particular situation. Once we controlled for access to the key complementary inputs, especially land and extension services, the gender of the farmer was no longer statistically signiﬁcant in explaining the adoption decision. This result suggests that it is not the gender of the farmer per se that affects adoption decisions, but rather that gender is important because it affects farmers’ ability to gain access to the necessary inputs. Further investigation indicated that women in female-headed households were less likely to adopt improved maize seeds, even after for controlling for other inputs, than male or female farmers in male-headed households. Thus, both male and female farmers in male-headed households were making similar decisions, based on their resources, but female-headed households were less likely to adopt new technologies in the same situation. This may be because male household heads have greater access to resources than do female household heads in ways that we could not account for in our empirical analysis. Many studies looking at the adoption of technology now include a gender variable. Yet, much of the time, it is simply a variable indicating the gender of the household head. In some instances, this may be appropriate, if the household head is the one making the agricultural decisions. In other instances, however, we may want information about the wife who is doing the farming. Examining only the gender of the household head leaves out of the analysis all of the women farmers living in male-headed households. This may be a signiﬁcant omission. Work on gender patterns of cropping in Ghana (Doss 2002) ﬁnds that the patterns

86 Cheryl Doss differ depending on how you deﬁne a “female farmer.” Three deﬁnitions were used. The gender of the farmer may be deﬁned as the gender of the head of household, the holder of the land, or the one who has the right to keep the proceeds. One important result is that women in female-headed households are more likely to be growing staple crops than women in male-headed households. Thus, the deﬁnitions matter.

Household decision-making Discussions of the costs and beneﬁts of new technologies to men and women implicitly assume that the costs and beneﬁts facing men and women are different. If we think of the household as a single entity, making decisions about how best to allocate the workload and the outputs, then we would expect that the costs and beneﬁts would be shared – and that anyone who faced the increasing costs would receive a proportionate share of the beneﬁts. Yet this doesn’t seem to be necessarily the case. Many factors determine who faces the costs and who reaps the beneﬁts of new opportunities. The household bargaining literature suggests women’s outside options affect how these costs and beneﬁts are distributed among household members. For example, recent work in Ghana indicates that when women own a greater share of household assets, household expenditure patterns differ (Doss 1997). A greater share of the household budget is spent on food and less on recreation and on alcohol and tobacco when women own more of the household assets. One interpretation of this is that women with more assets have a greater inﬂuence on household spending patterns. Others have found similar effects – when women own more of the assets, they have more power to determine outcomes (e.g. Quisumbing and Maluccio 2003). Intrahousehold analyses help us to understand when new agricultural technologies will beneﬁt women and when they will make their situations worse off. An extensive literature is developing whereby econometric tools are used to demonstrate the importance of bargaining power. This literature is extensively discussed in the World Bank report in their chapter on intrahousehold issues. The challenge with this empirical literature has been to find measures of bargaining power that are quantiﬁable. The measures that have been used, such as ownership of land, not only are measures of bargaining power but are also outcomes of individual power within their household and community. The studies are limited both by the availability of data and by the difficulty in conceptualizing and quantifying measures of power. Further limitations of these studies are discussed in Chapter 11 of this volume by Orgocka and Summerﬁeld. Much of the empirical intrahousehold literature has assumed that households make cooperative decisions and that the outcomes are efﬁcient, often using a measure of Pareto efﬁciency which assumes that no one individual can be made better off without making another individual worse off. The efﬁciency is often tested in the literature using consumption decisions as outcome measures. It is important to realize that, given the levels of bargaining power for each household

Engendering technology for Africa’s farmers 87 member, an outcome that is inherently inequitable may be efﬁcient. The one area of the empirical literature where noncooperative or inefﬁcient outcomes have been clearly identified is in agricultural production. In the following examples, the introduction of new opportunities resulted in household roles and responsibilities being contested and renegotiated. These examples demonstrate the importance of considering the dynamic impacts of new opportunities. In The Gambia, rice was traditionally grown on women’s plots. A project to increase women’s agricultural productivity introduced centralized pump irrigation. The net effect of this project was to shift control of rice production from women to men. Irrigated rice was grown under the authority of the male compound head and was no longer under women’s control (von Braun and Webb 1989). If the technologies require inputs from men, such as labor or cash, then women may have to give up some of the beneﬁts in order to gain access to these inputs. If the technologies don’t increase women’s outside options, women may not gain the beneﬁts from the new technologies. If women can’t obtain the other inputs unless they live within a male-headed household, their outside options will not increase. One ﬁnding that has implications for agricultural technology is that because individuals within households may be making independent decisions, the results may not always be efficient. For example, in Burkina Faso, one study found that by shifting inputs, especially fertilizer, from plots controlled by men to plots planted with the same crops but controlled by women, total level of household production would have increased (Udry 1996). Similarly, in Cameroon, labor was not allocated efﬁciently across men’s rice ﬁelds and women’s sorghum ﬁelds – by reallocating labor across the ﬁelds, the total production would have increased ( Jones 1993). A ﬁnal example looks at decision-making within pastoral households about where to locate the household and the impact of location on women’s milkmarketing opportunities (McPeak and Doss 2003). Men traditionally make the decisions about where to locate the pastoralist households among the Gabra of northern Kenya. These decisions have an impact on the ability of women to take advantage of new opportunities to sell milk in local markets. Empirical results suggest that men may make decisions about location speciﬁcally to limit women’s ability to market milk. This may be to limit women’s access to independently controlled income. In this case, a new market opportunity, rather than a new technology, challenges traditional roles and is contested within the household. These analyses do not sufﬁciently provide us with an understanding of why a household might reach an inefﬁcient outcome – an outcome where the same resources could have been used to obtain a higher level of production. One explanation that can be inferred from these is that men do not want to give up the power that they hold, even if in the short run there might be gains from doing so. Giving women increased access to resources (whether increased output on their plots or cash from milk sales) may have the effect of shifting the power

88 Cheryl Doss dynamics within the households. These potential shifts in power are hotly contested, even when there is the possibility for increased production for the household as a whole. This approach provides us with a starting point for understanding why technologies may have differential impacts on men and women. Yet, it is not sufﬁcient. Simply looking at outside options is only one way of conceptualizing bargaining power. The power relations within households may be more complex. Sen notes the importance of considering perceptions, well-being, and agency (1990), rather than just some of the measure of bargaining power and individualist outcomes. Folbre (2001), among others, critiques the models for their individualistic approach.

Impacts of agricultural technologies on women Technologies will have a variety of effects. Any given technology may have both positive and negative effects. For example, some improved varieties of seeds have greater yields but they may also require greater inputs, especially in terms of labor. New technologies may reduce the risk that a farmer faces (for example, irrigation technology may ensure an adequate water supply) or they may increase risks, making farmers more vulnerable to changes in weather or market conditions. Even technologies that are designed to beneﬁt women will have both costs and beneﬁts for women associated with them. Thus, we need to explore the circumstances under which technologies beneﬁt women farmers. Individual farmers may come to different conclusions as to whether the costs of a given technology outweigh the beneﬁts. It will matter who bears the costs and who receives the beneﬁts. When technologies increase women’s outside options, they are more likely to have a positive effect on women’s well-being. There are three key dimensions along which technologies affect farmers. The effects can vary along the three dimensions, which may make it difﬁcult to predict the net overall effect. These three dimensions provide a useful framework for looking at the potential effects of new technologies and asking how we can develop technologies that beneﬁt African women farmers. The first dimension is how new technologies affect women’s workload. Women are already contributing much of the labor for agriculture, in addition to their household work of preparing meals, fetching water, childcare, laundry, and cleaning. Technologies that reduce women’s workload – and especially technologies that reduce the drudgery of some tasks, such as weeding, fetching water or ﬁrewood, or pounding grain – will have a positive impact on women. Even technologies that increase the productivity of women’s work may have negative impacts on a woman’s well-being if they also require her to work longer or harder, or if she does not control the output. One of the challenges to developing agricultural technologies is that simply looking at the roles and responsibilities of women before the introduction of new technologies does not necessarily tell us how these will be allocated following the introduction of the technologies. These roles and responsibilities

Engendering technology for Africa’s farmers 89 respond to economic incentives. An interesting example was described in The Economist (2001). Women are usually responsible for collecting water in Africa. However, in South Africa a new technology, the Hippo Roller, has recently been introduced and is gaining wide acceptance. The simple technology involves a plastic barrel that can roll along with a handle that ﬁts into it. Women are able to transport signiﬁcantly more water per trip using this technology. Young men have apparently decided that this technology is “cool” and have begun collecting water themselves (The Economist 2001). It remains to be seen whether young men will continue to collect water after the novelty has worn off! Yet this somewhat trivial example indicates that the gender allocation of these tasks is not static. Reallocations are more frequently in the other direction. When opportunities for increased productivity open up for men as a result of a technological innovation, such as irrigation or improved seeds, women’s labor may be required in men’s activities. If the innovations don’t increase women’s outside options, women may not be able to take advantage of the benefits. Women may be required to contribute additional labor without receiving compensatory beneﬁts. The second dimension is how new technologies affect access to resources, especially land. Women farmers in Africa obtain access to land in a variety of ways. In some areas women have traditionally held their own rights to land. In other areas, they obtain rights to land through a male relative or through the village head. Many places are moving toward formal systems of titling land that create the possibility for formal land markets (see Lastarria-Cornhiel (1997) for a discussion of the effects of land-titling on women in Africa). With formal land markets, land can be bought and sold. Land may become more valuable as increasing market opportunities become available. Farmers in peri-urban areas are increasingly growing crops for sale in urban areas and sometimes even for export. Increasing productivity also makes the land more valuable. This suggests that policies to increase agricultural productivity may have an unintended effect of increasing the value of land and, thus, making it more difﬁcult for women to maintain control over land. It is important to consider the ways in which women have access to land in order to understand the potential impacts of new technologies. Women will beneﬁt more from technologies that increase agricultural productivity when they have secure access to the land and when the ownership of land is part of their outside option. Similarly, as other resources become more valuable, women may have a harder time obtaining access to them. For example, as male wages increase, it will be harder for women to get men to work on their fields. As more farmers are eager for extension services to learn about a newly available technology, it will be harder for farmers who are poorer and less powerful, including women farmers, to gain access to the information. Extension agents are often more likely to go to farmers who are more likely and more able to adopt the recommended technologies. Thus, it is important to have procedures in place at the time that technologies are developed to ensure that they are disseminated to the target audience.

90 Cheryl Doss The ﬁnal dimension is how new technologies affect women’s control over their labor and output. Technologies that are designed to beneﬁt women may have unexpected effects if they result in women losing control over activities once they become proﬁtable. When women’s labor is required for production of crops controlled by men, they may lose, even if the total output produced by the household increases. While women may have control over outputs when they are for household consumption, they may lose control over the outputs when they are destined for sale in the market. It is hard to predict a priori how new technologies will affect women’s workload, access to resources, and control over their labor and output. Certainly having women involved at all stages of the design and implementation of new technologies will help. They will have a more nuanced sense of their power within the household and communities and may have some idea of how these roles and responsibilities will be renegotiated. They may also have some sense of whether the new technologies will enhance or diminish their power.

Engendering Development and agricultural technologies In the World Bank report, Engendering Development, the authors lay out a threepart strategy: 1 2 3

Reform institutions to establish equal rights and opportunities for women and men. Foster economic development to strengthen incentives for more equal resources and participation. Take active measures to redress persistent disparities in command over resources and political voice.

These are fairly general strategies, not speciﬁcally related to women farmers. In fact, agriculture and women farmers are rarely mentioned in the report. Although agricultural extension services are listed in the index, agriculture is not. Agriculture is discussed in the context of women owning plots of agricultural land. It would not be clear from reading the report that a large proportion of the women in Africa live in rural areas and engage in agriculture. Yet, these general strategies are useful ways of thinking about how to ensure that African women farmers beneﬁt from the processes of growth and development. In order to increase agricultural productivity and have women share the beneﬁts of this increased production, women must have control over the land that they farm and their own labor. Otherwise, increasing agricultural productivity will result in the further loss of their control over their labor and land. Related to this, the institutions in the rural sector must be available to women. Women must have access to credit and extension services. Yet, several key things are missing from the report. In some cases, it is because these points are not clearly made in the extensive literature that they reviewed. In other cases, it was simply not emphasized in the report.

Engendering technology for Africa’s farmers 91 In examining the welfare effects of different policies, whether they are new technologies or new markets, it is important to look at both the absolute effects of the policies and on the relative effects across groups, such as between men and women or between urban and rural people. Both are important. They are often used interchangeably, without specifying which matters. The absolute effects clearly are important. But the relative ones matter as well. Much of the intrahousehold literature which examines the effects of bargaining power on household decisions is concerned with the relative bargaining power of men and women in the household. Changing this balance will affect the outcomes. Although quantitative studies have typically been static, we would expect these effects to magnify in the long run. With all of the focus on gender in the report, women are not differentiated. Although it may be important to keep the focus broadly on gender, it is key to note that from any given policy, there may be some women who beneﬁt and some who lose. Most of the issues of class and ethnicity are ignored. Yet, it is important to always ask which women gain and which lose from any situation. It is rarely the case that all women are similarly affected by a policy. This particularly comes out in the section of Engendering Development where the authors discuss household work. They accurately note that women spend much of their time in unpaid work within the household and on the farm. They suggest that, “where available, hired labor can substitute for family labor on farms or in care activities, thus easing the workload of household members” (World Bank 2001: 186). While it may be true that labor markets may ease some of the inefﬁciencies within households, it is likely that the person being hired to do this work will be another woman. Developing labor markets for household work simply transfers it to another group of women. Many different factors will affect whether this is to the beneﬁt or disadvantage of the women being hired. Understanding which women will have the power to renegotiate their labor allocation and control over the output when new opportunities are created is a challenge. The empirical analyses of these outcomes usually do not ask the question as to which women beneﬁt, other than to include some descriptive variables such as age, education level, or wealth.

Conclusion It is challenging to target agricultural technologies, such as improved seeds or fertilizer, to improve the well-being of Africa’s women farmers. Yet it is crucial that these new technologies do not simply bypass women farmers. It is hard to develop technologies, in the abstract, that will benefit women, although technologies that reduce women’s workload and increase their productivity are certainly a start. In addition, it is important to think about how a technology will increase women’s outside options and thus allow them to maintain control over their own labor and some of the beneﬁts from the new technology.

92 Cheryl Doss The strategy proposed in Engendering Development is a general overall strategy. In implementing it, it will be critical to remember that all of the parts must go together. In particular, it is critical to remember that growth and development are necessary, but not sufﬁcient conditions, to improve women’s well-being. The other aspects, such as increasing women’s control over their resources, will be important, especially in the agricultural sector. Translating this strategy into effective policies for African women farmers will continue to be a large challenge. We must ﬁnd ways to both increase women’s agricultural productivity and allow them to beneﬁt from it.

Notes 1 2 3 4 5

See, for example, Marglin 1996. Recent examples of the use of this quote are in UN Wire, Ag Journal, and FAO News Highlights. This study does not look at the share of output or the share of the value of output produced by women since the data was not available to do this. For a synthesis of the studies, see Doss et al. 2003. For a synthesis of these studies, see Doss et al. 2003. For a critique of the studies and the methodology typically use in adoption studies, see Doss 2003.

References Ag Journal (2003) “More Women Work the Land,” Friday, October 3. Bryson, J.C. (1981) “Women and Agriculture in Sub-Saharan Africa: Implications for Development (an Exploratory Study),” in N. Nelson (ed.), African Women in the Development Process, Totowa, NJ: Cass. Burﬁsher, M.E. and N.R. Horenstein (1993) Sex Roles in the Nigerian Tiv Farm Household and the Differential Impacts of Development Projects, Case Studies of the Impact of LargeScale Development Projects on Women 2, New York: Population Council. Cloud, Kathleen (1985) “Women Farmers and AID Agricultural Projects: How Efﬁcient are We?” in Women Creating Wealth: Transforming Economic Development, selected papers and speeches from the Association for Women in Development Conference, April 25–27, Washington, DC: AWID. Doss, Cheryl (1997) “The Effects of Women’s Bargaining Power on Household Health and Education Outcomes: Evidence from Ghana,” paper presented at the Population Association of American annual meetings, New Orleans, LA. —— (2002) Men’s Crops? Women’s Crops? The Gender Patterns of Cropping in Ghana,” World Development, 30(11): 1987–2000. —— (2003) Understanding Farm Level Technology Adoption: Lessons Learned from CIMMYT’s Micro Surveys in Eastern Africa, CIMMYT Economics Working Paper, No. 02–05, Mexico, DF: CIMMYT. Doss, Cheryl, Wilfred Mwangi, Hugo Verkuijl and Hugo de Groote (2003) Adoption of Maize and Wheat Technologies in Eastern Africa: Synthesis of Eastern African Adoption Studies, CIMMYT Economics Working Paper No. 03–06, Mexico, DF: CIMMYT. Doss, Cheryl and Michael Morris (2000) “How Does Gender Affect the Adoption of Agricultural Innovations? The Case of Improved Maize Technology in Ghana,” Agricultural Economics, 25: 27–39.

Engendering technology for Africa’s farmers 93 Economist, The (2001) “Keep it Simple,” Dec. 10–16: 13. Evenson, R. and D. Gollin (2003) “Assessing the Impact of the Green Revolution, 1960–2000,” Science, 3000 (May 2): 758–62. Ezumah, N.N. and C.M.D. Domenico (1995) “Enhancing the Role of Women in Crop Production: A Case Study of Igbo women in Nigeria,” World Development, 23(10): 1731–44. FAO News Highlights (1998) “Women Feed the World: FAO Announces Theme for World Food Day 1998,” March 9. Folbre, Nancy (2001) The Invisible Heart, Economics and Family Values, New York: The New Press. Gladwin, Christina (ed.) (1991) Structural Adjustment and African Women Farmers, Gainesville, FL: University of Florida Press. Jones, Christine (1993) “The Mobilization of Women’s Labor for Cash Crop Production: A Game Theoretic Approach,” American Journal of Agricultural Economics, 75: 1049–54. Kumar, S.K. (1987) “Women’s Role and Agricultural Technology,” in J.W. Mellor, C.L. Delgado and M.J. Blackie (eds), Accelerating Food Production in Sub-Saharan Africa, Baltimore, MD: Johns Hopkins University Press. —— (1991) Adoption of Hybrid Maize in Zambia: Effects on Gender Roles, Food Consumption, and Nutrition, Washington, DC: IFPRI. Lastarria-Cornhiel, Susanna (1997) “Impact of Privatization on Gender and Property Rights in Africa,” World Development 25(8): 1317–33. Marglin, Stephen A. (1996) “Farmers, Seedsmen, and Scientists: Systems of Agriculture and Systems of Knowledge,” in Frederique Apffel-Marglin and Stephen A. Marglin, Decolonizing Knowledge, Oxford: Oxford University Press. Mehra, Rekha (1991) “Can Structural Adjustment Work for Women Farmers?” American Journal of Agricultural Economics, 73(5): 1440–55. McPeak, John and Cheryl Doss (2003) “Are Household Production Decisions Cooperative? Evidence on Pastoral Migration and Milk Sales from Northern Kenya”, mimeo, Department of Public Administration, Maxwell School, Syracuse University. Poats, Susan V., Marianne Schmink and Anita Spring (eds) (1988) Gender Issues in Farming Systems Research and Extension, Boulder, CO: Westview Press. Quisumbing, Agnes and John Maluccio (2003) “Resources at Marriage and Intrahousehold Allocation: Evidence from Bangladesh, Ethiopia, Indonesia, and South Africa,” Oxford Bulletin of Economics and Statistics, 65(3): 283–327. Saito, K.A. (1994) “Raising the Productivity of Women Farmers in Sub-Saharan Africa”, World Bank Discussion Papers No. 230, Washington, DC. Sen, Amartya (1990) “Gender and Cooperative Conﬂicts,” in Irene Tinker ed. Persistent Inequalities, New York: Oxford University Press. Staudt, K. (1987) “Uncaptured or Unmotivated? Women and the Food Crisis in Africa,” Rural Sociology, 52: 37–55. Udry, Christopher (1996) “Gender, Agricultural Production, and the Theory of the Household,” Journal of Political Economy, 104(5): 1010–46. UN Wire (2003) “Bush Must Address Gender During Africa Trip, Experts Say,” July 3. von Braun, Joachim and Peter Webb (1989) “The Impact of New Crop Technology on the Agricultural Division of Labor in a West African Setting,” Economic Development and Cultural Change, 37(3): 513–34. World Bank (2001) Engendering Development, New York: World Bank and Oxford University Press.

Taking gender differences in bargaining power seriously Equity, labor standards, and living wages1 Stephanie Seguino The relative wages of men and women . . . are . . . largely determined by the structure of markets . . . Firms operating in competitive environments discriminate less against women in hiring and pay practice than do firms with signiﬁcant market power in protected environments . . . (World Bank, Engendering Development, p. 17) Making trade conditional on . . . [labour] standards may benefit many of the workers in the north and a few in the south but it is also likely to further marginalize large sections of the world’s poor, particularly in the south. A winwin solution in international trade . . . is unlikely to arise out of ad hoc, piecemeal and self-serving demand for universal labour standards. (Naila Kabeer, Power to Choose, p. 391)

Introduction The World Bank Policy Research Report that is the focus of this book takes up the laudable task of assessing the relationship between gender equality in wellbeing and economic growth. The Bank argues that gender equality in well-being is improved by growth, and in turn, equity is an unequivocal stimulus to growth. The Bank narrowly defines well-being as capabilities, measured in terms of education and life expectancy. But well-being is a much broader concept, and includes female access to resources and opportunities, such as the means to generate a stable and sufficient income.2 The World Bank report gives less attention to this aspect of well-being. This may not be surprising since the policies needed to promote equity in access to income would require a rethinking of the Bank’s support for neoliberal macroeconomic policies that constrain chances for gender wage equality. As this chapter argues, this tension suggests that there is an inverse relationship between gender wage equality and growth, implying that efforts to close wage gaps could slow economic growth. To advance gender equity, this constraint must be squarely faced and overcome with policy measures.

Gender differences in bargaining power 95 The Bank is silent on the policies that could make wage equality more compatible with economic growth. Instead, the Bank’s prescriptions for narrowing gender gaps in income are based on proposals to level the playing ﬁeld in institutions and human capital investment, and the promotion of economic growth based on principles of free markets and free trade. Despite the Bank’s optimism, several factors weaken the link between economic growth and women’s relative wages. Women’s segregation in “mobile industries” – that is, labor-intensive manufacturing or service export industries – limits their bargaining power vis-à-vis employers, making it difﬁcult for them to bargain for higher wages and job security, even as labor demand increases in the process of growth. This type of job segregation makes it particularly difﬁcult to raise female wages without dampening female employment, given the growing mobility of ﬁrms. This chapter provides evidence of women’s weaker bargaining power in labor markets that sustains gender wage inequality, and I propose ways to overcome this impediment. I discuss two proposals that have recently received a great deal of attention. Labor standards and a global living wage have been proposed as a response to growing global inequality, and are seen as a means to put a ﬂoor under worker compensation, overcoming capital’s increased bargaining power vis-à-vis workers. These strategies are likely to disproportionately beneﬁt women who have typically occupied the lowest paid jobs (including those covered by minimum wage laws). Such policies can promote gender wage equality since they provide some (but not complete) equalization of female and male bargaining power vis-à-vis employers, and effectively raise the minimum wage. These policy proposals are not uncontroversial. The Bank a priori opposes such efforts to interfere with market-determined wage outcomes, advocating instead that greater ﬂexibility in labor market institutions and more female education are sufﬁcient. But in so doing, the Bank appears not to recognize or take seriously women’s relative lack of bargaining power in the workplace. Some from the global South, while recognizing the constraints on women’s bargaining power in the workplace, argue that labor standards and living wage measures that are linked to trade agreements are a disguised effort to protect white northern male jobs. There is thus opposition to such policies on the grounds that, if enacted, they will lead inevitably to job losses for poor women from the global South (Kabeer 2000; Singh and Zammit 2000). The concern over the potential negative employment effects of improving women’s work conditions is related to a key question that this chapter explores: can enactment of labor standards and living wages be successfully implemented in export-oriented developing economies without producing negative effects on output and employment? The ﬁndings suggest that negative effects of improved labor standards and wages can be avoided, or at the very least, attenuated, through the adoption of development strategies that shift the export production mix to goods that are more price inelastic and income elastic. Under some conditions, higher wages may also stimulate increases in labor productivity, neutralizing the effect of wage hikes on unit labor costs, and thus leading to the avoidance of job loss. Further, if labor standards are enacted simultaneously among competitor

96 Stephanie Seguino countries, a country’s relative competitiveness is not likely to deteriorate, with the result that negative employment effects are avoided. The mechanisms necessary to ensure these happy outcomes are dependent on an important role for the state in shaping the development and growth process, in direct contradiction to the Bank’s policy proposals. To elucidate these issues, I begin with a discussion of labor standards and living wages, followed by a discussion of trends in foreign direct investment (FDI) and their gender-related effects. I then discuss the possible pitfalls of labor standards and living wages in yielding unintended outcomes, and ﬁnally address means to overcome those constraints.

Labor standards and living wages: some background Labor standards cover a variety of areas but for the purposes of this chapter, I refer to process-oriented core labor standards that can improve wages: freedom of association, abolition of child labor, and elimination of discrimination in respect to employment and occupation. Global living wages (an outcome-based standard) have also been advanced as a way to improve worker well-being, with living wages deﬁned as a wage level sufﬁcient for an adult to provide for the basic needs of self and dependants. While little has been said about viable mechanisms to enforce the payment of living wages, proponents argue such a policy can slow a “race to the bottom,” acting as a global minimum wage, with the added benefit of being able to reflect country-specific cost of living differences. A global living wage has the potential to narrow gender wage gaps since those most affected by such a wage ﬂoor are women, particularly those in export industries who face limited bargaining power to negotiate higher wages. The effect of such a policy is to narrow the wage dispersion, where women are concentrated in the lower tail of the distribution and men in the upper. This chapter focuses on the macroeconomic effects of both types of legislation, referring readers to a wide literature for consideration of the practical issues of implementation and enforcement.3 Previous research on the macro effects of labor standards has not taken a gender perspective although some studies have attempted to include tests of the effects of reduced gender discrimination in employment. In empirical analyses, Kucera (2002) uses female share of industrial employment while Busse (2002) relies on female share of the labor force as a measure of gender discrimination. In both cases, a higher female share is assumed to be indicative of an improvement in labor standards. Measured in these ways, neither Kucera nor Busse ﬁnd evidence of a negative effect of an improvement in labor standards on foreign direct investment (FDI) and comparative advantage, respectively. While this might appear to be good news at ﬁrst sight, the measures these authors use to capture improvement in gender equity are inadequate to reﬂect the way that investment, trade, gender, and labor standards interact. The problems that women face in improving their relative economic status is less a function of job access than of job segregation coupled with lack of bargaining power to

Gender differences in bargaining power 97 raise their wages in the jobs they can get. Those issues are discussed in the next section as they relate to the current environment of globalized production processes.

Firm mobility, bargaining power, and gender Efforts to promote gender equity via labor standards and living wages, to be successful, must take into account and overcome constraints imposed by the process of globalization. Of particular concern are the open-economy effects of higher production costs on investment and exports which inﬂuence output and employment. This section outlines some of these constraints and their relationship to gender-segregated labor markets. Mobile capital and foreign direct investment Investment liberalization has resulted in an increased ability of ﬁrms to respond to higher costs and more regulation (or the threat of these) by shifting production to countries with less regulation, lower costs, and in general, higher expected profits. The effect of reduced regulations on investment can be measured as the share of total foreign direct investment in gross ﬁxed capital formation, or what I term physical capital mobility. This effect can be captured by summing inward and outward FDI, which is a quantitative assessment of the ability of ﬁrms to relocate across borders, should they face local cost increases, and thus reﬂects changes in their bargaining power vis-à-vis workers. This, in other words, is a proxy measure for the size of the “threat effect” that ﬁrm relocation poses to workers in wage or other labor-management negotiations. The data in Table 7.1 indicate that physical capital mobility has been increasing in much of the developing world (see totals in bold italic type for each region) and indeed, the share of total FDI in gross fixed capital formation has more than doubled in the period 1987–2000. Of particular note is the rise in outward FDI, indicating that FDI is a more unstable capital inflow than it had been previously understood to be.4 The degree of physical capital mobility has implications for labor demand. As ﬁrms become more mobile across borders, they gain access to “substitutes” for domestic labor, and as a result, their demand for labor becomes more elastic. Except perhaps at high skill levels, labor has not become equally as mobile. Thus labor’s options have not expanded.5 The result of this asymmetry is an increase in capital’s bargaining power vis-à-vis workers, both on the front of wages as well as other components of work conditions.6 Mobility and gender This asymmetry suggests that investment now responds more strongly (negatively) to increases in labor costs than in the past. Further, capital’s increased bargaining power has differential effects by gender. This can be traced to the

98 Stephanie Seguino Table 7.1 Trends in FDI in developing countries (%) Region

1987

2000

Change from 1987 to 2000

5.4 0.7 6.1

20.7 2.4 23.1

15.3 1.7 +17.0

7.1 6.9 14.0

10.5 1.2 11.7

3.4 –5.7 –2.3

Asia Inward Outward Total

3.3 1.6 4.9

11.6 7.4 19.0

8.3 5.8 +14.1

Developing economies Inward Outward Total

3.9 1.4 5.3

13.4 5.8 19.2

9.5 4.4 +13.9

Latin America and Caribbean Inward Outward Total Sub-Saharan Africa Inward Outward Total

Source: United Nations (1999, 2002), World Investment Report.

practice of job segregation, with women in semi-industrialized economies typically concentrated in “mobile” industries and men in “immobile” or nontradable industries.7 (Mobile industries can be described as those for which sunk costs, including training costs, are limited and there is easy ﬁrm entrance and exit. These include labor-intensive manufacturing ﬁrms as well as services (such as informatics, data processing, and possibly tourism).8 These industries are more likely to engage in vertical FDI, which is driven by ﬁrm efforts to locate labor-intensive segments of production in least-cost labor sites and capitalintensive production in countries where capital is relatively cheaper. Such ﬁrms tend to produce goods for export rather than sale to the domestic economy. In contrast, horizontal FDI occurs in immobile industries, that is, where ﬁrms locate production in a country to facilitate sales to that economy, due perhaps to import barriers. In the latter case, labor costs have a smaller effect on investment decisions. Women are often the preferred labor force in mobile industries due to their lower wages and purportedly more limited resistance to poor working conditions, two factors which enhance ﬁrm proﬁts.9 The demand for female labor in mobile industries has risen, facilitated by trade and investment liberalization, leading Guy Standing (1989) to dub this period one of “global feminization.”10 Men, on the other hand, tend to be concentrated in non-tradable industries and capital-intensive industries, even if these latter produce for export.11 Men are also more concentrated in industries for which horizontal FDI is relatively

Gender differences in bargaining power 99 high – such as, for example, automobile manufacturing in China or Indonesia. In general these tend to be “immobile industries,” in part due to the limited role that labor costs play in affecting proﬁtability and sales. In the case of nontradables, the price elasticity of demand for such goods tends to be low, and higher wage costs can be passed on to consumers. In more capital-intensive industries, higher wages for men may reduce turnover, protecting the firm’s investment in training. Industry structure and the price inelasticity of product demand thus ratify relatively higher male wages. Firm mobility and capital–labor bargaining power These stylized facts help to explain the persistence of gender wage inequality in a global economic environment that otherwise might promote a closure of the gender wage gap, as the demand for female labor rises relatively faster than for male labor. Because women are located in mobile industries in which the threat effect of ﬁrm relocation to lower wage sites is credible, women’s bargaining power relative to that of capital does not rise, even as labor demand increases. Bhattacharya and Rahman (1999) have provided implicit evidence of this in the case of Bangladesh’s female-dominated garment industry. The demand for female labor there has increased sharply in recent years. Despite that, female wages in that sector have not kept pace with rising productivity, so that the wage share of income has fallen – women workers are worse off relative to capitalists. The mobility of garment ﬁrms is a likely explanation for the power differential that has led to this outcome, in addition to other social and economic constraints on women’s bargaining power. In contrast, workers in immobile industries have more bargaining power to demand higher wages and better working conditions. Downward pressure on the wages of workers in mobile industries (in this case, women) can spill over into other sectors of the economy, insofar as jobs are gendered and men and women are not perceived to be substitutes. Thus low wages for women in mobile manufacturing ﬁrms serve to hold down female wages in non-tradable industries, such as retail sales, as well as social and community services. Men’s wages are not similarly constrained. One possible outcome of this process is growing wage inequality. Indeed, the polarization of wages in recent years is a well-documented phenomenon, although the role of gender in this process has received less attention. With regard to gender, we can hypothesize that as physical capital mobility increases, gender wage inequality worsens, even as trade expands and growth continues. There is some evidence of this in the case of Taiwan as compared to South Korea (Seguino 2000a).12 During the period 1981–92, Taiwan liberalized rules on inward and outward foreign direct investment while, by comparison, South Korea, did not substantially alter rules on FDI. Figure 7.1 captures these trends, using the sum of inward and outward foreign direct investment, or total FDI, as a share of gross fixed capital formation. That figure also includes data on Singapore for comparative purposes.

100 Stephanie Seguino 30% 25% 20% 15% 10% 5% 0% 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 Taiwan

South Korea

Singapore

Figure 7.1 Total foreign direct investment as a percentage of gross ﬁxed capital formation.

Interestingly, over the period 1981–1992, gender wage differentials in South Korea narrowed, while in Taiwan, they widened. One of the factors that explains this divergent outcome is the increase in physical capital mobility in Taiwan, as compared to South Korea.13 The interpretation of these results is that women’s bargaining power vis-à-vis capital decreased in Taiwan as female-dominated ﬁrms became more mobile, contributing to a widening gender wage gap. In fact, female wages fell from roughly 68 percent of average male wages in 1981 to 62 percent in 1992. In contrast, gender wage differentials improved slightly in South Korea over this period. These results imply that growth is not sufﬁcient to close gender wage gaps (a point made in this volume by numerous authors, including Avin and Doss). The inverse relationship between capital mobility and wages, coupled with gendered job segregation, suggests that efforts to raise female wages in semiindustrialized economies might result in reduced output and growth, and thus declining employment, particularly for women. There is thus potentially an inverse relationship between gender equity, measured as higher relative female wages, on the one hand, and growth of output on the other. Empirical analyses are consistent with this, and one study (Seguino 2000b) indicates that, among a set of semi-industrialized economies, those with the greatest gender wage inequality grew the most rapidly during the period 1975–95. Higher relative female wages, then, appear to slow growth, at least for this set of countries, for two important reasons: (1) investment responds negatively to increases in female wages, and (2) exports fall as export prices rise, reducing the ability of an economy to import capital goods, thereby slowing productivity growth. The analytics of this problem are addressed in the appendix within the framework of a simple neo-Kaleckian model of an open developing macro economy (see also Seguino 2000b; Blecker and Seguino 2002).

Gender differences in bargaining power 101 These results stand in sharp contrast to the Bank study which argues that gender equity promotes economic growth. The Bank report focuses on educational gaps as a measure of inequality, and implies that those countries that treat women better grow faster. But as the preceding analysis suggests, the effect of gender inequality on growth depends on how inequality is measured. While closing educational gaps may potentially stimulate growth by improving the quality of the labor supply, wage inequality can and does co-exist with educational equality. The gap between educational attainment and wages signiﬁes the degree to which women fail to be sufﬁciently compensated for their productive characteristics – in a word, it is a measure of the degree of exploitation. It is that exploitation which is a stimulus to growth, via its effect on ﬁrm proﬁts and on prices that stimulate product demand. These results suggest that the enactment of labor standards that lead to wage increases might stimulate outward FDI and a decline in exports of labor-intensive goods. A possible result would be employment losses for women, at least partially negating the beneﬁcial effects of higher minimum wages. Kabeer (2000) argues that beneﬁciaries would be northern workers who would gain these jobs. The problem is more likely to be South–South job competition, however, since northern light manufacturing jobs all but disappeared a number of years ago, except in niche markets (Ross and Chan 2002; Polaski 2003). Thus the issue of concern is the effect of (and antidotes to) investment and export competition among a set of semi-industrialized economies. The next two sections consider those issues.

The potential for living wages to reduce gender wage inequality While there are constraints on efforts to improve the living standards of lowwage workers by raising wages, as the previous section shows, there is room to maneuver in achieving this policy goal. This maneuverability is not obvious in World Bank analyses, which assume perfect competition in goods and labor markets, and inﬁnitely price-elastic export demand. An important component of arguments against labor standards is an assumption that wages rise only in one country. Thus, as with the model presented in the appendix, wages are assumed to rise in country A, while they are held constant in competitor countries so that the real exchange rate appreciates in country A. The application of global labor standards and living wage rules, however, implies the possibility of a simultaneous increase in female wages that may leave relative export prices among competitor countries unchanged, e.g. the relative price of garments produced in Bangladesh, say, and Thailand, remains unchanged. The effects of this policy shift would then be quite different than if wage increases occur only in country A. I know of no research that examines the effects of a simultaneous wage increase among countries that are export competitors, but I will sketch what I think to be some important implications of such a strategy. First, insofar as wages rise in

102 Stephanie Seguino alternative or competitor production sites, there is little profit incentive for “footloose” capital to shift investment abroad (since, in effect, unit labor costs in competitor countries rise in tandem). Thus the negative employment effects, induced by higher wages on domestic economies, may be small. Second, because the price of “substitute” export goods rises also, then the negative demand-side effect on exports will be smaller, although presumably this will still be negative. Together, these two possibilities suggest that, if enacted simultaneously amongst competitor countries, living wage effects (or labor standards) on employment may not be negative or only mildly so. This is consistent with findings of a number of studies on employment effects of labor standards, including Kucera (2002) and Heintz (2002). In the latter study of a set of 49 countries from 1981 to 1996, Heintz ﬁnds that the elasticity of employment growth with respect to changes in the real wage is –0.61 – a 10 percent increase in real wages will lead to a 6.1 percent decline in employment, other factors remaining constant. Other factors may adjust when wages rise, however, including productivity, prices, and consumer demand. The effect of higher wages on productivity is taken up in the section below, but before that, it is useful to consider the effects of higher wages on prices and consumer demand. Anti-sweatshop campaigns advertising the harsh working conditions faced by export workers in clothing industries, for example, reveal a disdain for such work conditions in the global South. In one study, northern consumers indicated they would be willing to pay 2–6 percent more for goods produced under more humane conditions, which would cover 100 percent of wage increases needed to bring workers up to a living wage (Pollin et al. 2001). It is thus conceivable that ﬁrms advertising the payment of living wages would experience little decline in product demand if prices were raised. Of course, many goods of this kind are produced and marketed along a commodity chain, with multinationals subcontracting to smaller ﬁrms to produce small batches of goods. The subcontracting ﬁrm, often located in a developing country, is under great pressure to reduce costs. Labor standards and living wages that raise wages would force subcontractors globally to raise prices. But this may not have a large effect on product price if multinationals that market and distribute these goods were forced to squeeze their mark-ups, which have been shown to be very high in the case of goods in which brand name recognition is high (Heintz 2003). The result in both cases is a redistribution from the North (consumers or ﬁrms or both) to the South, in this case, southern female workers. Assuming that women are the primary beneficiaries of such a policy, this implies an increase in the female wage bill, deﬁned as the average female wage multiplied by female employment. Thus, total female income rises, and depending on the effect of living wages on male income, the female share of the total wage bill may also rise, contributing to greater gender income equality. Enactment of living wages and labor standards would also require coordinated exchange rate adjustments between competitor countries that have experienced

Gender differences in bargaining power 103 wage increases, so that real exchange rates remain unchanged. That is, a sufﬁciently coordinated monetary policy would be required so that no country’s exchange rate movements could offset the effects of a wage increase any more than in competitor countries. That is a big assumption, and suggests that any effort to implement living wages would also have to consider coordination of exchange rate policies among countries as well.

Wage-led productivity effects, gender equity, and growth: what are the chances? Another important feature of the negative commentaries on labor standards (as well as the model in the appendix) is the assumption that there is no link between wages and productivity.14 There is a possibility, however, that higher wages for workers in labor-intensive manufacturing industries could stimulate productivity growth, thereby neutralizing the effect of wage increases on unit labor costs and prices. Empirical research on wage-led productivity effects in labor-intensive manufacturing industries is scarce. This may be because we don’t often see ﬁrms in these industries using higher wages as a way to promote increased productivity or quality improvements. In part, this may be because the firms do not have to use wages as a tool to stimulate labor effort. Monitoring is easy; rapid turnover is not costly because of low levels of investment in training and worker skills; and the target labor force is relatively powerless – women have few job alternatives. Thus, ﬁrm strategy often involves a stick, rather than a carrot. Further, if ﬁrms can rely on low wages to achieve a cost advantage, they feel less pressure to raise productivity. Indeed, this might be described as a “low wagelow productivity” trap where wages that are too low slow improvements in productivity, output, and thus living conditions (Seguino 2004). In contrast, externally induced wage increases (via government policy or labor organizing) might, under the right conditions, spur ﬁrms to become more productive, to innovate, to adopt more sophisticated technology – all of which serve to attenuate the negative effect of higher wages on product price and therefore demand.15 Further, higher wages might induce improvements in product quality, again offsetting negative effects of wage increases on demand. Finally, higher wages may contribute to health and nutritional improvements that lead to improved labor productivity on the job. Unit labor costs and wage-led productivity effects One might question why employers fail to raise wages on their own, if higher wages, due to productivity effects, are cost-neutral and possibly even proﬁtable. One explanation is that the wage-led productivity effect described here suggests multiple equilibria, as shown in Figure 7.2. Deﬁning unit labor costs as: C = w/b

104 Stephanie Seguino where C is unit labor costs, w is the nominal wage, and b is labor productivity, then the rate of change of unit labor costs can be written Cˆ = wˆ − bˆ

where the “hat” signiﬁes rate of change. Over a certain wage range,16 productivity growth is a positive function of the rate of change of the nominal wage or bˆ = f (wˆ ).

Looking at Figure 7.2, the ray from the origin signiﬁes constant unit labor costs. Note that the two wage-productivity equilibria (E1 and E2) reflect identical unit labor costs. We can imagine a situation in which a developing economy ﬁnds itself at E1, representing a low wage and low level of productivity. This is a stable equilibrium with no tendency for the wage, and thus productivity, to rise (Seguino 2004). An increase in the nominal wage however, could stimulate labor productivity growth. If the wage rose to wˆ2, the wage-induced increase in productivity would be sufﬁcient to compensate for the higher wage such that unit labor costs remain constant. As a result, prices are unchanged and thus higher wages do not generate employment losses. What are conditions under which externally induced wage increases could induce productivity growth and quality improvements? While there may be several important conditions, of primary interest is the extent to which ﬁrms are “footloose,” that is, the degree of physical capital mobility. If ﬁrms have few alternatives to domestic labor as wages rise, that is, if they are not easily able to relocate to lower wage sites, then they are disciplined by the higher wages. The bˆ Cˆ = 0

E2 bˆ 2 ˆ bˆ = ƒ (w) E1 bˆ 1 ˆ1 w

ˆ2 w

Figure 7.2 Wage-induced productivity growth.

ˆ w

Gender differences in bargaining power 105 effects of this discipline might take the form of increased corporate initiative to innovate and to improve productivity. They may take the form of more energetic efforts to market goods, or to reorient the product to niche markets where quality matters more and price less. Or simply, ﬁrms that are immobile may observe improvements in productivity and quality as labor effort increases in response to the wage incentive. The limitations on capital mobility force ﬁrms to stay around long enough to observe the productivity increases that would otherwise not have become apparent, had the ﬁrm relocated to lower wage sites. South Korea and productivity growth in female-dominated industries The case of South Korea is instructive. During the period 1975 to 1990, a time of limited inward and outward FDI, real wages more than quintupled in the manufacturing sector. There is evidence that real wage increases led rather than lagged growth, spurring labor productivity and ﬁrm efforts to innovate (Seguino 1999–2000). This period then appears to have been one of wage-led growth, with rising wages stimulating ﬁrms to invest in order to overcome the potentially negative effect of higher wages on export demand. In a sense, firms were squeezed by higher wages. To regain prior proﬁt levels, ﬁrms were pushed to invest as a way to raise producivity. An interesting feature of this period is that wages and productivity rose rapidly in female- as well as male-dominated industries. This is surprising, since it is often argued that wages are unlikely to produce signiﬁcant increases in productivity in labor-intensive industries as compared to capital-intensive industries. The reasons advanced are that these industries simply do not lend themselves to greater mechanization, and most productivity gains resulting from process innovation have probably already occurred. The data from South Korea, however, suggest that wage increases stimulated productivity growth, even in labor-intensive female-dominated industries. The data in Table 7.2 are indices of labor productivity in selected female- and male-dominated manufacturing industries during Table 7.2 Indices of labor productivity in selected South Korean manufacturing industries, 1976–1990 Industries

1976

1990

Female-dominated Wearing apparel Footwear Electronics

100 100 100

481 562 808

Male-dominated Iron and steel Machinery Transport

100 100 100

483 592 317

Source: Korean Productivity Center.

106 Stephanie Seguino the period 1976–90. Note that productivity gains in female industries are similar to or exceed those in male industries. The case of South Korea suggests that living wage standards that raise the wages of female manufacturing workers can stimulate productivity growth, either through increases in labor effort or because ﬁrms are prodded to become more efﬁcient, and perhaps to increase investment in more sophisticated technology.17 This result is less likely to occur in an environment of footloose capital, it would seem, since firms can bargain down wages, using low wages as a cost advantage in lieu of productivity-enhancing investments. If that is the case, we would expect to see slower – not more rapid – productivity growth in countries for which physical capital mobility is high. That is, investment liberalization, because it reduces worker bargaining power and thereby depresses wages, can lead to a “low wage-low productivity growth” trap. In short, investment liberalization can make ﬁrms “lazy” in pushing for cost advantage via efﬁciency gains. The data presented in Figure 7.3 are consistent with this hypothesis, and show that those countries with the least physical capital mobility (total FDI as a share of investment) have had the most rapid productivity growth in recent years. This result has been conﬁrmed in an analysis of 37 semi-industrialized economies for the period 1970–2000 (Seguino 2004). These results suggest that global labor standards, which essentially act as a constraint on capital mobility by reducing the incentive for ﬁrms to run from higher wages, may also induce higher rates of productivity growth.

Growth in real output per worker

4% South Korea

Indonesia

Taiwan 3% 2%

Brazil

Singapore

1% Thailand

1% 0% 0%

10%

15%

20%

25%

30%

Total FDI as % investment

Figure 7.3 Average productivity growth and total FDI as percentage of gross ﬁxed capital formation, 1972–1990.

Gender differences in bargaining power 107

Taking bargaining power differentials seriously: policy implications The patterns of gender segregation and the differential bargaining power that arise in jobs in mobile vs. immobile industries underscore the need for intervention in labor markets as part of any effort to close gender gaps in well-being. A hands-off approach to labor markets will not sufﬁce. Focusing on labor standards and living wages as a means to close gender gaps, there are four related policies that can be adopted to make such a strategy work, in the sense that women’s wages rise and job losses are avoided.18 First, coordination among developing countries to ensure stable relative export prices would be necessary. In particular, monetary authorities would play a role in ensuring that no country’s exchange rate movements offset the effects of a wage increase any more than in competitor countries. Second, governments should develop industrial policies to move their laborintensive sectors up the industrial ladder to the production of goods that are income elastic and where price matters less, and quality more. When this occurs, the price elasticity of export good demand is lowered, and the negative employment effects of wage hikes are attenuated. The role of the state is particularly important in the case where exports are largely produced by small ﬁrms since they often lack the technological expertise, or access to marketing and distribution networks that would permit them to shift to the production of more price inelastic export goods. Indeed, one reason why ﬁrms may have been able to respond to the wage-push stimulus in South Korea is that they possessed the internal resources, as large conglomerates, to purchase new technologies and to implement new processes that raise productivity. Smaller ﬁrms may not be similarly equipped. Rama (2001) found, for instance, that when Indonesia doubled the real minimum wage in the early 1990s, productivity and employment rose in large manufacturing firms, but not in small firms. In fact, workers in small firms experienced substantial job losses as a result of the minimum wage hike. Why was this so? It may very well have been due to the technical and resource constraints small ﬁrms experience in attempting to raise productivity. This is a particularly important issue as regards the enactment of living wages and gender equity, since women workers tend to be more concentrated in small ﬁrms with informal work arrangements than men. Third, anti-sweatshop campaigns have advanced a variety of mechanisms to link better labor practices with rewards in product markets. Heintz (2002) discusses these in detail. I simply note here that these campaigns have gender implications and efforts to improve gender equity should include ways to integrate such northern campaigns with efforts to improve work conditions and wages for women in the global South. The World Bank’s support of such a strategy would be beneﬁcial. Finally, a major constraint on making living wage and labor standards work is ﬁrm mobility. There is increasing evidence that it is time to implement “speed bumps” for physical capital as has been done for ﬁnancial capital in such countries

108 Stephanie Seguino as Chile. Reductions in ﬁrm mobility can permit wage-led productivity effects to emerge, thus producing a gain for workers, particularly female workers, in terms of higher wages while keeping unit labor costs low. How is this to be done? This is a question that requires some study and debate. I will point out here that South Korea, Taiwan, and Japan, and, more recently, China are examples of countries that have managed and constrained foreign direct investment to their advantage.19 Controls on investment are not now in vogue. The current World Bank fashion, for example, emphasizes liberalization. But these winds of liberalization should not prevent those concerned with gender equality from arguing for more gender-equitable macro strategies and a debate on how to implement such policies. A ﬁnal note on the effects of labor standards on women employed in informal sector work: the effects of such standards on the work conditions and wages of those employed in the informal sector are ambiguous. Among informal sector workers are the self-employed and those working as home workers doing industrial piecework. The latter group may beneﬁt indirectly, if formal sector work conditions improve, thus raising the wage ﬂoor. Further, technological improvements induced by higher wages as formal sector ﬁrms attempt to hold down unit labor costs would make those industrial piecework activities less proﬁtable. An effort to organize home workers is likely to be necessary to prevent increased informalization of formal sector jobs. Because informal sector work is often residual employment, a more direct mechanism to improve the conditions of those who work as self-employed workers, such as street vendors, would beneﬁt more from macroeconomic stimulus, and would be harmed by policies that increase formal sector unemployment.20

Conclusion The World Bank PRR makes the important claim that gender equity is instrumental for economic development and growth. The Bank deﬁnes equity in terms of education, assuming that wage equality will automatically ensue, so long as liberalization of markets, trade, and capital ﬂows yields competitive conditions and a “level playing ﬁeld.” The Bank’s conclusion that equality unambiguously promotes growth suggests, however, the absence of a theoretical understanding of the linkages between gender, wages, economic structure, and macroeconomic outcomes.21 The uncomfortable facts are that gender wage gaps can persist despite educational equality, and that attempts to close those gaps can slow, not stimulate, economic growth. Uncomfortable though this may be, proponents of gender equity should not shy away from these conclusions. This chapter argues that gender wage gaps are in part the result of unequal female bargaining power vis-à-vis employers in the current globalized environment. The World Bank, in its analyses of gender, fails to note this constraint, in marked contrast to its willingness to acknowledge unequal female bargaining power within households, and the positive effect of outside options, such as human capital, earned income, and transfers, on women’s access to household

Gender differences in bargaining power 109 resources. As a result, the Bank’s report thus fails to mount a broader consideration of the factors that lead to unequal gender outcomes in labor markets. But if we are to close gender gaps in well-being, unequal bargaining power within labor markets must also be acknowledged, and policies must be designed to ameliorate this condition without producing negative effects on women’s access to income through employment. Cognizant of these constraints, this chapter lays out some of the important macroeconomic considerations related to closing gender wage gaps in labor markets via enactment of labor standards and living wages. Such policies can improve the wages of those at the bottom of the wage distribution, where women are concentrated. In so doing, gender wage gaps can be narrowed and thus such policies not only improve women’s absolute well-being but can also promote gender equality in incomes. The evidence on wage-induced productivity growths suggests that higher female wages that approach living wages can provide a stimulus for ﬁrms to innovate. The result will be an increase in productivity, making wage increases costneutral and possibly even profitable, thus protecting against job loss. The realization of this positive effect of higher wages requires constraints on ﬁrm mobility. Even without restraints on ﬁrm mobility, the potential negative effects of higher wages on investment, when adopted in an individual country, can be lessened if implemented as a global labor standard. Industrial policies that shift production to income elastic and price inelastic exports will also help. All of these policy proposals suggest an important role for the state in managing the process of development and growth. Despite the Bank’s resistance to state intervention in the production sphere of the economy, there is a clear connection between such intervention and improvements in gender equity. A virtuous cycle can be set in motion by reasoned state-level actions, with the resulting increase in gender equality further stimulating development and growth – a link the Bank does acknowledge.

APPENDIX

Wages in a two-sector macroeconomy with gender-segregated labor A stylized characterization of developing country economies with a signiﬁcant manufacturing or service export sector (that is, a semi-industrialized economy) is that these are two-sector economies, with a non-tradables sector producing “home” goods that employ male labor and a tradables (export) sector that employs female labor.22 Assuming complete gender segregation, the mark-up price equations for the “home” and export goods sectors, respectively, can be written: PH = τ (wm aH + eP *nH ), τ > 1

(1)

110 Stephanie Seguino PX = ϕ (wf aX + eP *nX ),24 ϕ > 1

(2)

In these equations, τ and ϕ are the mark-up factors; wm and wf represent male and female wages, respectively; H is home goods and X is exports; the ai (i = H, X) are the labor coefﬁcients; the ni are the intermediate input coefﬁcients; P* is the world price of intermediate inputs; and e is the exchange rate (domestic currency price of foreign exchange). The home sector mark-up (τ) is rigid due to high protectionist barriers, heavy subsidies, and a highly concentrated oligopolistic structure. The export-sector mark-up (ϕ) is assumed to be ﬂexible in response to international competitive pressures and in particular, changes in the real exchange rate.23 The primary macroeconomic effects of any increase in female wages are exerted on export demand as well as investment in export industries. To see this, consider ﬁrst the following expression for export demand: ψ

⎛ eP * ⎞ EX = A⎜ X ⎟ , 0

(3)

where EX is export demand, A is a shift factor (such as foreign income), PX* is the price of competitive foreign exports, and ψ is the price elasticity of demand. Equation 3 implies that export demand can be stimulated by a devaluation and a decline in PX. Further, it can be seen that a decline in the price elasticity of export demand will attenuate the negative effects of a rise in the domestic price of exports on export demand. A simple reduced form investment function, with investment expenditures a function of proﬁts in the two sectors, can be written: I = I 0 + b1RH + b2RX

(4)

where IH is home investment goods, and RH and RX are sectoral proﬁts. The left-hand side is desired investment spending and the right-hand side is the investment demand function, in which I0 is a constant term or shift factor (reﬂecting Keynesian “animal spirits”) and b1, b2 > 0 measure the responsiveness of domestic investment to proﬁts in the two sectors (call this the “proﬁtability” effect). We can assume that b2 > b1 since the X sector is more “footloose,” reﬂecting the greater ease with which such ﬁrms can move into or out of a country in response to changes in proﬁtability. These equations can be used to demonstrate the constraints on raising women’s relative wages in an open economy. The effects of raising the female wage operate through two major channels: (1) an export price effect – the export price rises, causing the exchange rate to appreciate; and (2) a class redistribution effect – the ﬁrm-mark-up in the export sector is squeezed by higher wages, reducing the profit share from that sector. The first implies that higher female wages will

Gender differences in bargaining power 111 cause export prices to rise, and export demand to fall (equations 2 and 3). Proﬁts may also be squeezed, resulting in a decline in aggregate investment (equation 4). The combined negative demand-side effect is larger the more price-elastic exports (the larger is ψ ). It is possible, of course, that a redistribution to women workers will induce an increase in consumption spending (if female workers’ propensity to consume is higher than that of capitalists).24 This effect would have to be quite large to overcome the negative effects on exports and investment, which are deflationary. This can be seen from the balance equation in an open economy (with the simplifying assumption that the government budget deﬁcit is 0): I WF + X WF

( −)

(5)

(?)

where S is aggregate saving and M is imports demand. The price elasticity of export demand will inﬂuence the size of the effect of higher female wages on export demand. The “footlooseness” of capital will determine the extent to which higher female wages that squeeze proﬁts also depress investment demand (as a result of physical capital ﬂight or outward FDI). The difference in the propensity to consume out of wage and profit income will influence the effect of higher female wages on consumption demand for both the export good and the “home” good. Higher female wages that lead to a decline in output and female employment is likely to occur if capital is footloose, if the price elasticity of exports is high, and if the spending propensities of workers and capitalists are similar. The first two conditions are likely to obtain in countries in the early stages of industrialization, and in that regard, the fears of Kabeer (2000) and others appear well-founded.25 An optimistic scenario – in which higher female wages result in little loss of export demand, little if any reduction in investment, and a boost to consumer demand, as income is shifted from those with a high propensity to save (capitalists) to those with a low propensity to save (female workers) – is likely to be possible only in those cases where capital is immobile (i.e. FDI faces restrictions), where export goods are price-inelastic (because they are skill-intensive, or because they are goods for which quality matters), or where export goods form a non-trivial share of wage goods. One can think of relatively few countries that fall into this category, but an important and instructive example is South Korea (discussed on pages 105–6). Nevertheless, under some conditions, higher female wages are consistent with increases in employment and output, even when there are no wage-led productivity effects.

112 Stephanie Seguino

Notes 1 I am grateful to astute and helpful comments from Christian Weller, Marilyn Power, and participants at the “Global Labor Standards and Living Wages” Conference at University of Massachusetts at Amherst, April 19–20, 2002. 2 Grown et al. (2003) and Malhotra et al. (2002) explore how to deﬁne and measure gender well-being in detail. They note that gender well-being also includes security and empowerment. I leave aside those issues here, focusing primarily on the contradictions evident in the Bank report between the goal of expansion of female opportunities while promoting economic growth under conditions of liberalized trade and capital ﬂows. 3 See, for example, Brenner (2002) on issues related to calculating global living wages, Heintz (2002), Polaski (2003), and Sengenberger and Campbell (1994) on implementation concerns. 4 This is consistent with results presented in Rand and Tarp (2002), whose empirical analysis of business cycle data for a set of developing countries reveals that FDI inﬂows are very volatile, much more so than aid ﬂows. 5 Thus, abstractly speaking, labor supply schedules have not become steeper to compensate as labor demand schedules have ﬂattened. 6 See Zhao (1998) for a demonstration of the negative effect of physical capital mobility on wages within a Nash bargaining framework. Epstein (2000) explores these issues in further detail, exploring the effect of mobility on public ﬁnances as well as wages. 7 Semi-industrialized economies are those with a signiﬁcant share of manufacturing in GDP and a correspondingly small share of agriculture. Manufacturing output tends to be labor-intensive as compared to industrialized economies, with a large share of manufacturing goods destined for export. Amongst this group can be included economies with a large export service sector, such as informatics and data processing. Examples of these include Jamaica and Barbados. 8 This phenomenon may not hold in countries of differing economic structures. For example, in the US, men have held the largest share of jobs in durable goods manufacturing industries where, in recent years, job losses have been high as ﬁrms have relocated to lower wage sites. We do see evidence of this pattern in service export-oriented economies, however, such as in the Caribbean where women are concentrated in “export” data processing jobs and in India, call centers. 9 For data on women’s concentration in export production, see, for example, Standing (1989, 1999), UNDP (1999), and, for a set of semi-industrialized economies, Seguino (2000b). On services see, for example, Freeman (2000). 10 I refer here to the global phenomenon, since in some countries the demand for female labor in labor-intensive industries has fallen, either due to technological upgrading that excludes female workers, or because ﬁrms have relocated to lower wage sites. This process can be observed in the US and Europe, as well as some of the East Asian economies, such as Taiwan and Hong Kong (Kucera and Milberg 2000; Berik 2000). 11 A digression on gender and economic structure in sub-Saharan Africa (SSA) is in order. Unlike other regions, in the agricultural sector of many SSA economies, women are concentrated in subsistence food production, arguably a non-tradable industry, due to lack of internal transportation networks, and the perishability and types of food crops grown. Men are more likely to be involved in “cash” crop food production. Thus, the dichotomization of women as employed in tradables and men in non-tradables breaks down in the SSA. Nevertheless, cash crop production is often labor-constrained, with men depending on wives to provide labor. Rural African women’s labor thus plays a role in generating foreign exchange, although in a different manner than in semi-industrialized economies or the manufacturing sector of African economies.

Gender differences in bargaining power 113 12 A widening wage gap has been noted in several other countries with strong growth as well – e.g. Chile, Mexico, and Hong Kong (UNIFEM 2000). Female relative wage trends may be faring worse in other countries as well, given the push toward subcontracting and home-based work where wages tend to be lower. But because most labor surveys cover establishments with a minimum of 10 workers, women’s low wages in unsurveyed small shops may not show up in the ofﬁcial data, leading to artiﬁcially high estimates of average female wages. 13 In contrast, in the US, increased mobility in the 1990s was more pronounced in male-dominated industries, such as autos and machine tools, leading to a decline in male relative wages. On the “threat effect” in the US, see Brofenbrenner (2000). For a different view, see Black and Brainerd (2002) who attribute the closure of the wage gap to increased competition, rather than the threat effect of ﬁrm relocation holding down male wages. 14 In the model in the appendix, this is measured as the inverse of the labor coefﬁcient (aX) in equation 2. 15 This line of thought is not new. Power (1999) describes the use of parasitic-industries analysis as a motivating argument behind the push for a living wage for women in the early twentieth-century US. Based on Beatrice and Sydney Webb’s analyses, activists argued that industries paying less than a living wage were parasitic on society, which was then required to take up the burden of providing for the additional costs of supporting women, via the family or through the net drain on women’s health and well-being. Demands for a living wage in sweatshops addressed those concerns. 16 At some point, the wage-led productivity function may turn negative (not shown here). Bowles and Gintis (1985), for example, posit that when full employment is approached, and the “cost of job loss” falls, higher wages exert a smaller positive effect on productivity, which may eventually turn negative, if workers “shirk” or the costs of monitoring rise. Female-dominated labor intensive industries lend themselves to monitoring, however, and thus the conditions under which the wageproductivity function turns negative is more likely to be linked to technological constraints. 17 In another interesting study on minimum wage effects in developing countries, Saget (2001) estimates the effect of real wage increases on employment and ﬁnds little evidence of negative effects. This may be explained by non-compliance. It may also suggest that wage-led productivity effects are signiﬁcant. 18 An obvious policy approach, not discussed here, would be to take steps to integrate labor markets. 19 A common reference on this topic is Alice Amsden (1989). See also Braunstein and Epstein (2002) and Ha-Joon Chang (2002). 20 This would not require direct policy action, if redistributions from capitalists to female workers led to a rise in the economy-wide marginal propensity to consume, thus stimulating aggregate demand. See Blecker and Seguino (2002) on this topic as well as the appendix. 21 This comment applies not only to the Bank report in question, but also to numerous Bank studies on the topic poverty and growth. Other authors in this volume address the Bank’s failure to fully address the links between gender and macroeconomic outcomes as well (see, for example, Rose-Marie Avin and Cynthia Wood). The Bank is not the only entity that lacks such an understanding, which extends to other international financial institutions, bilateral agencies, as well as many otherwise progressive macroeconomists. 22 This formulation is from Blecker and Seguino (2002); see that paper for full details of this model. 23 Modelling price formation as a mark-up pricing process (reflecting oligopolisitic power) can in part account for the different viewpoint on the employment effects of labor standards as compared to neoclassical analyses. A neoclassical approach (e.g.,

114 Stephanie Seguino Brown 2001), which assumes perfect ﬁrm competition, would imply an export price equation such as PX = waX + rk where PX is the export product price, w is the wage rate, aX is the labor coefﬁcient, r is the proﬁt rate, and k is the capital coefﬁcient (ignoring intermediate goods for simplicity). Higher wages immediately result in higher prices, and given infinitely elastic demand for exports, demand falls and along with it, employment in the export sector. 24 Thus there is a third potential effect – a consumption effect. There is, however, little empirical evidence on which to base such an assumption about spending propensities, other than Seguino and Floro (2003), which considers only gender differences in saving propensities among workers. 25 Gibson and van Seventer (2000) point out, based on dynamic CGE simulations of the South African economy, that the effects of real wage increases on employment also depend on the macro policy environment. When there is monetary “policy dominance,” efforts to raise wages can be frustrated by monetary authorities that respond to inﬂation (induced by higher wages) by raising interest rates, and thus engineering a contraction and a decline in employment. This is a point worth considering further in trying to anticipate the effects of enactment of labor standards and living wages.

References Amsden, Alice (1989) Asia’s Next Giant: South Korea and Late Industrialization, Oxford: Oxford University Press. Berik, Gunseli (2000) “Mature Export-Led Growth and Gender Wage Inequality in Taiwan,” Feminist Economics, 6(3): 1–26. Bhattacharya, Debapriya and Mustaﬁzur Rahman (1999) “Female Employment Under Export Propelled Industrialization: Prospects for Internalizing Global Opportunities in the Apparel Sector in Bangladesh,” Occasional Paper No. 10, United Nations Research Institute for Social Development (UNRISD). Black, Sandra and Elizabeth Brainerd (2002) “Importing Inequality? The Effects of Increased Competition on the Gender Wage Gap,” NBER Working Paper No. W 9110, Cambridge, MA. Blecker, Robert and Stephanie Seguino (2002) “Macroeconomic Effects of Reducing Gender Wage Inequality in an Export-Oriented Semi-Industrialized Economy,” Review of Development Economics, 6(1): 103–19. Bowles, Sam and Herb Gintis (1985) “The Production Process in a Competitive Economy: Walrasian, Neo-Hobbesian, and Marxian Models,” American Economic Review, 75(1): 16–36. Braunstein, Elissa and Gerald Epstein (2002) “Bargaining Power and Foreign Direct Investment in China: Can 1.3 Billion Consumers Tame the Multinationals?” Center for Economic Policy Analysis (CEPA) Working Paper 2002–13. Brenner, Mark (2002) “Defining and Measuring a Global Living Wage: Theoretical and Conceptual Issues,” mimeo, Political Economic Research Institute, University of Massachusetts at Amherst. Brofenbrenner, Kate (2000) “Uneasy Terrain: The Impact of Capital Mobility on Workers, Wages, and Union Organizing,” submitted to the US Trade Deﬁcit Review Commission. Brown, Drucilla (2001) “Labor Standards: Where Do They Belong on the International Trade Agenda?” Working Paper, Medford, MA: Tufts University. Busse, Matthias (2002) “Do Labor Standards Affect Comparative Advantage in Developing Countries?” World Development, 30(11): 1921–32.

Gender differences in bargaining power 115 Chang, Ha-Joon (2002) Kicking Away the Ladder: Policies and Institutions for Economic Development in Historical Perspective, London: Anthem Press. Epstein, Gerald (2000) “Threat Effects and the Impact of Capital Mobility on Wages and Public Finances: Developing a Research Agenda,” Working Paper, Amherst, MA: University of Massachusetts at Amherst and Political Economy Research Institute. Freeman, Carla (2000) High Tech and High Heels in the Global Economy: Women, Work, and Pink-Collar Identities in the Caribbean, Durham, NC: Duke University Press. Gibson, Bill and Dirk Ernst van Seventer (2000) “Real Wages, Employment and Macroeconomic Policy in a Structuralist Model for South Africa,” Journal of African Economies, 9(4): 512–46. Grown, Caren, Geeta Rao Gupta and Sahia Khan (2003) Promises to Keep: Achieving Gender Equality and the Empowerment of Women, Washington, DC, International Center for Research on Women. Heintz, James (2002) “Global Labor Standards: Their Impact and Implementation,” Working Paper Series, No. 45. Amherst, MA: University of Massachusetts at Amherst and Political Economy Research Institute. —— (2003) “The New Face of Unequal Exchange: Low-Wage Manufacturing, Commodity Chains, and Global Inequality,” mimeo, Political Economic Research Institute, University of Massachusetts at Amherst. Kabeer, Naila (2000) The Power to Choose: Bangladeshi Women and Labour Market Decisions in London and Dhaka, London: Verso. Kucera, David (2002) “Core Labour Standards and Foreign Direct Investment,” International Labour Review, 141(1–2): 31–70. Kucera, David and William Milberg (2000) “Gender Segregation and Gender Bias in Manufacturing Trade Expansion: Revisiting the ‘Wood Asymmetry’,” World Development, 28(7): 1191–210. Malhotra, A., S. Schuler and C. Boender (2002) “Measuring Women’s Empowerment as a Variable in International Development,” Working Paper, International Center for Research on Women. Polaski, Sandra (2003) “Trade and Labor Standards: A Strategy for Developing Countries,” mimeo, Carnegie Endowment for International Peace. Pollin, Robert, Justine Burns and James Heintz (2001) “Global Apparel Production and Sweatshop Labor: Can Raising Retail Prices Finance Living Wages?” PERI Working Paper No. 19, University of Massachusetts at Amherst. Power, Marilyn (1999) “Parasitic-Industries Analysis and Arguments for a Living Wage for Women in Early Twentieth-Century United States,” Feminist Economics, 5(1): 61–78. Rama, Martín (2001) “The Consequences of Doubling the Minimum Wage: The Case of Indonesia,” Industrial and Labor Relations Review, 54(4): 864–81. Rand, John and Finn Tarp (2002) “Business Cycles in Developing Countries: Are They Different?” World Development, 30(12): 2071–88. Ross, Robert and Anita Chan (2002) “From North–South to South–South: The True Face of Global Competition,” Foreign Affairs, 81(5): 8–13. Saget, Catherine (2001) “Poverty Reduction and Decent Work in Development Countries: Do Minimum Wages Help?” International Labour Review, 140(3): 237–69. Seguino, Stephanie (1999–2000) “The Investment Function Revisited: Disciplining Capital in Korea,” Journal of Post-Keynesian Economics, 22(2): 313–38. —— (2000a) “The Effects of Structural Change and Economic Liberalization on Gender

116 Stephanie Seguino Wage Differentials in South Korea and Taiwan,” Cambridge Journal of Economics, 24(4): 437–59. —— (2000b) “Gender Inequality and Economic Growth: A Cross-Country Analysis,” World Development, 28(7): 1211–30. —— (2004) “Is More Mobility Good? Firm Mobility and the Low Wage–Low Productivity Trap,” Working Paper, Burlington, VT: University of Vermont. Seguino, Stephanie and Maria Floro (2003) “Does Gender Have Any Effect on Aggregate Saving?: An Empirical Analysis,” International Review of Applied Economics, 17(2): 147–66. Sengenberger, Werner and Duncan Campbell (1994) International Labour Standards and Economic Interdependence, Washington, DC: Brookings Institution. Singh, Ajit and Anne Zammit (2000) The Global Labour Standards Controversy: Critical Issues, Geneva, Switzerland: South Centre. Standing, Guy (1989) “Global Feminization Through Flexible Labor,” World Development, 17(7): 1077–96. —— (1999) “Global Feminization Through Flexible Labor: A Theme Revisited,” World Development, 27(3): 583–602. UNIFEM (2000) Progress of the World’s Women, New York: UNIFEM. United Nations Development Programme (UNDP) (1999) 1999 World Survey on the Role of Women in Development: Globalization, Gender, and Work, New York: UNDP. Zhao, Laxiun (1998) “The Impact of Foreign Direct Investment on Wages and Employment,” Oxford Economic Papers, 50: 284–301.

World Bank discourse and World Bank policy in Engendering Development A comment Karin Schoenpﬂug

One of the difﬁculties feminists face when confronted with material such as the World Bank’s report Engendering Development: Through Gender Equality in Rights, Resource, and Voice (World Bank 2001) is a conﬂict between perceptions: the World Bank’s perception of gendered reality in developing countries versus the perspective feminists have developed from their sources of information. What makes this congregation of views – from my point of view – look more like a struggle than an enriching discussion is the reality of a multi-dimensional conﬂict of material interests behind the scenes of the report. The Bank shapes public discourse on development and gender, and its inﬂuence on policy-making has enormous consequences for the lives of millions of people in developing countries. Engendering Development is like a play being performed in order to conceal the economic drama behind the stage; now the show has gone on tour and is being performed in countries all over the world. When evaluating the World Bank’s descriptions of perceived realities and goals of social change, it seems beneﬁcial to use some sort of proven systematic theory to help disentangle the interwoven realities created by language, deﬁnitions and arguments. (Hofmann 2004). I use Michel Foucault’s discourse analysis to roughly categorize the arguments made by Avin, Doss and Seguino, and to outline the strategies necessary to deconstruct the Bank’s conclusions. According to Foucault, discourses are collective ways of constituting reality by means of rules and conventions (Foucault 1972). Within discourses, agents refer to shared knowledge about differences and similarities. Discourses are never neutral or homogeneous; they are always practices of in- and exclusion (Foucault 1978). As a consequence, discourses – like the one led by the World Bank in Engendering Development – legitimate speciﬁc ways of perceiving reality and exclude others. Most importantly, the power of discourses comes from the order they create and their function in sense-making and legitimization. Michel Foucault’s theoretical approach towards discourse can be understood in terms of four localities. The ﬁrst considers the role of discourse in the constitution of subjectivity and issues of voice. The second examines the processes of legitimization and the power of deﬁnition and brings to light the way power and knowledge are mutually constituted. The third notes that Foucault focuses on relations rather than dualisms, which permits a consideration of intersectionality.

118 Karin Schoenpﬂug The fourth explores the material or economic consequences of discourse. I will now consider these four localities in turn: 1. Foucault highlights the role of discourse in the constitution of subjectivity and authorizing certain voices while silencing others. (Who is talking (loudest)?) The ways in which the World Bank speaks for the (poor) people/women in developing countries has always been an issue of criticism. For instance, Joseph Stiglitz, in an interview by Greg Palast in the online London Observer, claims that “the Bank’s staff ‘investigation’ consists of close inspection of a nation’s 5-star hotels. The investigation concludes with the Bank staff meeting some begging, busted finance minister who is handed a ‘restructuring agreement’ pre-drafted for his ‘voluntary’ signature” (Palast 2001: online). In contrast, Rose-Marie Avin’s chapter asks the bank to “acknowledge that Third World women, and especially women’s organizations, have been and are agents of economic development”, who have a voice of their own (this volume, page 65). This issue is acknowledged by the Bank, since the issue of a “political voice” is the third part of the World Bank’s suggested three-pronged strategy: “take active measures to redress persistent disparities in command over resources and political voice” (World Bank 2001: 2). Nevertheless under closer scrutiny it turns out that the issue of “agentiﬁcation” through voice is used contrary to the feminist concept (as expressed in Virginia Woolf ’s A Room of One’s Own in 1929): in the European Union, for example, the feminist agenda of empowering women by giving them economic independence and a voice has been replaced with the concept of gender mainstreaming. This strategy also seems to be employed by the Bank as well, even though the terminology is not mentioned explicitly in the report. Gender mainstreaming is a process implanted in bureaucracies which demands that gender equality needs to “always be taken into consideration” by policy-makers. Contrary to this, in a feminist sense, empowerment means that subjects may speak for themselves and alter existing institutional structures. In other words, instead of just adding women and stirring, feminists may want to change the cake recipe altogether. Although gender mainstreaming purports to strengthen women in a patriarchal world by “considering them always” it fails to consider the conditions necessary for women to become agents and take responsibility for their own lives. In this way “the World Bank report does not directly address the . . . issues of power relations within households, communities, and countries” (Doss, this volume, page 80), the World Bank cannot “take active measures to redress persistent disparities” for someone as is suggested in the third strategy. Whether the Bank simply does not see the problem: “if only policy-makers and senior scientists understood that their clients were women, then . . . poverty would be reduced” (as Doss remarks cynically, page 81) or whether the Bank actually does not want to change women’s position in developing countries is the Gretchen Question in this regard.1

World Bank discourse and policy 119 2. Foucault focuses on processes of legitimization and the power of deﬁnition. This brings to light the way power and knowledge are combined in a certain society. (What is the language of the discussion?) The issue here is the power of deﬁnition in World Bank terminology. First consider the distinction between growth and development. Avin, Doss and Seguino all note the fact that the World Bank report does not distinguish between economic growth and economic development. As Rose-Marie Avin points out, there are definitions for the multi-dimensional process of development such as the ones by Michael Todaro, who lists three “core values of development” or by Amartya Sen, who deﬁnes development as “a process of expanding the real freedoms that people enjoy” that take into consideration an array of factors that are integral to a nation’s development (Avin, this volume, page 66). They are completely ignored (although ﬁve pieces of Sen’s work are nevertheless quoted in the literature section of the report) in favor of the simplistic and incorrect view that growth is equal to development. The focus on growth rather than development results in two distortions in the World Bank report. First, Avin points out that “the three-part strategy proposed by the report is inconsistent with the neoliberal agenda of international lending organizations, including the World Bank” (Avin, page 77) This means that the three-part strategy suggested does not seem to ﬁt the overall goals, or might even work contrary to them. Another problem lies in the Bank’s static world view that considers only movements or interactions used in standard macroeconomic theory (i.e. sinking interest rates will lead to increased (foreign) investment). It does not consider a ﬂuid reality nor the impacts of gender on macroeconomic processes. Seguino ﬁnds for instance that mobile industries (measured by FDI movements) make it more difﬁcult to close gender wage gaps and diminishes the bargaining power of workers. So to the extent that growth is dependent on FDI, it will exacerbate rather than ameliorate gender inequalities. 3. Foucault emphasizes relations rather than dualistic concepts. This makes it possible to consider not only gender, but its intersections with class, race, ethnicity, sexuality, age, nationality, and other categories of difference. (Who remains invisible?) One of the most severe criticisms of the World Bank report centers around the exclusion of certain categories. The Bank’s focus on dualistic concepts such as male–female, developed–developing, etc. reinforces and reproduces socially constructed binary differences. The intersectionality that is so central to feminist scholarship today is ignored. As Avin points out, the report “fails to take into consideration the issues of class, race, and ethnicity, a very important omission [. . .] It is important to acknowledge differences among women in terms of race and ethnicity” (Avin, page 65).

120 Karin Schoenpﬂug This leads us to seriously question their interpretation of the deﬁnition of “engendering.” Most surprisingly, the report does not offer a single hint on what the Bank exactly means by “engendering.” I am left to interpret engendering as gender mainstreaming. But instead of “taking gender into consideration” it seems that the Bank is (unconsciously?) strengthening binary gender categories; by not differentiating among women it reiﬁes gender difference. Avin, in contrast, demonstrates an example of a more nuanced approach towards gender when describing the effects of the Sandinistas’ institutional reform in Nicaragua of the 1980s. She explains how different modes of access for women from different classes and cultural backgrounds made it possible for more than the most privileged sub-group of the category “women” to become empowered. Women also played a signiﬁcant role in the political arena. . . . While many poor women entered the political arena at the grassroots level by joining popular organizations, middle- and upper-class women achieved high positions such as ministers or regional party coordinators . . . Working-class and peasant women increased their rights, their access to resources, and for the ﬁrst time had a voice in their community. (Avin, page 72) Doss provides a concrete example of the importance of differentiating within categories. She points out that there are three deﬁnitions for a “female farmer.” “The gender of the farmer may be deﬁned as the gender of the head of the household, the holder of the land, or the one who has the right to keep the proceeds” (Doss, page 86). The deﬁnition depends on particular legal and cultural norms that are highly contextual. The report’s missing differentiation within the category “female farmer” makes its ﬁndings for those women farmers rather useless, since policies’ consequences differ greatly for each sub-group of women working in agriculture. This takes us directly to the next locality: 4. Foucault is interested in the practices of inclusion and exclusion, the processes occurring between subjects, by which dominant groups control social discourse, policy-making and therefore distribution and economic well-being. This focus on the material consequences of discourses is, for economists, certainly the most important issue. The material or economic consequences are the manifestation of unjust social relations naturalized through discourse. (Who wins/loses materially?) The speciﬁc effects of development policies on sub-groups of considered groups is pointed out by Avin who states that some empirical studies . . . have shown that the poorest members of society may become worse off with economic growth . . . So there is no trickle

World Bank discourse and policy 121 down with economic growth as this report seems to imply . . . economic growth is necessary but not sufﬁcient to improve the lives of the vast majority of people and to improve gender equality. (Avin, page 67) She challenge the Report’s overall conclusions that “on balance the evidence suggests that females’ absolute status and gender equality improved, not deteriorated during the structural adjustment period of the 1980s and 1990s in Latin America” (Avin, page 65). She concludes that the neoliberal policies of the 1990s have not led to greater economic growth and gender equality. In her chapter Seguino goes even as far as to argue that not only have the predicted successes of World Bank policies failed to materialize, but that the outcomes are actually contrary to contrary to the Bank’s predictions. For example, she argues that “there exists an inverse relationship between gender wage equality and growth, implying that efforts to close wage gaps could slow economic growth.” As evidence she notes that “those [countries] with the greatest gender wage inequality grew the most rapidly during the period 1975–1995” (Seguino, page 100). Material consequences are easily uncovered when empirical tools are applied. Many feminist economists have so far concentrated on proving the effects of gender-blind or gender-adverse policy; now the task seems to have been expanded to demonstrate the adverse effects of allegedly gender-inclusive policy. The uncovering of material consequences allows alternative solutions with more favorable economic outcomes to be offered. For instance, empowering the silenced has always been large on the feminist agenda. Avin, Doss and Seguino offer various strategies ranging from empowerment through technology (in the African context), through institutional reform (in the Latin American context) or through restricting capital mobility (in the Southeast Asian context). All three authors stress the need for international cooperation among developing and developed countries when employing measures such as speed bumps for physical capital, or global actions, such as anti-sweatshop campaigns. In order to leave the dualistic framework of North–South, where the focus is on the World Bank “doing something to” developing countries, I will now briefly aim to demonstrate that this process of deconstruction of powerful institutions’ rhetoric and policy is also taking place in the European Union. In a recent project, Roswitha Hofmann uses a Foucauldian framework to deconstruct the process of knowledge creation and policy-making for the European Union’s Lisbon Process (Hofmann 2004).2 The Lisbon strategy, with its seemingly progressive agenda includes ten explicit commandments for employment reform, such as promoting gender equality in employment and pay and combating discrimination against disadvantaged groups.3 Unfortunately, the primary goal of the EU’s overall economic policy (the “Broad Economic Policy Guidelines”) is the Stability and Growth Pact (SGP), which demands balanced budgets and debt reduction (Euro Papers 2002). The SGP has been signed in order to enable European monetary union (the Euro) and it disables

122 Karin Schoenpﬂug member states’ ability to conduct either monetary or ﬁscal policy. Ironically, the Lisbon goals are organized in such a way that all measures for achieving those goals have to be taken nationally. This restriction makes it impossible to actually increase resources for socially and economically marginalized groups given the dominant goal of budget consolidation and shrinking power of national governments. The European Union situation with the formerly ten commandments is analogous to the World Bank’s three-pronged strategy goals. Both sets of goals sound progressive but are subordinated to larger, opposing goals. For the European Union this is the SGP, for the World Bank, it is SAPs. The same political context of neoliberalism is prevalent in both cases. The redistributive power of governments is turned upside down; tax systems are reversed from progressive to regressive, social assistance and public goods are being replaced by competing private enterprises selling their services (i.e. retirement schemes, water, and education) to those who can afford them. It is necessary for feminists to always bear in mind that policy-making in developing countries is connected to policy-making in “the ﬁrst world.” This connection demonstrates that a differentiated, inclusive feminist approach (one takes in as much of the whole picture as possible) is essential in order to challenge institutional operations such as those of the World Bank. As the feminist theorist bell hooks points out: “globally, women and children are the new proletariat and white women in the so-called developed countries support the enslavement of lower-class and poor women around the world” (hooks 1996: online). That’s an indictment that is hard to hear. But if we really want to talk about the liberation of women, then we’ve got to talk about the investment that bourgeois women of all races have in the social structure. Not only is the political practice of “ﬁrst world” institutions in developing countries post-colonial at best, disparities of wealth, income, and opportunity are increasing both between the developed and developing worlds, and also within geographic boundaries, between men and women and other (marginalized) groups. In the European Union, double talk results from the power struggle between labor unions, the “left” and their employment commandments versus banks/large businesses and their neoliberal agenda ( Jenson and Pochet 2002). Unions in the ﬁrst world are trapped into protecting their employed workers at the expense of the unemployed, protecting the wages of male workers at the expense of female workers, and protecting domestic workers at the expense of those in developing countries. Similar power conflicts may exist in the World Bank and this leads to the question, who within the Bank is interested in gender equality? Basically no one and everyone is the answer. The process of engendering development is only one device in the larger context of internationalizing suppression and enhancing the processes of neocolonization. Feminist deconstruction and analysis has revealed the World Bank report Engendering Development as a powerful piece of institutional sweet-talking aimed at creating its own sense-making and legitimization for World Bank policies.

World Bank discourse and policy 123 Similar processes can be observed worldwide, as has been pointed out in the context of the European Union. All three authors have concluded that the Bank’s policies have not had the positive effects on gender equality as has been claimed. The analysis of Engendering Development from a feminist point of view ﬁnds that gender equality cannot and will not be reached, given the conﬂicting strategies recognized in the report. To conclude with Seguino’s words: “the policies needed to promote equity . . . would require a rethinking of the Bank’s support for neo-liberal macroeconomic policies” (Seguino, page 94). Since this may not happen in the near future, a way to implement alternative feminist policy solutions will need to be found.

Acknowledgments Thanks to Nicky Imre for proofreading, Roswitha Hofmann for discussing and sharing her work on Foucault and the Lisbon strategy and to Drucilla Barker for her enormous help in wording this piece.

Notes 1

The “Gretchen Question” is a parable for questioning someone’s dubious religious or political convictions, using a slightly naïve slant. Asking the “Gretchen question” refers to a passage in Goethe’s Faust. Faust has long sold his soul to the devil, which his young and pure lover Gretchen certainly does not know. Then in one scene Gretchen innocently asks Faust if he really believes in God, a question poor Faust can only dodge. In 2000 the European Council decided that the EU should become the “most competitive and dynamic knowledge based economy in the world, capable of sustainable economic growth with more and better jobs and greater social cohesion” (European Parliament 2000). In 2005 the European Union has altered its failing Lisbon Strategy. There are currently seven guidelines for employment instead of ten and the promotion of gender equality has been taken off the (explicit) agenda. Growth and sustainability seem to be of equal importance in the new “Interpreted Guidelines for Growth and Employment (2005–2008)”; the Stability and Growth Pact has been extended by multiple cases of exceptions.

References Euro Papers (2002) Co-ordination of Economic Policies in the EU: A Presentation of Key Features of Main Procedures, Paper No. 45, Brussels: Directorate General for Economic and Financial Affairs. European Parliament (2000) Lisbon European Council 2000: Presidency Conclusions, Strasbourg. http://www.europarl.eu.int/summits/1is1en.htm (accessed October 11, 2005). Foucault, M. (1972) The Archaeology of Knowledge, London: Routledge. —— (1978) Dispositive der Macht, Berlin: Merve. Hofmann, R. (2004) “Divides of the European ‘Knowledge Society’,” Vienna: University of Economics and Business Administration Working Paper.

124 Karin Schoenpﬂug hooks, b. (1996) “Tough Talk for Tough Times,” in On the Issues, December 1, 1996, online, available http://static.highbeam.com/o/ontheissues/december011996/tough talkfortoughtimes/ (accessed August 24, 2004). Jenson, J. and P. Pochet (2002) “Employment and Social Policy since Maastricht: Standing up to the European Monetary Union” Paris: University of Notre Dame, online, available http://www.cevipof.msh-paris.fr/rencontres/colloq/palier/Full%20 paper/Palier%20FP.doc (accessed August 24, 2004). Palast, G. (2001) The Globalizer Who Came In From the Cold, in the London Observer, Wednesday, October 10, online edition: http://www.gregpalast.com/detail.cfm?artid =78&row=1 (accessed October 14, 2005). World Bank (2001) Engendering Development, New York: Oxford University Press.

Part III

Disciplinary paradigms/ development paradigms

Colonizing knowledge Economics and interdisciplinarity in Engendering Development Suzanne Bergeron

Setting the disciplines in motion at the World Bank: adding the social and cultural to the equation One of the features that distinguishes the emerging post-Washington consensus from the neoliberal agenda that proceeded it is the attention that has been given to social issues in development. The dislocations caused by two decades of economic restructuring, along with pressure from anti-poverty, women’s, environmental and indigenous people’s movements, have prompted development institutions such as the World Bank to be more responsive to social concerns such as poverty and inequality. Under President James Wolfensohn, the Bank has undertaken a self-proclaimed “paradigm shift” that includes a commitment to helping marginalized groups such as women and indigenous minorities, promoting an equitable distribution of income, working to promote democratic participation with the people directly affected by its projects, protecting the environment, and increasing its lending for health and education. This stated change in direction marks a break from the neoliberal, economistic approach taken by the Bank during the past 20 years, exemplified most famously in a memo signed by former chief economist Lawrence Summers on the efﬁciency of shipping toxic waste to Africa, which dismissed potential criticisms as likely to be based on some fuzzy set of “social concerns” and “moral reasons” that had no legitimate place in the development conversation (Summers 1992: 66). Now, social concerns seem to be taking center stage at the Bank. However, increased attention to social development at the Bank also needs to be placed in the context of the development community’s current fascination with theoretical innovations in economics that are adding cultural and social factors to the equation (Fine 2001; Bergeron 2002). These new approaches allow economists to take into account the impact of social and cultural forces on economic outcomes without sacriﬁcing the core ideas of economic analysis. For example, in contrast to the growth theories that dominated development thinking 25 years ago, today’s theories contend that social relations, culture, and institutions matter for economic growth. This creates a space for examining how formerly neglected social aspects of development, such as gender inequalities, might interfere with the goal of growth (Klasen 1999). Another innovation which has caught on in development circles is the concept of social exclusion. This

128 Suzanne Bergeron concept links poverty to social and political as well as economic factors, particularly in terms of how socially excluded groups are denied access to resources that would affect their economic well-being. The related concept of social capital has of late received significant attention from economists, sociologists, and political scientists working in development in general (Woolcock 1998) and gender and development in particular (Rankin 2002). Because of economic theory’s inﬂuence on development research and practice, these theoretical innovations have made it more legitimate for the “epistemic community” (Haas 1992) of development thinkers to take social and cultural factors into account, and embrace new models based on multidisciplinary approaches. It is not surprising that the Bank’s paradigm shift has been accompanied by the entry of increasing numbers of anthropologists, sociologists, and political scientists into an institution that until recently had been the almost exclusive territory of neoclassical economists (Cernea 1991). Even though these non-economists remain only a small percentage of the total research staff, a cursory examination of the Bank’s research output ten years ago versus today suggests that the disciplines are in motion at the Bank, with increased exchange of disciplinary knowledges creating fresh space for understanding the relationship between social institutions, culture, and economy. At ﬁrst blush, this recent paradigm shift at the Bank, and the increased dialogue among the disciplines that has accompanied it, seems to represent the kind of changes in development theory and practice which political economists, feminists, and others on the margins of the profession have long called for. The acknowledgement that development problems do not come in discipline-shaped boxes has allowed for a more multi-dimensional analysis of existing research issues than was allowed by the economistic framework that dominated the ﬁeld during the past 20 years. Inequality is a good example. The recognition that inequalities are based on differences which contribute to social exclusion, for instance by gender, has created openings for multidisciplinary approaches to understanding poverty, as mainstream economics by itself hasn’t much to offer with regard to theorizing social roles, relationships, and resistance to exclusion ( Jackson 2002: 500). Such analytical breakthroughs also support policy initiatives, for instance government programs to address exclusion issues such as discrimination in labor and credit markets. This acceptance of the existence of market failures, and the need for intervention to resolve those failures, can be contrasted with the freemarket orthodoxy of the Washington consensus years (Stiglitz 1998).

Comparing frameworks: multidisciplinarity at the World Bank and the critical transdisciplinary approach of women’s studies The focus on the social that characterizes feminist scholarship, combined with the ongoing debates on the cultural character of economic processes within feminist theory, would seem to place it in a position to make a signiﬁcant contribution to development theory and policy at this juncture ( Jackson 2002: 499).

Colonizing knowledge 129 However, the sort of critical and transdisciplinary approach that feminist theory can bring to the development conversation might not get much of a hearing in development institutions such as the World Bank due to a variety of structural, epistemological, and institutional factors. For one thing, attempts to bridge disciplinary divides are somewhat fraught because in addition to the constraints associated with the political economy of knowledge production in the academy (where many face pressures to hew to the line of disciplinary boundaries), institutions like the World Bank continue to be dominated by economics. These pressures are exacerbated by the fact that development economists at the Bank have historically not take seriously those knowledges that, in their view, fail to achieve the same standard of rigor as economic theory. As William Ascher puts it: Many professionals in the World Bank have been reluctant to incorporate new considerations in formulating development strategies if they require modes of analysis less rigorous than the traditional economic framework. Unless and until development strategies can be converted into decision making procedures acceptable to the professional norms of those entrusted with using them, the implementation of these strategies is bound to meet resistance. (Ascher 1983: 417) So what tends to get a hearing in the post-Washington consensus Bank are those insights about social institutions, cultural norms, and politics from noneconomics ﬁelds which can be ﬁtted to the existing model. The new emphasis on institutions, and thus the level of disciplinary interchange, relies not so much on contingent sociopolitical and cultural analysis as much as it does on theoretical notions dear to mainstream economists’ hearts such as individual optimization. For example, as one World Bank researcher notes: What distinguishes new institutional economics from previous “institutional” or “historical” schools is that the former has more theory behind it. That particular theory has two interdependent components: first, a reliance on competition as the condition that induces efﬁciency, and second, the belief that the right incentive structure is the best answer to the restrictions that arise from the neoclassical model. (Wiesner 1998: 114) Adding culture, social factors, and institutions to the equation, in other words, is within the bounds of acceptable practice only if it is framed in terms of preserving core ideas of neoclassical economics such as methodological individualism and a focus on market efﬁciency. A story told by Cecile Jackson in a recent World Development article gives one example of how the economics-centered approach of the Bank in the current conjuncture frames the ways that disciplinary encounters are conceived.

130 Suzanne Bergeron A well-known World Bank economist, when asked about his vision of interdisciplinary research, said that his ideal was to be able to call up a similarly wellknown social anthropologist when he came across questions that puzzled him ( Jackson 2002: 500). As Jackson explains, this perspective fails to value the modes of inquiry and insights of anthropology. It is at best an attempt to expand the tool kit of the discipline of economics, rather than fostering a true interdisciplinary engagement that would value and connect different ways of working and thinking. The view of other disciplines serving as handmaiden to economics, allowing for better explanations of problems as they are framed by the Bank’s pre-existing research agenda, provides a very limited sort of exchange, and one that falls far short of the ideal of feminist theory and women’s studies research. Different disciplines bring new issues, new questions, and different conceptual frameworks to the table, and feminist research at its best tends to take more of a transdisciplinary approach, one that not only broadens knowledge production but reorganizes it along new lines. Instead of viewing disciplinary border-crossing as adding to one’s disciplinary toolkit, or a happy exercise in bridge-building, an apt analogy for the approach most valued in women’s studies might, as Marjorie Pryse (1998) suggests, be Gloria Anzaldua’s idea of “mestiza conciousness” – the coming together of two or more self-consistent but habitually incompatible frames of reference that causes a cultural collision. Thinking about it in this way suggests connecting different ways of thinking and working, and acknowledges the different modes and disparities among disciplinary knowledges (versus a more utopic bridge-building approach which would wish these disparities away). Further, if certain disciplines have in some sense colonized knowledge – and what better example of this is there than economics, which has increasingly been characterized by imperialism, the effects of which have been felt quite keenly in academic ﬁelds of political science and sociology in recent years – then to the extent which women’s studies constitutes a challenge to disciplinary models we might consider it as a particular sort of postcolonial strategy. The World Bank’s approach to adding social and cultural factors to the equation, in contrast, does not challenge the core of economics or its imperialistic tendencies. The recent paradigm shift at the Bank is certainly multidisciplinary, in the sense that it adds insights from a variety of disciplines to expand the toolkit used to make sense of social factors in development while retaining the core of economic disciplinarity. The reframing of development suggested by the women’s studies approach would be more of a transdisciplinary method that brings the disciplines together in ways that challenge their cores, while reshaping the very boundaries that organize disciplinarity itself. Thus, while border-crossing is a usual characteristic of knowledge growth, one that has characterized the development of all disciplines (including economics), much women’s studies research is based on the more critical premise that disciplinary organization is contested (Klein 1993). This sort of critical transdisciplinarity, while likely quite productive in terms of generating knowledge

Colonizing knowledge 131 that can serve as a basis of social change and empowerment for women, might not be an easy sell for institutions such as the World Bank where expertise still tends to be constructed along disciplinary (economic) lines.

The multidisciplinary approach of Engendering Development The World Bank’s policy research report on Engendering Development reflects the more general paradigm shift of the post-Washington consensus Bank toward social concerns and multidisciplinary approaches. The goal of the report is to understand the links between gender, public policy, and development by bringing a gender perspective to bear on the analysis and design of development policies and projects. In pursuit of that goal, Engendering Development adopts a self-consciously multidisciplinary approach that draws upon literature from economics, anthropology, sociology, political science, and education disciplines. In its conclusions, the report proposes a strategy based on reforming institutions to establish the legal and economic conditions for equality, fostering economic development to strengthen incentives for equal pay, and improving women’s access to resources and political voice to reduce gender disparities (ibid.: 16–26, 231). The report’s emphasis on the social and cultural factors that inﬂuence women’s capabilities and agency, along with its focus on social development concerns such as education and health care, represent a real improvement over the earlier efﬁciency and growth-fetishized approach of the Bank. The emphasis on issues such as social norms, women’s autonomy, and voice, because of the complexity of the issues raised by these concepts, shows a recognition and concern for creating the conditions in which women can influence social change, and the importance of asking key questions about what voices will be heard, how participation is fostered, and whether women will be able to represent and articulate the interests of other women. This suggests a great salience for the sort of transdisciplinary approaches that the best of women’s studies scholarship can offer ( Jackson 2002: 501). Understanding the ways that social institutions impede gender equality, for example, would entail a close study of the following: how cultural norms about gender interact with the construction of the self and subjectivity; gender relationships in social institutions such as the family and education; political life such as community engagement, government policies, and the legal structure; economic processes including the globalization of markets and local provisioning. It would also entail an analysis of how all of these interact with each other. This is the sort of complex, multilayered analysis suited to creative and critical exchange across and between the disciplines. But while the report does make an attempt to bring these different factors to bear on the analysis of how gender disparities persist, there is in general not much analysis about how these factors work together to produce the outcomes they do. When there is such an attempt, disciplinary knowledges and insights that might challenge the Bank’s attachment to the core ideas of economics are pushed

132 Suzanne Bergeron to the margins. The multidisciplinary approach taken in the report, while a positive step compared to the narrowness of earlier economism, remains a limited perspective on gender and development by failing to pursue insights that a genuinely transdisciplinary approach would enable. Among other things, this would include drawing upon social constructivist and postcolonial insights to analyze the sources of difference among women and how these relate to women’s agency. While it would certainly challenge the core neoclassical story in which individual agents are born and not made – as Julie Nelson (1996) has observed, neoclassical theory tends to see economic agents as Hobbesian “mushroom men” who emerge fully formed – grounding the report in these aspects of feminist theory would help to avoid the obscuring of difference created by universalistic accounts of women’s behavior and experience which I examine below. Instead, the report’s view of women’s agency and its account of outcomes by gender (e.g. the idea that all women in a particular country benefit from economic growth) are a reﬂection of its continued attachment to rational choice theory on the one hand and macroeconomic representations which link GDP growth to national collective well-being on the other. This has implications for how policy solutions are framed as well. For example, when the question is ﬁrst raised regarding why gender disparities exist and are worse in some places than others (World Bank 2001: 97–9), the answer provided fails to incorporate insights from disciplines that have attempted to theorize social agency and social movements. In the absence of a richer theory of gender and agency, increased output and modernization are offered as the answers. Like earlier World Bank approaches to inequality, growth and marketization are consistently presented throughout the report as the great levelers, and deviant local “culture” as the constraint that keeps women from achieving empowerment through markets (see also Barriteau, this volume). Unlike the strict neoliberal approach taken by the Bank in the 1980s and early 1990s, the report does specify a role for government policy in improving both efﬁciency and equity. Many of these policy prescriptions (e.g. afﬁrmative action, equal pay laws) are grounded in theories of market failure that acknowledge such concerns as monopoly power and information problems which could result in statistical discrimination and impede optimal outcomes in markets (World Bank 2001: 126–42). This is at least in part due to the innovations in economics discussed in the ﬁrst section. In the past, there was little room to theorize the problems of discrimination in mainstream neoclassical thought, which made it difﬁcult for such concerns to have a voice at the Bank (Kardam 1991: 72). But as mainstream economics has embraced new models such as social exclusion theory, the discourse of the Bank has changed as well. Combining the individual-rationality approaches of neoclassical economics with sociological insights about discrimination by race and gender, a new information-theoretic approach to understanding labor market outcomes can explain not only why discrimination might exist, but also why it might be pervasive and in need of correcting via policy. Analyzing the social and cultural factors that contribute to peoples’ exclusion, and designing projects aimed specifically at including

Colonizing knowledge 133 these formerly marginalized groups in the development process (which, since this is the World Bank, means integrating them into the global capitalist economy) form an integral part of the Bank’s new anti-poverty initiative (World Bank 1998) and this is reﬂected in the Engendering Development report as well.

Some implications of the colonizing knowledge of economics in Engendering Development While this paradigm shift at the Bank can be seen as a positive step in that it is putting the question of gender inequality on the World Bank’s map, there is also a need for caution. This shift remains within the general framework of neoclassical thought as it maintains an analytical separation of the social and cultural from the economic, and therefore tends to make sense of social norms and culture as things that interfere with the universal (culture-free) market economy. In this sense, the insights gleaned from sociology, anthropology and political science do not challenge as much as expand the toolkit of an existing discipline. The report might have included some of the recent literature on gender and global economics, which argues that the economy itself is deeply cultural (e.g. Benería 1999). But this sort of story is unlikely to get a hearing in the neoclassically centered approach of the report, which tends, except in the most limited of cases, to see culture and social relations as the “traditional” constraints on women’s empowerment, which, in its modernization tale, occurs through integration into the global market. A story that underscores this point is told by Joanna Kerr, who attended a summit on gender and structural adjustment in Africa, at the end of which one World Bank economist said, “I’m still not convinced that structural adjustment has a negative impact on women, but I am convinced that women’s inequality has a negative impact on economic growth.” This frames the social and economic as separate, and the economic as universal. What needs to be changed is not the Bank’s relatively pro-market policy, but only the barriers that hinder women from contributing to the market. “Women are seen as an ‘untapped’ labor resource – as merely instruments of growth” (Kerr 1999: 197). The policy implications of maintaining an analytical separation between the economic and the social are in fact quite deep. For example, Elson and Çagatay’s analysis of recent attempts to integrate social content such as gender equity into development macroeconomic policies suggests that the promises of Bank President Wolfensohn to make gender central to such policies have not yet been met. In part, this is because the Bank continues to envision the social and the economic as two separate spheres, and gives priority to the latter. “Sound” monetary and fiscal policy are viewed by the Bank as the “essential backdrop” against which social goals must accommodate themselves (World Bank 2000: 21, cited in Elson and Çagatay 2000). Similarly, the analytical approach of the Engendering Development report continues to give priority to economic “fundamentals” such as expanding markets and growth, both of which it deems

134 Suzanne Bergeron to be beyond discussion. As Elson and Çagatay suggest, an approach aimed at truly integrating social concerns (including gender equity) into economic policy would include analyses that might call into question what should count as fundamentals, or “sound” economic policy (Elson and Çagatay 2000: 1353). Unfortunately, the new economics’ tendency to separate the economic from the social has given priority to economic fundamentals at the expense of a more transformative approach. While there is a tendency for analyses to remain within the general discursive frame of mainstream economics in the report, it is certainly not completely dominated by economics. There are, in fact, entire sections of the report that do rely speciﬁcally on non-economic knowledges to explain certain aspects of gender subordination. In chapter 3, for example, there is an extended discussion of how gender-related norms and customs affect women’s equality. This section describes differences in women’s experiences across cultures using an anthropological frame. It is a welcome addition to the analysis of this question in the report, being attentive to, among other things, the inner workings of kinship, religious systems, and social norms as factors that contribute to gender disparities. For example, the section discusses the custom of brides moving to their husband’s village upon marriage in parts of Southeast Asia, which explains their families’ unwillingness to send them to school, whereas in other parts of Southeast Asia, women are expected to play a more important economic role in the family, thus creating an incentive to send them to school (World Bank 2001: 110–11). It also discusses how machismo in Latin America contributes to occupational segregation by gender, as well as to gendered violence (ibid.: 111–12). This section provides an alternative to the general thrust of the Bank’s economic/social theorizing, which reduces everything to individualistic, rational, self-interested behavior. It opens the door to a more complicated historical and cultural narrative of women’s oppression that eschews the universalizing tendencies of neoclassical economics. Still, it remains problematic that a set of concerns related to the production of such knowledge about gender roles and traditional culture in anthropology are not included here, or in any other part of the report. For example, anthropologist Ruth Behar’s Translated Woman: Crossing the Border with Esperanza’s Story (1993) provoked a great deal of discussion among feminist anthropologists regarding the representation of the “other” in their work, and Behar’s inﬂuence has led to more nuanced attempts to define the subject position of Western anthropologists, particularly as they view themselves as the “liberated” modern other against whom the “traditional” non-Western woman is judged (see Mohanty 1991 and Marchand and Parpart 1995 for examples of how feminists working out of other disciplines have tackled this issue). Integrating these insights would be useful because they would contribute to an understanding that traditional social norms are not always constraints on gender equality. For example, a participant in an online discussion of a draft of the report suggested as an example the Islamic concept of repudiation which has been used by some women to empower themselves.

Colonizing knowledge 135 As Avin (this volume) points out, the theoretical framework of the report is not able to address the problem of work not empowering women. Its universalizing, naturalistic portrayals of the economy impede it from imagining the ways that increased integration into the formal labor market might interfere with women’s well-being. The ﬂip side of this is that the report can not easily imagine how non-market social processes might potentially act as sites of negotiation and women’s agency – with the exception of a handful of arguments that rely on essentialism, for example, the claim, discussed below, that women have more potential social capital because women as a group are naturally more cooperative. By maintaining the economy/culture dichotomy in which the universal market provides salvation from the backwardness of local culture, the report tends to entrench the image of the vulnerable, pre-modern Third World woman who is limited by social norms which are “slow to change,” and it pushes to the margins those narratives of strategies for empowerment in which these women may already be involved on a daily basis. This reﬂects a more general tendency of the report to present its ﬁndings and solutions within the growth-oriented modernization framework of neoclassical economics. One of the effects of taking this road is the report’s tendency to see women’s rights and empowerment in an instrumental manner: they matter only because inequalities by gender interfere with modernization goals such as growth. In chapter 3, the report states that equal rights for women are viewed as an important development goal because “legal social and economic rights provide an enabling environment in which women and men can participate productively in society, attain a basic quality of life, and take advantage of the new opportunities that development affords” (World Bank 2001: 116). Similarly, the argument in favor of providing more education to girls is instrumental. Educating girls leads to multiple “payoffs” including later marriages, increased contraceptive usage, reduced infant mortality and higher earnings (ibid.: 196). It is justiﬁed on the grounds that it will increase productivity and decrease birth rates. This sort of argument views gender equity as a means to an end, not as a goal unto itself (Baden and Goetz 1998: 24). Further, the objectives are rendered as technical (efficiency, birth rates) rather than political, which means that they may ultimately fail to address the systematic nature of gender and social injustices. The rhetorical strategy of tying the goals of women’s equality instrumentally to achieving the goals of economic growth needs to be made sense of in institutional context – it may, as Nukut Kardam (1991) suggests, give feminists some leverage within the growth-centered discourse of the Bank. The reticence of economists and other Bank staff to seriously consider the gender implications of their policies has caused some gender entrepreneurs inside and outside of the Bank to utilize arguments that it can understand. However, this approach, by staying inside the dominant disciplinary discourse of the Bank, may cost more than it is worth. As Diane Elson (1998: 155) puts it, “if we only work on the ‘inside’ we run the risk of merely achieving small improvements in the formulation of models or collection of statistics which do not actually transform

136 Suzanne Bergeron women’s lives.” While focused on the question of cross-cultural knowledge, David Crocker’s discussion of the relative merits of “insider” and “outsider” knowledge is relevant here as well, especially when we recognize that the World Bank has an institutional culture that in part frames its knowledge production. Crocker argues that insiders face social pressures that might inhibit their ability to develop alternative ways of thinking (Crocker 1991). This seems to be the case in Engendering Development, which does not address the more contingent sociopolitical and cultural dimensions associated with improving women’s lives and instead relies on the World Bank’s well-worn script of modernization and marketization as the key to improvement. Examples of more integrated, transdisciplinary approaches to these issues include the essays in the edited collections by Afshar and Barrientos (1999) and Chang and Ling (2000) among others. These authors view women’s struggles for rights and resources in a context that suggests a much more contested relationship with the growth of the market and modernization, as well as a complex renegotiation of their places in the household, workplace, and community. The implicit and explicit equating of women’s empowerment as economic empowerment via integration into the market economy can also be seen in Engendering Development’s discussion of reproductive labor. The report should certainly be commended on its inclusion of non-market labor as an important factor in women’s lives. However, its focus on reproductive labor as something that should be substituted with market labor is quite limiting. Here, among other things, there may be space for a postmodern critique, as the report’s analysis relies on a dualistic formulation of market versus non-market, reproductive versus productive labor, in which, like all dualistic frameworks, one term (in this case, the market) is the privileged term. I can imagine this dualism being exploded in a number of different ways, for instance by including the insight that it leaves out an important third term, community labor, which has become an important part of women’s resistance and survival strategies during the current era of economic restructuring (Pardo 1998). Still another aspect of the neoclassical framing of the report that would likely be eschewed by a more transdisciplinary women’s studies approach lies in its representation of progress in such a way that the North is continually held out as the highest stage of development and liberation for women. At the end of chapter 3, for instance, the authors suggest that Third World countries might take heart at the progress made by the United States, which experienced gender inequities that were even “more severe than developing countries today”: Take also the case of colonial America, where women were essentially the property of men, either their father or their husband. Women could not own property or enter contracts independent of their husband, and they had to relinquish property and wages to him. They had no custodial rights to their children, could not vote or hold political ofﬁce, and were routinely excluded from social and professional organizations. These restrictions, common in 1776 – and more severe than those in developing countries

Colonizing knowledge 137 today – no longer exist in the United States. Why? Increases in per capita wealth, women’s activism, and women’s education all contributed to broadening women’s rights. (World Bank 2001: 144) Similarly, the discussion of the gender gap in work on pages 184–5 contends that in countries that experience growth, the gender gap in total hours worked will fall. This is done by comparing hours worked by men and women in richer and poorer countries, and showing that in richer countries it is more equal. By taking this stance, the report suggests that all countries are marching along the same path, some are simply behind others and need to catch up to those countries where economic development has liberated women. This seems to replicate precisely the sort of “stages of development” modernization approach that fails to capture the contingent realities that different countries face. This approach has come under criticism from multiple quarters, including what was certainly one of the most vibrant debates in the gender and development literature in the 1990s (Connelly et al. 1995). In addition to its general neoclassical adherence to the view of development as growth, and attachment to outdated notions of linear progress, the report also tends to trade in essentialistic portrayals of women. There is a strong tendency in Engendering Development to use gender as a monolithic category despite the existence of multiple analyses of the complex realities and difference among women that exist in the gender and development literature more broadly. While some attempts are made to differentiate women in the report, usually via a listing of characteristics such as age, race, ethnicity, religion, etc. in different sections of the text (e.g. p. 112), most of the analysis seems to take a more ﬂattened approach. Some examples of this include the following. First, the general thrust of the book seems to posit that even while there are differences among women, women are, as a group, relatively worse off than men (see especially chapter 1). To be fair, one exception to this is found in the discussion of adjustment on page 211, where it is pointed out that men, especially young men, were hit harder by unemployment than women during Thailand’s economic crisis in the 1990s. Similarly, the general argument of chapter 5 is that development makes women as a group better off. This second formulation (relating to chapter 5) seems to come from the somewhat universal neoclassical economic assumptions regarding growth lifting all boats. But arguing that social arrangements affect women more adversely than men, or that development improves the lot of women generally, is problematic for several reasons. Such representations do not allow for differentiation among women and men in terms of class, race, ethnicity, age, nationality, and education. As such they do not provide a very useful basis for theorizing feminist activism or gender-sensitive policy-making. Some speciﬁc examples of essentialism in the report include the section on corruption and gender on pp. 92–6. Summarizing a growing body of research on the subject, the text states that “there may be intrinsic differences in the behavior of women and men” that cause governments to be less corrupt when

138 Suzanne Bergeron women are in power, and that “gender differentiated attitudes toward corruption seem to be more or less a worldwide phenomenon” (World Bank 2001: 93). While these differences might bear out empirically, their origins could be treated in a more nuanced way in the report. While the existing literature utilizes psychological, sociological, and criminal justice perspectives to explain these differences in terms of women’s role in reproduction and/or or gender socialization, the report simply offers them as “intrinsic.” This representation of women as if they embody a set of given characteristics is troubling not just because it obscures differences among women, but also because it suggests generalized gender differences in behavior that, as Jackson puts it, “can be relied on in gender differentiated policy” (Jackson 2002: 502). The tendency to theorize that women are universally disadvantaged vis-à-vis men and the reliance on generalized gender differences to imagine policy scenarios are carried over into the discussion of social exclusion and social capital. Women, it is argued, have less social and political capital than men, thus putting them at a disadvantage (World Bank 2001: 97, 101). Interestingly, in the past couple of years, the World Bank has taken a slightly different view from what is presented in the report, imagining on the one hand contexts in which women suffer from social exclusion, but increasingly making the argument that women are better at social networking, and thus have more (or particularly distinct and more useful) social capital than men. This, for instance, is the basis of the World Bank’s microcredit lending programs, in which women’s social cohesion is viewed as a substitute for collateral. As Robert Picciotto of the World Bank puts it, “cooperation is more of a female attitude,” making women a particularly good credit risk for these loans. Note that this essentialist characterization is one that depends on gender subordination, for the argument is that women are better at cooperating because they don’t have access to resources like well-paid jobs (World Bank 2002). Another example of essentialism occurs on page 186 of the report. In the discussion of how economic growth can contribute to the deepening of markets, which in turn will lighten women’s load, the report states that “where available, hired labor can substitute for family labor on farms or in care activities, thus easing the workload of family members.” It goes on to say that “a market for farm workers enables households to cope with peak season labor demand without pulling children out of school or adding excessive hours to women’s work.” In this section the clear implication is that the “women” beneﬁting from these labor market arrangements are the ones doing the hiring. This represents a combination of the “integrate, integrate, integrate” approach of liberal feminism along with the class-blindness associated with adherence to a more or less neoclassical framework that has been challenged in feminist circles (especially feminist development circles) in the past twenty years. The class, ethnic, age, and other differences that might underlie this market arrangement (what will be the impact on the well-being of the person – woman? – being hired, what circumstances might have brought her to the labor market as a household worker or “peak season” farm worker in the ﬁrst place?) are thus pushed to the margins

Colonizing knowledge 139 in such an analysis, which, as we know from many previous cases, has often resulted in policies that attempt to deal with gender inequalities but only serve relatively well-off and educated women, leaving poor women behind.

Conclusion The shift towards a greater appreciation of the social, cultural and political aspects of development, and thus towards interdisciplinary approaches, has led to some interesting and refreshing insights into the relationship between gender and development as presented in the World Bank’s Engendering Development report. However, much of the analysis of the report has remained within the disciplinary framework of neoclassical economics that continues to dominate institutions such as the World Bank. Here, the other social sciences serve primarily as handmaidens to the Bank’s already established assumptions and agendas. A more critical transdisciplinarity, one that builds on the knowledge-producing strategies of women’s studies research in general and recent debates in gender and development in particular, would create the conditions for less instrumental and more transformative approaches to both knowledge production and development policy.

Acknowledgments The author would like to thank two anonymous referees and the editors of this collection, Edith Kuiper and Drucilla K. Barker, for their helpful comments and suggestions.

References Afshar, Haleh and Stephanie Barrientos (eds) (1999) Women, Globalization and Fragmentation in the Developing World, New York: St. Martin’s Press. Ascher, William (1983) “New Development Approaches and the Adaptability of Internation Agencies: The Case of the World Bank,” International Organization, 37(3): 415–39. Boden, Sally and Ann Goetz (1998) “Who Needs (Sex) When You Can Have (Gender)? Conﬂicting Discourses on Gender at Beijing,” in Cecile Jackson and Ruth Pearson (eds), Feminist Visions of Development, London: Routledge. Behar, Ruth (1993) Translated Women: Crossing the Border with Esperanza’s Story, Boston, MA: Beacon Press. Benería, Lourdes (1999) “Globalization, Gender and the Davos Man,” Feminist Economics, 5(3): 61–83. Bergeron, Suzanne (2002) “Challenging the World Bank’s Narrative of Inclusion,” in Amitava Kumar (ed.), World (Bank) Literatures, Minneapolis, MN: University of Minnesota Press. Cernea, Michael (1991) “Using Knowledge from Social Science in Development Projects,” World Bank Discussion Paper Series, Washington, DC: World Bank. Chang, Kimberly and L.H.M. Ling (2000) “Globalization and Its Intimate Other: Filipina

140 Suzanne Bergeron Domestic Workers in Hong Kong,” in Marianne H. Marchand and Anne Sisson Runyan (eds), Gender and Global Restructuring, London: Routledge: 27–43. Connelly, M. Patricia, Tania Murray Li, Martha MacDonald and Jane M. Parpart (1995) “Restructured Worlds/Restructured Debates: Globalization, Development and Gender, Canadian Journal of Development Studies (special issue). Crocker, David (1991) “Insiders and Outsiders in International Development,” Ethics and International Affairs, 5: 149–73. Elson, Diane (1998) “Talking to the Boys: Gender and Economic Growth Models,” in Cecile Jackson and Ruth Pearson (eds), Feminist Visions of Development, London: Routledge. Elson, Diane and Nilufer Çagatay (2000) “The Social Content of Macroeconomic Policies,” World Development, 28(7): 1347–64. Fine, Ben (2001) Social Capital Versus Social Theory, London: Routledge. Haas, Peter (1992) “Introduction: Epistemic Communities and International Policy Coordination,” International Organization, 46 (Winter): 1–36. Jackson, Cecile (2002) “Disciplining Gender?” World Development, 30(3): 497–509. Kardam, Nükut (1991) Bringing Women In: Women’s Issues in International Development Programs, Boulder, CO: Lynne Rienner Publishers. Kerr, Joanna (1999) “Responding to Globalization: Can Feminists Transform Development?” in M. Porter and E. Judd (eds), Feminists Doing Development: A Practical Critique, London: Zed Press. Klasen, Stephan (1999) “Does Gender Inequality Reduce Growth and Development? Evidence from Cross-country Regressions,” Background paper for Engendering Development, World Bank, Washington, DC, available online at http://www.eldis. org/static/DOC8972.htm (accessed October 11, 2005). Klein, Julie Thompson (1990) Interdisciplinarity: History, Theory and Practice, Detroit, MI: Wayne State University Press. —— (1993) “Blurring, Cracking, and Crossing: Permeation and the Fracturing of Discipline,” in Ellen Messer-Davidow and David R. Shumway (eds), Knowledges: Historical and Critical Studies in Disciplinarity, Charlottesville, VA: University Press of Virginia. Marchand, Marianne and Jane Parpart (1995) Feminism/Postmodernism/Development, London: Routledge. Mohanty, Chandra Talpade (1991) “Cartographies of Struggle: Third World Women and the Politics of Feminism,” in C.T. Mohanty, A. Russo and L. Torres (eds), Third World Women and the Politics of Feminism, Bloomington, IN: Indiana University Press. Nelson, Julie (1996) Feminism, Objectivity and Economics, New York: Routledge. Pardo, Mary (1998) Mexican American Women Activists, Philadelphia, PA: Temple University Press. Pryse, Marjorie (1998) “Critical Interdisciplinarity, Women’s Studies, and Cross-Cultural Insight,” NWSA Journal, 10(1): 1–22. Rankin, Catherine (2002) “Social Capital, Microfinance, and the Politics of Development,” Feminist Economics, 8(1): 1–24. Stiglitz, Joseph (1998) “Towards a New Paradigm for Development: Strategies, Policies, and Processes,” Prebisch Lecture, UNCTAD, Geneva. Summers, Lawrence (1992) “Let Them Eat Pollution” (excerpt of Lawrence Summers’ internal memo to the World Bank), The Economist, February 8: 66. Wiesner, Eduardo (1998) “Transaction-Cost Economics and Public-Sector Rent-

Colonizing knowledge 141 Seeking in Developing Countries: Toward a Theory of Government Failure,” in E. Weisner and R. Picciotto (eds) Evaluation and Development: The Institutional Dimension, New Brunswick, NJ: Transaction Publishers for the World Bank: 108–23. Woolcock, Michael (1998) “Social Capital and Economic Development: Towards a Theoretical Synthesis and Policy Framework,” Theory and Society, 27(2): 1–22. World Bank (1998) “Social Development Update: Making Development More Inclusive and Effective,” Social Development Department, Paper No. 27, May 28, Washington, DC: World Bank. World Bank (2000) Entering the 21st Century: World Bank Development Report 1999/2000, New York: Oxford University Press. —— (2001) Engendering Development: Through Gender Equality in Rights, Resources, and Voice. Washington, DC: World Bank. World Bank (2002) “Social Capital” and “Social Capital and Gender.” On the World Bank website http://worldbank.org/poverty/scapital, accessed March 17, 2002.

10 Adjustment with a woman’s face Gender and macroeconomic policy at the World Bank Cynthia A. Wood The World Bank’s policies for structural adjustment have been consistently criticized for the unequal distribution of their negative effects, or “social costs.” Two major strands of this criticism are the poverty critique, articulated in UNICEF’s Adjustment with a Human Face (Cornia et al. (eds) 1987, 1988), and the gender critique, put forward in The Invisible Adjustment (UNICEF 1989; see also Commonwealth Secretariat 1989).1 However, the World Bank has responded to these critiques very differently. The Bank’s many published analyses of the social costs of adjustment since 1987 deal almost exclusively with the poor. Gender inequalities in the costs of adjustment have not been addressed. While the poverty and gender critiques have much in common, they are not identical. The Bank’s different responses to them thus have important implications for policy. Because the Bank has shown increasing commitment to issues of gender and development in other areas of policy, the persistent lack of attention to gender in its analysis of structural adjustment requires an explanation. Much of this explanation must come from institutional and political analyses of the Bank which are beyond the scope of this chapter (see O’Brien et al. 2000: 24–66; Women’s Eyes 1997).2 However, such analyses are not enough; conceptual issues must also be addressed. I argue in this chapter that criticisms of the social costs of adjustment are given credence by macroeconomic policymakers at the Bank only to the degree that they can be framed in terms of the neoclassical economic paradigm which has dominated Bank lending for so many years (Kapur et al. 1997; Murphy 1995: 35). Arguments in Adjustment with a Human Face which were so framed were quickly addressed in ofﬁcial discussions of adjustment. In contrast, the gender and adjustment critique is not framed in neoclassical terms, and suggests instead that the paradigm is problematic. Consequently, the Bank disregards this literature even in its discussions of social costs and of compensatory measures to ameliorate those costs. The few Bank documents which discuss gender in the context of structural adjustment generally present only issues compatible with the paradigm and do not refer to the gender critique. With continued pressure from critics, gender is beginning to enter Bank analysis of structural adjustment. However, many gender inequalities resulting from adjustment emerge in areas likely to be marginalized in neoclassical

Adjustment with a woman’s face 143 economics. I focus on one such area, unpaid domestic labor (or housework), to show that the failure to consider issues excluded by the paradigm makes it likely that future macroeconomic policy emerging from the Bank will continue to be gender-biased.3 Furthermore, without attention to unpaid domestic labor and other issues incompatible with neoclassical theory, compensatory programs are unlikely to counter many gender inequalities in the effects of structural adjustment, and may worsen them in some cases. As long as the neoclassical paradigm dominates macroeconomic policymaking at the Bank, institutional changes and political pressure which encourage greater attention to gender will not, by themselves, eliminate gender bias in adjustment policies. I base my argument on an examination of publicly available World Bank documents – formal publications, technical notes, and material on the Bank’s external web site – analyzing structural adjustment, the social costs of adjustment, and gender. I also analyze Bank documents that mention gender in the context of adjustment, and other publications which demonstrate the Bank’s position particularly well. My goal is to present an accurate picture of the Bank’s analysis of structural adjustment and social costs using documents which together constitute its public discourse on the topic.4 After a description of structural adjustment policies, I brieﬂy review the gender and adjustment literature. Next, I document the absence of gender in Bank analyses of structural adjustment and social costs, including a review of exceptions to this rule. I then illustrate the overall lack of discussion of adjustment in the Bank’s gender analyses until the publication of Engendering Development (World Bank 2001b). I explain the Bank’s failure to incorporate gender in macroeconomic policymaking through an exploration of the neoclassical paradigm, focusing on its exclusion of unpaid domestic labor. Finally, I discuss the success of the poverty critique in attracting Bank attention and implications for the future of gender analysis in Bank macroeconomic policy.

Structural adjustment and compensatory programs From a neoclassical economic perspective, the problems faced by less-developed countries beginning with the debt crisis in the early 1980s were due to longterm interference with the market by individual governments (Mosley et al. 1991: 4–9; Commonwealth Secretariat 1989: 19–22). “Structural adjustment” was the process of correcting these market imbalances (Mosley et al. 1991: 23). Successful adjustment was signaled by market criteria: low inﬂation, competitive exchange rates, reductions in government spending, and a “microeconomic environment that is favorable to new investment” (World Bank 1990: 11). Policies to achieve macroeconomic “stabilization” devalued exchange rates, cut consumption and government spending, and tightened monetary policy. The World Bank made loans conditional on changes in national economies which increased the role of the market through trade liberalization, privatization, reducing state involvement in the economy, and eliminating exchange rate controls and subsidies on consumer goods and services (Mosley et al. 1991: i, 34).

144 Cynthia A. Wood From the Bank’s perspective, short-term recessional effects resulting from adjustment policies were transitional costs necessary to achieve the conditions necessary for long-term growth (World Bank 1987: 6). Rising unemployment, falling wages, widening income inequality, and increasing morbidity and mortality rates devastated countries which applied structural adjustment policies during and after the 1980s. Many argue that the long-term effects of these policies will be equally damaging (see Cornia et al. (eds) 1987, 1988; Oxfam Policy Department 1995; SAPRIN 2001). In response to criticisms such as these, the Bank developed “compensatory programs” such as social emergency funds, which were designed to accompany adjustment and compensate vulnerable groups through targeted provision of income and social services (Vickers 1991; Benería and Mendoza 1995). The Bank eventually made social protection a condition of further loans (World Bank 1996). However, gender is not yet a factor in Bank evaluations of the social costs of adjustment, despite a substantial literature on the topic.

Gender critiques of adjustment Numerous presentations, articles, and books on gender and adjustment have appeared since 1987, several published by respected international agencies.5 Most authors of these works are development practitioners or Ph.D. economists at universities, international agencies, or non-governmental organizations (NGOs). Overviews of the literature highlight a few major points (see Çagatay et al. (eds) 1995; Sparr (ed.) 1994; Blackden and Morris-Hughes 1993; Vickers 1991: 15–42): 1 2

Structural adjustment policies are biased against women in conception, implementation, and evaluation; The invisibility of this bias to policymakers at international agencies such as the World Bank is part of the problem and is reﬂected in the paucity of statistical data on the issue; Evidence consistently shows that women suffer more and gain less than men from adjustment policies.

At the conceptual level, many argue that the neoclassical underpinnings of adjustment policies are a major source of gender bias. There are many threads of this argument which I shall be unable to discuss here. I focus on unpaid domestic labor because I believe that in deﬁning economic analysis in market terms, neoclassical theory is particularly resistant to the inclusion of such labor as an economic activity (see Wood 1997). Critics argue that because neoclassical economics does not include unpaid domestic labor in its deﬁnition of economic activity, the effects of adjustment on such labor are invisible to Bank policymakers. Yet this labor is an important household resource and must be considered if the effects of adjustment are to be understood fully. The substitution of unpaid domestic labor for market income

Adjustment with a woman’s face 145 and social services, in particular, is a major factor in the “success” of structural adjustment policies as defined by neoclassical policymakers (see Elson 1989: 57–8). This substitution occurred routinely as a direct result of structural adjustment policies in the 1980s and after; poor households used all available resources to adapt to worsening economic conditions. Since unpaid domestic labor is socially deﬁned as “women’s work,” this particular effect of adjustment was not gender neutral. Women’s workloads increased more than men’s. Quantitative data is difﬁcult to ﬁnd on unpaid domestic labor. Since it is not considered an economic activity, it is not measured in conventional economic surveys.6 However, case studies and anecdotal evidence from NGOs and other sources suggest that adjustment increased the quantity and intensity of unpaid domestic labor. Women compensated for falling income by spending more time on day-to-day management of the household and by substituting unpaid work for commodities they could no longer afford. Searching for lower prices and cheaper substitutes to make a smaller income go further, women in Argentina, Mexico, and elsewhere went to more shops or markets, shopped more frequently, traveled further from home, and spent more time checking prices (Feijoó and Jelin 1989: 39; Benería 1991: 175). Time spent seeking out loans or selling assets such as jewelry in response to lower income also increased (Moser 1989: 149; Stewart 1991: 1853). Women worked longer hours producing what they could once buy, with a lower quality of domestic equipment (Feijoó and Jelin 1989: 39; Benería 1991: 175; Waylen 1992: 158). They bought food that was less processed, which required that the processing be done at home, and spent more time gathering ﬁrewood and water (Commonwealth Secretariat 1989: 65). Repairs not made on items such as refrigerators increased cooking time (Benería 1991: 175; Moser 1989: 148). Many adjustment policies directly increased women’s domestic labor by cutting government funds to social services. Decreasing public expenditures on water, sanitation, and sewerage made the search for potable water and the disposal of waste more difﬁcult (Rocha et al. 1989: 27). As a result, illness became more common. Since many countries also cut spending on health care and subsidies on food as part of their adjustment program, women spent more time caring for the sick (see Cornia 1987: 29, 39; Commonwealth Secretariat 1989: 42; Waylen 1992: 159–60). Cutting clinic hours or shutting down clinics meant that women spent more time traveling or waiting in line with sick children. Raising or introducing charges for clinic visits meant that many could not afford to go at all. Poor women found innovative ways to cope with reduced income and social services, such as communal kitchens in Peru, but these also often required increases in unpaid domestic labor ( Jacobs 1991; Waylen 1992: 170–1; Commonwealth Secretariat 1989: 42). Even outside assistance to address the effects of structural adjustment on the poor increased women’s unpaid work (Moser 1989: 158). Increases in adult women’s workload overall resulted in girls staying home from school to take on domestic labor (Moser 1989: 154–5; Cornia 1987: 40).7

146 Cynthia A. Wood While the breadth and complexity of gender critiques of adjustment cannot be pursued here, this summary is enough for an examination of the World Bank’s treatment of gender in its evaluation of structural adjustment policies. Almost without exception, Bank policymakers evaluating structural adjustment and their social costs do not measure, or even recognize, the effects of adjustment on women described in these critiques.

Gender in the Bank’s analysis of structural adjustment The Bank has not addressed gender critiques of its policies in publications devoted to adjustment or social costs. It fails for the most part even to acknowledge the existence of these critiques, in stark contrast to its response to Adjustment with a Human Face. Gender itself is marginal to the Bank’s discussions of structural adjustment, as even some within the Bank have noted (Blackden and MorrisHughes 1993: 16). Consider a paper on the “Structural-Adjustment Debate” given in 1993 by Lawrence Summers, then Chief Economist of the World Bank. According to Summers and his co-author, the paper “distill[s] the main lines of criticism levied against structural-adjustment programs to four critiques and assess[es] the validity of each” (Summers and Pritchett 1993: 383). Summers discusses the poverty critique, but does not refer either to gender or to the literature criticizing adjustment’s effects on women. Since Summers was a major ﬁgure in the development of Bank policies for structural adjustment at the time, the absence of gender in his presentation is especially signiﬁcant. His lack of attention indicates the Bank’s general position. A review of more recent Bank publications on structural adjustment and social costs demonstrates the continued failure to address gender in this area of analysis. In the overview Structural and Sectoral Adjustment: World Bank Experience 1980–92, there is only one brief reference to women ( Jayarajah and Branson 1995: 177). The study Social Dimensions of Adjustment: World Bank Experience, 1980–93, supposedly a review document on social costs, “looks at all the evaluated adjustment operations supported by the Bank . . . and tracks what happened to poverty and income distribution” ( Jayarajah et al. 1996: xi). The review never refers to gender in the context of structural adjustment, nor does it analyze gender in terms of social costs. It mentions women only brieﬂy in isolated comments (see ibid.: 107, 111, 114, 118). While it is difﬁcult to document the absence of attention, I have found no Bank analysis of adjustment or the social costs of adjustment which makes more than a passing reference to gender, with the exception of the documents discussed below. The first exception is Paul Collier’s work, which explains the impact of adjustment on women in terms of the constraints they face in earning income and differences in their consumption of certain public services (Collier 1993). Collier’s essay is signiﬁcant in deﬁning a legitimate space for gender in the World Bank’s structural adjustment literature, as demonstrated by its appearance in a

Adjustment with a woman’s face 147 publication edited by senior economists, but this space is telling for what it excludes. He does not refer to the literature on gender and adjustment, which is important because it allows him to deﬁne both the problem and the solution to it in neoclassical terms, as I discuss below. The second exception is an Africa Region Technical Note by C. Mark Blackden and Elizabeth Morris-Hughes entitled Paradigm Postponed: Gender and Economic Adjustment in Sub-Saharan Africa (Blackden and Morris-Hughes 1993). Blackden and Morris-Hughes not only refer to gender critiques of adjustment, but take this literature (and feminist economic theory) as their primary frame of reference in analyzing policies for structural adjustment. Research and policy recommendations of the multi-donor Special Program of Assistance for Africa (SPA), including a pilot Structural Adjustment and Gender in Africa (SAGA) initiative funded by the World Bank, are based in part on Paradigm Postponed (World Bank 1996; Blackden 1998). However, no other publication in the Bank’s literature on adjustment and social costs to date refers to Paradigm Postponed, nor does it appear in a collection or series indicating general acceptance among Bank economists responsible for macroeconomic policy, as Collier’s work does (see also O’Brien et al. 2000: 56). Even Adjustment in Africa, a formal Bank publication produced by the ofﬁce of the Chief Economist in “close collaboration with the World Bank’s Africa Region,” fails to cite Blackden and Morris-Hughes, and makes only two brief comments on gender in the context of structural adjustment (World Bank 1994a: xv, 167–8, 174). In light of these indications of marginality, I consider Paradigm Postponed an “outlier” in the Bank’s analysis of structural adjustment. Nevertheless, it is important as a published alternative to Collier’s approach to gender and adjustment appearing within the Bank, and the application of this work by the Africa Region and the SPA models innovative macroeconomic analysis responsive to the gender critique of adjustment (see Blackden 1998). Examining the exceptions to the general absence of gender in the Bank’s macroeconomic analysis runs the risk of overstating their importance. Bank overviews of adjustment and social costs cite neither Collier’s work nor Paradigm Postponed (e.g. Jayarajah and Branson 1995; Jayarajah et al. 1996; World Bank 2001a). Even combined with the work of the Africa Region and the SPA, these documents were not enough to bring about the incorporation of gender in the Bank’s overall analysis of structural adjustment. At the United Nation’s 1995 Fourth World Conference on Women held in Beijing, James D. Wolfensohn, President of the World Bank, stated: A priority concern must be to ensure that women are not hurt by structural adjustment programs. I am well aware of the wide criticism of the Bank on this subject. I believe that a macroeconomic plan is crucial to development, but I will be vigilant and more sensitive to arguments which relate to disproportionate adverse social impacts on women. (Wolfensohn 1995)

148 Cynthia A. Wood Wolfensohn is known to be concerned with the social costs of adjustment and “is regarded by gender equity advocates both within and outside the Bank as a tremendous positive resource for change” (O’Brien et al. 2000: 53). Nevertheless, some years after Beijing there is little evidence that macroeconomic policymakers are paying serious attention to the effects of structural adjustment on women, except in the Africa Region, where interested and committed staff responded very early to the gender critique (see Women’s Eyes 1997; O’Brien et al. 2000: 24–6).

Structural adjustment in the Bank’s analysis of gender Corresponding to the absence of gender from the Bank’s analysis of structural adjustment is the lack of discussion of adjustment in its now extensive work on gender. Bank staff working on gender are not directly involved in macroeconomic policymaking, so this is the less important of the two in the design of policies for structural adjustment. However, pressure exerted by staff interested in or mandated to promote gender issues at the Bank could, at least theoretically, affect macroeconomic policy.8 Unfortunately, serious pressure to incorporate gender in the Bank’s analysis of structural adjustment is not obvious in Bank publications on gender. While unpublished internal pressure certainly exists, to date these efforts have failed to “engender” macroeconomics at the Bank. The recent publication of the Policy Research Report Engendering Development (World Bank 2001b) actually undermines the impetus to do so. The Bank’s literature on gender is devoted almost exclusively to projectbased lending in areas of “social development,” such as education, population, and health, so much so that adjustment policy must be seen as outside the purview of gender analysis as defined by the Bank, at least until Engendering Development (see Buvinic et al. 1996: 2–4). In two World Bank studies, Josette Murphy reviews the Bank’s “record of incorporating gender issues in the operations it supports” and discusses a total of 800 projects either implemented or approved between 1967 and 1996 (Murphy 1995: xi, 1; 1997: 1). Neither study considers macroeconomic lending, and other than a brief comment on the work of the SPA, Murphy does not mention structural adjustment (1995: 54; 1997: 15–16). Indirectly, she suggests that gender and macroeconomics are mutually exclusive: “The Bank’s shift toward macroeconomic policy during the early 1980s made such issues as poverty alleviation and gender less immediate Bank concerns” (1995: 35). A few exceptions to the general absence of structural adjustment in Bank discussions of gender exist before Engendering Development. In 1994, the first World Bank policy paper on women appeared, Enhancing Women’s Participation in Economic Development (World Bank 1994b). Policy papers must be approved by the Board of Executive Directors and represent the Bank’s ofﬁcial position, so it is especially signiﬁcant that this document explicitly addresses the effects of structural adjustment on women: “In some countries and situations, . . . shortterm adverse effects [of adjustment policies] may fall disproportionately on

Adjustment with a woman’s face 149 women” (World Bank 1994b: 66–7). It also suggests that “the beneficial effects of adjustment may be slow in reaching women” (ibid.: p. 68). These remarks constitute the Bank’s first acknowledgment in an official document that adjustment may have had adverse effects on women. Coupled with the statement that “[g]ender issues will be systematically addressed . . . in the design and implementation of lending programs, including adjustment operations” (ibid.: p. 13), they justify and even mandate analysis of gender and adjustment. Unfortunately, follow-through on this particular aspect of the policy paper has not been forthcoming. Even the Operational Policy statement accompanying the paper summarizes recommended policy changes without mentioning structural adjustment (World Bank 1994c). Two Bank publications prepared for the UN’s Beijing conference mention macroeconomic policy in the context of gender, but neither comments on actual or potential negative effects (see World Bank 1995a, 1995b). A progress report following Beijing devotes a section to “integrating gender into adjustment operations and economic reform” (World Bank 1996). This report acknowledges the gender critique of Bank policies, and suggests that “the design and implementation of [economic reform] programs . . . should take gender differences in needs, constraints and resources into consideration.” However, the SPA’s work is the only concrete example offered to support the assertion of Bank progress in this area. The majority of this section discusses poverty. Again, documenting exceptions to the general absence of adjustment in the Bank’s gender analysis makes them appear more important than they are. The Bank itself does not take these exceptions seriously. In Mainstreaming Gender and Development in the World Bank, Caroline Moser and her co-authors do not refer to structural adjustment in their comments on the documents analyzed above (Moser et al. 1999).9 The most recent publication on gender, Engendering Development, presents several pages on gender differences in the effects of structural adjustment, ﬁnally breaking the silence on this issue in gender analysis at the Bank (World Bank 2001b: 212–19). However, despite several promising suggestions for engendering macroeconomics and a call for more research, it seems likely that policymakers will ﬁnd ample reason to ignore gender in the design and evaluation of adjustment policies in this statement: “While there is evidence to support both sides of the debate about the impact of structural adjustment, on balance the evidence suggests that females’ absolute status and gender equality improved, not deteriorated, over the adjustment period” (ibid.: 215). Further investigation is necessary to understand how the authors of the report come to a conclusion so contrary to the negative effects “exhaustively documented” in the existing gender and adjustment literature (World Bank 1998). It seems to be based primarily on the Bank’s own research conducted for purposes of the report.10 Despite acknowledging the existence of a debate and the legitimacy of many of its arguments, the report does not actually respond to speciﬁc gender critiques of adjustment. This is certainly due in part to the Bank’s consistent dismissal of qualitative research as legitimate evidence (see World Bank

150 Cynthia A. Wood 2001b: xiv). The conclusion bears scrutiny, in any case. For example, while the report ﬁnds that changes in women’s life expectancy and girls’ school enrollments over the adjustment period were more positive for adjusting than for nonadjusting countries, it assumes that these positive outcomes resulted from adjustment policy. Other potential explanations would suggest that these outcomes occurred despite rather than because of adjustment.11

Explaining the absence of gender in the Bank’s analysis of structural adjustment Why does the Bank fail to consider gender or the gender critique in its analysis of structural adjustment and social costs? My answer to this question is based on an analysis of the economic paradigm dominating macroeconomic policymaking. Neoclassical theory excludes or marginalizes issues central to a complete understanding of the effects of adjustment on women. As a result, even Bank economists committed to the elimination of gender inequality will produce gender-biased macroeconomic policies. Further, they will be unable to recognize that their policies are gender-biased. This not only explains the absence of gender in the Bank’s discussions of structural adjustment throughout the 1990s, but also predicts the outlines of gender bias in future macroeconomic policies which recognize gender as a topic relevant to discussions of adjustment. According to the World Bank charter, “it must lend only for productive purposes and must stimulate economic growth in the developing countries where it lends” (World Bank 1985: 3). The apparent neutrality of this goal obscures the implicit assumption of market criteria for deﬁning legitimate boundaries for economic analysis and policy evaluation. In practice, for policymaking and lending at the Bank, “economic growth” refers to growth in the market sector, and market indicators are the only evidence necessary for measuring economic advance or decline. Neoclassical approaches to macroeconomics assert that free prices are the most efficient and desirable means to policy goals (World Bank 1991: 1). This approach has not been altered by the Bank’s recent attention to poverty reduction and sustainable development, both of which it sees as deﬁned by and best achieved through the market. In light of gender critiques of the neoclassical paradigm, an evident problem with this focus on the market is that it ignores or marginalizes nonmarket economic activity such as unpaid domestic labor in the creation and evaluation of policy. Since the Bank deﬁnes productivity and economic growth in market terms, attention to unpaid labor in Bank policy would be incompatible with its goals. The emphasis on “getting prices right” also ensures that such labor remains invisible to policymakers. Macroeconomists do not look at unpaid domestic labor in their analyses of economic policy except in a few cases as a noneconomic factor explaining market behavior, because the paradigm does not allow them to do so. Institutionalized attention to women in the analysis of development and policy does not prevent this exclusion. According to Nüket Kardam, policymakers

Adjustment with a woman’s face 151 within the Bank who wish to promote policy beneﬁting women have “taken care to make WID issues acceptable by framing them within economic rather than social welfare and equity arguments” (Kardam 1991: 79; see also Razavi and Miller 1995: iv–v, 40; Murphy 1995: 44; Buvinic et al. 1996: 73; Women’s Eyes 1997; O’Brien et al. 2000: 48). Gender research at the Bank is thus devoted to proving economic (market) beneﬁts of projects targeting women or taking gender into account. For example, in 1992, Chief Economist Summers spoke at the Bank’s annual meeting on the importance of girls’ education in the process of development (1994). Summers suggests that “investment in girls’ education may well be the highest return investment available in the developing world” (Summers 1994: 1).12 However, the focus of this analysis is still the market, since “education is positively correlated with overall economic growth, with one year of schooling . . . leading to as much as 9 percent increase in GDP for the ﬁrst three years of schooling” (ibid.: v). A primary argument in Engendering Development is that “[g]ender inequalities undermine the effectiveness of development policies” (World Bank 2001b: xiii) and that gender inequality “diminishes an economy’s capacity to grow” (ibid.: p. 11). If gender achieves legitimacy at the World Bank by being made acceptable to neoclassical economists, it follows that gender analysis which is not framed in neoclassical terms will be excluded. The theoretical framing of much of the gender and adjustment literature requires analysis of unpaid domestic labor as an economic activity. Since the neoclassical paradigm assumes a market deﬁnition of economic activity, discussions that analyze unpaid domestic labor in economic terms (i.e. as itself an economic activity rather than subsidiary to economic activity) will be marginalized.13 Both because it is critical of the neoclassical paradigm and because it focuses on nonmarket issues, the gender critique is necessarily excluded from the Bank’s analysis of structural adjustment. More telling, perhaps, is the absence of gender itself. This absence cannot be completely explained by hostility to the gender and adjustment literature, because the topic could be discussed without reference to the literature. But if, as many gender critiques suggest, understanding the effects of structural adjustment on women requires a look at unpaid domestic labor, economic analyses of adjustment based on the neoclassical paradigm are hard-pressed to address the topic at all. The few Bank documents which look at gender in the context of adjustment do not refer to unpaid domestic labor. In keeping with neoclassical economic analysis, these documents focus almost exclusively on the market repercussions of adjustment, differentiated by gender. According to Enhancing Women’s Participation in Economic Development, for example, “[a]djustment policies typically remove price distortions and restore proﬁtability of certain crops and activities, but women may not be able to take advantage of such beneﬁcial changes unless their particular constraints are removed” (World Bank 1994b: 68). Similarly, most of Paul Collier’s work analyzes the impact of adjustment on women in terms of the market (Collier 1983: 185–93). He looks at unpaid domestic labor

152 Cynthia A. Wood only as it affects the allocation of women’s time to market work, and gives no indication that structural adjustment has any effect on unpaid domestic labor itself (ibid.: 193–6). In its discussion of structural adjustment, Engendering Development refers to domestic labor only in terms of the impediments such labor imposes on women’s ability to “participate fully in longer term economic opportunities associated with adjustment” (World Bank 2001b: 214). While it is important to look at the market effects of structural adjustment policies on women, gender critiques highlight the centrality of unpaid domestic labor to understanding gender bias in adjustment. In this context, the absence of any reference to such labor in the Bank’s few analyses of gender and adjustment is striking. Even if Bank policymakers believe these critiques to be ﬂawed, their analyses should at least address the effects of adjustment on unpaid domestic labor, if only to give solid reasons for dismissing the importance of these effects. Paradigm Postponed suggests that for the Bank, “[t]he critical question is whether the distribution of costs (and beneﬁts) of adjustment between men and women . . . is sufﬁciently important to make a difference to economic outcomes and prospects” (Blackden and Morris-Hughes 1993: 16). In other words, the Bank will show serious concern for gender inequality in the effects of adjustment only if an economic justiﬁcation for such concern is provided. Since “economic” in a neoclassical context refers to the market, Bank statements to the effect that “the effective integration of gender concerns into the design and implementation of reform measures can lead to improved performance outcomes” do not imply that the Bank will consider the effects of adjustment on unpaid labor, except perhaps as instrumental to improving market indicators (World Bank 1996). To the degree that the gender bias of structural adjustment policies derives from their effects on unpaid domestic labor or some other topic excluded from neoclassical economics, future policy will thus remain gender-biased.

Adjustment with a human face The World Bank addressed the poverty critique of structural adjustment policies almost immediately, and in most ways the social cost literature can still be seen as deﬁned by that critique. How can this difference in reception be explained? According to Joan Nelson, Adjustment with a Human Face “focused and legitimized concerns that had been mounting in many quarters” (Nelson 1989: 95). For the Bank’s macroeconomic policymakers, legitimization entailed making concerns over the effects of adjustment policy on the poor consistent with neoclassical economic theory. Adjustment with a Human Face was intentionally framed to appeal to a neoclassical perspective: “From the start, it was clear that practical advance toward adjustment with a human face would only be possible if it attracted the understanding and support of those in the mainstream of economic policy-making” (Cornia et al. 1987b: 3). The authors did not question the need for structural adjustment, and they highlighted the “economic” reason for their concern – the

Adjustment with a woman’s face 153 long-term and irreversible damage to “human resources” which occurs with increasing poverty (ibid.: 5–6). Concern over the social costs of adjustment from this perspective is legitimized by the potential for falling rates of growth resulting from lost (human) resources. However, criticisms of structural adjustment policy which were not translated into the neoclassical economic paradigm were not addressed. The Bank never took seriously, for example, the argument that “the appropriate adjustment may not be . . . in the national economies where the major and unsustainable imbalances emerge, but in the international conditions which give rise to them” (Cornia et al. 1987c: 133). From the Bank’s perspective, Adjustment with a Human Face was a warning about the dangers of disinvestment in human capital which demanded no more than minor changes in the design of adjustment policy. This is the only aspect of its analysis that had “economic” signiﬁcance within the neoclassical paradigm. Many policymakers see no problem with discussing the social costs of adjustment exclusively in terms of poverty, since policies which benefit the poor necessarily beneﬁt women (e.g. World Bank 1996). This argument ignores issues crucial to understanding gender inequality in the effects of structural adjustment. An analysis of gender bias in compensatory programs will illustrate this point further.

Compensatory programs Because the World Bank’s analysis of social costs does not currently incorporate gender, compensatory programs are designed without attention to gender inequality and are themselves gender-biased.14 The neoclassical paradigm establishes theoretical constraints on the evaluation of policy which make it likely that gender bias in structural adjustment will persist despite greater emphasis at the Bank on addressing social costs, even as this comes to include attention to gender (e.g. see World Bank 2001b: 227). Safety net and social emergency programs have reduced the negative effects of adjustment on vulnerable groups, but while women tend to benefit from income and employment programs designed for the poor, they do not necessarily beneﬁt equally. Lourdes Benería and Breny Mendoza argue that gender inequality in Emergency Social Investment Funds (ESIFs) in Honduras and Nicaragua can be explained in part by (neoclassical) conceptual biases which ignore gender issues in income distribution (1995: 65; see also Buvinic et al. 1996: 64).15 The exclusion of unpaid domestic labor from economic analysis is a less obvious source of gender inequality in compensatory programs. In accordance with the neoclassical paradigm, the Bank’s analysis of social costs is limited to market effects. Protecting the Poor During Periods of Adjustment asserts that “[a]nalyzing the effects of a policy change on the poor involves assessing the effect of that change on real purchasing power by determining the impact on the disposable income of the group and the prices and quantities of the goods and services the group consumes” (World Bank 1987: 36). From this perspective,

154 Cynthia A. Wood decreases in market consumption deﬁne the possible negative effects of adjustment, and directing monetary income to households is the best compensation. Again, changes in unpaid domestic labor (or in nonmarket consumption derived from such labor) resulting from structural adjustment policies or compensatory programs will not be visible to policymakers in this context. Nor will they be compensated for in future policy. In one World Bank Living Standards Measurement Study (LSMS) study, Paul Glewwe and Gillette Hall report that during the late 1980s, female-headed households in Peru were no more vulnerable to macroeconomic shocks produced by structural adjustment than male-headed households (Glewwe and Hall 1995). These results are “contrary to the common assertion that female-headed households, and women per se, are one ‘vulnerable group’ in periods of . . . structural adjustment” (ibid.: 31). It follows that attention to gender in adjustment policy is not justiﬁed even on the basis of equity. But this report, like most LSMS studies, measures household welfare strictly in terms of market consumption (ibid.: 5). Female-headed households in Peru may have maintained their market consumption by increasing paid labor time but also increasing the intensity of time spent on unpaid domestic labor. Or time spent on food preparation, repair of clothes, and childcare at home may have increased in intensity or been shifted on to young daughters. Other than the increase in paid labor time, none of this would be captured by an LSMS survey. How valid, then, is Glewwe and Hall’s conclusion? Evidence of structural adjustment’s effects on unpaid domestic labor suggests that it is at least problematic. Policy will nevertheless be based on this conclusion, which ignores a persistent source of gender inequality. Glewwe and Hall argue that while “similarity in structural adjustment policies across many countries may produce the same ‘vulnerable’ groups in those countries, . . . in principle, these groups can be made ‘invulnerable’ by corresponding policy changes” (ibid.: 3). This is true only if neoclassical policymakers recognize and measure all aspects of vulnerability. Since they do not, it is unlikely that women will be made “invulnerable” to the negative effects of structural adjustment by future policy changes. The relationship between unpaid domestic labor and social services is another potential source of gender bias in Bank policies unlikely to be addressed in compensatory programs, again because issues outside the neoclassical paradigm are not considered. For example, a cut in subsidies decreases the number of days spent at the hospital by increasing costs to the consumer. According to the neoclassical paradigm, fewer scarce resources are used as a result of the increased price. However, Diane Elson points out that “in reality there has been a transfer of the costs of care for the sick from the paid economy to the unpaid economy of the household. The financial costs fall but the unpaid work of women in the household rises. This is not a genuine increase in efﬁciency . . . ” (Elson 1991: 178). Due to the effects of policy on unpaid domestic labor, compensatory programs which increase household income will not necessarily compensate women for

Adjustment with a woman’s face 155 cuts in social programs that do not have an affordable and readily available market substitutes. Cutting clinic hours in rural areas is likely to increase women’s unpaid domestic labor (by increasing waiting at the clinic, travel time to another clinic, or time spent caring for the sick), whether or not her household has been compensated with increased income to cover the higher cost of health care elsewhere. Supplementary feeding programs, a common component of compensatory programs, may also increase unpaid domestic labor by requiring women’s direct involvement in providing food (see Vilas 1996: 24). Structural adjustment policies shaped by neoclassical economics produce gender inequalities likely to be invisible to policymakers. Compensatory measures, which also derive from neoclassical theory, fail to address these gender inequalities and may actually compound them whether or not gender is accounted for in implementation. Conformance with the neoclassical paradigm limits the potential both for adjustment policies to be gender neutral and for compensatory programs to correct the policies’ gender biases.

Conclusion There is no doubt that over the past ten years the World Bank has increased its commitment to gender issues in development. That its attention to the social costs of adjustment increased simultaneously only highlights the lack of intersection between these two areas of policymaking. Macroeconomics at the Bank remains devoid of gender analysis, despite gender critiques of adjustment. Because Bank lending continues to focus on policies for structural adjustment, the failure to incorporate gender in its analysis of adjustment and social costs is likely to affect women in economies in the process of being “reformed” for years to come. The Bank is likely to include gender as a component of macroeconomic evaluations of “reform” relatively soon.16 However, my analysis of the reasons for the absence of gender in the design and evaluation of policies for structural adjustment suggests that the terms of this incorporation are predictable and will be limited by the neoclassical paradigm. The Bank will continue to consider only market aspects of gender differences in the effects of structural adjustment and ignore or marginalize nonmarket effects, especially those on unpaid domestic labor. Other issues incompatible with neoclassical economics but vital to gender analysis of macroeconomics will also be excluded, and any conflict between market growth and women’s welfare will be decided in favor of the market. Under these conditions, it is doubtful that “engendering” macroeconomics on the Bank’s terms will eliminate gender bias in adjustment policies. The implications of this analysis go beyond the Bank. The shared approach to policy defined by the “Washington consensus” of mainstream economists in the US government and international financial institutions suggests that gender is similarly absent in macroeconomic policies promoted by the IMF, the US Agency for International Development (USAID), the Inter-American Development Bank (IDB) and other agencies afﬁliated with, inﬂuenced by, or

156 Cynthia A. Wood dependent on these institutions (see Williamson 1996). It follows that economic policies prescribed by these agencies are no less gender-biased than the Bank’s, and for similar reasons.17 If the neoclassical paradigm is a primary source of gender bias in policies for structural adjustment, then improved training for policymakers and increased accountability on gender issues will not be enough to eliminate gender inequality from macroeconomics at the Bank and elsewhere. The economic theory on which policy is based must be transformed. Feminist economists underscore the sweeping changes necessary for new economic paradigms to include gender as a fundamental category of analysis. Women’s nonmarket activities must be recognized as economic and valued equally with market aspects of the economy. Differences in household organizations of production and conﬂicts over distribution not governed by the market must be built into the theory as well. Method can challenge or reinforce fundamental concepts, so traditional approaches to how economics is done must also be interrogated. These ideas are difﬁcult to grapple with, but essential to explore if gender equity in economic policy is to be achieved.

Notes 1 UNICEF published the ﬁrst edition of The Invisible Adjustment in both English and Spanish in 1987, the same year as the ﬁrst volume of Adjustment with A Human Face. Note that while this chapter emphasizes academic and development practitioner critiques of Bank policies, grassroots opposition stimulated these critiques and has been vital in bringing pressure to bear on the Bank to reevaluate structural adjustment. 2 See Kapur et al. (1997) and Mosley et al. (1991) for histories of the World Bank. 3 I use the term “unpaid domestic labor” to refer to those unremunerated activities performed in maintenance of households and their residents, such as cooking, cleaning, and childcare. Other activities such as childbirth, emotional caretaking, and sex should possibly be included as well. However deﬁned, unpaid domestic labor is predominantly women’s work. See Wood (1997) for an analysis of the importance of deﬁnitions and First World bias in how such labor is conceptualized. 4 The World Bank is a large and complex institution, with internal conﬂicts which are not necessarily reflected in public documents. There are Bank professionals currently working to promote gender analysis in macroeconomic policy. There is also external political pressure to do the same. While these are very important, I believe that the Bank’s public documents reﬂect who is “winning” struggles over the direction of its policies for structural adjustment. 5 In addition to UNICEF (1989) and Commonwealth Secretariat (1989), see Afshar and Dennis (eds) (1992), Palmer (1991),Vickers (1991), Benería and Feldman (eds) (1992), Sparr (ed.) (1994), Bakker (1994), Elson (1989, 1991, 1998), and Çagatay et al. (eds) (1995). There are differing approaches among these authors, which should not be overshadowed by my brief literature review. 6 Feminist activists and development practitioners working on gender have sought the inclusion of unpaid domestic labor in ofﬁcial data for well over two decades, with little success (see Benería 1992). 7 Women’s paid labor force participation also increased more than men’s (Cornia 1987: 22; Stewart 1991: 1853; Moser 1989: 144–7, 153; Benería 1991: 172–3). 8 See Murphy (1995), Women’s Eyes (1997) and O’Brien et al. (2000) for accounts of the evolution of Bank approaches to gender.

Adjustment with a woman’s face 157 9 This is especially signiﬁcant because before her tenure at the Bank, Moser was a prominent critic of Bank policies. Attributing many of the negative economic changes in a barrio of Guayaquil, Ecuador directly to structural adjustment, her work focused on the effects of these policies on women (1989). In another study done for the Bank, Moser discusses the economic changes in Guayaquil only in terms of “external shocks” and “economic crisis,” with no comment on the relationship between structural adjustment policies and the worsening economic environment (Moser 1996). 10 The Bank often uses its own data to support arguments, and ignores critical external research. For example, the Structural Adjustment Participatory Review Initiative (SAPRI) was a joint World Bank/civil society project. A Bank SAPRI report was positive in its evaluation of adjustment policies (World Bank 2001a). The ﬁnal SAPRI summary by the NGO SAPRIN, in contrast, “present[s] evidence from studies in which the Bank was involved of the negative impacts of its policies” and suggests that systematic Bank resistance to the SAPRI process demonstrates “the disingenuousness of . . . claims that they are interested in engaging civil-society groups in meaningful endeavours to attend to the issues that have driven so many to the streets” (SAPRIN 2001: 7). 11 Less positive results for life expectancy in non-adjusting South Africa, for example, might result from civil war, transitional governance, or the escalating AIDS epidemic among women during the “adjusting period.” 12 Summers’ speech on girls’ education was given the year before his talk on structural adjustment, which omitted discussion of gender. This reinforces my argument that the Bank’s failure to analyze gender in the context of adjustment cannot simply be explained as a lack of knowledge or commitment to gender issues generally. 13 Economists do discuss some forms of unpaid labor such as subsistence agriculture. However, the System of National Accounts, which ostensibly measures all forms of economic activity, explicitly excludes unpaid domestic labor. See Wood (1997) for an analysis of causes and implications of this. 14 Issues of implementation are also important to understanding gender bias in compensatory programs. See Benería and Mendoza (1995). Note that a rising commitment to social protection has not altered structural adjustment policies substantively. Compensatory programs are meant to counter policies which are otherwise unproblematic (World Bank 1987; 1994b: 68; Razavi and Miller 1995). 15 For example, 75 percent of jobs created through the ESIF funded by the World Bank in Honduras went to men (Benería and Mendoza 1995: 58). Income directed toward men is unlikely to be distributed equally within households or to compensate fully for an equal amount of income lost to women due to adjustment policies, since men tend to keep a proportion of their income for personal use (see Dwyer and Bruce 1988; Benería 1992). 16 There are many indications of this. In July 1999, for example, Mark Blackden ran a session on “tools for engendering adjustment” for the Bank’s Poverty Reduction and Economic Management (PREM) network, which is directly involved in macroeconomic policymaking (personal communication, June 18, 1999). The Bank’s recent SAPRI report commented on the negative effects of adjustment cuts in public sector employment on women (World Bank 2001a: 27). In addition, an upcoming World Development Report will be devoted to gender, in anticipation of the ﬁfth UN Conference on Women to be held in 2005 (O’Brien et al. 2000: 47). Pressure on the Bank to demonstrate some change in response to demands made at Beijing to address the disproportionate effects of structural adjustment on women will be very high. 17 The same argument applies to the rising “post-Washington consensus,” which is only a reﬁnement of neoclassical approaches to policy and will not contribute to the incorporation of gender in macroeconomics (see Stiglitz 1998).

158 Cynthia A. Wood

Acknowledgment Thanks to Derek Stanovsky, Kay Smith, Susan Eckstein and Jay Wentworth for comments, and to Daryl Sink for assistance in researching this chapter.

References Afshar, Haleh and Carolyne Dennis (eds) (1992) Women and Adjustment Policies in the Third World, London: Macmillan. Bakker, Isabella (ed.) (1994) The Strategic Silence: Gender and Economic Policy, London: Zed Press. Benería, Lourdes (1991) “Structural Adjustment, the Labor Market and the Household: The Case of Mexico,” in G. Standing and V. Tokman (eds), Towards Social Adjustment: Labor Market Concerns in Structural Adjustment, Geneva: International Labour Ofﬁce: 161–83. —— (1992) “Accounting for Women’s Work: The Progress of Two Decades,” World Development, 20(11): 1547–60. Benería, Lourdes and Shelley Feldman (eds) (1992) Unequal Burden: Economic Crises, Persistent Poverty, and Women’s Work, Boulder, CO: Westview Press. Benería, Lourdes and Breny Mendoza (1995) “Structural Adjustment and Social Emergency Funds: The Cases of Honduras, Mexico and Nicaragua,” European Journal of Development Research, Spring: 53–76. Blackden, C. Mark (1998) “Integrating Gender Into Economic Reform Through the Special Program of Assistance for Africa (SPA),” presented at the Symposium on Economie et Rapports Sociaux entre Hommes et Femmes, January 28–29, Geneva, Switzerland. Retrieved January 11, 2002. (http://www.unige.ch/iued/new/ information/publications/pdf/yp_silcence_pudique/8-Eco-Blackden.pdf). Blackden, C. Mark and Elizabeth Morris-Hughes (1993) Paradigm Postponed: Gender and Economic Adjustment in Sub-Saharan Africa, Africa Technical and Human Division (AFTHR), Technical Note No. 13, Washington, DC: World Bank. Buvinic, Mayra, Catherine Gwin and Lisa M. Bates (1996) Investing in Women: Progress and Prospects for the World Bank, Policy Essay No. 19, Washington, DC: Overseas Development Council. Çagatay, Nilüfer, Diane Elson and Caren Grown (eds) (1995) “Gender, Adjustment and Macroeconomics,” Special Issue, World Development, 23(11). Collier, Paul (1993) “The Impact of Adjustment on Women,” in L. Demery et al. (eds), Understanding the Social Effects of Policy Reform, Washington, DC: World Bank: 183–97. Commonwealth Secretariat (1989) Engendering Adjustment for the 1990s, Report of a Commonwealth Expert Group on Women and Structural Adjustment, London: Commonwealth Secretariat. Cornia, Giovanni A. (1987) “Economic Decline and Human Welfare in the First Half of the 1980s,” in Cornia et al. (eds), Adjustment with a Human Face, Vol. I, Protecting the Vulnerable and Promoting Growth, Oxford: Clarendon Press: 11–47. Cornia, Giovanni A., Richard Jolly and Frances Stewart (eds) (1987a) Adjustment with a Human Face, Vol. I, Protecting the Vulnerable and Promoting Growth, Oxford: Clarendon Press. —— (1987b) “Introduction,” in Cornia et al. (eds), Adjustment with a Human Face, Vol. I, Protecting the Vulnerable and Promoting Growth, Oxford: Clarendon Press: 1–8.

Adjustment with a woman’s face 159 —— (1987c) “An Overview of the Alternative Approach,” in Cornia et al. (eds), Adjustment with a Human Face, Vol. I, Protecting the Vulnerable and Promoting Growth, Oxford: Clarendon Press: 131–46. —— (eds) (1988) Adjustment with a Human Face, Vol. II, Country Case Studies, Oxford: Clarendon Press. Dwyer, Daisy and Judith Bruce (eds) (1988) A Home Divided: Women and Income in the Third World, Stanford, CA: Stanford University Press. Elson, Diane (1989) “The Impact of Structural Adjustment on Women: Concepts and Issues,” in B. Onimode (ed.), The IMF, The World Bank and African Debt, Vol. 2, Atlantic Highlands, NJ: Zed Books Ltd: 56–74. —— (1991) “Male Bias in Macro-economics: The Case of Structural Adjustment,” in D. Elson (ed.), Male Bias in the Development Process, New York: Manchester University Press: 164–90. —— (1998) “The Economic, the Political and the Domestic: Businesses, State and Households in the Organisation of Production,” New Political Economy, 3(2): 189–208. Feijoó, María del Carmen and Elizabeth Jelin (1989) “Women from Low Income Sectors: Economic Recession and Democratization of Politics in Argentina,” in UNICEF, The Invisible Adjustment: Poor Women and the Economic Crisis, Santiago: UNICEF, The Americas and the Caribbean Regional Ofﬁce: 29–58. Glewwe, Paul and Gillette Hall (1995) “Who Is Most Vulnerable to Macroeconomic Shocks? Hypotheses Tests Using Panel Data from Peru,” World Bank Living Standards Measurement Study No. 117, Washington, DC: World Bank. Jacobs, Beryl K. (1991) “Lima’s Communal Kitchens,” Women in Action, 2: 16–17. Jayarajah, Carl and William Branson (1995) Structural and Sectoral Adjustment: World Bank Experience, 1980–92, Washington, DC: World Bank. Jayarajah, Carl, William Branson and Binayak Sen (1996) Social Dimensions of Adjustment: World Bank Experience, 1980–93, Washington, DC: World Bank. Kapur, Devish, John P. Lewis and Richard Webb (eds) (1997) The World Bank: Its First Half Century, Vols 1 and 2, Washington, DC: Brookings Institution Press. Kardam, Nüket (1991) Bringing Women In: Women’s Issues in International Development Programs, Boulder, CO: Lynne Rienner Publishers. Moser, Caroline O.N. (1989) “The Impact of Recession and Adjustment Policies at the Micro-Level: Low Income Women and Their Households in Guayaquil, Ecuador,” in UNICEF, The Invisible Adjustment: Poor Women and the Economic Crisis, Santiago: UNICEF, The Americas and the Caribbean Regional Ofﬁce: 137–66.. —— (1996) Confronting Crisis: A Comparative Study of Household Responses to Poverty and Vulnerability in Four Poor Urban Communities, Washington, DC: World Bank. Moser, Caroline O.N., Annika Tornqvist and Bernice van Bronkhorst (1999) Mainstreaming Gender and Development in the World Bank: Progress and Recommendations, Washington, DC: World Bank. Mosley, Paul, Jane Harrigan and John Toye (1991) Aid and Power: The World Bank and Policy-based Lending, Vol. 1, Analysis and Policy Proposals, New York: Routledge. Murphy, Josette L. (1995) Gender Issues in World Bank Lending, Washington, DC: World Bank. —— (1997) Mainstreaming Gender in World Bank Lending: An Update, Washington, DC: World Bank. Nelson, Joan (1989) “The Politics of Pro-poor Adjustment,” in J. Nelson (ed.), Fragile Coalitions: The Politics of Economic Adjustment, Washington, DC: Overseas Development Council: 95–113.

160 Cynthia A. Wood O’Brien, Robert, Anne Marie Goetz, Jan Aart Scholte and Marc Williams (2000) Contesting Global Governance: Multilateral Economic Institutions and Global Social Movements, Cambridge: Cambridge University Press. Oxfam Policy Department (1995) A Case for Reform: Fifty Years of the IMF and the World Bank, Oxford: Oxfam Publications. Palmer, Ingrid (1991) Gender and Population in the Adjustment of African Economies: Planning for Change, Geneva: International Labour Ofﬁce. Razavi, Shahra and Carol Miller (1995) Gender Mainstreaming: A Study of Efforts by the UNDP, the World Bank and the ILO to Institutionalize Gender Issues, Geneva: United Nations Research Institute for Social Development. Rocha, Lola, Eduardo Bustelo, Ernesto López and Luis Zuñiga (1989) “Women, Economic Crisis and Adjustment Policies: An Interpretation and Initial Assessment,” in UNICEF, The Invisible Adjustment: Poor Women and the Economic Crisis, Santiago: UNICEF, The Americas and the Caribbean Regional Ofﬁce: 9–27. SAPRIN (Structural Adjustment Participatory Review International Network) (2001) “The Policy Roots of Economic Crisis and Poverty: A Multi-Country Participatory Assessment of Structural Adjustment,” Washington, DC: SAPRIN, retrieved January 11, 2002 (http://www.saprin.org/SAPRIN_Synthesis_11-16-01.pdf ). Sparr, Pamela (ed.) (1994) Mortgaging Women’s Lives: Feminist Critiques of Structural Adjustment, London: Zed Press for the United Nations. Stewart, Frances (1991) “The Many Faces of Adjustment,” World Development, 19(12): 1847–64. Stiglitz, Joseph (1998) “More Instruments and Broader Goals: Moving Toward the PostWashington Consensus,” retrieved August 13, 1999 (http://www.wider.unu.edu/ stiglitx.htm). Summers, Lawrence H. (1994) Investing in All the People: Educating Women in Developing Countries, EDI Seminar Paper No. 45, Washington, DC: World Bank. Summers, Lawrence H. and Lant H. Pritchett (1993) “The Structural-Adjustment Debate,” The American Economic Review, 83(2): 383–9. UNICEF (1989) The Invisible Adjustment: Poor Women and the Economic Crisis, Santiago: UNICEF, The Americas and the Caribbean Regional Ofﬁce. Vickers, Jeanne (1991) Women and the World Economic Crisis, London: Zed Books. Vilas, Carlos M. (1996) “Neoliberal Social Policy: Managing Poverty (Somehow),” NACLA Report on the Americas, 29(6), May/June: 16–25. Waylen, Georgina (1992) “Women, Authoritarianism and Market Liberalisation in Chile, 1973–1989,” in H. Afshar and C. Dennis (eds), Women and Adjustment Policies in the Third World, London: Macmillan: 150–78. Williamson, John (1996) “Lowest Common Denominator or Neoliberal Manifesto? The Polemics of the Washington Consensus,” in R. M. Auty and J. Toye (eds), Challenging the Orthodoxies, New York: St. Martin’s Press: 13–22. Wolfensohn, James D. (1995) “Women and the Transformation of the 21st Century,” Address to the Fourth UN Conference on Women, retrieved July 8, 1999 (http:// www.worldbank.org/gender/how/womenand.htm). Women’s Eyes on the World Bank – US (1997) “Gender Equity and the World Bank Group: A Post-Beijing Assessment,” Washington, DC: Oxfam America. Wood, Cynthia A. (1997) “The First World/Third Party Criterion: A Feminist Critique of the Production Boundary in Economics,” Feminist Economics, 3(3): 47–68. World Bank (1985) The World Bank Annual Report, Washington, DC: World Bank.

Adjustment with a woman’s face 161 —— (1987) Protecting the Poor During Periods of Adjustment, Washington, DC: World Bank. —— (1990) World Development Report 1990, Washington, DC: World Bank. —— (1991) World Development Report 1991, Washington, DC: World Bank. —— (1994a) Adjustment in Africa: Reforms, Results, and the Road Ahead, Washington, DC: World Bank. —— (1994b) Enhancing Women’s Participation in Economic Development: A World Bank Policy Paper, New York: Oxford University Press. —— (1994c) “The Gender Dimension of Development,” World Bank Operational Policy 4.20, retrieved July 8, 1999 (http://www.worldbank.org/gender/how/ termr2.htm). —— (1995a) Advancing Gender Equality: From Concept to Action, Washington, DC: World Bank. —— (1995b) Toward Gender Equality: The Role of Public Policy, Washington, DC: World Bank. —— (1996) “Implementing the World Bank’s Gender Policies,” retrieved July 1, 1999 (http://www.worldbank.org/gender/how/report.htm). —— (1998) “Selected Comments on the Concept Note from External Readers,” retrieved July 8, 1999 (http://www.worldbank.org/gender/know/comment.htm). —— (2001a) “Adjustment from Within: Lessons from the Structural Adjustment Participatory Review Initiative,” retrieved January 7, 2002 (http://www.worldbank. org/research/sapri/). —— (2001b) Engendering Development, Washington, DC: World Bank.

11 Gender and intrahousehold decision-making International migration and other frontiers for development policy Aida Orgocka and Gale Summerﬁeld For more than two decades feminist economists have criticized neoclassical concepts of the household and recognized the unequal gender relations in intrahousehold decision-making and allocation of resources (see Woolley 1999; Summerﬁeld 1998; Benería 1995). Negotiation within the household affects the well-being of all its members and inﬂuences the effectiveness of development policies. Engendering Development: Through Gender Equality in Rights, Resources, and Voice, a Policy Research Report (PRR) of the World Bank, includes an entire chapter on these dynamics (2001: 147–80).1 Our chapter addresses whether the PRR presents sufﬁcient material on gender and intrahousehold decision-making for policy-makers to consider in making, implementing, and revising development policies. Our main critique of the PRR is that it does not push ahead to reﬂect cutting-edge approaches to work being done now or look toward the future. It also is rather selective in what it considers even of work from the recent past; it stresses econometric studies rather than including other approaches or the theoretical literature. Our critique is organized into two main sections. The ﬁrst section examines key factors for the analysis of intrahousehold decision-making including functions and structure of households, policy areas that could contribute to a more equitable allocation of resources within the household, the need to include theoretical and conceptual contributions to the study of intrahousehold dynamics, and use of qualitative methods as part of an interdisciplinary analysis of these dynamics. The second section utilizes an example from our research on international migration to illustrate how factors overlooked in the discussion in the PRR are needed to understand some frontier issues in intrahousehold aspects of development policy. In particular, the section examines the context of cross-border development issues that exist in the North (in this case, through immigrants from Muslim countries to the USA) and intergenerational interactions within the household regarding young women’s sexuality education, a part of health policy.

Gender and intrahousehold decision-making 163

Key factors for analysis of intrahousehold decision-making A summary of gender aspects of intrahousehold decision-making needs to include relevant theoretical material as well as empirical studies. It should provide an understanding of the dynamics of households and families and the importance of embedding the studies in particular contexts. Understanding gender aspects of power in the household requires studying not only spouses or adult partners but also relationships between generations, between siblings, and within the same sex (since those also reﬂect and reproduce the biases of society). Taking some material from the work on families in ﬁelds outside economics can build an interdisciplinary analysis that is useful in understanding how development policies affect individuals within families. Functions and structures of households The PRR chapter on intrahousehold power, incentives, and allocation of resources begins with a description of the functions and structures of households around the world. It notes that households are context-speciﬁc and dynamic and that members do not necessarily share responsibilities and resources equitably. Households reproduce gendered relations and outcomes by socializing children into gender-specific roles, including making gender-specific investments in human capital and perpetuating a gendered division of time and tasks. The PRR authors then discuss resource control, bargaining power, and household allocations in detail. They criticize the conceptualization of the household as a unitary model and pinpoint the pitfalls of (1) looking at a household as one in which common preferences are the basic characteristic of household decisionmaking, and (2) concentrating only on the total level of household resource allocations and investments instead of including the distribution of resources within households as well. The lengthy discussion of intrahousehold issues in the chapter is a refreshing change from the traditional, economic/survey deﬁnition of a household as a stable unit over time and space with a cooperative, unitary approach to decision-making. While the PRR devotes some attention to bargaining power and the divergent interests within households, its main theoretical/conceptual discussion is a brief presentation of the inadequacy of the unitary model of the household and some empirical tests of the model’s implications (starting on p. 154). A wealth of material exists that elaborates on alternatives to the unitary model and would add to understanding the gendered distribution of resources within the household. Familiarity with these ideas is especially important for formation and implementation of development policy. Especially notable is the absence of any reference to capability approach concepts, especially Sen’s discussion of the family as a site of cooperation and conflict (Sen 1990, 1999; Nussbaum 2000). These concepts have informed numerous studies dealing with intrahousehold bargaining power in recent years

164 Aida Orgocka and Gale Summerﬁeld (and they implicitly seem to inform the PRR also because of the stress on going beyond the unitary model). The conceptual framework of the family as a unit that embodies both cooperation and conﬂict (as opposed to a completely cooperative family of the unitary model) assumes that the relationships between individuals in a household are based both on common interests and areas of conﬂicting goals (Sen 1990). Agency, status, and bargaining power at home are inﬂuenced by opportunities outside the family and a complex of other factors including the different perceptions of individuals regarding their position in and contribution to the family. In making decisions about resource allocation, individuals have to factor in what the outcomes will do to their relative positions in the household. Bargaining is partly shaped by how much risk individuals are willing and able to undertake in the process. Relationships within the household in addition to the ones between wives and husbands are important in determining household resource allocation. The PRR discusses the relationships between parents and children brieﬂy, but the discussion is incomplete and raises some questions about the implications of other sections (see PRR: 164). Allocation of children’s work is determined not only by traditional gender roles but also by age, birth-order, and sibling composition (Punch 2001; Blanchet 1996; Adagala 1991). Moreover, generational and gender relationships are constantly changing and negotiated against a backdrop of children’s acquiring skills and shaping a sense of autonomy from their family as well as other socioeconomic conditions (Finch and Mason 1993). The PRR misses a discussion on children’s agency in shaping the allocation of resources. They are seen mainly as recipients of beneﬁts coming from the parents (and investments for future support of the parents). Examination of these dynamics would contribute to a deeper understanding of household power distribution and use of resources. The relationships between mothers and daughters provide a good illustration of these dynamics. Studies have shown complex and diverse relations among female family members, from mothers or grandmothers passing down hurtful traditions such as foot-binding in China before it was banned in the 1950s (Papanek 1990) to mothers being more likely to spend their income on assuring the education of daughters as well as sons (as the PRR discusses). In conveying gender roles to their daughters, mothers are usually caught in the double bind of assuring daughters’ economic and social well-being while adhering to patriarchal gender norms. Such contradictions appear in both developing