A WORLD BANK COUNTRY STUDY (*aa 20830 June 2000 Public Disclosure Authorized Ecuador Gender Review Issuesand Recommendations Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized

A WORLD BANK COUNTRY STUDY

Ecuador Gender Review Issuesand Recommendations

The World Bank Washington,D. C. Copyright i 2000 The International Bank for Reconstruction and Development/THE WORLD BANK 1818H Street, N.W. Washington, D.C. 20433,U.S.A.

All rights reserved Manufactured in the United States of America First printing June 2000

World Bank Country Studies are among the many reports originally prepared for internal use as part of the continuing analysis by the Bank of the economic and related conditions of its developing member countries and of its dialogues with the govermnents. Some of the reports are published in this series with the least possible delay for the use of governments and the academic, business and financial, and development communities. The typescript of this paper therefore has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. Some sources cited in this paper may be informal documents that are not readily available. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of the data included in this publication and accepts no responsibility for any consequence of their use. The boundaries, colors, denominations, and other information shown on any map in this volume do not imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. The World Bank encourages dissemination of its work and will normally grant permission promptly. Permission to photocopy items for internal or personal use, for the internal or personal use of specific clients, or for educational classroom use, is granted by the World Bank provided that the appropriate fee is paid directly to Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, U.S.A.,telephone 978-750-8400,fax 978-7504470. Please contact Copyright Clearance Center prior to photocopying items. For permission to reprint individual articles or chapters, please fax your request with complete information to the Republication Department, Copyright Clearance Center, fax 978-7504470. All other queries on rights and licenses should be addressed to the World Bank at the address above, or fax no. 202-522-2422. ISBN:0-8213-4778-0 ISSN: 0253-2123

Library of Congress Cataloging-in-Publication Data has been applied for. CONTENTS

Abstract ...... iv Preface ...... v Acknowledgments...... vi List of Abbreviations and Acronyms...... vii Executive Summary...... viii 1. INTRODUCTION ...... 1

2. CONTEXT ...... 3 MACROECONOMIC SETTING ...... 3 POVERTY ...... 5 INSTITLrTIONALFRAMEWORK ...... 6 3. OVERVIEW OF GENDER ISSUES AND TRENDS .10

DEMOGRAPHICS.10 REPRODUCTIvEHALTH .12 GENERAL HEALTH ...... 16...... 16 VIOLENCE .. 20 EDUCATION .. 23 LABOR...... 29 ECONOMIC CRISIS AND SOCIAL PROTECTION ...... 36 4. THE RURAL SECTOR.40

GENDER ROLESIN AGRICLTURE .40 Sierra.40 Cossta.41 Odente.42 LAND ...... 42 Sierra.43 Costa.44 MIGRATION...... 44 5. SUMMARY AND IMPLICATIONS OF FINDINGS .48

6. PROPOSED OVERALL STRATEGYAND PRIORITY ACTIONS .52

OVERALL STRATEGY .52 PRIORITYACTIONS .53 SECToRAL REcOMMENDAONS. .55 Health.55 Violence.56 Education.S7 Labor.58 Rural Dewlopment.t59 EconomicFluctuations and Social SaJey Nets.60 PolticalD.adenb.. and Pa.ia.ation.61...... ,6 Institutionalne...... 561

References.62 Pofial1-aerbip n P rrdb d n ...... 6 Appendix 1: Matrix of Gender Issues by Project.68 Appendx 2: PreliminaryLessons from t.e National Rural Development Project (PRONADER). 76 S... - ABSTRACT

This paper presents findings of a review of gender trends and issues in Ecuador in demographics, health, violence, education, labor markets and social protection. It also briefly describes the current macroeconomic context and poverty situation as well as recent gender-related institutional and legal advances and includes a more comprehensive section on gender issues in rural development.

The report finds that Ecuador has witnessed important advances in reducing gender disparities and addressing gender-related development issues. Fertility has decreased, education inequalities have diminished, disparities in labor force participation have been reduced and advances have been made in the area of domestic violence. But issues related to gender remain. Reproductive health care remains a problem for women. As for men, accidents and violence are by far the most serious health risks among adolescent males and alcoholism and substance abuse are a serious social concern. In education, as a result of gender socialization, girls continue to follow different academic paths than boys, which affects their opportunities in the job market, their earning potential and over time, their well being. For boys, poverty has led them to enter the labor market at an early age thereby affecting their education prospects and human capital acquisition.

As a long term strategy, gender policies and programs in Ecuador should address socialization processes that reinforce traditional gender roles. Short term measures, however, need to focus on the acute and immediate needs of the population. And this short term and long term gender work should also not be limited to programs targeting girls and women only but rather should also focus on boys and men. The report identifies six priority action areas. First, both female and male gender issues need to be considered when designing and implementing social safety nets and emergency assistance programs. Second, indigenous people's initiatives should be strengthened so that they address gender differences and disparities. Third, issues such as maternal mortality, AIDS and STDs, teenage pregnancy, and family planning need to be addressed in an integrated fashion under the rubric of reproductive and sexual heaith. Fourth, initiatives need to be developed to address at-risk youth. Fifth, the institutional framework on gender at the national level needs to be defined and strengthened. And sixth, an evaluation of ongoing work on gender should be carried out to examine lessons from past gender interventions.

iv PREFACE

Over the years, Ecuador has made considerable strides in addressing gender issues. Gender gaps in educational enrollment and attainment have decreased significantly; access to and use of family planning methods has increased broadly; and, while still inferior to men's, women's labor force participation has grown steadily over the last two decades. In terms of institutional and legal advances, the Government created the National Council for Women (CONAMU) as an autonomous entity under the Presidency, enacted the Law on Violence Against Women and the Family, and began to establish sheriff's offices for female victims of family violence. Ecuador also has a strong civil society movement, which has been largely responsible for putting gender on the public agenda.

But gender continues to be an important development issue in Ecuador. Access to contraceptives remains limited for the poor; thus decreases in fertility rates have not been uniform and teenage pregnancy is highly prevalent. Maternal mortality also remains high despite efforts to improve health care services. And gender roles have also been changing thereby presenting a new set of challenges. For example, single mothers now head nineteen percent of families. In terms of the labor market, gender-based wage gaps persist even after taking into account as many variables related to education and skills as possible. As for the rural sector, even as men have migrated, development programs and policies have been slow to recognize the importance of women in the agriculture sector. Similarly, land distribution continues to be biased towards men, particularly in the Costa region.

And as the report indicates, gender issues are not limited to women. Among men, accidents and violence are by far the most serious health risks among adolescent males. Alcoholism is also an important social concern in Ecuador. For boys, poverty leads them to enter the labor market at a very early age, thereby affecting their educational opportunities. And the current economic crisis has exacerbated boys' school dropout, particularly in the coastal areas. In the reproductive sphere, health programs typically exclude men and fail to consider their key family planning roles. Indeed, men's issues in Ecuador - as gender issues in and of themselves - have yet to receive the attention they deserve given their important social consequences.

This report aims to bring to light the most salient gender issues affecting Ecuador's social and economic development today. It is the second in a series of country gender reviews having being carried out in the Latin America and the Caribbean Region over the last few years with this objective in mind. It is important to note that the report represents a rapid review of issues across key sectors and is in no way comprehensive nor does it attempt to capture the richness of all the gender work carried out in Ecuador over the years by capable practitioners and researchers. It does, however, provide a picture of gender issues of the day, as a basis for discussion, debate and action. It can also serve as a tool for strengthening current development interventions and or designing future policies, programs and services with a gender focus.

v ACKNOWLEDGMENTS

This report was written by Maria Correia (LCSPG) and Bernice Van Bronkhorst (consultant) based on a background paper written by Elizabeth Katz (consultant) and inputs from Olympia Icochea, Deborah Leyh, Wendy Cunningham and Elizabeth Shrader (LCSPG), and Patricio Marquez (LCSHD). Janet Entwisde, Pilar Larreamendy, Anita Villaquiran and Maria Arbolera from the Ecuador Resident Mission, Eduardo Wallentin (LCC4C), and Selpha Nyairo and Geoffrey Chambers (LCSPG) also contributed to the production of this report. Peer reviewers were Carolyn Winter (HDHED) and Lucia Fort (PRMGE). In Ecuador, numerous government agencies, civil society organizations, academics and international agencies added important insights and ideas. In particular, Lolita Villaquiran and all the staff from the ConsejoNational de la Mujer in Ecuador were instrumental in facilitating and assisting to carry out the review. In the World Bank, the Vice President is David de Ferranti, the Country Director is Andres Solimano, the Sector Director is Guillermo Perry, the Lead Economist is David Yuravlivker and the Sector Manager is Maria Correia.

vi LIST OF ABBREVIATIONS AND ACRONYMS

AIDS Acquired Immune DeficiencySyndrome CDC Center for Disease Control CENAISE National Center of Socialand Educative Research CentroNacional de Investigationes Socialesy Educativas CEPAR Center for Studies on Population and Social Development Centrode Estudios de Poblacidny Desarrollo Sodal CEPLAES Center of Planning and Social Studies Centrode Planificaci'ny Estudios Soeiales CONAMU National Council for Women ConsejoNacional de la Mujer DINAMU National Directory of Women DireccidnNational dela Mujer' EAP EconomicallyActive Population PoblaiodnEcondmicamente Activa ECD Early Childhood Development ENDEMAIN Demographic and Maternal and Child Health Survey EncuestaDemo,grdficay deSalud Materna e Infantil FLACSO Latin American Faculty of SocialSciences FacultadLatinoamericana deCiencias Sociales HIV Human ImmunodeficiencyVirus 1DB Inter-American Development Bank IECAIM Ecuadorian Institute of Training and Research on Women InstitutoEcuatoriano de Investigadonesy Capaitadidn de la Mujer IICA Inter-American Institute of Cooperation for Agriculture InstitutoInteramericano para la CooperacidnenAgricultura ILO International Labor Organization INIAP National Institute for AgriculturalResearch InstitutoNadonal de InvestigadoinAgropecuaria LAC Latin America and the Caribbean Region LSMS Living Standards Measurement Survey Encuestade Condidiones de Vida MSW Ministry of SocialWelfare MOH Ministry of Health NGO Non-Governmental Organization PAHO Pan-AmericanHealth Organization Or,ganiZatidnPanamericana de Salud PRONADER National Rural Development Project SIMUJER Situation of Women and Gender Inequality Indicators IndicadoresSobre la Situadidnde las Mujeresy las Desigualdades de Ge'nero SECAP Ecuadorian Service of Professional Training ServidoEcuatorianopara la CapadtacidondeProfesionales STDs SexuallyTransmitted Diseases UNDP United Nations Development Programme UNFPA United National Fund for Population Activities UNICEF United Nations Children's Fund USAID United States Aid Agency

I The predecessor of CONAMU, which was located in the Ministryof SocialWelfare.

vii EXECUTIVE SUMMARY

INTRODUCTION.

This paper presents findings of a review of gender trends and issues in Ecuador in demographics, health, violence, education, labor markets and social protection. To put the report in context, it also briefly describes the current macroeconomic context and poverty situation as well as recent gender-related institutional and legal advances. Given that almost half of Ecuador's population is rural-based the report includes a more comprehensive section on gender issues in rural development.

The purpose of the review was to identify gender issues across sectors with a view to reducing gender inequalities in wellbeing and improving the effectiveness of Ecuador's social and economic development programs. Gender in this report pertains to both men and women and refers to the different experiences, preferences, needs, opportunities and constraints men and women face because of socially ascribed gender roles and expectations. Thus where possible, and considering information deficiencies on men, the report discusses both male and female gender issues. The report also disaggregates by ethnicity when data permits.

The report draws from a desk review of a large number of published and unpublished materials; official statistical information; interviews with governmental officials, academics, representatives of non-governmental organizations and women's organizations, in-country counterparts of World Bank projects and World Bank staff; a review of the World Bank's portfolio for Ecuador; and field visits to sites of various World Bank-financed projects in Ecuador.

FINDINGS.

Ecuador has witnessed important advances in terms of reducing gender disparities and addressing gender-related development issues. For example:

Access to and use of family planning methods has increased broadly resulting in a drop in the fertility rate from 5.0 in the 1980s to the present rate of 3.6.

Ecuador has a strong civil society movement, and women's non-governmental organizations in particular have been very successful in putting women's issues on the agenda. The existence of the autonomous ConseioNational de la Mujer (CONAMU is due largely to the efforts of civil society organizations.

Gender gaps in educational enrollment and attainment have decreased significantly. The latest figures indicate that women on average have about 7.0 years of education compared to 7.4 years for men. However, boys have higher drop-out and repetition rates than girls.

viii Women's labor force participation has steadilyincreased over the last two decades, whereas the proportion of men in the workforce has declined. But men still predominate in the work force. In 1998, 46 percent of urban women wete economicallyactive, compared to 72 percent of men. In the rural sector, women have increasinglyimportant roles in both subsistenceand commercial agricultural production.

' There has been progress on the institutional and legal front with regard to gender. For example, the Government has established 19 comisariasde la mujer(police stations for women), enacted a law against familyviolence, and created an autonomous agency dedicated to women's affairs (CONAMU). Relative to its counterpart agencies in other countries, CONAMU is a strong performer. It has been successful in buildingimportant alliancesin the public sector, which have resulted in innovative gender initiatives. Also, relative to other countries, gender appears to be much more mainstreamed in public programs.

While these advances are indeed noteworthy, gender continues to be an important development issue in Ecuador. For example:

' Whilst having decreased substantially,reductions in fertility rates have not been uniform. Women with little education, for example, continue to have very high fertilityrates (6.2 children on average compared to 2.1 for women with higher education). Rural women - particularlythose in the Oriente - also have higher fertilitylevels.

-- Access to contraceptivesis a problem, particularlyamong the poor. According to ENDEMAIN-94, 70 percent of low-educatedwomen surveyeddesire fewer children than they had. The recent withdrawal of donors of familyplanning programs has raised serious concerns about the public health care system's capacity to provide reliable and accessiblecontraceptives.

' Pre-natal care is inadequate and the proportion of pregnancies that are at-risk remains very high, as do maternal mortality rates. In Ecuador, 25 percent of women do not receive prenatal care during pregnancy. The maternal mortalityrate was 159 per 100,000 live births from 1992-97.

' Reproductive health programs tend to exclude men. As a result, contraceptive use amongst men is minimalin Ecuador, which puts them and their partners at a high risk of contracting sexuallytransmitted diseases(STDs) and acquired immune deficiencysyndrome (AIDS). Only 3.0 percent of men use condoms and 0.3 percent are sterilized. AIDS is on the rise in Ecuador, but the Ministryof Health has yet to develop a coherent strategy to address the problem.

' Accidents and violence (male on male violence) are by far the most serious health risks among adolescent males and these are on the rise. From 1994-96, for example, homicide rose from the fourth main cause of death among men to the second main cause. Male violence against women and violence against children of both sexes also remains high - between 42 and 60 percent of women are victims of household

ix physical violence. Alcoholism and substance abuse are serious social concerns in Ecuador and have been linked to violence. Yet research on this issue is almost non existent.

In terms of education, girls and boys are being treated differently in the school system. As a result of gender socialization - both inside and outside the school system - girls continue to follow different academic paths than boys, which affects their opportunities in the job market, their earning potential and over time, their well being (and that of their families). For boys, poverty has led them to enter the labor market at an early age thereby affecting their education prospects and human capital acquisition. About 44 percent of boys who drop out of school claim work as the cause. In terms of early education, coverage of early childhood care and development programs is limited and quality remains a concern. Because of gender roles, lack of childcare affects poor women in particular and their ability to enter into and advance in the labor market.

Gender-based wage gaps persist even after taking into account as many variables related to education and skills as possible. Whereas previously women were earning more than men in industries typically considered female (such as food, textiles, clothes production, medical and domestic service industries), by 1997 these trends had turned in favor of men.

Rural development programs have been slow to recognize the increasing 'feminization' of agriculture. Land distribution - particularly in the Costa - has been skewed towards men. Moreover, population pressure and land scarcity are undermining equitable land inheritance customs in the Sierra.

Macro-econornic crises have gendered impacts, with women tending to enter into the labor force during these periods. This implies greater pressures on their time given that adjustments in the division of household work have not taken place. For men, unemployment threatens their role of family provider and creates problems of self esteem and depression - which may have other possible negative effects such as violence. In terms of health, women are more vulnerable than men during these periods because of their maternal roles. Preschool children (boys and girls) are affected by nutritional deficiencies, but boys from poor families are more likely to drop out of school to enter the work force (particularly in the Costa).

Gender is a factor in household coping strategies and capacity to confront crises. For example, based on LSMS-95, the proportion of women with a secure job is greater than that of male workers. However, it may be more difficult for women to secure new income sources during economic downturns. A greater number of women are illiterate, for example, which affects their employment opportunities. Also, women are less mobile than men because of domestic responsibilities. Rural households maintained by a single woman may be more vulnerable than (male) conjugally-headed households because they have less assets. For example, 44 percent of female single-headed households do not have land title, compared to 28 percent of male conjugally-headed households.

x Despite some modest advances, political participation and representation continues to be heavily skewed towards men.

+ Discussion of gender equality within the larger struggle of indigenous peoples' rights has largely been absent.

INIPLICATIONS OF FINDINGS.

The aforementioned gender issues have efficiency, equality, and social welfare implications. For example, boys' early entry into the labor force and the need for girls to assume domestic work in the house affects their educational attainment, human capital acquisition, their earning potential in the labor market, and ultimately their social well being. Lack of attention to the particular needs of boys and girls affects the success and effectiveness of programs designed to keep them in school.

In addition to affecting well being, poor health reduces economic productivity and is taxing on the country's health care system. Providing women with low cost pre-natal care at the community level, for example, could reduce health costs due to comphcations in child birth. Family planning programs are potentially less effective because they only target women and fail to consider men, who also have a role in family planning decisions. Reproductive health programs directed at men could also help to quell the spread of STDs and AIDS. High levels of violence not only affect well being, but are costly to the health care and criminal justice systems, reduce work productivity, and deter economic investment. Violence also hampers the accumulation of human capital. Alcoholism, which has been linked to violence, is a drain on household resources and reduces worker productivity.

In the rural sector, agriculture policies and programs that continue to treat men as the principal beneficiaries fail to optimize agricultural production and earnings. And in the case of land, gender may influence the effectiveness of land programs given that property rights tend to affect men's and women's control of income, their bargaining power within the household and corollary household expenditure patterns. Moreover, men and women may have different incentives and interests in conservation and resource use, thereby possibly affecting the outcome and sustainability of environmental programs. Also for equity reasons, both women and men are entitled to land because it increases their status, credit worthiness, well being and security in old age.

PROPOSED STRATEGY AND PRIORITIES.

As a long term strategy, gender policies and programs in Ecuador should address socialization processes that reinforce traditional gender roles. These gender roles, stereotypes and expectations can influence the opportunities men and women have and the constraints they face as well as their choices and decisions over their lifetime. Ultimately, rigid gender roles can negatively affect social and economic well being. The effects of gender roles start early. Among the poor, boys are often forced to abandon their studies so that they can enter the work force and contribute to family income - consistent with the male role of 'family provider'. In line with their maternal and caregiving roles, on the other hand, girls are expected to work in the home, where they are less likely to gain human capital and which, in turn, influences their labor market opportunities and earning potential later in

xi life. The effects of gender roles extend into old age. Women's reproductive roles lead them to participate less in the labor matket, to have more transient employment relationships, and to work in the informal sector in larger proportions than men. Hence, they are also less likely to receive social security or to have savings in their old age, and are more likely to be poor.

But gender work should not be limited to programs targeting girls and women, as has been the tendency in the past. Experience from Ecuador demonstrates that segregated women's programs have been ineffective because they fail to consider the views of men and dynamics between the genders. Moreover, men also have issues related to gender roles and masculinity, such as alcoholism and violence. The Ecuadorian Government has already made important strides by supporting, for example, programs to counter gender stereotypes in education. Over the medium and long term, these types of programs should be strengthened. But efforts should also be made to address gender socialization that takes place through other channels such as the family, the media and peer groups. Programs directed at children and young adults are of particular importance, given that their attitudes and behavior patterns are not fully developed and they are still in a position to accumulate valuable human capital.

Short term measures, however, need to focus on the acute and immediate needs of the population, given the current economic and financial crisis and the corollary deterioration in the standard of living of so many Ecuadorians. In terms of gender, the Government's social safety net and recovery programs should continue to consider gender- differentiated needs and impacts. But even these programs - which have tended to target women because of their role as mothers - could be strengthened so that they consider gender socialization and promote men's roles as fathers and caregivers, particularly among unemployed men. Broadening male gender roles could benefit men as well as women and their families, given that substance abuse, violence and depression among men have been linked to gender roles and the limited ways men have to affirm their identity. Programs to promote men as fathers are still very new in the Region and elsewhere but models exist and could be piloted in Ecuador. Similarly, any temporary work program being considered for Ecuador - including social infrastructure projects that generate employment - should include targeting mechanisms to reach women as well as men.

The report identifies six priority action areas. First, both female and male gender issues need to be considered when designing and implementing social safety nets and emergency assistance programs, so that, inter alia, programs strengthen the role of fathers and provide income generating opportunities for women. Second, indigenous people's initiatives should be strengthened so that they address gender differences and disparities. Measures being taken by the World Bank financed PRODEPINE Project to review country- wide participatory assessments from a gender perspective are an example of what can be done. Third, issues such as maternal mortality, AIDS and STDs, teenage pregnancy, and family planning need to be addressed in an integrated fashion under the rubric of reproductive and sexual health. Fourth, initiatives need to be developed to address at-risk youth following on the experiences of other countries in Latin America and the Caribbean. These have introduced, for example, 'life skills' programs into school systems, which include components on self-awareness/knowledge, anger-management, conflict resolution, family planning, and parenting. Fifth, the institutional framework on gender at the national level

xii needs to be defined and strengthened. And sixth, an evaluation of ongoing work on gender should be carried out to examine lessons from past gender interventions, to identify how sectoral recommendations can be incorporated into existing development programs and to learn how issues identified in this report might shape future investments.

SECTORALRECOMMENDATIONS.

Health. The report makes three recommendations: First, health issues such as maternal mortality, AIDS and STDs, teenage pregnancy, and family planning should be dealt with in an integrated way under the broader area of reproductive and sexual health. Current public health programs are narrowly centered on maternal and child health with other issues getting much less or no attention. In the short term, the Government urgently needs to put in place systems to make contraceptives available to women and men - particularly the poor - given the link between poverty and family size, the demand for family planning voiced by poor women and the imminent decline in family planning donor funding. Also, reproductive health services need to target men given the low incidence of condom use and the cost effectiveness of male sterilization. Given budget pressures, community organizations and NGOs could be mobilized to provide reproductive and sexual health services for men and women. Second, alcoholism needs to be placed firmly on the public agenda given the pervasiveness of the problem and its gendered nature, its effects on productivity, and the link between alcoholism to violence. A first step would be to identify the roles of different actors in contributing to and mitigating alcoholism. And third, as a long term objective, health care providers need to better capture and analyze epidemiological data by gender in order to understand the different diseases and health risks faced by the male and female population and improve disease control and prevention.

Violence. Actions related to violence - all forms of violence, not just domestic violence - should focus on prevention. Schools are an important vehicle for addressing and reducing violent behavior given that they influence the cultural values that promote violent behavior. The report recommends a number of actions including: retraining teachers to ensure they do not promote violent behavior among boys and submission among girls; eliminating gender stereotypes in textbooks and other pedagogical materials; and developing innovative programs to teach children nonviolent resolution skills and to promote civic values. Community level programs can be used to provide informal education programs, teach citizens about legal sanctions against violence, establish violence prevention strategies, and ptovide social services for victims of violence. And lastly, media can be used to promote nonviolence. Mass media education campaigns represent a potentially effective response by providing nonviolent role models. Educational programs in the media can produce and disseminate more positive images of interpersonal interactions (male-female and adult-child). Lastly, over the long term the current famnilyviolence program that establishes comisariasde la mujer should be strengthened by, interalia, including preventative programs for male aggressors, particularly young men.

Education. Initiatives designed to increase school attendance and enrollment among the poor during the crisis need to consider how gender roles affect the ability of girls and boys to go to school (the need to work influencing boys and domestic work influencing girls). Also, to the extent possible, initiatives in the short term aimed at improving the

xiii quality of instruction and providing textbooks to poor schools should avoid transmitting gender stereotypes. Over the long term, education policies should focus on: ensuring that gains achieved over the decades in gender equity reach all sections of the population, including indigenous groups; continuing to address the issue of gender stereotyping in schooling; and maintaining a gender focus in school drop out and repetition initiatives. In addition, given the serious concern with gendered issues such as violence, alcohol and substance abuse, the spread of STDs, and teenage pregnancy in Ecuador, strengthening school-based programs to address these kinds of issues is important over the medium and long term. And lastly, given that early childhood education is a long term investment in the country's human capital, these programs should be a priority.

Labor. Ecuador needs to address existing gender and labor issues such as higher female unemployment and a persistent wage gap. The report makes three recommendations: first, Ecuador should examine the differentiated impact of labor legislation and regulation/deregulation on men's and women's employment patterns and opportunities. In addition to the reported wage gaps, consultations in Ecuador suggest that gender discrimination exists in job advancement, coverage of benefits, and job security, such as the firing of women because of pregnancy. A regulated or non regulated informal labor market may also affect men and women differently. Second, support to entrepreneurs and the self employed should take gender into account, and be tailored to the different circumstances in which self-employed men and women work. And third, community organizations and other civil society groups that are well known in the community should act as a clearinghouse for jobs as well as provide training on grooming and preparing oneself for a job. Women would likely benefit more from these services because household responsibilities constrain them from leaving their homes to search for jobs, women have less experience presenting themselves to potential employers, and women have less access to job related information.

Rural Development. The report makes recommendations in four areas. First, given the important gender division of labor in the rural sector as well as the feminization of agriculture in some areas, agricultural research, technical assistance and training services need to ensure that both men's and women's needs are considered, as well as their constraints to participating in and benefiting from rural development initiatives. Extracting lessons from rural development projects in Ecuador would be an important step given that some of these projects have made important efforts on gender already. Second, gender equality in land titling should be promoted and monitored by, inter alia, requiring that both partners be listed on applications for land, and requiring INDA to generate statistics that disaggregate the title holder by gender as well as the type of title (joint, individual or communal). Third, rural development policies should consider gender dynamics in migration. Evidence from the Sierra, for example, suggests that agricultural intensification and the improvement of off- farm employment would act to deter female migration, while improving access to land would keep more men from leaving. And fouth, non-farm employment and enterprises should be supported, paying particular attention to the differentiated sectors in which men and women operate, as well as their respective productivity and growth constraints.

Social Protection. Given that new types of income transfers are currently being considered for Ecuador, it would be important to design these with gender considerations in nind. Emergency assistance and social protection programs can have different impacts

xiv based on who receives the income transfer or service. Men and women also have different needs, priorities and demands in crisis situations. Current programs such as the BonoSo;idario have primarily targeted women because of their immediate maternal needs and caregiving roles and because they are considered to be more responsible with household resources than men. To the extent possible, however, these programs should also attempt to involve men as fathers and caregivers. Men are often underemployed or off work during economic downturns and therefore could share the burden of household responsibilities. In contrast, women often enter the workforce to compensate for household income losses during periods of economic crisis and have less time to engage in domestic chores. Furthermore, any program designed to provide the poor with temporary employment needs to address gender, as women may be excluded because they are not perceived as the main family breadearner. Supply and/or demand factors that contribute to these types of outcomes would need to be identified and taken into consideration. In terms of future research, data collection and analysis needs to better reflect the heterogeneity of household structures, their decisions, dynamics, coping strategies and responses to incentives, by composition of the household and roles of its members.

Political Leadership and Participation. Both government and civil society should make efforts to promote women's participation in public life. Indigenous people's development projects, such as the World Bank financed PRODEPINE, provide a logical entry point for promoting women's leadership, empowerment and participation in political and economic life. Decentralization and municipal development programs also offer opportunities to enhance gender equitable leadership, participation and inclusion. However, these efforts should consider gender relations with a view to avoiding possible resistance among men and spouses, who face cultural and social pressures to dominate in the public sphere.

Institutional Framework. Given pressures to reduce public expenditures and modernize public services, the report recommends carrying out a diagnostic with a view to informing the Government on the most efficient and effective institutional arrangement for addressing gender related issues. The proposed analysis would involve: identifying the institutional implications of existing gender inequalities on public programs and policies as well as the status of gender mainstreaming in public agencies; exarninng the role and effectiveness of CONAMU in facilitating the mainstreaming process and addressing gender inequalities; and identifying corollary institutional options given resource constraints.

xv

"Son 'mandarinas' los hombres que dejan que las mujeres participen." 2

1. INTRODUCTION

This paper reviews gender trends and issues in Ecuador in demographics, health, violence, education, labor markets and social protection. The report summarizes issues in these sectors and is descriptive rather than analytical, relying almost solely on secondary sources. The paper also reports on institutional and legal advances made by government and civil society in addressing gender issues in Ecuador and describes the current macroeconomic context and poverty situation.

Gender issues in Ecuador must be considered in the context of other variables such as race, class, region, and level of urbanization. Thus to the extent possible, the report attempts to take these variables into account. Where possible, it also considers other factors such as marital status and household structure with a view to understanding differences in the diverse conditions of men and women in Ecuador. The report includes a more in-depth section on gender issues in the rural sector, given that more than 40 percent of Ecuador's population is rural-based (World Bank, 1997b) and the poor are concentrated in rural areas. To the extent possible it also disaggregates the gender analysis by geographic region (Sierra, Costa and Oriente).3

OBIECTIVES. The objective of the review was to identify gender issues across sectors with a view to reducing gender inequities and inequalities - and as a corollary enhancing wellbeing - and improving the effectiveness of Ecuador's social and economic development programs. Gender in this report pertains to both men and women and refers to the different experiences, preferences, needs, opportunities and constraints men and women face because of their socially ascribed gender roles and expectations (see Box 1). The report discusses both male and female gender issues on the premise that both men and women are affected by socially dictated gender roles in negative ways. But because it relies primarily on secondary sources and previous gender work that has largely focused on women, information on male issues is limited.

DATA SOURCES. The report draws from: (a) a desk review of a large number of published and unpublished materials; (b) official statistical information - including Censuses, Urban Employment Surveys, Family Planning Surveys, and statistical information such as the SIMUJER (IndicadoresSobre la Situacidnde las Mujeresy las Desigualdadesde Genero)database -

receive 2 Literaly men are "mandarines" or softies, which is unmanly. The comment explains the sarcasm and ridiculemen from other men when they aDowor encourage their wives or partners to participate in development activitiesthat are typicallyconsidered male (Marlene Barba, FASBASEProject, personal communication). (Lanjouw,1998). 3 While 40 percent of the population resides in rural areas, about 55 percent of the poor are rural-based This is confirmed in the World Bank's poverty study for Ecuador (WorldBank, 1996).

1 as well as the Living Standards Measurement Survey (LSMS) for 1995 and 1998; (c) interviews with the ConsejoNatonal de la Mujer (CONAMU) and other governmental officials, academics, representatives of non-governmental organizations (NGOs) and women's organizations; (d) a review of the World Bank's portfolio for Ecuador;4 Box 1: The Scope of Gender and (e) interviews with World Bank staff and in-country counterparts and While sex refers to the biological state of being male or beneficiaries of Bank projects. female, gender refers to the socially defined aspects of being a man or a woman. Gender is a variable - like class, ethnicity, nationality, and religion - used to 'identify and ORGANIZATION. The report is measure differences in the roles of men and women, the organized in six sections. The first activities they undertake, the responsibilities they have, the section sets the stage of the report by opportunities and constraints they face, and ultimately, their well-being. Gender is dynamic because men's and women's describing macroeconomic events that roles - as dictated by changing social and cultural norms took place in Ecuador at the time the and values - also vary over time'(Moser, Tornqvist, van report was written and recent gender- Bronkhorst, 1998). related legal and institutional advances. Thus, gender in this report refers to aspects of social The second provides an overview of life and poverty which are experienced differently by men gender issues and trends in and women because they have different ascribed roles. demographics, reproductive health and These result in: other health indicators, violence, E men and women manifesting different needs, education, labor markets and social preferences, interests and priorities; 0 inequalities and inequities based on whether one is protection. A third section presents an male or female; analysis of gender issues in the rural o men and women facing different opportunities, sector. The fourth section of the constraints, and challenges; and report providesconclusionsreport providesoliclusions andcontributingand E men and inwomen different differently ways to, affected social and by, economic and implications of the main findings. A development - through their actions and participation fifth section recommends priority (Correia, 1998). actions and strategies related to gender. And finally, the sixth section makes specific sectoral recommendations.

4 A preliminary cursory review of the portfoho resulted in the identification of those projects with likely gender differentiated impacts and/or those projects relevant in terms of addressing gender disparities and inequalities. No other selection criteria were used.

2 2. CONTEXT

MACRO ECONOMIC SETTING

Any issue examined in Ecuador today must take into account the country's current critical macroeconomic and financial situation. At the time this report was written in July 1999, Ecuador was undergoing an unprecedented macroeconomic and financialcrisis, encompassing swollen fiscal and external-accountsdeficits, intensified exchange-rateand price-levelinstability, a systemic banking crisis,and deepening recession. Real gross domestic product (GDP) was 4.1 percent lower in the first quarter of 1999 compared to the same period of 1998 and was expected to decline as much as seven percent in 1999. The immediate cause of the crisis was a confluence of external and climatic shocks beginning in late 1997. These included (i) sharply decliningpetroleum export prices; (ii) extensive damage from El Nino rains in the first months of 1998; and (iii)the effects of the crises in East Asia, Russia, and Brazil, including recession in export markets, intensified competition from economies whose exchange rates had depreciated sharply, and virtual closure of international financial markets as they reevaluatedemerging economies' risk. Ecuador was particularly hard hit by the crisis because its economy had performed poorly since the early 1980s and because its structural reform efforts have lagged far behind those of some of its Andean neighbors. Inadequate structural adjustment is a large part of the reason why Ecuador proved so vulnerable to the shocks of 1998.

The macroeconomic and financial effects of the shocks were aggravatedby an unbalanced policy response during 1998. Declining oil revenue, diminished tax revenue as a consequence of El Niiio, and heavy reliefand reconstruction expenditure widened the fiscal deficit, with virtually no offsetting fiscal policy measures to stabilize the situation. The 1998 non-financial public-sector deficit swelledto an unprecedented 5.9 percent of GDP, compared with to approximatelytwo percent in 1997. The monetary authority attempted to carry the burden of stabilization,raising interest rates sharply in an effort to sustain the exchange rate. This was followed by a severe crisis of the banking sector in 1998 due to the effects of the shocks on the banking system's asset quality;reductions in external credit lines to commercialbanks. the effects of restrictive monetary policy;and an exchange-rate depreciation. Undisciplinedbanking practices, including connected lending and abusive use of off-shore banking, as well as ineffectiveregulation, also contributed to the problem. Since December 1998, ten banks and financialinstitutions, including the two largest banks (there were a total of 40 at the end of 1998), have failed.

In February 1999 the Central Bank floated the exchangerate to protect its diminishinginternational-reserve position. The ensuing depreciation,roughly 35 percent over three weeks, led to a burst of high inflation. Consumer prices rose 13.5 percent in

3 March, Ecuador's highest monthly rate ever. The depreciation also worsened the fiscal accounts: not only the public external debt, but the bulk of the US$2b public domestic debt are now dollar-denominated.

The diminishing reserve position raised questions about Ecuador's ability to meet its external payments obligations, particularly Brady-bond payments of about US$1 00m each due in February and August 1999. International markets valued Ecuador's Brady issues at levels that indicated expectations of default. These concerns were eased somewhat as the deepening recession turned Ecuador's external accounts around sharply over the first months of the year. The trade surplus ran at about US$200m over the first quarter of 1999, about five percent of GDP at an annualized rate (compared with approximately --6 percent for 1998). This was even before petroleum export prices began rising and before the bank deposit and credit freeze had their effect on import demand. The trade surplus reached US$190m in April 1999 for the month alone. Nevertheless, it was clear that the essential debt problem was not the balance of payments, but rather the fiscal burden, and the continuing fiscal weakness contributed to growing anxiety regarding Ecuador's capacity to continue servicing its external debt.

Ecuador's crisis has had a prolonged and massive human cost. For example:

(a) The damage caused by El Nino had a variety of devastating consequences for poorer people living in affected coastal areas, including the destruction of homes and agricultural property and unemployment.

(b) Large segments of the population have suffered significant income losses. One of the main reasons is that unemployment has increased sharply in urban areas. According to the Central Bank, production and sales volumes by 200 companies declined sharply in April 1999, the first full month following the banking freeze. Loss of employment has also resulted from companies losing access to bank credit or being overwhelmed by rising debt service. During March and April 1999, 155 firms in Guayaquil and 100 in Quito went out of business.

(c) Central Bank of Ecuador (BCE) data indicate the growing magnitude of the unemployment and under-employment problem in urban areas, based on surveys carried out in Quito, Guayaquil and Cuenca. Combined data for the three cities indicate steady increases in unemployment and under- employment from mid-1998 through May 1999. Unemployment nearly doubled from 8.5 percent of the three cities' economically active work force in May 1998 to 15.6 percent in May 1999. Underemployment rose from 47 to 51 percent over the same period. Guayaquil has suffered the most, with unemployment rising from nine to 18 percent over this period, while Cuenca has suffered relatively less, with unemployment rising from seven to nine percent.

4 (d) Unemployment falls heavily on the poorest people: among the 20 percent poorest people in the population, the unemployment rate in November 1998 was 21.4 percent.

(e) Workers have lost income for reasons other than unemploymnent. One notable example is public-sector workers, whose pay has been running in arrears of up to two to three months. Workers affected include school teachers, health workers, and even police.

() The cost of living has risen in several surges since the onset of the crisis, and incomes have generally lagged considerably. In real terms, the income of a typical worker earning the minimnumwage in the organized sector has declined 10 percent from March 1998 to March 1999.

(g) While social sector spending was fairly stable until 1995, it was predicted to dechine five percent for health and 13 percent for education in real terms from 1998-99.

POVERTY5

According to a recent World Bank social sector study, poverty levels and income inequality worsened in Ecuador prior to the crisis and have since increased. The Gini coefficient for income increased nearly four percentage points, from 0.54 in 1995 to 0.58 in 1998, and in 1998 the lowest decile of the population arranged by income received less than 1 /10thof one percent of total income. Reduced incomes caused by El Nifno, lower prices for agricultural products, increased earning differentials by level of education and, more recently, increased unemploymnent have all contributed to growing inequality.

Based on the headcount of people able to buy a basket of basic goods, poverty increased from 34 percent in 1995 to 46 percent in 1998. While poverty has affected all regions, rural areas and the Costa Region - where El Nifio was most destructive - witnessed the greatest poverty increases. As of 1998, the bottom consumption quintile and the population in extreme poverty were largely synonymnous,with 17 percent of the population below the extreme poverty line. Besides being unable to purchase a basket of basic goods, the poor are also characterized by having a greater ratio of dependents to economically active household members, by living in dwellings without electricity and running water, and by having limited access to social services such as education and health.

For some, the effects of poverty are permanent. For example, one-quarter of all preschool children have stunted growth and the proportion increases to 39 percent for the bottom consumption quintile. Only 12 percent of children in the bottom consumption quintile complete lower secondary school. Children from poor famiihes - particularly boys - are also more likely to drop out of school to work. This is the case for 24 percent of chcilren in the bottom quintile.

5 This section is based on the World Bank's study on the social sector in Ecuador (World Bank, forthcomidng).

5 INSTITUTIONALFRAMEWORK

LEGAL AND CONSTITUTIONAL ADVANCES. Ecuador has made considerable strides in institutional and legal reforms related to gender. Creating CONAMU as an autonomous entity under the Presidency, and enacting the Law on Violence Against Women and the Famnilyare a few noteworthy accomplishments. Other institutions established include: the Comisidnde la Mujer, el Niioy la Familia, under the Congress and the Defensorlade la Mujer Adjunta a la Defensoriadel Pueblounder which the Comisanasde la Mujer (police stations for women) operate. A chronology of the most significant events over the last two decades is provided below:

1980 National Office for Women established in the Ministry of Welfare

1986 National Office for Women becomes the National Directory for Women (DINAMU) in the Ministry of Social Welfare 1994 Women's Police Stations created 1995 Law on Violence Against Women and the Family 1996 Equal Opportunities Plan approved 1997 DINAMU becomes CONAMU under the Presidency of the Republic 1998 Free Maternity Law established 1998 New constitution approved which guarantees equal rights to women and men in areas such as education, health, work and political participation. It also provides a legal framework, which facilitates institutionalizing public policies on gender equity.

POLITICAL PARTICIPATION AND REPRESENTATION. Ecuador was the first Latin American country to grant women the right to vote, through the 1929 constitution. Women have been less likely than men, however, to exercise their voting privileges. Although voting was made compulsory in 1967, less than half of the voters were women in recent elections, even though there are more eligible women than men in the country as a whole.

Although progress has been slow, the gender balance in political participation and representation has been improving in recent years. In particular, the quota of at least 25 percent of women on electoral lists - which was the initiative of the Parliamentary Commission on Women, Children and the Family - has promoted and facilitated a greater gender balance in political participation. Despite this measure, the gender balance in positions of political leadership remains heavily skewed towards men as the following statistics demonstrate:

a) Executive Branch: The first woman minister in Ecuador was appointed in 1979 to head the Ministry of Social Welfare. In 1990, none of the twelve existing ministries was headed by a woman, but women filled five sub-secretarial posts. In August of

6 1998 women held three of the 15 cabinet posts; in February 1999, four out of 15 were women, including the Minister of Finance (Fundaci6n Mujer y Sociedad, 1999, cited in CONAMU/INEC, 1999).6

b) Legislative Branch: In 1998, 20 percent of national congressional representatives and 12 percent of provincial representatives were women (ibid.). This represents an increase from 1996 when all national representatives were men and only six percent of provincial representatives were female (ibid.)

c) Local Government: Figures from 1996 indicate that the following percent of women were elected as representatives at the local level: one out of 21 mayors (Prefecto/a),six percent of provincial council members (Consejeros/asProuinciales), three percent of Alcalde/sa PresidentaConsejo, and eight percent of local councils members (Conseal/laMunicipal) (ibid.).

d) Judiciary: Very few women are represented in the hierarchy of the courts. In 1999, one of 31 supreme court judges (Ministros/as de la CorteSuprema dejusticia) was a woman (CONAMU, personal communication). Their representation at other levels vary. All Fiscal Judges for example are men and 14 percent of Civil Judges are women (Consejo Nacional de la Judicatura, Direcci6n Nacional de Personal, cited in CONAMU/INEC, 1999). The largest propottion of female judges are Jueras/es de Inquilinatoof which 56 percent are female (ibid.).

CONAMU. The first time women's development concerns were addressed by the Ecuadorian government was in 1970 when the Government created the National Office for Women in the then Ministry of Social Provision and Work (Ministeriode PreuisionSodaly Trabajo). In 1980 a women's department was created within the Ministry of Social Welfare, which in 1986 became DINAMU, and later evolved to become CONAMU, an independent council under the Presidency. The main goal of CONAMU is to institutionalize public sector policies for women's development and gender equity and to promote women's participation in development processes. Under its current structure, CONAMU enjoys a certain level of budgetary and administrative autonomy. CONAMU has also been able to supplement its very limited budgetary allocation with financing from the IDB and the Dutch Government.

According to a World Bank cursory review carried out in 1998, CONAMU appears to be a strong performer, not only when contrasted to other women's agencies but also compared to other public sector organizations. Its strengths are as follows:

a) CONAMU is apolitical and has no official link to the first lady of Ecuador. The position of director, for example, does not coincide with the national electoral cycle.

b) Having made technical competence an explicit and priority objective, CONAMU clearly values technical excellence, in contrast to focusing on lobbying or implementing politically motivated token projects for women. A clear example of CONAMU's focus on technical competency is its SIMUJER gender disaggregated data base which compiles gender statistics from numerous sources in the area of

6 The Minister of Finance subsequently resigned from her post in August 1999.

7 demographics, health, education, violence, political representation, etc. The data base is the first of its kind in the Region.

c) The Council has demonstrated the capacity to collaborate with other government agencies in a constructive manner. Its innovative work with the Ministry of Education and Culture on gender stereotypes is a case in point.

d) CONAMU exhibits positive entrepreneurial characteristics and capacity in attracting significant external donor support and resources.7

e) The Council has a strong relationship with civil society: three of CONAMU's seven- member directive council are representatives of NGOs.

Like most of its counterpart agencies in other countries, CONAMU continues to target most of its activities on women. This is understandable given that its main constituency are women's groups, which would like to see CONAMU play an even more proactive role in furthering women's rights. There is an increasing recognition, however, that effective gender programs need to target men as agents of change. There are two main justifications for this changing paradigm on gender. First, through gender relations, men influence women's lives and well being, hence it is insufficient to target women only. And second, men are 'gendered' beings too and as such have their own issues related to gender. For example, male violence can be at least partially attributed to the roles and expectations that society - including women - place on men.

At present CONAMU finds itself in an uncertain situation. Due to budgetary pressures and for reasons of efficiency and effectiveness, the Government is taking steps to modernize it public sector. Among other actions, the Government is reviewing the function and viability of numerous councils that were created during the last administration and placed under the Ministry of the Presidency. CONAMU is among the agencies under threat of elimination. The Government has been considering, for example, absorbing CONAMU into the Ministry of Social Welfare.8

MAINSTREAMING GENDERIN PUBLIC PROGRAMS.9 A cursory review of public sector programs indicates that, relative to other countries, Ecuador is quite advanced in mainstreaming gender into government programs (for example see Annex 1 for a review of how gender issues have been addressed in projects financed by the World Bank). Land and agriculture programs, for example, have a long history of promoting women's participation, which is one element of gender mainstreaming. In the area of justice, gender initiatives have included contracting civil society organizations to manage the recently established women's

7 CONAMU has also been exploring the possibility of establishing a fund with the objective of making their activities sustainable over the long term. 8 The institutional assessment is being financed by the British Embassy in Quito, but the World Bank has been assisting the Government with technical support. 9 AS defined by the Economic and Social Research Council of the United Nations, 'Mainsteaming a gender perspective is the process of assessing the implications for women and men of any planned action, including legislation, policies or programmes in any area and at all levels. It is a strategy for making women's as well as men's concerns and experiences an integral dimension in the design, implementation, monitoring and evaluation of policies and programmes in all political, economic and societal spheres so that women and men benefit equality and inequality is not perpetuated. The ultimate goal is to achieve gender equality'(cited in UNIFEM/CIDA-SEAGEP, 1998:6.

8 police stations.10 For its part, the Ministry of Education and Culture's stereotypes program is novel in the Region. Water and sanitation is yet another sector that has made efforts to identify and address gender issues.

This is not to say that objectives in mainstreaming gender have been met. Many rural development policies and programs, for example, have encountered difficulties in dealing with the constraints and backlash of women's participation efforts, which in some cases has led to domestic violence and resistance among community members. Moreover, despite progress, both government extension workers and rural development NGOs tend to treat men as the principal beneficiaries of technical assistance and information in the agriculture sector and reserve training in health, nutrition, craft production, household gardens and small animal production for women. This continues despite a growing recognition of women's incteasingly important roles in subsistence and commercial agricultural production as well as the daily management of natural resources such as soil, water and fuelwood.

An area of possible concern relates to the treatment of gender and indigenous issues. By its own admission, the Consejode Desarrollode las NacionalidadesyPueblos del Ecuador (CODENPE) - which is the main government agency charged with indigenous policies and programs in Ecuador - does not have a clear definition of gender and its importance in indigenous peoples' development. To a certain degree this oversight is understandable. Public indigenous structures are only now gaining ground. And much effort has been placed on resolving internal inconsistencies and differences between ethnic groups. However, those leading the indigenous movement in Ecuador would be seen as hypocritical if they excluded gender disparities and inequities from their human rights agenda.

THE ROLE OF CIVIL SOCIETY. The civil society movement in general, and women's organizations in particular, have played a central role in putting women's issues firmly on Ecuador's social and political agenda. In recent years, a number of measures to improve the situation of women have been put in place. For example, in 1995 Congress approved a law on violence against women and the family. The creation of DINAMU under the auspices of the Ministry of Social Welfare is another example of the success of the women's movement and the willingness of Ecuador's political establishment to institutionalize public sector policies for women's development and gender equity. In March 1996, the Movimientode las Mujerespresented 'El Plan de Igualdadde Oportunidades1996-2000' which focused on citizenship and political participation and representation, the feminization of poverty, the gender division of labor, labor market issues, and gender issues in health and education. Moreover, it is clear that non-governmental agencies have significant implementation capacity. At the national level, organizations such as the CentroEcuatorianopara la Promocidny Acci6n de la Mujer (CEPAM), the Centrode Estudios e Investigacidnde la Mujer Ecuatoriana (CEIME) and the Centrode Estudios de PoblacidnyDesarrollo Social (CEPAR) have made important contributions on gender in Ecuador. In contrast to other countries in the Region, Ecuador also enjoys constructive and collaborative relations between Government and civil society organizations.

1'Nationwide, 19 women's police stations have been established to date.

9 3. OVERVIEW OF GENDER ISSUES AND TRENDS

DEMOGRAPHICS"

POPULATION. According to the last census carried out in 1990, Ecuador's total population was 9.6 million people, having tripled since the 1950 census. The LSMS-95 estimates the population to be about 11.5 million people (World Bank, 1997b). The female/male ratio continues to indicate a slight majority of women over men at 50.3 versus 49.7 percent, according to the 1990 census. Increasingly, Ecuadorians live in urban areas. According to census figures, the proportion of the urban-based population increased from 29 to 55 percent from 1950 to 1990. While the urban population continues to rise-with the estimated proportion being about 58 percent in 1995-Ecuador remains more rural than most countries in the region, compared for example with almost 80 percent of the population living in urban areas in Brazil. The population is still very young, with just under 40 percent under the age of 14 years old. But a continuing decline in fertility and mortality rates is slowly changing this demographic picture. Life expectancy has increased from 50.8 years for men and 54.2 years for women in 1974, to 62.3 and 66.1 respectively in 1990 (CONAMU/INEC, 1999). The infant mortality rate decreased from 67 per 1000 live births in 1980 to 36 in 1995, whilst the incidence of malnutrition is still very high at 45 percent of children under the age of five (World Bank, 1997 b). Table 1 provides some basic gender indicators for Ecuador.

Table 1: Basic Gender Indicators [ndicatori 00 00_040 Mei Womeni0000 Illiteracy (O/)1 8.3 12.1 Functional Illiteracy (0/)2 19.0 21.8 Matemal Mortalityper 100,0003 - 150 Unemployment Rate (%0)4 8.4 16.0 Earnings Inequality' 100.0 67.5 Landless Household Heads (%)2 27.5 40.8 Women as Mayors (%)5 - 3.0 Sources:1 LSMS-98;2 LSMS-95,3 PAHO, 1998,4 INEC (Nov. 97), 5 Fundaci6nMuier y Sociedad (-) Not applicable

ETHNIC AND REGIONAL DISTRIBUTION. Like its neighboring countries, Ecuador has not compiled demographic data on the indigenous and black population; hence, estimates of the ethnic population vary. According to a recent Interamerican Development Bank (IDB) study on indigenous and black populations, about 25 percent of Ecuador's

11This section and the section on reproductivehealth rely heavilyon the Demographicand Maternaland Child Healtl Survey (ENDEMAIN) datafor 1994 found in the publicationENDEMAIN-94 (CEPAR, et al., 1995). Unless otherwise stated, dataand statisticsfrom this and the followingsection are basedon ENDEMAIN-94.

10 2 population is indigenous/black, of which 80 percent is indigenous and 20 percent is black." CODENPE, which is the official source of information in Ecuador, estimates the indigenous population to be about 10 percent. In terms of regional population breakdown, approximately 50 percent of the population lives in the Sierra, 46 percent reside in the Costa 3 and four percent live in the Amazonia (also referred to as the Oriente)." The distribution of men and women is markedly different in the three regions of the country. The Sierra is characterized by a larger female population; men on the other hand predominate in Coastal areas (Secretaria Tecnica del Frente Social, 1998). The Amazon region is more heterogeneous, however, with men outnumbering women in petroleum producing areas (ibid).

MORTALITY. As indicated in Table 2, mortality rates are higher for boys than for girls in all age groups. Higher male mortality is more significant for the 20-59 year age group as a result of external causes such as traffic accidents, homicides, and suicides.

Table 2: MortalityRates by Age and Sex 1985-1990 Ae Male :=- - -emk -Rat perP1000 Ratepet 1000 0-4 18.0 8.4 14.6 6.9 5-19 1.2 1.7 1.0 1.4 20-59 5.5 18.2 3.7 12.9 60 and above 55.0 71.7 49.4 78.8 Total 15.4 100.0 14.4 100.0 Source: ECLAC, based on CELADEs DemographicBulletin Year 27, No. 53. (Cited in FLACSO, 1995)

FERTILITY RATES. The national fertility rate in Ecuador has declined substantially over the decades, from 5.3 in 1982 to 4.0 in 1990 (according to national census) and to the current rate of 3.1 (according to PAHO, 1998a). This compares to 3.0 for Peru, 4.3 for Bolivia, 3.0-for Venezuela and 2.7 for Colombia (ibid.). However, the national fertility rate hides a high variance that is strongly correlated to years of education, rural/urban differences and ethnicity. For example, according to ENDEMAIN-94, women with higher education had a fertility rate of 2.1 compared to those with no education at all, who had a fertility rate of 6.2. Regional differences are also very important. Amazonian families in the Oriente typically pursue high fertility strategies, with total fertility rates being more than two children per woman higher than for either the Sierra or Costa (CONADE/FNUAP 1996a:56, Table 24). An analysis of the 1990 census data suggests that, even after controlling for individual factors such as age, education, marital and migrant status, regional characteristics - especially the family-based nature of the agricultural economy in the Oriente - contribute to higher-than-average birthrates.

EARLY PREGNANCY. Women become sexually active and have children at an early age. According to ENDEMAIN-94, about 18 percent of women age 19 have either been pregnant or already have had a child. Rural areas report higher rates (21 percent) than urban areas (15 percent). Education is another important factor. For example, an adolescent with secondary education is four times less likely to become pregnant than one with little or no education (ENDEMAIN-94).

calculatesthat 12 INEC estimatesthat four percent of the populationis indigenousor black,whereas CONAIE indigenousgroups comprise 35-40 percent of the population. 13 Less thanone percent of the populationlives in the Galapagos.

11 According to 1990-95data, adolescent pregnancy in Ecuador is higher than the average for the Region (FLACSO, 1992). During that period, Ecuador recorded 79 births per 1000 adolescent women compared to 68 for the Region. Among the Andean countries, Boliviahad a higher rate, at 83 (ibid.). Early marriage is probably one of the main reasons for adolescent pregnancy. According to INEC 1997vital statistics, 32 percent of female adolescents are married (the proportion of males is 13 percent).

HOUSEHOLD STRUCTURE.Ecuadorian familystructures are changing,with households declaringfemale headship now constituting about 19 percent of all households (LSMS-98).14At 21 percent, the proportion is greater in urban areas than rural areas, where the proportion is about 15 percent (LSMS-98). Female-headedhouseholds also have fewer dependents than those headed by men - only 45 percent of female headed households have more than three children, compared to 70 percent of households declaringmale headship (LSMS-98).Also, female-headedfamilies tend to be single-parented compared to male headed households, which typicallyhave two spouses. For example, 89 percent of households that declare themselves male-headedhave a spouse present, whereas the proportion is only six percent in the case of female headed households (LSMS-98). As shown in Table 3, female headship is more common among poorer households. According to LSMS-95,18 percent of non-poor households are headed by women, compared to 22 percent of poor households. Lastly,female-headed households tend to be less educated than their male counterparts. According to LSMS-98,16 percent of female heads and seven percent of male heads have no education.

REPRODUCTIVEHEALTH

AT-RISKPREGNANCIES. In Ecuador about 25 percent of pregnant women do not receive prenatal care during pregnancy and are thus considered at-risk (ENDEMAIN-94). Less educated and rural women have the least access to maternal health care. For example, as Table 4 shows, 50 percent of rural women have less than five prenatal care visits during pregnancy and 60 percent have not received an anti-tetanus vaccination. Anti-tetanus vaccinations can prevent serious infections and reduce maternal mortality. Similarly,more than 60 percent of women with less than primary education report less than five visits, and half of these women have not been vaccinated.

Women who are pregnant before the age of 20 and those over the age of 35, as well as women who have more than four children or who have spaced their pregnancies less than 24 months, are also considered at-risk. Unfortunately, data on these groups were limited, but taking into account availableinformation on teenage pregnancy and high fertilityrates, particularlyin rural areas and among women with little or no education, the proportion of these at-risk groups is probably high.

TYPE OF MATERNAL HEALTH CARE.According to LSMS-98data, 68 percent of all deliveriesin Ecuador are assisted by a doctor, 11 percent by an obstetrician and 12 percent

14 According to Rosenhouse (1989),problems exist in the concept of headship because it assumes a hierarchical relationship between household members. It also implies that the head is the most important member in the household, is present in the household, has overriding audhorityin imnportanthousehold decisions, and provides consistent and central economic support These commonly made assumptions between dte head and the household can result in a biased portrayal of the intemal dynamics of the household.

12 by a midwife (CONAMU/INEC, 1999). In urban areas, 94 percent of expectant mothers are attended to during childbirth by either a doctor or obstetrician, but in rural areas the proportion is 56 percent (ibid.). Women with little or no education and lower income women have the lowest levels of professionally assisted childbirths (ENDEMAIN-94). Professional care during childbirth is provided mainly by the Ministry of Health (33 percent) and by private clinics (17 percent). In the Costa region, a considerable share of births are attended by obstetricians.

Table 3. Household Headship and Principle Earner by Sex in Urban Areas Tye fHoueold lIncome 1'8 1991-=191 1995 199

Head of Household Poor MaleIC 81.5 81.3 79.8 78.9 77.6 Female % 18.5 18.7 20.2 21.1 22.4

Head of Household Non-Poor Male % 82.4 83.9 82.8 82.5 82.5 Female % 17.6 16.1 17.2 17.5 17.5

Principal Earner Poor Male % 79.7 80.5 77.9 76.7 73.7 Female % 20.3 19.5 22.1 23.3 26.3

Principal Earner Non-Poor Male % 77.1 77.9 77.9 76.5 76.5 Female % 22.9 22.1 22.1 23.5 23.5 Source:LSMS-95 in SIISE-SIMUJER 1999

MATERNALMoRTALITY. Although the maternal mortality rate has been steadily declining over the last few decades, it still remains relatively high. According to ENDEMAIN-94, the maternal mortality rate declined from 302 per 100,000 live births during 1981-87 to 159 during 1988-94. Comparative figures in the Region for the same period were 390 for Bolivia, 265 for Peru, 56 for Venezuela, 87 for Colombia and 114 for Brazil (PAHO, 1998a) But again, regional variations exist. From 1981-94, for example, the Sierra region had a higher maternal mortality rate (250 per 100,000) than the Costa region (197 per 100,000). Also, maternal mortality is greater among adolescent women. The Pan American Health Organization (PAHO) statistics from 1998 indicate that out of 480 annual cases, 230 cases were adolescent women (cited in CONAMU/INEC, 1999).

CAUSESOF MATERNALMORTALITY. Maternal mortality statistics suggest serious problems in quality of health care. About half of maternal deaths occur in health institutions, 36 percent at home, four percent on their way to hospital and seven percent elsewhere (ENDEMAIN-94). Regional differences in health care are also an important factor. In the Sierra region, the proportion of pregnant women who died at home was 44 percent, whereas in the Costa a higher number of maternal deaths took place in a health center. According to INEC 1996 statistics, among preventable maternal deaths, 38 percent were from toxemia,23 percent were from hemorrhages, and six percent were from abortion (CONAMU/INEC, 1999). Recognizing the depth of the problem, the Ministry of Health has put in place the National Micronutrient Program, which is an innovative program

13 designed to reduce maternal mortalityand morbidity and increase nutrition among infants and children(see Box 2).

ABORTION.Like other countries in the Region, abortion is illegalin Ecuador and, as such, is most probably under-recorded because of fear of sanctions. According to ENDEMAIN-94, eight percent of maternal deaths occur due to complications resulting from abortion. Spontaneous abortion (miscarriage)accounts for 85 percent of cases and induced abortion for 15 percent (ENDEMAIN-94). Women from urban areas and the Costa region who are in their latter productive years, as well as higher educated women, have the highest rate of induced abortions (ibid.). There is little difference between the abortion rates of teenagers and those of the general population.

Table4. PrenatalCare by Area,Education, and SocioeconomicStatus Cc~etieaic Nrnbe ogfVisit (% Ani -tewwR:g:g:jg Vacc;04XS00000ignatio %

Total 35.9 42.2 21.4 0.5 64.3 35.0 0.7 Urban 25.1 46.8 27.9 0.1 68.5 30.7 0.8 Rural 50.0 36.1 12.9 1.0 59.9 39.6 0.5 Lessthan primaryeducation 61.1 31.4 6.9 0.6 50.6 48.8 0.6 Lessthan secondaryeducation 38.1 43.0 18.5 0.4 70.6 28.6 0.7 Secondaryeducation or more 15.4 47.6 36.4 0.6 63.7 35.6 0.6 Low socioeconomic 60.4 31.0 7.9 0.8 - - - High socioeconomic 16.2 48.7 34.8 0.2 - - - Source: ENDEMAIN-94 (-) Not available

CONTRACEPTIVEUsE. According to PAHO (1998a),contraceptive use in 1996was 57 percent in Ecuador (proportion of women, all methods). This compares to 45 percent for Boliviaand 64 percent for Peru, and 66 percent for Colombia and 75 percent for Costa Rica. The most common form of familyplanning is female sterilization (32 percent), followed by the IUD (24 percent) and the pill (22 percent), according to LSMS-98 (CONAMU/INCEC, 1999). Male contraceptive methods are not frequentlyused, which confirms that familyplanning in Ecuador is a female responsibility. For example, only 3.0 percent of men use condoms and 0.3 percent have been sterilized according to the same source.

ACCESSTO FAMILYPLANNING METHODS AND INFORMATION. ENDEMAIN-94 data suggestproblems in access to familyplanning methods and information. For example, 33 percent of young women between the ages of 15-24 reported not wanting more children, and this proportion increased to 70 percent for low-educatedwomen. Most women would like to conceive fewer children. The estimated 'desired' fertilityrate is 2.9 children, which implies a reduction of almost 20 percent in current fertilitylevels. In rural areas, the gap between the actual and desired fertilityrate is even greater (4.6 vs. 3.6). According to the same source, men tend not to oppose familyplanning but rather leave this responsibilityto their spouses.

14 ENDEMAIN-94 statistics indicate that access to information about family Box 2. The National Micronutuient Program planningplanning methods varies by region, levelslevels oof The Ministry of Health's National Micronutrient education and income. Only 60 percent of Program, which is supported by the World Bank's women in Ecuador say they have heard of or FASBASE Project and other intemational orgatizations seen family planning messages. This such as PAHO/WHO, USAID and UNICEF, has been a proportion declines to 54 percent for women successful mechanism to reduce matemal mortalityin Ecuador. To address anemia, for example,the Program livingin rural areas, 53 percent for women supports: (i) provision of iron supplements to pregnant who have primary education only, and 28 women and children who are less than one year old as part p forwomen who are illiterate of regular pre- and post-natal check-ups; and (ii) a Ministry percet .of Health/private industry partnership for the fortification Nationwide, radio and television are the most of wheat flour with iron, folic acid and Vitamin B. The effective media to reach women, according to Program also provides Vitamin A supplements to infants respondents. Some 60 percent also receive and children between 12 and 36 months, supports repnensoe0oreceive fortification of salt with iodine and fluorine to improve oral information through newspapers and health, and provides information, education, and magazines, clinics (58 percent), printed communication campaigns and programs to modify health materials (55 percent), and through health and nutritional habits and practices. These nutrition interventions are an integral part of the Ministryof Health's workers (53 percent). However, for less- basic health care package. Integrating health and nutrition educated and rural women, radio is the most interventions is a cost-effective approach to address accessible form of media, with almost 91 Ecuador's dual problem of poor health and malnutrition percent reporting having access to radio. among children and pregnant and lactatingmothers because children who are born malnourished are more susceptible to About 80 percent of youth reported having childhood illnesses like measles, whooping cough, access to family.planning information respiratory infections, and diarrheal disorders. (CEPAR, 1989). Main sources of information I are school and friends (ibid.). For young men, friends and family are the most important source of information (ibid.).

According to numerous sources, the Ministry of Health (MOH) has not been active in providing contraceptives in Ecuador. Rather, external donors such as USAID and UNFPA have established programs to provide this service. ENDEMAIN-94 survey data indicates that public sector coverage is 38 percent. Private clinics or doctors are the main providers (26 percent), followed by the MOH (23 percent) and pharmacies (16 percent) according to the same survey. Given the important role played by USAID in distributing contraceptives, and its intention to reduce the size of its programs in Ecuador, there is concern that access to family planning will decrease.

MALEROLES IN REPRODUCTIVEHEALTH. According to a survey conducted in urban marginal areas of Quito by CEPAR (1992), men become sexually active at an earlier age than women. The average age of men's first sexual experience is 15 compared to 17 for women. Among young urban males, 17 percent used condoms during their first sexual encounter, which often takes place with a sex worker (30 percent) (ibid.). This puts these young men at a particularly high risk of being afflicted with the Human Immunodeficiency Virus (HIV) or other sexually transmitted diseases (STDs). Among adult men, fewer than half use any contraceptive method, either modern or traditional. Those who do use contraceptives most often use the rhythm method or condom. Most men do not consider male sterilization an option-4ess than one percent of men are sterilized (CONAMU/INEC, 1999). The number of sterilizations increases with age, higher level of education, years of marriage, and number of children (CEPAR, 1992). Main reasons for

15 men choosing sterilizationinclude preventing pregnanciesand limiting fertility(ibid.). Less than 20 percent of men use contraception for spacing,and a only a very smallgroup uses condoms for protection against HIV and other STDs."5

Data from other countries in the Region suggestthat men lack knowledgeon reproduction. For example, a 1996-97qualitative study carried out in an urban poor area of Argentina found that men lack information on sexualreproductive health and that the information that is availableto them tends to be partial and incomplete (Meglioli,1998, cited in Correia, 1998). According to the study: (a) men's knowledge of contraceptive choice is limited; (b) contraceptive use by men and their partners is low; and (c) lack of accurate information on reproductive health leads to the perpetuation of myths around contraception and sexuallytransmitted diseases.There is little to suggest that the situation is much different in Ecuador, particularlyamongst men with low levels of education.

It is now increasinglyrecognized in the Region and elsewhere that men need to share the responsibilityfor family-planningand for the protection against HIV and other STDs. Furthermore, teenage pregnancy is no longer exclusivelyregarded as a female problem, and increasinglyadolescent men are the targets of safe sex, familyplanning and responsible parenthood programs."6 More emphasis is also now being placed on conceptuallyunderstanding male roles in reproductive health. This has led reproductive health programs to focus on men in order to protect women's health, to attend to men's own health needs, and to engage men practicallyin reproductive health decisions and behaviors.

GENERAL HEALTH

GENDER ROLES AND MALE HEALTH. While from 1994-96the primary cause of overall mortality for both men and women was pneumonia, there are important gender- based differences in mortality. As Table 5 and 6 show, men are much more likelyto die from different types of accidents and violence,with car accidents and homicide resultingin the most number of deaths. In 1996,men made up 73 percent of fatalitiesas a result of accidents. Also noteworthy are homicides among men, which rose from the fourth cause of death among men in 1994 to the second in 1996. According to statistics compiled by FLACSO in 1992,92 percent of all victimsof homicide were men. Accidents and homicides are by far the most serious health risks for young men aged 15-24 years (FLACSO,1992). The National Police note that alcohol negligence and abuse are important causes of accidents and violent behavior.

According to researchers, aggressiveand risky behavior is linked to male gender roles and expectations. Studies have suggestedthat the inabilityof low income men to live up to societal and familialexpectations of what 'real' men should do-particularly low income men-is associatedwith stress, substance abuse, risk-takingand violence (Barker, 1998). According to Barker's 1997study conducted in the ghettos and favelasof Chicago and Rio de Janeiro, "low income young men, who may lack other, more mainstream ways of

15Survey of Information and Reproductive Experiences of Ecuadorian Youth of Quito and Guayaquil-- Decemnber1989. 16 See, for example nmelviwgMala in PvmwtingTeen Pregnangy A PracicalGmidefor Program Manager, The Urban Institute, 1998.

16 affirming their identity or of achieving manhood-in the workplace, or in school, for example-may use shows of force, fights and other forms of violence (including violence against women), experimentation with drugs, or acting recklessly to gain prestige within their peer group and to affirm a sense of self' (ibid.). Depression, according to PAHO (1998b), is on the rise in Ecuador and in 1996 was one of the primary reasons for seeking medical care. In 1995, 9.2 per 100,000 male inhabitants were hospitalized for depression (ibid.).

CANCER.Malignant tumors are an important cause of death for both men and women. According to PAHO 1998b, in 1995, stomach cancer claimed the lives of 14 and 11 per 100,000 male and female inhabitants respectively. For men, prostate cancer was the ninth most important reason of hospitalization, causing 333 deaths, and for women, cervical and breast cancer caused 676 and 243 deaths respectively (ibid.). At the national level 57 percent of women in fertile age have never been screened for cervical cancer according to LSMS-98 data (CONAMU/INEC, 1999). And in rural areas, this level rises to an alarming 70 percent (ibid.).

Table 5. Principle Causes of Female Mortality (grouping of 307 illnesses), 1994-96 -Order oflmportaie and Number of Death-s - CAlUgSE 9519 Order. O NO. Order Pneumonia 1 1,370 1 1,518 1 1,287 Psychosis without senility 2 1,311 2 1,137 2 1,130 Diabetes 3 1,090 3 1,011 5 718 Malignant stomach tumor 4 657 4 644 6 560 Heart attack 5 594 16 - 14 - Nephritis, Nephritic syndrome 6 588 6 530 7 559 and nefrosis Hypoxia/Neonatal asphyxiation 7 552 10 418 5 561 Intestinal infection 9 492 4 743 9 398 Heart attack, acute miocardium 8 494 8 528 8 533 Traffic accidents 7 495 10 377 9 431 Tuberculosis 13 - 10 378 12 - Cardiac arrhythmia 20 - 14 - 3 813 Source:INEC, Anuariosde nacimientosy defunciones,1994,1995, 1996. (-) Not available

ALCOHOLISMAND SUBSTANCEABUSE. According to various sources, alcohol and substance abuse is an important social problem facing Ecuador today. Despite concerns over the prevalence and consequences of alcohol consumption, however, it remains a relatively unresearched problem. Statistics from 1996 compiled by PAHO indicate that about eight percent of the population over age 15 are alcoholics (PAHO, 1998b). Other estimates suggest greater consumption levels. Twenty percent of respondents of a PAHO survey stated that they had drank in excess the month prior to being surveyed (ibid.).

Latin America has one of the highest incidences of alcohol consumption, with a rate three times greater than the rest of the world (Londono, 1996, as cited in World Bank, 1997a). Moreover, regional trends indicate an increase in alcohol consumption rates; these are expected to continue to rise due in part to aggressive marketing efforts by beverage

17 companies that have lost markets in industrialized countries (Cercone, 1994). In Ecuador, the beer company CMAde Cene:as Naconales recorded the highest company earnings in Ecuador in both 1997 and 1998, with profits exceeding those of all Banks, exploration and construction companies and other industries (Gesti6n, 1999). CenveiaAndina was in ninth place in 1998 (ibid.). Worldwide, alcohol use leads to two million deaths per year worldwide (World Bank, 1997a).

Alcohol consumption in Ecuador is more prevalent among men than women. According to a 1998 study of 220 poor urban-based adolescents, 45 percent of respondents consume alcohol (Programa del Muchacho Trabajador, 1998). However, as indicated in Table 7, boys are twice as likely as girls to consume alcohol (61 percent versus 29 percent). The use of alcohol nearly doubles from the 14-16 age group to the 17-18 age group (33 percent versus 60 percent), whilst seven percent of respondents started to consume alcohol before the age of 12. This is consistent with regional patterns of alcohol use. An important proportion of the respondents in this study reported using alcohol when they were depressed, had problems, felt sad or abused, or wanted to cheer themselves up (33 percent), whilst 51 percent consume alcohol at social occasions.

Table 6. Principl Causes of Male Mortality (grouping of 307 illnesses), 1994-96

Order Order N*~~~~~~~~ ~~~~~~~~~~~~~~~...... N...

Pneumonia 1 1,566 1 1,590 3 1,299 Homicides and lesions 2 1,501 3 1,395 4 1,176 Traffic accidents 3 1,347 2 1,408 2 1,398 Psychosiswithout senility. 4 985 4 825 7 862 Heart attack, acute miocardium 5 865 7 724 5 1,120 Malignantstomach tumor 6 842 5 804 8 724 Diabetes 7 804 6 750 11 - Hypoxia and neonatal 8 710 9 650 12 - asphyxiation Nephritis, Nephritic syndrome and 9 631 ? 10 614 nefrosis Intestinal infection 10 622 10 609 9 635 Hyperintensivecardiac illness 11 597 ? ? Tuberculosis 12 - 8 674 13 - Vascular cerebral illnesses ? 1 1,477 Cardiac Arrhythmia 19 6 925 Source: INEC, Anuarios de nacimientos y defunciones, 1994,1995, 1996. (?) Not available

A number of studies suggest that boys are more likely to use illicit drugs than girls. According to PAHO (1998 b), gender differences in illegal drug use are noteworthy, with 10 percent of men compared to less than one percent of women using illicit drugs. The ProgramaMuchacho Trabajador study suggests that males are more likely to use drugs. It finds that 13 percent of young people use illegal drugs, with 16 percent of boys and ten percent of girls using drugs, including marijuana, inhalants, cocaine, and injected drugs such as heroin.

18 Young people have a particularlyhigh exposure to both alcohol and drugs, and access to both is easy and widespread. There are no legal age restrictions for the purchase of alcohol, and many young people are exposed to alcohol use (and abuse) at home (74 percent accordingto the MuchachosTrabajadores study). Illegal drugs are not difficultto obtain, and use amongst youth is more widespread than commonly believed. Nearly 10 percent of the MuchachosTrabajadores study group reported that someone else in their home uses illegaldrugs (siblings40 percent, parent 20 percent and 40 percent would not say).

Table 7. Alcoholand Drug Use AmongYouth Age 14-19Years in Pocr Areas of Quito Chrct$4c f SpbwscKkus - p in Proor axto Consumealcohol % 45.4 60.6 28.5 Friendsconsume alcohol % 76.2 80.3 71.5 ConsumeIllegal Drugs % 13.3 16.3 10.0 Friendsconsume illegal drugs % 24.7 29.6 19.2 Alcohol is consumed by others in the 73.5 73 74 Household%O. Source: Programa Muchacho Trabajador, 1998

ACQUIREDIMMUNE DEFICIENCY SYNDROME (AIDS) AND STDs. While its incidence is relativelylow compared to other countries, AIDS/HIV is on the rise in Ecuador. The incidence of AIDS/HIV has increasedat a greater rate for men than for women. Reported cases for men went from five to 265 cases and from three to 70 cases for women from 1984-98. Over the last decade, infections attributed to heterosexual contact have increased. And while AIDS/STD has affected men to a greater extent than women, women have a greater chance of contracting AIDS for biologicalreasons. Gender-based differences in positions of power also make women more vulnerable to AIDS and other sexuallytransmitted diseases.

Male migrant workers are reportedly an important risk group for both contracting and spreading the HIV virus in urban and rural areas (UNFPA, personal communication). The magnitude of the problem is augmented-particularly in rural areas-by a limited knowledge of the diseaseand by the social stigma that is associated with it. This prevents many from seeking appropriate medical care or from protecting their sexual partners. This also leads to a considerable underreporting of HIV/AIDS cases, hindering an adequate epidemiologicalanalysis and design of effective preventative measures. According to the United Nations Development Programme (UNDP), initiativesto address AIDS/STDs have been partial, most having stemmed from the private sector. The public sector appears to lack HIV/AIDS prevention and treatment strategies.

OCCUPATIONALHEALTH AND HAZARDOUSWORK. The Government has not been vigilantin monitoring occupationalhealth hazards and illnesses,but several sources suggest that problems exist and that these are gender differentiated. For example, the flower industry, which employs about 60 percent women, requires intensive use of chemical substances. Similarlyin Coastal areas, the mining,shrimp and banana industry have been identified as work with gender-differentiatedhealth hazards. In rural areas women often work longer hours than men due to their domestic as well as productive activities,causing chronic fatigue,headaches and other symptoms. Although the International Labor

19 Organization (ILO) has set a regulatory framework to prevent occupational diseases due to these types of industries, their implementation in Ecuador is weak because of financial, educational and cultural constraints.

SMOKING. The use of tobacco - which in developing nations is much more prevalent among men than women - has become an increasing health concern in the developing world. The World Bank estimates that by 2030, ten million people a year will die as a result of cigarette smoking and 70 percent of these deaths will be in middle and low- income countries (World Bank, 1999). According to LSMS-98 data from Ecuador, nationwide about 16 percent of men smoke at least one cigarette daily compared to only three percent of women (CONAMU/INEC, 1999). The proportion is 17 and four percent in urban areas for men and women respectively. PAHO data confirm this tendency. According to PAHO, mortality attributed to smoking is about six percent, with about 21 percent of the population smoking habitually and male smokers outnumbering female smokers by a factor of 2.4:1 (PAHO, 1998b). Research indicates that in low-income countries, the smoking epidemic is spreading among low-income men (World Bank, 1999). In contrast, the 1990s saw smoking among men decline and among teenagers and young women increase in high-income countries (ibid.).

VIOLENCE

Violence - a growing concern in Ecuador which has been linked to increased poverty and inequality - has important gender dimensions. Latin America is the most violent region in the world, with a homicide rate twice the world average. According to PAHO statistics, Ecuador's homicide rates was 10.3 per 100,000 people in the late 80s and early 90s, having experienced an increase over the previous decade from a rate of 6.4 (World Bank, 1997). Homicides rates are low in Ecuador compared to neighboring Venezuela and Colombia, and are about the same as Peru (rates in those countries were 15.2, 89.5 and 11.5 respectively in the late 1980s) (ibid.). However, concern over violence is mounting in Ecuador. For example, in 1998 following El Nifio and the influx of migrants from affected areas, Guayaquil was forced to impose an extended state of emergency.

Both men and women are victims of violence in Ecuador, but homicide data fail to capture types of violence by gender and in particular intra-familial violence. Despite the scarcity of data, available information indicates that men are typically the perpetrators of violence.17 In terms of the victims, men are much more likely to be involved in labor conflicts, crime, and street and gang violence, whilst women are more likely to experience domestic and sexual aggression.

MEN'S ExPosuRE TO VIOLENCE. Men's exposure to violence begins at an early age. Through the school system, the family, the media and the streets, boys are socialized to be violent, thus making male aggression an acceptable, if not expected, behavior for men. As was previously discussed, mortality rates linked to external factors - including traffic accidents, homicide and suicide - are very different according to gender. In 1989, mortality rates were 92 and 22 per 100,000 (in the 15-24 years age group) and 145 and 17 (in the 25-44

17 It should be noted that in the case of physical,non-sexual child abuse, it is conunonplacefor women to physically punish theirchildren.

20 years age group) for men and women respectively. Ecuador's pattern of male mortality as a result of violence is similar to that of other countries, and anecdotal evidence suggests that, 8 if anything, levels of violence have been increasing in the last decade." Some of the consequences of violence include poor educational attainment, lack of employment alternatives, involvement in gang and criminal activities and Box 3. Challenges in Measuring Intra-housebold Violence delinquency, as well as the readyquency,a ofwellmalthe a Because they tend to be based on homicide statistics, crime reports ready acceptance of male and victimization surveys, conventional measures of crime and violence have violence within and outside the methodological biases which result in the underreporting of domestic family. violence. These biases are generally a result of one or more methodological constraints: the environment in which the questionnaire is applied; the context and wording of violence-related items when embedded in WOMEN'S EXPOSURE questionnaires measuring public security that cue respondents to only certain TO VIOLENCE. Women also types of criminalized violence; and the inadequate consideration of research experience high levels of ethics, particularly the inability to guarantee confidentiality, obtain informed consent and provide appropriate interviewer training. In response, violence, but more often as researchers are employing three main methodologies to assess domestic victims rather than violence: population-based prevalence surveys; service statistics from public perpetrators. Whereas boys institutions and private agencies; and opinion polling and knowledge, attitude to be aggressive, and practice studies. are socialized Prevalence surveys randomly sample national or subnational girls are encouraged to be populations to assess what portion of the population that has been passive, accepting and non- victimized by violence. Point prevalence surveys, which are the most questioning, thus making common, report the percentage of the survey population victimized by an act of violence over a fixed period of recent time, usualy over the past one, victimization of women an six or twelve months. Lifetime prevalence, in contrast, reports the acceptable behavior. Sexual percentage of the survey population victimized by a violent act ever in their violence mostly affects women lives. Hence the latter prevalence rates would obviously be considerable more elevated than the former. - particularly young women Service statistics refer to those data collected by service institutions, and girls - but remains widely usualy from health, education, social welfare, police, and criminal justice unreported and unprosecuted. agencies, both governmental and non-governmental. The limitation of this source is that the quality of data is highly variable. Many public and non- Similarly, domestic violence profit sector institutions, for example, do not have integrated information affects women systems, established intake protocols, or the capacity for interinstitutional disproportionately but is still cross-referencing. Moreover, prevalence data derived from service statistics widely accepted and remains are often biased, based as they are on study populations seeking violence intervention services who are more likely than the general population to have largely invisible (Gomez, been victimized by violence. Similarly, people seeking services from public 1993). or non-profit entities are more likely to be low income, thereby overrepresenting the poverty profile of the general population. Opinion polling and KAP studies are population-based surveys that But measuring ask respondents for their opinions, atttudes and beliefs regarding violence. domestic violence has proven These responses are close-ended, with response options often split into a 3-, to be challenging as explained 4-, or 5-point scale. While random sampling does aBow for generalizations in Box 3. Two surveys on to the population, opinion data and cultural beiefs are often too complex and nuanced to be adequately captured by scaled items. Another drawback domestic violence carried out is that surveys compile opinion data rather than actual behavior. For by the Center of Planning and example, 52 percent of men who agree or strongly agree that wife abuse is Social Studies (CEPLAES) in justified is not the same as reporting that 52 percent of men beat their wives. areaspopu/are of Quito Misreporting of opinion data may also inflate prevalence rates. involving 237 women found that 50 percent (1990 survey) and 60 percent (1991 survey) of female respondents had been physically abused at least once over their lifetimes by their partners (Camacho, 1996).

18These increases could be an artifact of better reporting. There is some evidence that in Latin America family violence has declined slighdy.

21 About 37 percent had been abused frequently, that is, at least two to three times weekly or 19 monthly (ibid.). These statistics are high if contrasted with those collected for other Latin American countries. Sample surveys found that in Chile 11 percent, in Colombia 20 percent, in Costa Rica 10 percent and in Paraguay nine percent of women had experienced physical violence.20

The CEPLAES study in Ecuador found violence to be fairly evenly divided between those of rural and urban origin, those with primary and secondary schooling, male partners with different occupations, and families with different levels of income. Violence reduced somewhat (10 percent) for women with post secondary education. Women's independent source of income was found to be the only statistically significant factor influencing domestic violence. About 71 percent of women with their own income declared having been physically abused compared to 47 percent of women without their own source of income. A similar study in Lima Peru, however, produced mixed results. The study found that violence unleashed or magnified psychological and sexual violence but not physical violence and that employed men inflict more physical and psychological violence than unemployed men, most probably because men are more dependent on female earnings (cited in Morrison and Biehl, 1999). 21

Violence makes women vulnerable and insecure and restricts their mobility and livelihood. Women, and young women in particular, are especially affected by perceptions and fear of violence in public spaces, which can prevent them from participating effectively in educational, economic, community, and political activities. For example, in many violent communities the threat and fear of violence makes women afraid to go out after dark unaccompanied, whether to work, visit a friend, attend evening classes or a community meeting (van Bronkhorst, 1999).

CAUSES OF VIOLENCE. Causes of male violence - including street violence and sexual and domestic aggression - have been linked to masculinity and gender roles. According to Barker (1998), to be a man in Latin America is equated with working hard, eaming well, being responsible, and providing financially for the family. When these goals become difficult to achieve, men regularly assert their masculinity through violence (ibid.). The incidence of family violence has also been linked to the use of alcohol, which is not necessarily the cause of, but may act as a facilitator and remove barriers that would otherwise exist. Moser (1997) observed these patterns in her community-based research in Cisne Dos, Guayaquil. She noted that escalating crime and violence is a result of increasing unemployment, particularly among young men. Women in the community identified a direct link between declining male earnings and increasing domestic violence, which was often associated with alcohol abuse.

Anti-violence initiatives in Ecuador have focused on reducing aggression against women. Women's organizations in particular have been active in this issue. The movement

19It should be noted that the study took place before the law on family violence took effect and the Comisariasde laMajer were in operation, both of which have served to raise awareness levels on the criminality of familyviolence. 20 Heise, Pitanguy and Germain, 1994 and cited in Morrison and Biehl, 1999. Samplesizes were 1000 women (Chile), 5390 women (Colombia),1312 (Costa Rica) and a national representative sample (Paraguay). 21 The study, which was carried out by Gonzales de Olarte and Gavilano Llosa,used a subsample of the 1996 National LivingStandards Suvey in Peru comprising 359 women living in metropolitan Peru.

22 against domestic violence culminatedin 1995 with the Congress passing the Law on Violence against Women and FamilyViolence. For the first time in Ecuadorian history, a law permits victims of abuse to report spousal abuse or abuse by familymembers and to obtain protection and immunity. Establishing women's police stations (comisarias)is another important advance in Ecuador. These comisaniasare operated with the assistance of NGOs; to date 19 comisariasare operating. However, while the comisaniassignify progress, a criticism is that they do not deal effectivelywith the abuser, thereby allowing him to enter into another relationship and repeat his violent behavior with other women.

VIOLENCE/SEXUAL ABUSE OF CHILDREN. The practice of physical punishment of children by parents and relatives- both male and female - is widespread throughout the region, and Ecuador is no exception. It affects both boys and girls, and its long-term effects include an acceptance of the use of violence (both in terms of using and receiving violence) and the production of intergenerationalcycles of violence. Child sexual abuse, on the other hand, mostly affects girls. Of sexual abuse reported in Quito and Guayaquil,69 percent involved girls between the ages of 10-19 years. Furthermore, 80 percent of these girls were abused by a father, relative, friend, or someone known by the victim (Camacho, 1996). In another study (CEIME, 1994), 22 percent of adolescent women reported having been the victim of some kind of sexual abuse in an educationalsetting. Although sexual abuse is mostly directed at girls,boys are also victims. Less help is availablefor sexually violated boys or young men, however, because culturalnorms surrounding manhood make it difficult for boys and young men to come forward and report sexual violence.

ALCOHOLISM AND VIOLENCE. The literature suggestingan association between violence and alcohol abuse is increasing. World-wide studies of battered women and men's treatment programs, for example,show the alcohol-violencelink. In the Region, research in Mexico found that almost 50 percent of those convicted of homicide admitted to having consumed alcohol prior to committing the crime (Cercone, 1994). In Colombia, a study found that homicide rates in cities decreased considerablyon days that the sale of alcohol was prohibited (World Bank, 1998). Interventions in Colombia to reduce alcohol saleswas based on findings that a significantproportion of homicide victims had elevated blood alcohol levels and that alcohol use was a risk factor for victimization. No empirical information exists on the link between violence and alcoholism in Ecuador, but several community-based NGOs consulted suggest a relationship. For example, in one community just outside of Quito, the incidence of domestic violence decreased after the community began to sanction and detain intoxicated males.22

EDUCATION'

ILLITERACY AND FUNCTIONAL ILLITERACY. Illiteracyrates have been decliningin Ecuador, although there is a concern that the current economic crisiswill alter this trend. National illiteracyrates are 12 and eight percent for women and men respectively according to LSMS-98data, and substantiallyhigher in rural areas (21 percent for women and 15 percent for men). And indigenous groups are more likelyto be uneducated: 53 percent of

22 Communitymembers of TunshiSan Javier, Riobamba, personal communication. 23 Figuresprovided in this section arebased for the most parton LSMS-95tabulations.

23 indigenous women are illiteratecompared to 35 percent of men.24 Illiteracyalso varies by region. The largest gaps are in the Sierra region where more than a quarter of women do not know how to read or write compared to 14 percent for men. In the Costa region, men - who have one of the highest illiteracyrates in the country - are at a disadvantage compared to women in both urban and rural areas. Gender gaps in functional illiteracyare smaller. Amongst the indigenous,65 percent of women are functionallyilliterate compared to 52 percent of men.

EARLY CHILDHOOD DEVELOPMENT. Research has consistentlyshown that early childhood development (ECD)" interventions are associatedwith positive outcomes such as improved school and labor market performance, lower crime rates, and more stable personal relationshipsas adults. A more immediate benefit is that child day care allows mothers and daughters,who have traditionallybeen the primary care providers, the chance to participate in income-generatingactivities or go to school. This is especiallyimportant for poor and single mothers. In a complementaryway, the market for childcare provision creates additional employment opportunities, primarily for women. In two studies carried out in low-income barrios in Guayaquil,Moser (1988 and 1997) found that women faced increasingpressures to earn an income which, in turn, resulted in less time to dedicate to child care and domestic responsibilities.She also found that when women with young children went to work, they often resorted to lockingtheir children in the house while they were away,or having older daughters miss school to care for younger siblings. This data suggestthat the positive externalitiesof ECD programs justifiestheir relativelymodest costs.

Ecuador has three main public childcare programs, which are operated by the Ministry of SocialWelfare, and the National Institute of the Child and the Family (INNFA). Municipalities,civil society organizationsand the private sector also provide childcare. The public programs include: (a) the Child Development Program (PD1),which is operated by INNFA and provides servicesin urban marginalneighbourhoods; (b) the Child Rescue Program (ORI) which is run by the Ministry of SocialWelfare; and (c) the Ministryof Education's Alternate Preschool Program (PRONEPE), which comprises two urban-based schemes (community homes for children and community centres for children) and a rural- based community promoter scheme.26 According to a review of social programs currently being carried out by the Bank, all three programs have the advantage of facilitatingfemale labor force participation; ORI and PDI also have the advantage of providing nutrition for children and PDI provides pre-school education. However, programs have low coverage.

24 The rates will likely decrease as younger generations bave much lower iOliteracy rates than their older counterparts - for example, the iliteracy rate for people between 15 to 24 years old is 3.2 percent compared to 17.2 percent for those over 40. 25 ECD refers to proper care, stimulation and education for children aged 0-5 years. Ideally, at the end of this period, the children are well prepared to enter primary school. 26 The two urban models have important elements in common a concern for incorporating families and the comnmunity and an emphasis on enhancing children's health, nutritional, and psychological development. One model provides care for children of working mothers within the community in the home of a female neighbor. Children stay at the care centers between eight and nine hours a day, five days a week. The child care provider is a person selected by the community, she and an assistant are trained by the Program. The children - generally 15 per home - range in age from three months to six years. The second model offers care for a maximum of 100 children. The community provides a community center, church, or sports club as location. Children from three to six years of age attend these centers five days a week, from three to four hours a day, during the morning or afternoon. Children are cared for by young people selected by the community and trained by the program, who serve as community promoters.

24 Also, overlapping servicescombined with the relativelysmall scale of programs translate into high administrativecosts, poor planning, and weak management controls (World Bank, forthcoming). All three programs are also continuallyunder the threat of budget cutbacks and in the case of PRONEPE, elimination(ibid.).

A new IDB funded program called 'NuestrosNiffos' offers the potential to increase coverage and improve the coordination and management of ECD services. The program will fund improvements and expansion of public, private and NGO programs on a competitive basis. The aim is to improve the existingservices of child care to 2000 centers, increase the number of children enrolled to 250,000,and improve community participation as well as the management of sector institutions. The Program includes training 8,000 mothers to care for children ('madres cuidadoras) and proposes integrated care for enrolled children. For example, it provides nutritional and medical care to children, carries out education programs at the familylevel, and gives specialattention to children suffering from violence.27

According to labor laws, firms with over 25 employeesare required to have childcare facilities;however, enforcement mechanismsare not in place to ensure adherence to the law. The law can also act as a double-edged sword, causingwomen's labor costs to rise and thus causinga disincentive for employers to hire women.

EDUCATIONAL ATTAINMENT RATES. Low levels of education are a problem for both males and females in Ecuador, which has one of the lowest educationalattainment rates in Latin America. Consistent with other countries in the Region, however, global educationalattainment levels are similarfor both genders. Overall, women stillreceive less education than men (7.0 years versus 7.4 years) according to the latest LSMS-98data (CONAMU/INEC, 1999). Differences are more significantin rural areas (4.1 years for women compared to 4.7 years for men) according to LSMS-95.28 Regionaldifferences in education are also noteworthy. In the Sierra region, 58 percent of men and 48 percent of women do not complete primary education, and less than ten percent of both men and women have access to secondary education. In the Oriente, women on average complete 3.5 years of education compared to 5.9 years for men. Education levels are particularlylow among the indigenous. For example, in the Sierra region, indigenous women have an educationalattainment rate of 1.4 years compared to a rate of 2.4 years for indigenous men.

SCHOOL ENROLLMENT RATES. At the national level,girls enrollment rates are higher than those of boys at all levels except the age 12-17 level,but gender differences are very small (see Table 8). Availabledata, however, suggestsan important gender difference in enrollment rates among the indigenous in the Oriente (Amazonia)at the secondary level where rates for girls are almost double that of boys (see Table 9). What is particularly striking, however, is the low enrollment levels overall for both girls and boys in the Oriente. Only 17.5 percent of girls and 8.9 percent of boys are enrolled in secondary school for example.

27 Consultant's report;Jorge Mejia for the Consultative Group on Early Clildhood Care and Development with support from UNICEF, New York, April 1990. 28 About 70 percent of men and women have had primary education and only 25 percent have had secondary educationL

25 SCHOOL DROPOUT AND REPETITION. In Ecuador, as in other countries in the LAC Region, boys have higher repetition and dropout rates than girls. At the national level, girls make up about 46 percent of students who have repeated a grade and only 38 percent of school dropouts (see Table 10 and 11). There are some variations, however, with female dropouts exceeding those of males in urban areas in the 25 and over age category, presumably because of marriage. Also, girls outnumber boys who repeat at the primary level in urban areas. But dropout and repetition for boys is clearly more of an issue. In urban areas over 75 percent of school dropouts at the primary level are male. And anecdotal information indicates that in the Coastal region, boys are being forced to abandon their studies to enter the paid work force at a rapid rate due to the current economic crisis.

Table 8. Enrollment Rates by Level of Education, Age Group, Gender and Urban/Rural Residence (%) Urban Rural Natio~~~~~nal AgeLvlFml Mae Fml Male Femaeal 6-11 98.0 95.3 92.8 88.9 95.7 92.5 12-17 83.7 82.4 57.2 58.2 71.1 71.4 18-24 31.7 33.1 16.9 14.2 26.3 25.2 25 + 4.1 3.6 0.7 1.2 2.8 2.5 Source: LSMS-98in CONAMU/INEC, 1999

Table 9. Indigenous Enrollment Rates by Region and Gender (%) ...... -..Secoad . IAryLVt, _ g in Leve ggio FealetImale Male; Male M Sierra 77.7 81.3 89.3 84.8 Oriente 17.5 8.9 22.4 26.0 Source: LSMS-98

Overall, gender appears to be less of a determinant than urban/rural residence,with drop out rates being significantly higher in rural areas than urban. As Table 12 shows, over 70 percent of girls drop out of sixth grade in the Sierra and Oriente. The rate for boys is 64 percent and 62 percent in the Oriente and Sierra respectively. Consistent with the anecdotal information, boys' drop out rates (70 percent) exceed those of girls' (64 percent) in the Coastal region.

Table 10. Number of ChildrenWho Repeat Grades by Age Group, Gender and Urban/Rural Residence UrbT RalNational Age:ieale S Me % Female Male % Fematle MA.l

6-11 14,816 12,831 53.6 23,392 27,970 45.5 38,208 40,802 48.4 12-17 18,380 23,570 43.8 14,302 20,493 41.1 32,682 44,063 42.6 18-24 10,523 9,806 51.8 1,138 1,764 39.2 11,661 11,570 50.2 25+ 4,003 4,815 45.4 641 1,758 26.7 4,643 6,573 41.4 Total 47,721 51,023 48.3 39,473 51,984 43.2 87,194 103,007 45.8 Source: LSMS-98in CONAMU/INEC, 1999

26 IMPACT OF MOTHER'S EDUCATION. The effects of mothers' education appear to be especially strong in Ecuador. A recent cross-national study found the relationship between mothers' education and the probability of children entering secondary school was larger for Ecuador than any other country in Latin America (IDB, 1998 cited in World Bank, forthcoming).

Table 11. Number of School Dropouts by Age Group, Gender and Urban/Rural Residence Age FmaleUrbn RurlNtoa ge,~: :Male 4% ele Ml % Fele Ml

6-11 4,434 13,510 24.7 1,850 2,271 44.9 6,284 15,781 28.5 12-17 6,183 11,563 34.8 2,629 3,810 40.8 8,813 15,374 36.4 18-24 4,906 5,289 48.1 457 422 52.0 5,362 5,710 48.4 25+ 2,308 1,320 63.6 - - - 2,308 1,320 63.6 Total 17,831 31,682 36.0 4,936 6,503 43.2 22,767 38,185 37.4 Source: LSMS-98in CONAMU/INEC, 1999 (-):Not available

Table 12. Dropout Rates by Region and Gender, Sixth Grade Age l6Wemale Male Coast 25.7 22.7 Urban 5.1 2.7 Rural 63.5 69.9 Sierra 35.4 28.1 Urban 7.3 1.2 Rural 71.7 61.7 Oriente 41.1 29.1 Urban -36.529 -14.2 Rural 71.1 64.0 Source: LSMS-98

REASONS FOR NOT ATTENDING SCHOOL BY GENDER. Reasons for not attending school differ for boys and girls because of gender roles. According to the LSMS-98 data, work, followed by age and cost are the main reasons that boys do not enroll in school (see Table 13). For girls, domestic work, followed by age and other forms of work, are the main factors. These figures suggest that while overall enrollment and attainment rates are not that different, programs designed to prevent school dropout and improve school performance would possibly need to target boys and girls differently if they are to be effective. The proportion of girls and boys who drop out because of disinterest also suggests that the relevance of education vis-a-vis the needs of children is a problem.

STEREOTYPES TRANsNETD THROUGH EDUCATION. In Ecuador as in other countries in the region, textbooks and teaching practices tend to transmit and reinforce gender stereotypes, linking men to public life - and all its dimensions of work, leisure, politics, wealth and power - and women to the private sphere of the household (Negrao

29 The negative figure signifies children who transferred into schools.

27 Table 13. Four Main Reasons Given for Not Enrolling in School (%) by Gender, National and Rural/Urban Residence3O

Cost 14.1 10.7 14.2 13.3 14.2 11.9 Work 20.6 55.8 13.5 46.0 17.6 51.3 Domestic Work 28.3 0.3 28.8 0.5 28.5 0.4 Age 15.7 12.8 22.0 19.0 18.4 15.6 Lack of Interest 5.8 6.6 7.1 8.4 6.3 7.4 Source: LSMS-98in CONAMU/INEC, 1999

and Amado, 1989). A 1992 study of 31 textbooks confirms the presence of gender stereotypes in school texts (see Box 4). From the analysis, the study finds: that men predominate in textbook images; that women are typically shown in maternal and caregiving roles compared to men who tend to be depicted in productive roles; and that men predominate in science texts (Robalino et al., 1992). Images of male and female children were also distinct, with girls In 1992, UNICEF/DINAMU/Ces*t Naionald typically portrayed playing indoors with Inssxsoers Socialkry Educatns (CENAISE3 sponsored a study of 31 schooltextbooks (which are used in Ecuadorin grades1- dolls and boys playingoutdoors or In 6). About two-thirds of the textbooks had been written by sports activities (ibid.). In recent years, independent authors and one-third were official books which however, several Ecuadorian NGOs had been edited by the Ministry of Education and Culture. The in education have developed study found the following: specializing . . Male images dotminatein text books; of 5,347 images with sound methodologies to address people, over half did not include women. stereotypes in textbooks and teaching o Women are typically depicted in maternal roles, as child practices at different education levels. care givers, and in education. A man did not appear once camingfor children, changing their clothes or giving them The Ministry of Education and Culture, medicine. with the technical support of 0 Wile men would typically be portrayed in productive CONAMU, has taken advantage of roles-cultivating, cutting wood, driving a car-women were shown to be sweeping, washing dishes and cleaning. these developments by integrating a0 Women workers were typically in jobs that represented gender stereotype module in teacher extensions of their reproductive roles, for example, nurses- training and other education prograrns. and never doctors-cleaners and cooks. O In science books, women as scientists, for example However, there is a need to support and handling precision equipment, were almost completely non- evaluate the effectiveness and impact of existent Men appear as the intellectuals. these new gender-aware educational 03 Women are shown as symbols of purity, justice, morals, practices on students' educational equality and fairness. O Gids play with dolls or do housework whereas boys are achievement, values, and self-esteem. typically shown as playing in sports, playing outdoors, or playingwith electronic toys. SOCIALIZATION OF BoYS AND Source: Robalino et al., 1992. GIRLS. Teachers can play an important role both in the socialization of children, as well as in their performance in school. Specifically, their values, attitudes and biases, and their different expectations and treatment of boys and girls - including promoting and condoning different forms of behavior - all affect how girls and boys are socialized. Abramowicz (1991) analyzes the influence of schools and teachers in particular in school performance. He observes that teachers tend to

30 For the 1997-98 school year, 6-24 year olds.

28 see boys' failure as a result of rebellion and 'boyish' behavior, and that of girls as resulting from low ability/intelligence and from their 'natural' traits.

GENDER DIFFERENCESIN ACADEMICAND VOCATIONAL CAREER PATHS. Despite reduced gender gaps in education, males and females continue to choose academic paths, and subsequently careers, consistent with their traditional roles. For example, women typically tend to take courses in the humanities, leading to lower wage 'caring' professions typically considered female (for example, teaching and nursing). Men, on the other hand, tend to select more technical courses and occupations. These trends are confirmed when one examines vocational training enrollments. For example, according to Ecuadorian Service of Professional Training (SECAP), which is the main provider of vocational and professional training in Ecuador, only a small gender gap exists in the student body, with about 43 percent of enrolled students being women (SECAP, 1997). Men are concentrated in programs such as mechanics, construction and electronics, however, whereas women make up the majority of students in administration, textiles and garment manufacturing and handicrafts (ibid.). Relative to men, women also specialize more in commerce, services and information technology (ibid.). Gender-based occupational training patterns affect labor market opportunities. For example, because of the technical skills women acquire, they may be better equipped than men to find a job in the tertiary sector, which is one of the most dynamic economic sectors in Ecuador.

LABOR

GENERALLABOR MARKET TRENDS. From 1970 to 1990, the female proportion of the economically active population (EAP) increased from 14 to 19 percent, whereas the male proportion decreased from 74 percent to 69 percent. Comparing Ecuador to other countries in the Region, women's labor force participation in 1990 was the second lowest in Latin America after Guatemala (FLACO, 1995). According to INEC data, by 1998 46 percent of urban women were economically active, having increased from 44 percent in 1993 (see Table 14). Over the same period, the proportion of men who were economically active decreased from 74 to 72 percent. Employment surveys show that the average rate of growth of the female EAP was greater than that of males from 1988-97, these being 5.7 and 4.2 percent, respectively (Samaniego, et al, 1998).

UNEMPLOYMENT. While women are a smaller proportion of the EAP, they are disproportionately represented among the unemployed. This is consistent with trends in other Latin American countries. In the 1990s, both male and female unemployment increased. The rate of unemployment for women increased from 12 to 16 percent from 1993-98 (CONAMU/INEC, 1999). For men, the increase was from six to eight percent over the same time period (ibid.). Examining the education levels of the unemployed using 1997 data, 56 percent of unemployed women had secondary education and 24 percent had superior education; the figures for men were 52 and 17 percent respectively. For men, the largest proportion of the unemployed (48 percent) was in the 15-24 age category, whereas for women it was among the 25-54 year olds (48 percent).31 Central Bank data indicate that unemployment and under-employment continued to rise from 1998 to the present as a result of the crisis. As previously mentioned, from mid 1998 to May 1999, global

31 Calculationsbased on Samaniegoand Femindez (1998) analysisof householdsurvey data

29 unemployment nearly doubled in Ecuador's three major urban centers, rising from nine to 16 percent (data were not disaggregated to permit a gender analysis).

IMPACTOF EDUCATIONON LABORFORCE PARTICIPATION. While women comprise a smaller proportion of Ecuador's EAP, they make up about half of all workers with no education or with superior and graduate level education (see Table 15). Their participation is less among workers with primary or secondary education. This pattern suggests that individuals enter the labor market if the marginal benefit exceeds the marginal cost. An educated woman has the ability to make high earnings in the labor market and hire domestic worker inexpensively, so her "cost" of working is low while the benefit is very high. An uneducated woman has low earnings ability but her economic need is so great (i.e. the "benefit" of earning income is very high) so even the benefits of earning a low income exceeds the value of her leisure and some homecare (cost).

Table 14. Proportion of Urban Population by Labor Market Category and Gender, 1998 Stat~~~rn Woanen Men Wo~~~~men"ofGou Urban population 100.001 100.002 51.1 Working age population 79.8 76.8 52.1 Unremunerated domestic work* 25.6 0.1 99.5 EAP* 46.1 71.8 41.1 Exmp,yed** 84.0 91.6 39.0 Modem Sector*** 37.8 46.0 34.5 InfornalSector*** 41.8 42.5 38.7 Agriculture*** 2.3 10.5 12.2 DomesticService*** 18.0 0.9 92.7 Usnempolod* 16.0 8.4 57.1 1 Number of Women:3,979,473 2 Number of Men:3,800,813 * Basecategory is workingage population ** Basecategory is EAP ***As a % of Employed Source:1998 INEC Employment,Unemployment and UnderemploymentSurvey, in CONAMU/INEC, 1999

Table 15. Proportion of Occupied EAP by Level of Instruction and Gender (%)

No education 58.0 42.0 46.5 53.5 49.9 51.1 Primary 39.9 60.1 34.0 66.0 36.4 63.6 Secondary 41.0 59.0 34.1 65.9 39.3 60.7 Superior or Graduate 44.1 55.9 45.9 54.1 44.2 55.8 Source: LSMS-95compiled by SIMUJERES-SISE

SALARIEDAND UNPAIDWORKERS. The largest proportion of women who are in the workforce are unpaid family workers. According to LSMS-95 data, 29 percent of the female EAP is in this category, followed by salaried workers (28 percent) and self-employed (26 percent). In contrast and as shown in Table 16, the greatest proportion of men are in the salaried worker category (41 percent), followed by the non remunerated worker category (16 percent) and the self-employed category (15 percent).

30 SELF-EMPLOYED WORKERS.As Table 16 shows, while fewer women are in the workforce, they outnumber the number of men who are self-employed (excluding agriculture workers). Women comprise 54 percent of all self-employed workers at the national level; in rural areas this figure reaches almost 57 percent. Data comparing incomes from self-employment versus salaried work are not available but research from other countries suggests that women who are self-employed tend to earn less than men. According to a study conducted on the self-employed in Argentina by Gallart et al. (1991), for example, one of the reasons for women's lower earnings is that they tend to juggle domestic and work activities. Among low income groups and the poorly qualified, both men and women operate under conditions of subsistence, but women face greater restrictions than their male counterparts because of their domestic responsibilities, which limit their time, mobility and flexibility and hence their opportunities for better paying jobs. The isolation of women's work - which is often done in the home - also limits their ability to interact with others and to acquire human capital. In the case of the better educated, motivations for pursuing self-employed work are different for men than for women. Men tend to choose self-employment over salaried work to increase their earning potential whereas women become self-employed to reinforce family income or juggle domestic responsibilities.

Table 16. EmpIoyed Male and Female EAP by Occupational Category, 1995 W.mn .AP in -this Cat Proporion of Women C .%) in the Category

'Category - - n~iat rual- urban :=iiona- natal. urban nidou ruralurln Salaried 27.8 12.2 37.3 41.0 19.9 58.5 30.6 25.S 31.9 Day worker 2.4 5.5 0.4 10.6 20.3 2.5 12.6 13.1 9.3 Owner or partner 2.5 0.3 3.8 5.9 2.7 8.5 21.5 5.7 24.8 Self-employed 26.0 19.4 31.3 14.6 8.4 19.7 53.7 56.5 52.7 Agriculture worker 3.9 9.0 0.7 11.6 22.8 2.3 18.0 18.0 17.4 Unpaid family 29.3 50.1 15.7 16.0 25.8 7.9 54.3 52.1 59.4 worker Domestic 8.1 3.7 10.8 0.3 0.0 0.5 94.6 97.8 93.9 Employee Total Aggregate 100.0 100.0 100.00 100.0 100.0 100.0 Total 39.4 35.9 42.3 TOTALSAMPLE 1,994,154 778,338 1,233,816 3,068,180 1387,950 1,680,230 Source: LSMS-95compiled by SIMUJERES-SISE

OCCUPATIONAL SEGREGATION. LSMS-95 data suggest segregation of the workforce by gender. For example, among the female EAP, 33 percent have jobs in commerce, hotels and restaurants, 29 percent work in personal and social services and 23 percent work in agriculture (see Table 17). In contrast, 37 percent of male EAP work in agriculture, 19 percent in commerce, hotels and services and 16 percent in personal and social services. Occupational segregation in itself may not be an issue if men's and women's average wages are equivalent when average respective amounts of human capital (both education and experience) are taken into account. Also, women's or men's dominance in one occupation or another per se may not be important as long as they are spread across occupations at large. For example, women may represent the majority of domestic workers

31 but may also predominate in professional occupations and other moderate to high paying jobs. PROFESSIONAL LEVELWORKERS. As indicated in Table 18, women have made significantstrides at the professional/technical/manageriallevel. At the national level,they comprise 35 percent of managers, 49 percent of professionals/scientists, and 42 percent of professional/technical workers.

In Latin America, rural non-farm activitiesare particularlyimportant for women. Hazell and Hagblade (1993) estimate that 79 percent of women in the rural wage-labor force in Latin America are employed in non-agriculturalactivities (cited in Lanjouw, 1998). This form of employment may also have significantequity and social welfare benefits. According to Lanjouw, even low productivity non agriculturalactivities - which are common particularlyamong rural women - may serve to reduce aggregateincome inequality,offset season agriculturalunemployment, and offer economic security to those who are unable to participate the agriculturallabor force (most notably women).

Table 17. Occu ied Male and Female EAP by job Category, 1995

Agriculture/hunting/f 23.35 54.52 3.41 36.54 69.49 9.30 29.3 30.5 30.9 ishing Mining 0.09 0.06 0.11 0.95 1.12 0.81 5.9 3.1 8.8 Manufacturing 12.77 10.50 14.22 11.54 6.80 15.42 41.8 46.4 40.0 E-lectricity/gas/water 0.07 0.04 0.09 0.48 0.15 0.75 8.3 125 7.6 Construction 0.28 0.07 0.42 8.14 6.14 9.76 2.2 0.6 3.0 Commerce/hotels/ 33.01 20.00 41.32 19.14 7.25 28.90 52.8 60.8 50.8 restaurants Transport/ 0.44 0.13 0.63 6.28 236 9.74 4.3 2.9 4.6 communications Financial services 1.15 0.16 1.79 0.82 0.06 1.45 47.8 60.4 47.2 Personal/social 28.84 14.53 38.01 16.10 6.63 23.87 53.8 55.1 53.5 services Total Aggregate 100.00 100.00 100.00 100.00 100.00 100.00 Total 39.4 35.9 41.9 TOTAL SAMPLE 1,994,154 778,038 1,215,816 3,068,180 1,387,950 1,684,230 Source: SIMUJERES-SISE (basedon LSMS-95)

RURAL NON-FARM WAGE-LABOR. According to an analysiscarried out on the rural non-farm sector by Lanjouw (1998), just under 50 percent of all rural women in the non wage-labor force have either a primary or secondary occupation in the non-agriculture sector. The corresponding figure for men is only 37 percent. His analysis indicates that the probability of primary employment in the non-agriculture sector rises from eight to 21 percent for a woman. Also, women are significantly less likely to be employed in the relatively high productivity occupations. According to his analysis, a man is twice as likely to enter a high productivity activity than a woman.

32 In the rural non-agriculturalsector, commerce is the most important sector for women: about two-fifths of all women compared to less than one-fifth of all men who work in the non-agriculturesector are engaged in commerce (see Table 19). Other sectors of importance for women's employment are community work and textiles/garment. For men, the most important sectors in order of importance are: construction, commerce, manufacturing and fishing.

Table 18. OccupiedMale and FemaleEAP by Job Level, 1995 F*ial*.,, .AP;t ia , Male Eii.in Cv v pii

Director,public sector 23.55 0.36 3.85 3.02 1.02 4.68 34.9 16.6 37.3 Professional/scientific 6.97 1.42 10.52 4.82 0.87 8.08 48.5 47.7 48.5 Professional/technical 4.95 263 6.43 4.42 1.22 7.06 421 54.8 39.7 Officeworker 5.93 1.04 9.06 1.85 0.39 3.06 67.5 59.7 68.2 Serviceor salesworker 0.03 19.32 31.23 10.62 4.64 15.55 61.9 70.0 59.2 Agricultureworker 20.53 49.48 1.99 27.29 54.01 5.21 328 33.9 21.7 Craftsperson 11.17 11.01 11.26 20.07 1293 25.97 26.6 323 23.9 Machineoperator 0.42 0.27 0.52 7.91 3.54 11.53 3.4 4.1 3.2 Non qualifiedworker 20.96 14.47 25.12 19.89 21.36 1867 40.7 27.5 49.3 Unspecified * * * 0.12 0.03 0.19 TotalAggregate 94.51 100.00 100.00 100.00 100.00 100.00 Total 39.4 35.9 420 TOTALSAMPLE 1,994,154 778,038 1215,816 3,068,180 1,387,950 1,680,230 Source: SIMUJERES-SISE(based on LSMS-95) -This wasnot reportedforwomen

DOMESTICEMPLOYEES. Domestic service remains a very important source of female employment, which is both good and bad. On the positive side, it enables poor women - and in particular young women with low educationallevels and no previous work experience - to enter into the paid labor force and represents a stepping stone to finding other jobs. The availabilityof domestic service has also paved the way for middle and upper-class women to enter the paid labor force. On the downside, it remains one of the lowest paid occupations and does little to encourage men to assume household responsibilities. In this sense, domestic service performed by women works to reinforce traditionalgender roles. A number of LAC countries, including Ecuador, have recently introduced labor legislationthat gives specialcoverage and benefits to domestic workers. Argentina and Brazilare other examples. This is a positive step provided that the law is enforceable and does not raise the economic cost of domestic workers to the level where employers lay them off. There is insufficientinformation to comment on the particular situation of domestic employees in Ecuador in this regard.

THETEACHNG PROFESSION. More than two-thirds of the teaching labor force are women. A study conducted by the National Center of Social and Education Research finds

33 workingconditions of teachersto be seriouslydeficient and recommendsthat the problem be urgentlyaddressed to enhancethe qualityof education(Robalino et al, 1992). Accordingto the report, one in two teacherssuffers from severestress, four out of ten sufferfrom respiratoryillnesses, and three out of ten sufferfrom varicoseveins. The study also highlightsthe poor socioeconomicconditions of femaleteachers. Accordingto the study,25 percent of teacherslive awayfrom their families,and for almosthalf, their salariesrepresent the only sourceof householdincome.

Table 19. Type of Non-Farm Wage Employment in Rural Ecuador (Principal and Secondary Occupations), By Gender

Fishing 4.6 12.5 1.6 3.2 Extraction 1.2 3.3 0.1 0.2 Manufacture 4.9 13.3 6.2 12.5 Textiles/Garment 0.2 0.5 2.7 5.5 Wood/Straw/Leather 1.9 5.1 1.8 3.6 Utilities 0.2 0.5 0.0 0.0 Construction 7.2 19.5 0.1 0.2 Commerce 7.0 19.0 18.3 37.0 Restaurant/Hotel 0.4 1.1 2.7 5.5 Transport 3.0 8.1 0.2 0.4 Finance 0.0 0.0 0.2 0.4 Property/Management 0.5 1.4 0.2 0.4 Administration 1.9 5.1 1.3 2.6 Teaching 1.2 3.3 3.5 7.1 Social Services 0.3 0.8 1.0 2.0 Community Work 2.4 6.5 6.3 12.7 Domestic Work 0.0 0.0 3.3 6.7 Total 36.9 100.0 49.5 100.0 Source: Lanjouw(1998) based on LSMS-95data

WAGE GAPS. According to the most recent LSMS-98 data, women's average salaries are about 68 percent of men's (see Table 20). At 54 percent, however, the gender gap is considerably bigger in rural areas. According to estimates for other countries, Ecuador presents a moderate salary gap. For instance, the female-male wage ratio in 1994 for the other countries in the region such as Argentina and Brazil was 75 percent and 60 percent respectively. The modem sector presents the smallest gap (89 percent) compared to the informal sector (73 percent) and the domestic service sector(64 percent).

One explanation for the wage gap differential is that men and women bring different levels of education and skills to the labor market. An analysis carried out by Somaniego et al (1998), however, found that even after removing as many of those differences as possible, differentials still remained between men and women. These may be due to unmeasurable productivity factors (for example, differential treatment in schooling, family expectations, "old boys networks", etc.) which could not be statistically removed from the numbers or from "market induced" differences, i.e. discrimination. When the researchers looked at industries that were typically female (food production, textiles, clothes,

34 small commerce, restaurants, public sector, teachers, medical, and domestic labor), women actually earned more than men with similar, observable characteristics in the food, textile, teaching, and domestic services industries (see Table 21). However, these trends turned around such that by 1997, women were earning less than men in all the 'female' industries. The increased disparity was the most severe in the food, textiles, clothes production, medical and domestic service industries, while the disparities, while still in favor of men, decreased in small commerce and the restaurant business.

Table 20. Wage Gap by Gender < > t 0 0; Avrage Earninji< Area W-omen i Men Wo>men1 Ean e Men~~~~~~~~~~~~~~a%o e' Urban 881,077 1,337,489 65.9 Rural 328,911 613,253 53.6 National 700,040 1,036,779 67.5 Source: CONAMU/INEC (1999) based on LSMS-98

Table 21. Percent by which Women's Wages are Above/Below Men's Wages Sector - 1988-- = 1993 1997 Food production 19 -36 -14 Textile production 22 -12 -23 Clothes production 15 -12 -37 Coommerce -28 -30 -18 Food services -37 -16 -19 Public sector -6 65 -13 Education 3 16 -6 Medical services -9 -9 -24 Domestic services 32 -11 -13 Source: Compiled from Samaniegoel al (1998) and based on Employment Surveys for 1988, 1993 and 1997

JOB ADVANCEMENT. Women often find it difficult to climb to positions of r esponsibility for a variety of reasons, ranging from discriminatory practices, to attitudes regarding the appropriateness of men and women holding positions of power, to women's own reluctance to pursue such positions. For example, a 1988 survey carried out in Quito and Guayaquil found that both men and women considered men to be more appropriate for supervisory positions than women (IECAIM, 1991). Moreover, 80 percent of those interviewed judged the work of male supervisors to be good; whereas only 58 percent of men and 78 percent of women considered women supervisors to be performing well (ibid.). Given that women continue to bear the burden of care giving and domestic tasks, they are also likely to seek flexible job situations, which makes it less likely that they will advance to higher positions.

SOCIAL SECURITY. Social security coverage is extremely limited for the population as a whole. In 1998, only 18 percent of Ecuador's population was covered by any form of social security (La Otra Mitad, 1999). Of those covered, 56 percent were men and 44 percent were women (see Table 22). The gender gap is most probably due to the fact that women are over represented in the informal sector or in part-time jobs, while men are predominantly in the formal sector (see Table 16).

35 DIVISION OF LABOR IN THE HOUSEHOLD. According to INEC data for 1998, only 0.5 percent of those who declare housework as their main occupation are men (SIMUJER/INEC, 1999). While this does not tell us anything about who carries out household activities on a part time basis, sources in Ecuador indicate that domestic work and child care continue to be mainly a female role. This is consistent with studies from other countries in the region which show that even among middle-class educated families with both partners in the labor force, housework and caregiving continue to be women's work.

Table 22. Social Security Affiliation by Progrm and Gender, 1998

SecuAity Aff,Uate General 1,077,982 48.2 679,129 54.3 398,853 40.4 Farmer's 982,388 43.8 502,000 40.0 480,388 48.7 Crafts 34,821 1.6 17,478 1.4 17,343 1.8 Voluntary 105,968 4.7 40,192 3.2 65,776 6.7 Domestic 38,124 1.7 14,106 1.1 24,018 2.4 Total Affiiates 2,239,283 100.0 1,252,905 56.0 986,378 44.0 Source: IESS, Direcci6n Matematica Actuarial cited in Fundaci6n Muier y Sociedad (1999)

HOUSEWORK. The problem of the gender division of labor in the household is universal and is not limited to Latin America or to Ecuador. In the United States, which is considered to be a more gender egalitarian society, housework and childcare tend to be one of the most contentious issues of families today, according to Deutsch (1999). Her studies show interesting differences among socioeconomic classes, with working class families being more openly conservative but 'practicing more than they preach', and middle class families being more liberal but 'preaching more than they practice'. Her studies also show that men who are taking on new family roles face light chiding and teasing from other men because they consider housework and childcare to be a women's domain. While these findings were found in the United States, they would also be relevant in other countries such as Ecuador.

CHILD LABOR. Child labor is both a result and a cause of poverty, to the extent that working reduces the time that children spend in school and this affects the accumulation of human capital. As indicated in Table 13, at the national level 44 percent of boys and 17 percent of girls report work (excluding housework) to be the main reason for not enrolling in school. An additional 22 percent of girls mentioned the need to work at home. Moser (1997) also finds that poor households use children's labor to reduce vulnerability in periods of economic crisis. Boys are more likely to eam income directly, whereas girls tend to assist indirectly, taking on childcare responsibilities to free women to work (ibid.). While work does not necessarily lead to school drop-out, it obviously has an effect on educational performance.

ECONOMIC CRISIS AND SOCIAL PROTECTION

Available information suggests that Ecuador's crises and adjustment policies have had the following gender-differentiated impacts:

36 (a) Unemployment has affected both men and women with rates increasing by about 33 percent for each group. LSMS-95 data, however, suggest that women are less likely to have a complete loss of income than men because they have more diversified sources of income. About 30 percent of women earn salaried income and 29 percent are self employed or business owners. In contrast, about half of all men are salaried workers and only 20 percent are self employed or own businesses. Women also appear to have more stable jobs than men. For example, among salaried workers, 58 percent of men compared to 49 percent of women report not having permanent employment. Arrears in paying public sector workers are also likely to have gender-differentiated effects because of the segregated nature of the labor force. For example, women are affected if public school teachers and health workers salaries are delayed, and men will feel the effects if police salaries are in arrears. And lastly, in the private sector, gender-differentiated effects in job losses would depend on the industries most affected by the crisis. For example, men have predominated in traditional industries, whereas women have concentrated in new sectors such as agriculture for exports.

(b) Poor households typically respond to declining income by mobilizing additional labor, usually women's as well as children's labor (Moser, 1997). In the case of women, petty trade and services such as domestic work offer easy entry and flexibility but tend to be lower paying. With regard to children, school drop-out among boys has been reported during the current crisis in the Coastal areas of Ecuador due primarily to boys entering the paid labor force (CONAMU, personal communication).

(c) Because the gender division of labor at the household level remains unchanged, an increase in women's participation in paid work implies new pressures on their time (Beneria, 1999). For example, as public health provision deteriorates, women spend more time caring for sick relatives, in addition to their regular domestic responsibilities. A recent World Bank study, however, suggests that while women's workload does indeed increase, domestic tasks decrease overall during economic downturns due to women's increased participation in the paid labor force, with corollary negative consequences for the upkeep of the family (Cunningham, forthcoming). Research conducted in Ecuador by Moser (1997) found that pressures on parents - and particularly mothers - to work outside the home result in them spending less time supervising their children, with resulting negative consequences such as sons spending more time on the streets, drinking, etc.

(d) For men of all ages, unemployment threatens their role as family providers, which often leads to problems of self-esteem, depression, violence and substance abuse (Barker, 1998, Correia, 1998). These tendencies have been observed in Ecuador as well as other countries in the Region such as Brazil and Argentina (ibid.)

37 HOUSEHOLD COPING STRATEGIESAND GENDER. Household strategies to cope with crisis can include: reducing household expenditures to match the new level of income; creating new income sources and/or intensifyingexisting income earning strategies;using savings or other types of assets to compensate for the loss of income; and increasingthe number of income earners. Possible implicationsof these coping strategies by gender and household composition include:

(a) In terms of reducing expenditures, households declaring female headship earn 77 percent as much as households with male heads. In rural areas the earning differential is greater (women earn 69 percent of male earnings). However, households declaring female headship have fewer members, which could increase their per capita income. For example, about 40 percent of male headed household have five or more members compared to 30 percent of female headed households.

(b) Level of human capital, mobility and other factors such as access to credit and networks - which influences one's capacity to seek other income sources - can vary by gender. In terms of human capital endowments, female household heads have less education than male heads - almost 50 percent of female heads have primary or no education in urban areas, compared to 41 percent of male heads. Also, 32 percent of women are illiteratecompared to 13 percent of men, and less than ten percent of women have secondary or higher education in contrast with 15 percent of men. Moreover, domestic and childcare responsibilitiestend to make women less mobile than men.

(c) Regarding the strategy to finance the crisiswith household assets, rural households maintainedby women may be more disadvantagedthan those headed by men. In the case of urban female heads, almost no difference exists between the number of female and male heads who own their house, the proportion being 55 percent and 56 percent respectively (according to LSMS-95). In rural areas, on the other hand, the proportion of female- headed households that rent is double that of male heads. Even more significant,however, in rural areas 41 percent of female heads do not have access to land, compared to 28 percent of men. Furthermore, among those who have access to land, women on averagehave smaller plots than men.

(d) Female household heads may be more vulnerablein periods of crisis because they tend to have only one adult present in the household and less chance to combine their income with other adult household members. About half of female-headedhouseholds are comprised of one adult and children. In contrast, most male-headed households (63 percent) have a spouse and children. Also, according to studies in other countries, children from single mother households are more likelyto abandon school (Geldstein, 1997).

38 SOCIALSERVICES FOR THE POOR. Government-funded social programs aimed at assisting the poor and the vulnerable have important gender dimensions, with women (and children) being the target of many of these programs. Of the myriad of Government funded social programs, the most important is the Bono de So/ida,idad(referred to as the Bono 32 Solidano),which provides direct cash transfers to poor women and the elderly. The Government introduced the Bono in September of 1998 in part to compensate for the 33 elimination of a number of long-standing subsidies (including cooking-gas and electricity). The Bono has provided a direct cash subsidy of the equivalent of about US$15 per month for women and US$7.50 for the elderly.34 Other social programs funded by the Government can be categorized into the following groups: those targeting women and children under age two, those targeting children under age six, those targeting school-age children and those targeting the elderly. Social infrastructure projects provide temporary employment as well as generating social benefits for the poor. The Bono is the largest program, reaching about 1.3 million people, followed by the School Meals Program with about 650,000 beneficiaries. The four main early childhood development programs benefit 35 about 120,000 children, of which the Child Rescue Operation (ORI) is the largest. In March 1999, the Government announced the Integrated Social Plan, which identified as its main targets, inter aka, street children, child workers, adolescents, pregnant and nursing women, persons with disabilities and victims of domestic violence.

3 2 Although not designed for this purpose, the Bono Solidariohas become the backbone of the Government'sSocial Protection Program. It is the only program currendy availableto many of the poor, but targeting problems and the low level of benefits constrain its overall poverty impact (World Bank,forthcoming). and regressive price 33 Subsequent to the validation of this report in July 1999, the Government re-established expensive subsidies for cooking gas and electricity. person per mondt (World 34 The bonorepresents about half of the amount required to meet the calorific needs of one Bank, forthcoming). The program represenitsnearly eight percent of the Central Governmuent'snon-debt spending and about 1.3 percent of Ecuador's GDP (ibid). (PREONEPE), the School 35 Other main programs include: the National Program of Alternative Pre-school Education Lunch Program and PACMI.

39 4. THE RURAL SECTOR

TRANSFORMATIONSIN GENDER DYNAMICSIN THE RURAL SECTOR. Throughout the 1990s, rural Ecuador has undergone substantial transformation, with important ramifications for women, men, and the households in which they live. Three gender-related phenomena in particular characterize contemporary rural Ecuador: (a) high rates of temporary male migration to urban labor markets, especially in the Sierra and Costa regions; (b) an increasing 'feminization' of agricultural production and marketing, especially in the Sierra and Oriente; and (c) population pressure and land scarcity undermining equitable land inheritance customs, especially in the Sierra and Costa. Changes in gender dynamics in the rural sector have varied in the Sierra, Costa and Oriente.

GENDERROLES IN AGRICULTURE

Gender analysis of the agricultural sector of the Ecuadorian economy is strongly differentiated by region and ethnicity. Substantially more published research exists on gender roles in Sierran agricultural systems, although some work has also been done in coastal and Amazonian settings. Most evidence suggests that the gender division of labor is significantly more restrictive among mesti.?yas compared to indigenous rural households.

Sierra

DIVISION OF LABOR. Studies of indigenous small farmer households in the Ecuadorian Sierra indicate extremely high levels of women's participation in all phases of agricultural production and natural resource management (Hamilton 1998). Over 90 percent of women in one indigenous highland community report that they participate equally with their husbands in the planting, cultivation (including pesticide and fertilizer spraying), and harvest of crops, as well as in the care of livestock. Almost 70 percent devote as much time as men to the marketing of their farm products. The only task to which women generally do not contribute their labor is plowing. Women average 47 hours a week in agriculture (ibid.), and are also active leaders and participants in community labor efforts (mingas).

VARIATION IN THE GENDER-BASED DIVISION OF LABOR. The gender division of labor in this region-both in agriculture as well as in domestic tasks such as cooking, cleaning and childcare-is characterized by mutual substitution and joint or alternate task performance. Regression analysis as well as ethnographic work suggest that women's participation in farming does not vary greatly with the level of household wealth, number of small children, degree of agricultural commercialization, or by the amount of household and hired labor available. This challenges the notion that the feminization of agriculture is

40 driven exclusively by poverty and male migration, which in turn may lead development institutions to exclude women from commercial agricultural technical assistance. As Hamilton (1998: 165) argues, "Although gender bias in regional labor markets does result in the investment of off-farm labor by men rather than by women, women's agricultural work is not determined by the absence of their husbands."

DECISION MAKING BY GENDER. Women also report that decision-making, as well as labor, is egalitarian in nature: more than 80 percent of a Sierran sample considered that they had equal control over land use, financial management, and product use decisions within their households. Seventy percent of the same sample said that they participated equally in technology selection. "In contrast with the expectations of many development professionals, women are equal partners in decisions regarding agricultural technology on farms of all sizes and among market-oriented producers using higher levels technology, independent of whether their husbands are absent fromnthe farm" (Hamilton 1998: 180).

HEADSHIP. Female headed households make up between ten and 20 percent of rural households in the Sierra (DeGraff and Bilsborrow, 1993; World Bank, 1996). Apptoximately half of these families are headed by widows, who are relatively well off in terms of land and income, but whose children have low school enrollment rates, indicating a division of labor in which children are replacing husbands' work on the farm (DeGraff and Bilsborrow, 1993). De facto female household heads, whose husbands have migrated into urban areas, are among the poorest in the population, although they have better access to male-linked networks than widowed or divorced women, and therefore find it less necessary 36 to draw on their children's labor for carrying out agricultural tasks (ibid.).

Costa

DIVISION OF LABOR. Although far less gender research has been conducted in the rural Costa region, the limited evidence available suggests that, due in part to a stronger "domestic ideology" that confines women to home-based activities, the gender division of labor in agriculture is significandy more pronounced than in the Sierra. Women's role in agricultural production-which is more commercially-oriented than in the Sierra-is concentrated in post-harvest processing such as degraining corn and drying rice and cacao (Phillips 1989). An exception appears to occur in the province of Manabi, where women are active participants in the coffee harvest (ibid.). Women also engage in small-scale income- generating activities such as raising chickens and pigs for market, attending small stores in their homes, and taking in washing. Domestic production includes collecting and chopping wood for cooking, fire building, cooking, sewing, childcare, clothes washing, and cleaning the house and solar (patio). Importantly, women play an essential role in building and maintaining the social networks that are essential to household reproduction.

HOUSEHOLD COMPOSITION. In terms of household composition, it is relatively common for rural men on the Costa to simultaneously maintain more than one common law union (compromiso). So-called "visiting unions", in which the male partner resides only temporarily with one or several women-who may themselves live permanently with extended natal family members-are a frequent residential pattern in this region. The

36 See Table 2 from DeGraff and Bilsborrow 1993:323.

41 implications of these types of unstable and multiple unions for economic activity patterns and income distribution are not well understood.

Oriente

The Ecuadorian Amazon is characterized by a tropical rainforest ecosystem, low but climbing population density, and rapid in-migration. Family-based farming systems combine the production of subsistence food crops (maize, rice, yucca and plantains), cash crops (mainly coffee), and pasture for livestock. Some household members may earn off-farm income in the oil or tourist industries, and many families supplement their incomes with bunting, fishing and gathering activities. The population is mnixedindigenous and mestizo, with the latter likely to be beneficiaries of the government colonization program or spontaneous migrants from other regions of Ecuador.

DIVISION OF LABOR. In this labor-scarce environment, women are active participants in agriculture and livestock production. In a sample survey in Napo and Sucumbios provinces in the northeastern Amazon, over 70 percent of women reported caring for small animals, while 63 percent take part in harvesting and 49 percent in planting (Thapa et al. 1996). Data from the same household survey suggest that women's participation in planting and harvesting increases with the amount of land dedicated to crops and with the number of children under 12 years of age in the family, and decreases with the level of off-farm income and the amount of hired labor employed on the farm (ibid.).

THE EFFECT OF FERTILITY RATES. In terms of regional agricultural and natural resource policy, the high fertility rates in the rural areas imply that agricultural intensification, including greater reliance on labor-intensive perennial crops and mixed cropping systems, could create more demand on women's labor time, and more incentive for having large families. In this sense, amplification of off-farm employment for both women and men, along with culturally appropriate family planning and reproductive health services, should form an integral part of a gender-aware policy for poverty reduction and environmental sustainability.

LAND

GENDER. LAND AND INCOME SECURITY. Gender analysis of land issues is important because, for rural households, access to land is a key determinant of poverty (World Bank, forthcoming) and is a crucial asset for production and collateral purposes. Land ownership and access is also a determinant of individuals' ability to bargain over the allocation of labor, income and other inputs into household welfare (cf. Doss 1996). Finally, land is imnportant as a source of income security during old age, both directly (as the basis for agricultural production and/or rental income) and indirectly (insofar as adult children are more likely to assist their elderly parents if they can expect an inheritance transfer). In the particular case of Ecuador, the agrarian structure was fundamentally altered by the land reforms of the 1960s and 70s, with important implications for gender-specific property rights and agricultural production. As is the case generally for understanding rural development in Ecuador, it is most useful to disaggregate the analysis by region. Unfortunately, gender-disaggregated data

42 are only available for the Sierra and Costa.

Sierra

Prior to the enactment of the Agrarian Reform Law of 1964, Sierran land and labor relations were dominated by the huasipzungosystem, in which large, privately-owned farms retained semi-tied labor to engage in generally low-technology crop and livestock production. The typical hlaszjpungerohousehold was extended in nature, and "a key characteristic of the extended huasipun,gofamily was the substitution of labor categories within the family, such that if one family member could not fulfil particular labor obligations to the hacienda, another would, irrespective of gender or age" (Phillips 1987:109). Under the 1964 law, huasipun,gueroswho had occupied the same plot for at least ten years became entitled to ownership, and haciendas larger than 800 hectares (plus 1000 hectares of pasture) were subject to expropriation. Land was adjudicated to household heads only, and although gender-disaggregated data on land reform beneficiaries do not exist, it is clear that the vast majority who received land-in both individual (parcela)and collective (comuna/cooperativa)forms-were men.

LAND OW'NERSHIPBY GENDER. Despite this initial male bias in the distribution of land, evidence suggests that traditional forms of equal male and female inheritance - especially among the indigenous Quechua-speaking population - have begun to equalize ownership of land reform properties, which are now being passed on to a second generation (Hamilton 1998; Belote and Belote 1988; Stark 1984; Alberti 1986). A detailed study of Canton Salcedo in the northern Sierra found that women were as likely as men to own land, either via inheritance or purchase. Moreover, men's and women's parcels were roughly equal in size (about two hectares on average) and deeds of purchase were generally jointly registered in the names of both husband and wife (Hamilton 1998). A larger survey of over 32 rural Sierran communities confirms the norm of equal inheritance to all children, regardless of age or gender: 90 percent of property owners claim that they intend to bequeath their land equally to sons and daughters. However, data from the same set of households indicate that inheritance practicesmay tend to favor male heirs: of the 158 parcels acquired through inheritance, two-thirds came from men's families and only one-third from women's (Katz 1999).

LAND OWNERSHIP AND HOUSEHOLD DECISION MIAKING. There is some evidence for the Sierra that women's joint or independent land ownership positively influences women's participation in household decision-making. Drawing on a small sample from rural Salcedo, an innovative multivariate analysis of the determinants of women's control of economic resources (land use, household financial management, selection of agricultural inputs, and the disposition of agricultural products for sale or consumption) suggests that this control increases significantly with women's ownership of land (Hamilton 1998:227 and Table C.3). This implies that such ownership conveys a degree of decision-making authority on women farmers in this area.

SHARECROPPING. Besides ownership, sharecropping is an extremely important means of land access for highland peasants in Ecuador. In her study of Canton Salcedo, Hamilton (1998) found that natal kin were particularly important sources of sharecropped

43 land, especially for younger couples who had not yet come into their inheritance. In this sense, both women's and men's ties to their extended families represent a productive resource in and of themselves.

Costa

MALE BLSES IN LAND DISTRIBUTION. The agrarian history of this region is substantially different than that of the Sierra. Coastal agriculture was dominated by cacao until the 1920s, after which rice, sugar and bananas became important crops. Land rental (precarismo) and a relatively mobile, temporary agricultural labor force, with little direct female participation in crop production, characterized the region prior to agrarian reform. Land reform legislation affecting the Costa was not enacted until 1973, and it took a different form than in the Sierra (Phillips 1987). Much larger properties (up to 2500 hectares plus 1000 hectares of pasture) were allowed to remain intact, and the emphasis was on the formation of cooperatives as opposed to individual parcelization. Since the reform was geared towards former precatistasand precluded both spouses from becoming members of the same cooperative (as well as requiring husbands' permission for wives to become members), over 90 percent of beneficiaries of coastal agrarian reform were men (ibid.). Evidence suggests that even women soias rarely work cooperative land themselves, and that there is a preference for male hired labor over female family labor during the peak seasons. Moreover, unlike the Sierra, it does not appear that gender egalitarian inheritance customs have reversed the male bias of the 1970s land distribution: an examination of the property registry in canton Vinces in Los Rios province showed that only 26 percent of all agricultural production units were owned by women, and in the majority of these cases, brothers or spouses actually managed the land (Phillips 1989:299).

MIGRATION

NATIONAL LEVEL RURAL/URBAN MIGRATION. In Ecuador, rural-to-urban migration has come down from its peak in the 1970s, but continues to be high by Latin American standards, contributing 42 percent of total urban growth during the 1980s (United Nations 1996). Nationally, Ecuadorian women made up 56 percent of all rural-urban mnigrantsin the 1960s and 50 percent in the 1970s (Singelmann 1993). In the particular case of the Sierra region, unequal land distribution and small average farm size, combined with a highly imperfect capital market and limited off-farm employment opportunities (especially for women), has led poor rural families to use both temporary and permanent migration as part of their "survival strategies." The relationship between poverty and migration is clear: according to the LSMS-95, nine percent and eight percent of households in the two poorest consumption quintiles surveyed in the urban areas of the Sierra had migrated from the countryside in the past ten years, while only five percent and one percent of households from the richest two quintiles had done so (World Bank. 1995). A rural qualitative assessment undertaken alongside the LSMS-95 found that temporary migration of young men and household heads ranged from 20-55 percent in the four Sierra communities included in the study (ibid.). Weiss (1985) argues that migration from the region has been spurred by delayed land inheritance due to longer life expectancy of parents, combined with the limited ability of the migrant generation to enter into traditional labor arrangements due

44 to their extended dependent position vis-a-vis their patents.

OUT MIGRATION FROM THE SIERRA. In the 1950s and 1960s, most out-migration from the Sierra went to the rapidly growing Costa region, but industrialization and centralization of government functions in Quito following the oil-led export boom in 1972 redirected rural-urban migration to the cities of the Sierra. In the particular case of Quito, the largest city in the Sierra, it is estimated that rural-urban migration contributed 55 percent of overall population growth between 1974 and 1982 (Palacios 1990). The most recent data from the 1990 population census suggest that women make up over half (52 percent) of the total inter-provincial migrant stream into Pichincha, where Quito is located.

FACTORSAFFECTING OUT-MIGRATION BY GENDER. Ecuador is relativelyunique in having a history of gender-disaggregated migration data collection and analysis. Evidence from the 1970s suggests that the forces driving women from the rural Sierra into the cities were quite different from those influencing male migration during the immediate post-oil boom period. A study based on ILO data from 1977-78 found that rural Sierran women migrating for economic reasons (i.e. in search of work) were on average younger, less educated, and more likely to be single than their male counterparts (Bilsborrow et al. 1987). Moreover, the likelihood of female migration was found to be positireAyassociated with (non- gender-disaggregated and possibly male biased) off-farm rural employment opportunities, and not significantly affected by either household farm size or distance to Quito - results exactly the opposite of those found for male economic migrants (ibid.). Bravo-Ureta et al. (1996), combining data from the 1974 and 1982 censuses, likewise demonstrate significant gender differences in the effects of macro-structural variables on internal migration in Ecuador during this same period: women are found to respond more strongly than men to the degree of urbanization and extent of agrarian reform activities, and less strongly to origin-destination gaps in literacy rates, population densities and mean income levels.

The analysis by Katz (1999), drawing on 1997 data from five Sierran provinces, strongly suggests that young men are important economic providers to their younger siblings, and that households with relatively large numbers of small children are more reluctant to allow these providers to migrate than they are to release adult women. The gender composition of the household, on the other hand, does not significantly affect migration probabilities, supporting the evidence that the gender division of labor in highland Ecuador is relatively flexible. The research also suggests a relationship between the marriage market and migration patterns, with a likely scenario that some marriages are followed by a period of separation in which men "commute" between their origin community and a destination area with better employment prospects; once a "beachhead" is established in the city, wives rejoin their husbands as tied movers. With regard to the labor market, the regression results suggest that the development of off-farm employment opportunities has disproportionately benefited rural men, while technical change in agriculture is a more important source of employment (and therefore deterrent to migration) for women.

Land market impacts also appear to be highly gender-differentiated, since larger endowments of privately owned land are more likely to restrict male migration, while access to common property resources acts as a migration deterrent for women. In more general terms, it seems that women are more likely to remain in communities with high levels of

45 local organization, suggesting that investments in social capital have a relatively high payoff in terms of stemming potential female migration. Table 23, as found in Katz (1999) presents comparative descriptive statistics on male and female migrant and non-migrant individuals from the 1997 survey.

Table 23: Individual and Household Characteristics, Descriptive Statistics37 Variable Full Sample Migrants Non-Migrants All F M All F M All F M N 1929 918 1011 872 399 473 1057 519 538 Dependent Variable NMigrant .28 .26 .30 Individual Characteristics Age 25.3 28.2 22.4 22.0 21.9* 22.0 26.5 30.4 22.5 Years of Education 8.2 7.9 8.5 9.1 9.2 9.0 7.8 7.5 8.3 Civil Status (=1 if single) .88 .86 .91 .76 .70 .82 .93 .91 .95 Gender (=1 if man) .50 .54 49 Employed in Origin .44 .35 .53 .36 .24 .45 .47 .39 .56 Household Characteristics Years of Education of 4.9 5.0 4.7 4.4 4.5 4.4 5.0 5.2 4.9 Household Head Dependency Ratio" .44 .43 .45 .40 .41 .38 .46 .44 .47 Gender Ratio39 .52 .65 .39 .50 .64 .38 .53 .66 .40 4 0 Standard of Living Index 4.6 4.6 4.5 4.2 4.241 4.1 4.7 4.8 4.7

Variable Rural Sample Rural Migrants Rural Non-Migrants All F M All F M All F M N 1210 568 642 550 248 302 660 320 340 Dependent Variable Migrant .33 .29 .36 Individual Characteristics Age 25.8 28.4 23.0 21.5 21.5** 21.4 27.9 31.3 24.0 Years of Education 6.4 6.2 6.7 7.8 7.8" 7.7 5.8 5.5 6.1 Civil Status (=1 if single .89 .87 .91 .80 .75 .84 .93 .92 .95 Gender (=1 if man) .49 .55 .47 Employed in Origin .47 .36 .58 .40 .30 .48 .51 .39 .64 Household Characteristics Years of Education of Head 3.6 3.5* 3.6 3.642 3.6** 3.6++ 3.5 3.5 3.6 Dependencv Ratio .50 .50 .51 .44 .49 .41 .53 .50 .57 Gender Ratio .52 .65 .38 .49 .63 .37 .53 .66 .39 Standard of Living Index 3.2 3.2- 3.2 3.1 3.0 3.2++ 3.3 3.3 3.2 43 Hectares of Owned Land 2.4 2.1 2.8 2.7 2.5 2.8++ 2.3 2.0 2.7 ** Not statisticallysignificantly different (at p<=.05) from male mean within migrant status category. ++ Not statisticallysignificantly different (at p<.05) from non-migrant mean within gender category. Source: Katz, 1999 (Table 2)

37Cell values are means weighted by census sector and household migration status. Individual and household characteristics refer to 1993values for non-migrants and values at time of migration for migrants 35Number of children (S 15 years old) / Number of adults. 39Number of women / Number of adults. 40 Calculatedas per Section 4. Minimum= 0; Maximum = 8. 41Not statisticallysignificantly different (at p < .05) from male mean within migrationstatus category. 42Not statisticallysignificantly different (at p < .05) from non-migrant mean within gender category. 43Eight outlier cases of households owning more than 50 hectares of land were excluded from the analysis.

46 MIGRATION AND INTRAHOUSEHOLDISSUES. There is some concern that temporary male migration undermines women's household position in the Sierra, insofar as men maintain control of their cash incomes and return with more 'patriarchal' (machista)values than those held in most indigenous Sierran communities (cf. Weismantel 1988). On t-he other hand, male migration contributes to the feminization of agriculture, leaving women largely responsible for the management-including hiring day wage laborers-and cultivation of the family parcels and the marketing of crops and livestock. On balance, it is not clear whether these "commuter" marriages, in which husbands specialize in urban wage labor and wives in agricultural production, are detrimental or beneficial to gender equity in the Sierran households.

47 5. SUMMARY AND IMPLICATIONS OF FINDINGS

Ecuador has witnessed important advances in terms of reducing gender disparities and addressing gender-related development issues. For example:

+ Access to and use of family planning methods has increased broadly, resulting in a drop in the fertility rate from 5.0 in the 1980s to the present rate of 3.6.

' Ecuador has a strong civil society movement; women's non-governmental organizations in particular have been successful at putting gender - and specifically women's issues - on the agenda.

' Gender gaps in educational enroUlment and attainment have decreased significantly. While years of education is still greater on average for men than for women, boys have higher drop-out and repetition than girls.

' Women's labor force participation has steadily increased over the last two decades, whereas the proportion of men in the workforce has declined. Nevertheless, men still predominate. In the rural sector, because of male migration, women have increasingly important roles in both subsistence and commercial agricultural production.

* There have been significant institutional advances on gender. For example, the Government has established 19 women's police stations, enacted a law against family violence, and created an autonomous agency dedicated to women's affairs (CONAMU). Relative to its counterpart agencies in other countries, CONAMU is a strong performer. It has been successful in building important alliances in the public sector, which have resulted in innovative gender programs. Also, relative to other countries in the Region, gender appears to be much more mainstreamed in public programs, with most key govemment agencies having made some progress on gender.

While these advances are indeed noteworthy, gender continues to be an important development issue in Ecuador, as demonstrated in the examples below.

' Whilst having decreased substantially, reductions in fertility rates have not been uniform. Women with little education, as well as those living in rural areas, continue to have very high fertility rates. Teenage pregnancy is also high, with 18 percent of 15-19 year olds already having had a child.

48 Access to contraceptives is limited among the poor. According to ENDEMAIN-94, 70 percent of low-educated women surveyed desire fewer children than they had. The recent withdrawal of donors of family planning programs has raised serious concerns about the public health care system's capacity to provide reliable and accessible contraceptives.

The reproductive health care system fails women in particular. Pre-natal care is inadequate and the proportion of pregnancies that are at-risk remains very high, as do maternal mortality rates.

Contraceptive use amongst men is minimal, which puts them and their partners at a high risk of acquiring STDs and AIDS. AIDS is on the rise in Ecuador, and the Ministry of Health has yet to develop a coherent strategy to address the issue; as a result, AIDS programs are partial and inadequate. Reproductive health programs typically exclude men and fail to consider their family planning roles and responsibilities.

The composition of the family is changing. Overall, 19 percent of families are characterized as single female-headed; this figure rises to 25 percent in urban areas. Single headed households face greater challenges than conjugally-headed households. For example, since domestic responsibilities can not be shared with other adult family members, these households typically spend a greater proportion of resources to purchase goods and services. Also, because they are responsible for economic support as well as taking care of the household and children, single mothers dedicate fewer hours to remunerated work. They also seek jobs that offer greater flexibility but not necessarily the best pay and conditions.

' * Accidents and violence (male on male violence) are by far the most serious health risks among adolescent males, and these are on the rise. Male violence against women and violence against children of both sexes also remain high. Alcoholism and substance abuse are serious social concerns in Ecuador and have been linked to violence. Yet despite their importance, little has been done to address these problems. Research on these issues, for example, is almost non- existent.

' In terms of education, girls and boys are being treated differently in the school system. As a result of gender socialization - both inside and outside the school system - girls continue to follow different academic paths than boys, which affects their opportunities in the job market, their earning potential and, over time, their well being (and that of their families). For boys, poverty leads them to enter the labor market at a very early age, thereby affecting their educational opportunities. And the current economic crisis has exacerbated boys' school drop-out, particularly in the coastal areas. In terms of early education, coverage of early childhood care and development programs is limited and quality remains a concern. This situation affects women to a greater extent than men by influencing their opportunities to enter into and advance in the labor market.

49 Gender-based wage gaps persist even after taking into account as many variables related to education and skills as possible. Whereas previously women were earning more than men in industries typically considered female (such as food, textiles, clothes production, medical and domestic service industries), by 1997 these trends had turned in favor of men.

In terms of rural development, policies and programs continue to downplay the importance of women's role in the agriculture sector. Moreover, population pressure and land scarcity are undermining equitable land inheritance customs, particularly in the Sierra. Land distribution has been biased towards men in the Costa region.

Macro-economic crisis and adjustment policies have gender differentiated impacts. Women tend to enter the labor force during these periods, thus implying greater pressures on their time given that adjustments in the division of household work have not taken place. For men, unemployment often threatens their role as family provider and creates problems of self esteem and depression, which may have other possible negative effects, such as violence.

Gender is a factor in household coping strategies and capacity to confront crisis. For example, available data indicate that the proportion of women with a secure job is greater than the proportion of men with a stable job. However, women may encounter greater difficulties in securing new income sources during economic downturns. A greater number of women are illiterate, for example, which affects their employment options. Also, women are less mobile than men because of domestic responsibilities. Rural households maintained by a single woman may be more vulnerable than conjugally-headed households because they have less assets.

Despite some modest advances, political participation and representation continues to be heavily skewed towards men.

Programs to promote indigenous peoples' tights have thus far failed to consider gender equity issues.

In reference to the report's objectives of identifying the efficiency, equality, and social welfare implications of gender differences:

In addition to affecting well being, poor health reduces economic productivity and is taxing on the country's health care system. Health costs due to complications from child birth, for example, could be avoided in some cases by providing women with low cost basic pre-natal care at the community level.

Family planning programs are potentially less effective because they only target women and fail to consider men, who also have a key role in family planning decisions. Reproductive health programs directed at men could also help to

50 reduce the spread of STDs and AIDS.

No studies were available on the prevalence of and the social and economic consequences of alcoholism among males in Ecuador, but consultations revealed that alcoholism is widespread. Studies from other countries have linked alcoholism to male violence and reduced worker productivity.

High levels of violence not only affect well being, but are costly to the health care and criminal justice systems, reduce work productivity, and deter economic investment. Violence also decreases the accumulation of human capital.

Low levels of educational attainment - due mostly to early entry into the labor force by boys and household work and other issues for girls - affect human capital and economic potential. Lack of attention to the particular needs of boys and girls can affect the success and effectiveness of programs designed to keep them in school.

The agricultural sector is very important both to the economy and to the survival of a large share of the rural population. Agricultural research and extension services that do not recognize the key role of women in agriculture fail to optimize production and earnings.

LandL ownership - which is skewed towards men - defines who is poor and who is not. Having property rights affects control of income, bargaining power and expenditure patterns in the household. It also influences incentives and interests in conservation and resource use, thereby affecting the sustainability of environmental programs. Land holders have greater status, credit worthiness, well-being and security in retirement.

Promoting the non-farm rural sector is viewed by many as a strategy for reducing rural poverty, absorbing the growing rural labor force and slowing rural-urban migration. But non-farm rural strategies need to take into account the different sectors in which men and women operate, as well as causes for gender differences in earnings and productivity.

51 6. PROPOSED OVERALL STRATEGY AND PRIORITY ACTIONS

OVERALLSTRATEGY

As a long-term strategy, gender policies and programs in Ecuadot should address the socializationprocesses that confine men and women to narrowly-definedgender roles. These gender roles, stereotypes and expectations can influence the opportunities men and women have and the constraints they face, as well as their choices and decisions over their lifetime. Ultimately,rigid gender roles affect social and economic well being. The effects of gender roles start early. Among the poor, boys are often forced to abandon their studies so that they can enter the work force and contribute to familyincome - consistent with the male role of 'family provider'. In line with their maternal and caregivingroles, on the other hand, girls are expected to work in the home, where they are less likelyto gain human capital and which, in turn, influences their labor market opportunities and earning potential later in life. The effects of gender roles extend into old age. Women's reproductive roles lead them to participate less in the labor market, to have more transient employment relationships, and to work in the informal sector in larger proportions than men. Hence, they are also less likelyto receive social security or to have savingsin their old age, and are more likelyto be poor.

Gender work should not be limited to programs targeting girls and women as has been the tendency in the past. Experience from Ecuador demonstrates that segregated women's programs have been ineffectivebecause they fail to consider the views of men and dynamics between the genders. Moreover, men also have issues related to gender roles and masculinity,such as alcoholism and violence. The Ecuadorian Government has already made important strides by supporting programs such as confronting gender stereotypes in education. Over the medium and long-term, these types of programs should be strengthened. But efforts should also be made to address gender socializationthat takes place through other channels, such as the family,the media and peer groups.

Short-term measures, however, need to focus on the acute and immediate needs of the population, given the current economic and financial crisis and the corollary deterioration in the standard of livingof so many Ecuadorians. In terms of gender, the Government's social safety net and recovery programs should continue to consider gender- differentiated needs and impacts. But even these programs - which have tended to target women as caregiversand mothers - could be strengthened so that they consider gender socializationand promote men's roles as fathers and caregivers,particularly among unemployed men.44 Broadening male gender roles could benefit men as well as women and

44 The World Bank is contemplating a study to determine how unemployed men spend their time, with a view to learning if men are taking on new tasks in the household, if they are engaging in non productive activities such as drinking, etc.

52 their families, given that substance abuse, violence and depression among men have been linked to gender roles and the limited ways men have to affirm their identity. Programs to promote men as fathets are still very new in the Region and elsewhere, but models exist and could be replicated in Ecuador. Sinriilarly,any temporary work program being considered for Ecuador - including social infrastructure projects that generate employment - should include targeting mechanisms to reach both women and men. Programs such as the TRABAJAR Program in Argentina, for example, have benefited mosdly men because no steps were taken to promote female participation.

PRIORITYACTIONS

As discussed in the previous section, the first priority relates to social safety nets and emergency assistance programs. These should ensure that both female and male gender issues are considered when designing and implementing social safety nets and emergency assistance programs, so that, inter alia, programs strengthen the role of fathers and provide income generating opportunities for women. The report recommends five other priority actions: first, strengthening indigenous people's initiatives so that they address gender inequalities; second, investing in integrated reproductive and sexual health programs that encompass issues such as maternal health and access to family planning; and third, developing youth programs that address at-risk youth, alcohol and substance abuse, and violence with a gender perspective; fourth, defining and strengthening the institutional framework on gender at the national level; and fifth, evaluating past and ongoing gender work in Ecuador to learn operational lessons in adding a gender perspective into different sectors. To this end, the Government should continue its attempts to work with Ecuador's strong civil society organizations. The other priority actions are briefly described below.

INTEGRATE GENDER INTO INDIGENOUS PEOPLE'S PROGRAMS. Integratinggender into the indigenous people's movement in Ecuador would be the first priority measure. Significant gender gaps in Ecuador persist amongst the indigenous, for example in educational attainment, yet to date gender has largely been absent in the discourse on indigenous rights. Given that 127 participatory assessments were carried out for the World Bank financed Indigenous People's Development Project, a first step would be to analyze this data from a gender perspective. Other priorities include: identifying mechanisms to ensure that both women and men have a voice in indigenous peoples development planning and programs, that investments reflect both men's and women's needs and interests, and that Ecuador's Indigenous Council improves its gender balance.

+ PROMOTE REPRODUCTIVEAND SEXUAL HEALTH PROGRAMS. Integrated reproductive and sexual health programs and policies is another priority given the high levels of maternal mortality, lack of access to reproductive health and family

45 'Mainstreaminga gender perspective is the processof assessingthe implicationsfor womenand men of anyplanned action,including legislation, policies or programmesin anyarea and at all levels.It is a strategyfor makingwomen's as well as men's concernsand experiencesan integraldimension in thedesign, implementation, monitoring and evaluationof policiesand programmes in all political,economic and societalspheres, so that womenand men benefitequally and inequalityis not perpetuated.The ultimategoal is to achievegender equality.' Definition of the Economicand Social ResearchCouncil of the UnitedNations (cited in Chantand Gutmann,1999).

53 planning services among the poor, the prevalence of AIDS and STDs, and the incidence of teenage pregnancy. At present, public programs focus on maternal and child health. While these are important in meeting women's basic needs, they are restrictive because they fail to consider women's requirements prior to pregnancy, as well as men's reproductive health requirements and their roles in the reproduction process. They also serve to perpetuate gender stereotypes. Programs should be reoriented to improve coverage of services for women that go beyond maternity and infant care, to develop services for men in line with their needs, and to promote more active male participation in childcare and parenthood. Given the important role played by civil society in Ecuador, local and community based programs that work close to beneficiary groups could be contracted to provide these services. In terms of integrating men into reproductive and sexual health initiatives - which is a relatively new focus - a number of lessons can be gleaned from successful programs in countries such as Mexico and Colombia.

. DEVELOP YOUTH PROGRAMS. Among the growing concerns in Ecuador are the various issues facing its young people, for example: high levels of school drop out; lack of employment opportunities; rising levels of violence; alcohol and drug abuse; and high rates of teenage pregnancy. Youth programs in Ecuador are scanty. Youth 'life-skills' programs, which have been successfully introduced in school systems of other countries such as Colombia and Trinidad and Tobago, are one alternative. These programs have included components on self-awareness/knowledge, anger- management, conflict resolution, family planning, and parenting skills. Youth programs also provide an important space to directly address the issue of gender roles and gender-differentiated expectations and identities.

STRENGTHEN INSTITUTIONAL FRAMEWORK ON GENDER. Public sector reform processes in Ecuador have included deliberations on how to cut costs and make government services more responsive to the needs of the people. Eliminating the numerous councils created under the Presidency - of which CONAMU is one - is an option being discussed. An assessment of the Government's advances and challenges in mainstreaming a gender perspective in public policies and programs, and the effectiveness of CONAMU in facilitating this mainstreaming process, would help the Government decide on suitable institutional arrangements related to gender.46 More details on the proposed institutional assessment are provided in the section that follows.

'-* IDENTIFY LESSONS FROM PAST AND ONGOING GENDER WORK. Many public programs and projects in Ecuador already have a gender perspective, although the level of effort varies. Much could be learned from the experiences of these programs. For example, some targeted programs for women reportedly have backfired because they failed to consider gender relations and men's views. In some

46 'Mainstreaming a gender perspective is the process of assessing the implications for women and men of any planned action, including legislation, pohcies or programmes in any area and at all levels. It is a strategy for making women's as well as men's concerns and experiences an integral dimension in the design, implementation, monitoring and evaluation of policies and programmes in all political, economic and societal spheres, so that women and men benefit equally and inequality is not perpetuated. The ultimate goal is to achieve gender equality.' Definition of the Economic and Social Research Council of the United Nations (cited in Chant and Gutmann, 1999).

54 cases, domestic violence and family break-up have resulted. Other programs such as the Ministry of Education and Culture's stereotype initiative appear to have generated positive results. Hence, evaluating programs that have taken steps to incorporate a gender perspective - with a view to identifying what has worked and what has not - would be a strategic action.

SECTORALRECOMMENDATIONS

Health

REPRODUCTIVEAND SEXUALHEALTH. Health care practitioners stress the importance of encompassing issues such as maternal mortality, AIDS and STDs, teenage pregnancy, and family planning under the rubric of reproductive and sexual health. Current public health programs are narrowly centered on maternal and child health, with other issues getting much less or no attention. But the problems are obviously connected and thus need to be dealt with in an integrated fashion. This shift in focus would require training of health practitioners in counseling and other techniques, given that health care workers are generally unaccustomed to providing these types of services.

In the short term, the Government urgently needs to put in place systems to make contraceptives available to women and men - particularly the poor - given the link between poverty and family size, the demand for famnilyplanning voiced by poor women and the imminent decline in family planning donor funding. Also, reproductive health services need to target men given the low incidence of condom use and the cost effectiveness of male sterilization. Similarly, information on family planning and reproduction needs to be made available to youth (both male and female).

Given budget pressures, community organizations and NGOs could be mobilized to provide reproductive and sexual health services for men and women. Training nurses and midwives to attend to home deliveries would be an important measure for indigenous woomenwho, for a variety of reasons in addition to cost, prefer to stay home to give birth. Throughout, attempts should be made to include men in programs on family planning, childcare and nutrition which would recognize and promote their roles as husbands/partners, parents and care givers.

GENDER-DIFFERENTIATEDEPIDEMIOLOGICAL DATA. As a long term objective, health care providers need to better capture and analyze epidemiological data by gender in order to understand the different diseases and health risks faced by the male and female population and improve disease control and prevention. This measure is important given that demographic and epidemiological profiles have been changing in Ecuador as in the Region. For example, mortality due to external causes - violence, accidents and risky behavior - are important male health issues, yet public health programs have yet to capture these tendencies and translate them into preventative programs.

ALCOHOLISM.Alcoholism is recognized as an important social problem in Ecuador. Yet there is very little discussion about the health and economic consequences of the

55 problem, and limited programs exist to address the issue. The Government has also been slow to react to this problem. A first step, therefore, would be to put the issue on the public agenda and begin discussingthe roles of different actors in contributing to and mitigating the problem. Key actors include the Government in terms of regulating liquor sales as well as civil society and community level programs in terms of developing preventative programs and initiatives to work with alcoholics. Because of its gender dimension, research examining how gender roles might contribute to alcohol consumption would be another priority.

Violence

Actions related to familyviolence - male/female and adult/child - should focus on prevention. According to Morrison and Biehl (1999),important vehicles for addressing and reducing violent behavior include the education system, community level programs, and the media. For example, the education system can influence cultural values that promote violent behavior. Possible actions include: (a) retraining teachers to ensure they do not promote violent behavior among boys and submission among girls; (b) eliminatinggender stereotypes in textbooks and other pedagogicalmaterials; and (c) developing programs to teach children nonviolent resolution skillsand to promote civic values. Schools and health clinicsalso represent an important vehicle for identifyingat-risk women, children and families. Community level programs can be used to provide informal education programs, teach citizens about legal sanctions against violence, establishviolence prevention strategies,and provide social services for victims of violence. And lastly,media can be used to promote nonviolence and counteract the media's role in reinforcingviolence. Mass media education campaignsrepresent a potentially effective response by providing nonviolent role models and by promoting nonviolent responses. Educational programs in the media can produce and disseminate more positive images of interpersonal interactions (male-femaleand adult- child). Moreover, 'telenovelas'(soap operas)can be a powerful medium for illustrating nonviolent conflict resolution. In terms of sanctioningviolence, ensuring that men and women are aware of laws on violence and that women have access to legal services,are other important measures.

The report also recommends that the current familyviolence program which establisheswomen's police stations be strengthened by, interalia, considering preventative programs for men, particularlyyoung men. This could be done on a pilot basis initiallyto capture insights into the types of programs that might be effective as well as costs and benefits. Other recommendations include supporting research on the specific causal factors of familyviolence as well as evaluating the compliancewith and effect of Ecuador's new domestic violence law. Experiences from other countries suggestthat in order to reduce levels of familyviolence, greater sensitivityamong the judiciaryand the police is a requirement, as well as increased public awarenessand changes in cultural attitudes on domestic violence.

56 Education

Gender gaps in educational attainment have been reduced significantly. However, anecdotal evidence indicates that during the current economic crisis,boys more than girls have been forced to abandon their studies to provide income for the family. Thus short- term efforts to increase school attendance and enrollment among the poor need to consider the implications of gender roles on education. Also, to the extent possible,initiatives in the short-term aimed at improving the qualityof instruction and providing textbooks to poor schools should avoid transmittinggender stereotypes. Over the long term, education policies should focus on ensuring that gains achieved over the decades in gender equity reach all sections of the population, includingindigenous groups. The report makes the following specific recommendations related to gender.

' EDUCATIONALATTAINMENT. Given regional gender differences, programs designed to keep children in school (such as the proposed scholarshipprogram) need to be designed to consider the different reasons girls and boys abandon their studies. Programs may need to target boys in some cases (for example in the Costa) and girls in others.

' TEACHER TRAINING AND EDUCATIONAL MATERIALS. Looking to the future, the Ministry of Education and Culture should determine the impact of its pilot gender stereotypes program, looking specificallyat how the program has influenced teachers' attitudes, students' performance (includingboys' and girls' career choices, motivations and life expectations) and parents' attitudes. Also, plans to improve teacher training should include gender modules to reduce gender socialization through education.

SCHOOL-BASED'LIFE SKILLS'PROGRAMS. Given the concern over violence, drug use, the spread of STDs, and teenage pregnancy, the Ministryof Education and Culture should look at introducing 'life-skills'programs for youth. Successful programs in other countries have included teaching on self-awareness/knowledge, anger-management,conflict resolution, familyplanning, and parenting skills.

' DAY CARE/EARLY CHILDHOOD EDUCATION. Early childhood education programs are associated with improved school performance and hence should be a priority over the long term. A more immediatebenefit is that child day care allowswomen or their daughters - who have traditionallybeen the primary care providers - the chance to participate in income-generatingactivities or stay in school. In a complementary way, the market for child care provision also creates additional employment opportunities, primarily for women. Community level organizations have been found to be a good vehicle for strengthening and expanding early childhood education and care; thus the NuestrosNinos represents an important step in supporting and expanding community childcare programs. To the extent possible, programs should also attempt to promote fathering.

57 Labor

Ecuador needs to address existing gender and labor issues such as higher unemployment among women and a persistent wage gap. Gaining an understanding of why the gender gap is increasingin all occupations, even female occupations, while it is decreasing across skill groups and education levels would also be imnportant.The report makes the following recommendations:

' LABOR LEGISLATION. The Government should examine the impact of its labor legislationon employmentpatterns and opportunities by gender, as well as wage gaps, and identify what measures need to be taken to further promote gender equalityin the labor force. Policy recommendations include: (a) systematically reviewing existinglaws to ensure that they do not impose higher costs on employers for hiring women than for hiring men (which leads to disincentivesto female employment);(b) strengthening enforcement mechanisms, judiciary processes, and institutional support; and (c) making information on job rights accessible to female workers.

I-0 SELF EMPLOYMENT. Services to support the self-employed and entrepreneurs need to take into account, and be tailored to, the different circumstances in which self-employed men and women work. Mexico, for example, is working to develop innovative financial services that include making visits to women entrepreneurs in their home-based workplace. Lending services should also ensure that both men and women are in sectors for which there is a demand for their services/products, and that entrepreneurs have a comparative advantage in the sector. Other services such as training, participation in tradeshows, networks, exchanges, etc. have also been identified as activities benefiting small scale producers, who are disproportionately women. Women entrepreneurs in Mexico, for example, complained of being excluded from "boys clubs" and requested similar types of support mechanisms for women to help them succeed in their business.

JlOB MATCHING. NGOs and other civil society groups that are well known in the community could act as a clearinghouse for jobs as well as provide training on grooming and preparing oneself for a job. Women in particular would be likely to benefit from these services because they have difficulties leaving their homes to search for jobs (due to household constraints), presenting themselves suitably to potential employers, and getting access to information on jobs. These types of services could help to address the higher rates of female unemployment. They are also important during crises when women who normally are not in the labor force are forced into the labor market. Because these individuals are not accustomed to working, they are particularly unprepared for the job search process.

DOMESTIC WORK AND CHILDCARE. According to Deutsch's seminal research on how shared parenting works in the United States, three conditions need to be in place if gender equality in the household is to be achieved: (a) men need to learn new skills; (b)

58 women need to give up the control they have had over the household; and (c) men and women need to have flexible work schedules. Short-term efforts in Ecuador should focus on the first two conditions, which can be promoted by civil society organizations working at the local level, for example, through youth programs, community water programs, adult education programs etc. The last condition - which involves the reorganization of work - would be a long-term objective given the pressing nature of unemployment in Ecuador today.

Rural Development

AGRICULTURE AND LIVESTOCK PRODUCTION. Agricultural research, technical assistance and training, whether provided by the government, the private sector, or NGOs, should ensure that both men and women receive services according to the crops and livestock they manage. Experimenting with different initiatives to assist female farmers to overcome constraints to participating and benefiting from rural development services would be one priority. This could include, for example, working with male farmers so that they understand that supporting women's participation does not mean they are "mandannas" and training men to share domestic chores and childcare. As a first step, the Government should ensure that lessons are disseminated from projects such as the National Rural Development Project (PRONADER), given that it has significant efforts to include women in project activities (see Appendix 2 for a preliminary analysis carried out during the gender review). As for a research agenda, much less is known about trends in the rural gender division of labor in the Costa region than in the Sierra and Amazonia, hence new research could focus on this topic.

LXND OWNERSHIP. Land is a critical asset, particularly in periods of economic shocks, hence efforts to address gender differences in land ownership should continue. The report makes four recommendations with regard to land titling. First, the Government has experienced some success in joint land titling by making modifications to titling request forms.47 This measure which, was supported by the PRONADER project, should be strengthened. Second, Ecuador would benefit by comparing its experiences with those of other countries such as Colombia, Costa Rica, Honduras, and Nicaragua, which also have taken steps to improve gender equity in land titling. Colombia, for example, has given priority to female household heads and victims of social violence in the allocation of land (Deere and Le6n 1998);and Honduras has used information campaigns to inform women and men of gender land rights and to increase demand for joint titling. Third, INDA should be required to generate statistics that disaggregate the title holder by gender as well as the type of title (joint, individual or communal). Such a measure would allow for better analysis and evaluation of trends and patterns of land ownership by gender. A fourth measure would be to learn more about Ecuador's equitable inheritance patterns in the Sierra region.

MIGRATION. Rural development policies need to take gender dynamics into account if they are to contribute to more sustainable development and stem the inflow of migrants into large cities. For example, report findings indicate that many young women and men are leaving rural areas to pursue educational and labor market opportunities in the cities. But

47 Changes in the form involved requiring both the marital status of the person requesting the land and the name of the requester's partner.

59 evidence from the Sierra suggests that agricultural intensification and the improvement of off-farm employment would act to deter female migration, while improving access to land would keep more men from leaving.

NON-FARM RURAL PRODUCTION. Rural development programs should examine ways to promote the rural non-farm sector given the evidence suggesting the importance of this sector in reducing rural poverty, creating employment and decreasing migration to urban zones. But interventions to enhance rural non-farm activities should consider gender dynamics such as women's reliance on this sector as a key income source for themselves and their families. The World Bank's Ecuador Poverty Report recommends facilitating women's employment in this dynamic sector via credit, training and investment (World Bank, 1996). In the Sierra, expansion of non-agricultural rural employment could be related to food and flower processing; for example, cheese production has been successfully promoted. Resistance to female employment is not likely to be a problem in the Sierra. Because of anticipated opposition in the Costa, however, off farm employment for women in the Costa could start with micro credit for small, home-based industries but ensuring that men are also brought in to deal with the male resistance issue. Lanjouw (1998) makes two recommendations to improve the effectiveness of the rural non-farm sector: (a) improving access to secondary education, which increases the likelihood one will engage in home enterprises and as a corollary improve one's access to income sources; and (b) developing infrastructure services, which in particular would have important implications on women's commercial activities. In the Oriente, transportation is a serious issue. Thus many of the off-farm viable opportunities for the Sierra and Costa would not make sense until the infrastructure problem can be dealt with.

Economic Fluctuations and Social Safety Nets

Emergency assistance and social protection programs need to take into account gender-differentiated needs and impacts of program targeting by gender. Current programs mostly target women because of their immediate maternal needs and caregiving roles. To the extent possible, these programs should also attempt to involve men as fathers and caregivers given that men are often underemployed or off work during economic downturns, and women typically enter the workforce to compensate for income losses. Furthermore, any temporary employment program needs to address gender, as women may be excluded because they are not perceived as the main family breadearner. Supply and/or demand factors that contribute to these types of outcomes would need to be identified and taken into consideration.

In terms of future research, data collection and analysis needs to better reflect the heterogeneity of household structures, their decisions, dynamics, coping strategies and responses to incentives, by composition of the household and roles of its members. In particular, attention should be paid to how households distribute work (paid and unpaid) to men and women during economic crises. A better understanding of the dynamics of unemployment by gender for the different sectors of the economy would be another priority, as it is likely that a segregated labor market will differentiate the gender impact of austerity measures.

60 Political Leadership and Participation

Continued efforts need to be made in Ecuador to increase women's participation in public life given men's predominance at all levels of politics and the judiciary. Trends in decentralization provide one window of opportunity for promoting women's leadership at the local level, for example through the Bank-funded municipal development projects (see Annex 1). Indigenous people's initiatives such as the Bank financed PRODEPINE Project represent yet another important entry point. Ecuador, like other countries in the Region, has put in place an affirmative action policy that establishes a minimum 20 percent quota of female candidates. To understand the impact, however, the policy needs to be reviewed. And other constraints to a more egalitarian political power structure, such as the political 'old boys networks' also need to be identified. As for increasing the number of female voters, programs need to be put in place to ensure that women get an identity card as well as to inform women and men of their civic rights. Civil society organizations could be employed to carry out these roles.

Institutional Framework

Given pressures to reduce public expenditures and modernize public services, the report recommends carrying out a diagnostic study with a view to informing the Government on the most efficient and effective institutional arrangements for addressing gender related issues. The proposed analysis would involve: identifying the institutional implications of existing gender inequalities on public programs and policies as well as the status of gender mainstreaming in public agencies; examining the role and effectiveness of CONAMU in facilitating the mainstreaming process and addressing gender inequalities; and identifying corollary institutional options given resource constraints. The Government has expressed a great deal of interest in the proposed institutional study given that it must decide on the fate of numerous councils that were created under the Presidency during the previous administration.

61 REFERENCES

Arboleda, Maria, 1998. Genero,Pobregay Municipios en el Ecuador.Unpublished manuscript. Quito: Sociedad Alemana de Cooperaci6n T&nica GTZ.

Asociaci6n Solidaria y Acci6n (A.S.A). Undated materials: El Club de las Familias con Problemas Relacionados con el Alcohol; Carta Europea sobre el Alcohol, Principios, Objetivos y Estrategias; mElAlcohol esta Causando Problemas a Ti o a Tu Fanilia?

Barker, Gary, 1998. Boys in the Hood, Boys in the Barrio: Exploratory Research on Masculinity, Fatherhood and Attitudes Toward Women Among Low Income Young Men in Chicago, USA, and Rio de Janeiro, Brazil. Paper Prepared for the IUSSP/CENEP Seminar on Men, Family Formation and Reproduction, 13-15 May 1998, Buenos Aires, Argentina.

Beneria, Lourdes, 1992. "Accounting For Women's Work: The Progress of Two Decades." WorldDevelopment. Vol. 20, No.11: 1547-1560.

Bilsborrow, Richard E., Thomas M. McDevitt, Sherrie Kossoudji and Richard Fuller, 1987. "The Impact of Origin Community Characteristics on Rural-Urban Out-Migration in a Developing Country." Demography24(2): 191-210.

Bravo-Ureta, Boris E., Ricardo E. Quiroga, and Jorge A. Brea, 1996. "Migration Decisions, Agrarian Structure, and Gender: The Case of Ecuador." Journalof DevelopingAreas 30 (4): 463-475.

Camacho Z., Gloria, 1996. MujeresFragmentadas, Identidady Violenmiade Genero.Quito: CEPLAES.

CEDATOS, 1998. Tabaquismo,Alcoholj Drogas: Consumoen losJdvenesde Areas Marginalesde Quito. Unpublished manuscript. Quito: CEDATOS.

CEPAM, UNFPA, Municipio de Quito, 1997. Ley Contrala Violenciaa lI Mujer y la Familia. Quito: CEPAM.

CEPAM, UNICEF and DINAMU, 1991. Educandoparala Igualdadentre los Ge'neros.Como Prevenirel Sexismoen la EducacidnPre-Escolar. Quito: CEPAM.

CEPAR, 1985. La Mujer, el Empleoy la Fecundidaden el Ecuador. Quito: Grupo Esquina.

CEPAR, 1989. Encuestade InformacidnyExperiencia Reproductiva de losJo'venesEcuatorianos en Quitoy Guayaquil. Quito: Ediciones Culturales UNP S.A.

CEPAR, 1992. Encuestade Saludy Plan/icacidnFamiliar en Barios Urbanos Iarginalesde Quito. Quito: Grficas Aviles.

62 CEPAR, CDC, and USAID, 1995. Ecuador:Encuesta Demograficay de Salud Materna e Infantil. Quito: CEPAR.

CEPLAES and UNFPA, 1990. Mujery Trabajo.Quito: CEPLAES.

Cercone, James, 1994. Alcohol-RelatedProblems as an Obstacleto the Developmentof Human Capital. World Bank Technical Paper Number 219. Washington D.C.: The World Bank.

Chant, Sylvia and Matthew Gutmann, 1999. MenstreamingGender? Questioning New Currentsin Genderand Development.Unpublished Manuscript. Washington D.C.: The World Bank.

CONADE/FNUAP. 1996a. FecundidadyAMortalidad. Serie Poblaci6n y Desarrollo. Ecuador.

CONADE/FNUAP. 1996b. MizgracidnyDistribucidn Espacial. Serie Poblaci6n y Desarrollo. Ecuador.

CONAMU/INEC, 1999. Mujeresy Hombresdel Ecuador en CiJras. Ecuador.

CONAMU, 1998. ProyectoPoliticas de Generohafia elAno 2000, Perdodo1999-2000. Quito: CONAMU.

CONAMU, 1998. Plan de Igualdadde Oportunidades1996-1998. Quito: CONAMU.

CONAMU, 1996. Mision del CONAMU, InstitucionalizarPoliticas Puiblicaspara el Desarrollode la Mujery la Equidad de Ge'nero.Quito: CONAMU.

Correia, Maria C., 1998. GenderPortfolio Review: Argentina, Chile, UruguayProgram. Unpublished manuscript. Washington D.C.: The World Bank.

Correia, Maria C., Valeria Junho Pena, 1999. BraZil GenderReview: Issues and Recommendations. Draft. Washington D.C.: The World Bank.

Cunningham, Wendy, forthcoming. WhoseDouble Shift? HouseholdAllocationof DomesticLabor in MacroeconomicFluctuations. Washington D.C.: The World Bank.

DeGraff, Deborah S. and Richard E. Bilsborrow, 1993. "Female-headed Households and Family Welfare in Rural Ecuador." Journalof PopulationEconomics 6(4): 317-336.

Deere, Carmen Diana and Magdalena Le6n, 1998. "Mujeres, Derechos a la Tierra y Contrareformas en America Latina." DebateAgrario 27.

Deutsch, Francine M., 1999. Halving itAll. How Equally SharedParenting Works. Harvard University Press.

Doss, Cheryl, 1996. "Women's Bargaining Power in Household Economic Decisions:

63 Evidence from Ghana." Staff Paper P96-11, Departmnent of Applied Economics, University of Minnesota.

FLACSO, 1992. MujeresLatinoamericanas en Cifras:Ecuador. Santiago de Chile: FLACSO.

FLACSO, 1995. Latin American Women Compared Figures. Santiago de Chile: FLACSO.

Floro, M., 1994. "Work intensity and women's time use." In G.Young and B. Dickerson, eds., Color Class and Country: Experiences of Gender: London and New York. Zed Press: 162-181.

Gallart, Maria Antonia, Martin Moreno and Marcela Cerrutti, 1991. Los Trabajadorespor Cuenta Propuiadel Gran BuenosAires: Sus EstrategiasEducativasy Ocupationales.Cuaderno del CENEP 45. Buenos Aires: Centro de Estudios de Poblaci6n

Geldstein, Rosa, 1997. MujeresJefasde Hogar Familia,Pobrenay Genero.Buenos Aires: UNICEF.

Gesti6n, Economia y Sociedad, 1999. "Las 50 de Gesti6n." Junio de 1999, No. 60, Ecuador, pp. 12-30. Published by Multiplica.

Hamilton, Sarah, 1998. The Two-HeadedHousehold: Gender and Rural Developmentin the EcuadorianAndes. Pittsburgh: University of Pittsburgh Press.

Hazell, P. and S. Hagblade, 1993. "Farm-Nonfarm Growth Linkages and the Welfare of the Poor." In Lipton and van der Gaap (eds), Including the Poor: Washington D.C.: The World Bank.

Htun, Mala N, 1998. Participaci6n,Representacidny Liderazgo Politico de la Mujer enAmerica Latina. Washington D.C.: ICRW.

IECAIM, 1991. La Situacidnde la Mujer Empleada bajoRegimen de Dependenda.Quito: Nueva Editorial.

Katz, Elizabeth G., 1998. Genderand Ejido Reform. Unpublished manuscript Washington, D.C.: The World Bank.

Katz, Elizabeth G., 1999. Individual,Household and Community-LevelDeterminants of Migrationin Ecuador Are thereGender Differences? Unpublished manuscript. New York: Department of Economics, Barnard College.

Lanjouw, Peter, 1998. Ecuador'sRural Nonfarm Sectoras a Route Out of Poverty.Policy Research Working Paper 1904. Washington D.C.: The World Bank.

Le6n, Guadalupe, 1995. Del Encubrimientoa la Impunidad,Diagn6stico sobre Violencia de Genero, Ecuador, 1989-1995. Quito: CEIME.

64 Lobao, Linda M.; Brown, Lawrence A. 1998. "Development Context, Regional Differences Among Young Women, and Fertility: The Ecuadorian Amazon." SocialForces 76(3):819 - 850.

Meglioli, A., 1998. Hacia Una Sexualidad Responsable:Actitudes MasculinasSobre Salud Sexualy Reproductivaen La Matanza, Provinciade BuenosAires. Robert McNamara Fellowships Program. Unpublished manuscipt. Washington D.C.: The World Bank.

Morrison, Andrew R. y Maria Loreto Biehl (Editors), 1999. Too Closeto Home. Domestic Violencein theAmericas. Washington, D.C.: Interamerican Development Bank.

Moser, Caroline O., 1992. "Adjustment from below: low income women, time and the triple role in Guayaquil, Ecuador." In H. Afshar and C. Dennis, eds., Women and Adjustment Policies in the Third World, London, McMillan.

Moser, Caroline O., 1997. HouseholdResponses to Poveryand Vulnerabiiiy: Volume I, Confronting crisisin CisneDos, Guayaquil,Ecuador. Urban Management Program. Washington D.C.: The World Bank.

Moser, Caroline O., Annika Tornqvist and Bernice van Bronkhorst, 1998. Mainstreaming Genderand Developmentin the World Bank: Progressand Recommendations.Washington D.C.: The World Bank.

PagainJose and Susana Sanchez, 1999a. GenderIssues in WorkforceParticipation and Self- Employmentin Rural Mexico.Unpublished manuscipt. Washington D.C.: The World Bank.

PaganJose and Susana Sanchez, 1999b. Explaining GenderDiferencesin Earnings in the MicroenterpriseSector. Unpublished manuscipt. Washington D.C.: The World Bank.

PAHO,1998a. HealthSituationoftheAmericas. BasicIndicators1998. WashingtonD.C.: PAHO.

PAHO, 1998b. La Salud en lasAmericas. Edici6n de 1998, Volumen II. Washington D.C.: PAHO.

Palacios, Diego, 1990. "Ecuador." In InternationalHandbook on InternalMigration, Charles B. Nam, William J. Serow and David F. Sly, eds. Westport: Greenwood Press.

Phillips, Lynne, 1987. "Women, Development and the State in Rural Ecuador." In Rural Women and State Policy:Feminist Perspectives on Latin American AgriculturalDevelopment, Carmen Diana Deere and Magdalena Le6n, eds. Boulder and London: Westview, pp. 105-123.

Phillips, Lynne. 1989. "Gender Dynamics and Rural Household Strategies." CanadianReview of SociologyandAnthropology 26(2):294 - 310.

65 Reed, Carolina, Carlos Larrea and Mercedes Pinto, 1997. IndicadoresSocialespara elAnalisis de las Desiguadadesde Genero:Educacidny Empleo en el Ecuador.Quito: Graphus.

Robalino, Magaly, Marco Villarruel and Edgard Isch, 1992. MujeresInvisibles: Un Estudiodel Sexismo en la Doceniay en los Textos EscolaresEcuatorianos. Quito: La Huella Impresores.

Saldafia B., Patricia, Paz y Mifio C., Isabel, 1998. Hacia la Equidad, Proyectosde Muj1#eryGenero en elEcuador. Unpublished manuscript. Quito: CONAMU.

Samaniego, Pablo, Gabriela Fernindez y Camilo Pinz6n, 1998. Situacidnde laslMujeresen el Empleoy el Ingreso:Analisisy PropuestaparaMejorar elAccesoy Reducirla BrechaSalarial. Unpublished manuscript. Quito, Ecuador.

Shrader, Elizabeth (2000). "Enhancing the 'Hot Spots' Index: Complementary Methodologies to Measure the Gender Dimensions of Crime and Violence." Unpublished manuscript. Washington D.C.: The World Bank.

Siguenza, Marcia and Pedro Zeas, 1998. GobiernosLocalesy Plan/ifcacidn de Genero.Cuaderno de Trabajo Niimero 5. Quito: CONAMU.

Singelmann, Joachim, 1993. "Levels and Trends of Female Internal Migration in Developing Countries, 1960-1980." In InternalMigrationof Women in DevelopingCountries, United Nations, Department for Economic and Social Information and Policy Analysis. New York: United Nations.

United Nations, 1996. World Economicand SocialSurvey 1996. New York: United Nations.

UNICEF, 1992. Women, Work and the Needfor Cbild Care,Opportunities for Programmatic Collaboration, A Review of UNICEF-Supported Programmes in Nepal, Ecuador and Ethiopia. New York: UNICEF.

Urrutia, Rafael, Ed., 1994. Empleo:Cuello de BotelladelAjuste. Quito: CEPLAES-CEOSL.

Van Bronkhorst, Bernice, 1999. Youth Violencein Latin America and the Caribbean:Causes, Costs and Interventions.Forthcoming LCSES technical paper. Washington D.C.: The World Bank.

Weismantel, Mary J., 1988. Food, Genderand Povertyin the EcuadorianAndes. Philadelphia, PA: University of Pennsylvania Press.

Weiss, Wendy A., 1985. "The Social Organization of Property and Work: A Study of Migrants from the Rural Ecuadorian Sierra." American Ethnologist12 (August): 468-488.

World Bank, 1995. World DevelopmentReport 1995. Washington, D.C.: The World Bank

World Bank, 1996. EcuadorPovertyReport. A World Bank Country Study. Washington, D.C.: The World Bank.

66 Wotld Bank, 1997a. Alcohol Poligyas a Means of ControllingCrime and Violence:A reviewof the research.Unpublished Manuscript. Washington D.C.: The World Bank.

World Bank, 1997b. World DevelopmentReport 1997. Washington D.C.: The World Bank.

World Bank, 1998. CountryAssistance Strategyfor Ecuador. Draft. Washington D.C.: The World Bank.

World Bank, 1999. Curbingthe Epidemic. Governmentsand the Economicsof TobaccoControl. Washington D.C.: The World Bank.

World Bank, forthcoming. Crisis,Poveryv and SocialServices in Ecuador. Washington D.C.: The World Bank.

Younger, Stephen, Mauricio Villafuerte and Lily Jara, 1997. IncidenciaDistribuitiva del Gasto Puiblicov Funcionesde Demanda en elEcuador.Nutmero 3. Quito: FLACSO.

67 APPENDIX 1: MATRIX OF GENDER ISSUES BY PROJECT

Indigenous Peoples Project components include: (a) indigenous and Afroecuatorian organizations * One consultancy carned out to * Follow gender analysis being carried Development strengthening, such as evaluating the institutional capacity "secondary" broadly identify gender issues in out on participatory assessments to (PRODEPTNF.) organizations and providing training accordingly; (b) land tenure and water rights the project produced limited see what results it yields. regularization such as distributing 20,000 hectares in community titles in insights and information. Esmeraldas; (c) legal capacity building, including training for community-level * Project will analyze from a indigenous groups and assisting 160 local organizations in 288 parroquias to gender perspective 127 prepare a "participatory local development plan"; (d) mral investments, which participatory assessments funds community project proposals; and (e) strengthening of the Indigenous conducted during project Affairs Secretariat. Gender issues vary greatly by ethnic group but little has been preparation in Afroecuatorian documented about how gender differences contrast across groups. Gender issues and indigenous areas; research include differences in access to land, credit and leadership positions. will begin in early 2000. Judicial Reform Project components include: (a) court administration; (b) legal and judicial * I.egal and judicial training * Design pilot preventative programs (PROJUSTICL\) training; (c) access to justice; and (d) alternative dispute resolution. Possible component includes training of for male aggressors. genider issues include: judges in gender issues, such as * D)omestic violence (women and children); domestic violence programs do the differential treatment of njot address the aggressor. Heiice, male aggressors are likely to repeat their lawyers and clients by gender. violent behavior in subsequent unions. * Access to justice componient oo * Delinquency and violence among (male) yotrth. includes legal services for * Legal issues in labor iniclude wage discrimtination and discrimination against wotmietnadministered by NGOs pregnatnt women. in Quito, Cucenca and Guayaquil. * Access to jUstice and legal process limited for both men and women duLeto A nationwide grant program institutional weaknesses of the judicial system. But women appear to be whicth finiances legal research anid more disadvantaged because they have more limited access to information information, judicial access and and resources to hire legal assistaice, court reform projects is * L.ack of sensitivity of judges to gender-differeitiated legal constraints. supporting an inative to collect - ~~~~~~~~~~~~~~~~~~~nationalstatistics oni social (inicluding domestic) violenice. * Alternative dispute resolution component includes mediation for domestic disputes. Current Gender-Related Recomneudaions Projects Possible Genider Issues Interventions ....______preparationincluded a * Follow-upto see how project design The project,which is currentlyunder preparation,includes two components: (a) * 'roject Rural and SmallCity to examinethe and subsequendyproject institutionalreform at the localand nationallevels' and (b) increasedcoverage for consultancy Water Supplyand dimensionsof the project implementationincorporates results of drinkingwater and sanitation.According to project staff,gender is consideredof gender Sanitation referencefor the consultancy. importatncein the projectsfor the followinigreasons: (a) womenprimarilv (terms of (PRA\GtA.\S) preparedin demand water;,b) women are the primaryusers of water resources;(c) with men consultancy withiWorld Bank migrating,women are left behind in rural areasto asstumeleadership roles in the consultation family;and (d) women are thieprimary agents of changein terms of hygieneand staft). wateruise because of their domesticand caregivingroles. Other gender issues include: * Possibleconflict in water use given that wornenare Likelyto prioritize domesticwater and men may be more concernedabout water for irrigation. * NMaintenanceand sustainabilityof water servicesmay be dependent on those groups who most value water availability,which mayvary by gender. * Mlenare likelyto dominatein terms of decisionson willingnessto pay for water services,although water is more likelyto be valuedby women; migrationmeans men are often not present to makedecisions oni water. e l'Techniciaiiswho work in the sector tend to have engineeringor other technicalbackgrounds and lack trainingon socialaspects of delvcring water and sanitationservices. workingwith CONAMIT * The Bank should facilitateinternational Tlhs technicalassistance loan willadminister a 20-30 percent nationalsample * INEC is 01) Agricultural in 1974).The pilot will to identifyhow best to integrate examplesof census modulesthat NO Informationand surveyin the agriculturesector (the last census took place the largercensuis, as examinegender in the agriculturesector, be completedat the end of the yearand the cenisuswill subsequently be carried gender into CensuLs a module on wiuch CONANIUhas requested. out in 2000. But becausethe census is so broad, the executingagency, INEC, will wellas establish carryout yearlymodules to cover specifictopics, of which genderwould be one. gender. Biodiversity The project involvescollaborating with indigenousorganizations to establishand * Project has attempted to work * Strengthen"women's offices"that Conservation managea "ethnobotanical"park near Puyo. Project activitiesinclude: (a) both with men and women are located withinindigenous (OMAERE) conservation,including training on managingbiodiversity; (b) educationon accordingto their roles. Project organizations. botany,information systems, human resources,project formulation, has also tried to promote women * Encourageboth women's and men's administration,etc.; and (c) scholarshipsfor preparinggraduate theses. The in leadershippositions, for participationin the scholarship project involvesworking with four ethnic groups in the project area. Gender examplethe ConsgioDirectivo del program,and be sensitiveto their issuesidentified by project staff include: Parque,and encouragedwomen's problemsof adapting to external * Womenare responsiblefor collectingmedicinal and forestplants and participationin productive environments. managingtraditional gardens. Men'sprimary responsibilties are activities. * Developactivities to promote men transportingplants and workingin agriculturalproduction, fishing and participatingin domesticchores. hunting. Men alsowork as guidesand guards.Only women have * Increase the availabilityof family demonstratedan interestin botanicalnursefies. planningto the indigenous. * Whilewomen's participationhas been promoted, culturaltaboos often * Addressalcoholism and violence hinder women and men from workingtogether. Other problemsindude amongmen. women'sinability to communicatein Spanishand to write reports, women's time constraintsdue to reproductiveroles, and women's limited leadershipand project managementexperience. * Alcoholismwas cited as a health problem affectingmen's productivity. * Accessto contraceptivesis limited,thus familiesarc largerthan desired; _1______familysize affects povertyand constrainswomen in particular. O> El NinioEmergency Joint World Bank/lDB project that financeshousing for resettledpopulations * Project requiresthat both men More informationrequired to make Project and infrastructureimprovements, the latter being the largerof the two and women be registeredas recommendationson this project. components. The project is currentlyunder supervision. Accordingto project propertyholders but most staff,project activitiesare too narrow. The CHONE resettlementplan includesa couplesare not legallymarried range of women'sdemands in social services,including child care,which the and the project includesno project can not finance(CHONE was the onlyapproved plan at the time that provisionsfor common law interviewstook place). Other possibleissues include: genderwage disparities, marriages. with men reportedlyearning five times more than women for work performed * In the ClIONE resettlement outsidethe home. Socialproblems identified include: familyplanning-in plan, about 90 percent of survey project areas such as Porto Viejo fertilityrates are reportedly6-8 children;and respondents werewomen; few singlemotherhood- the proportion of familiesled by singlemothers is as high as men participated. 50 percentin some areas. Recommendations Projects Possible Gender lswies Current Gender-Related i ______. . ; _ _ . .:____ . . . . - . . ______.I.t. venton s supported by the project Document the experiences of PRONADER Ecuador National I'roject components include: (a) adaptive research and extension services to Actions with respect to gender to learn what worked Rural Development validate and transfer technology to small farmers; (b) land tenure regularization; included: * Forming a team of 12 social under what circumstances. Project (PRONADER) (c) a community development fund to finance small local projects; and (d) counterparts also indicated a need for this training for beneficiaries and beneficiary organizations. Gender issues in the promoters to organize members and target evaluation of experiences. project include: community * Men migrate to the cities thus women are responsible for agricultural women. Specialist in production and orgarization. Male migration has led to the feminization of * Retaining a Gender the agriculture sector. the IICA project team. self-training * Over the life of the project, agriculture or livestock technology packages * Developing a generated and validated have corresponded to male crops. While the manual for project staff on rural project financed a study on adapting technologies for women's crops, no planning and implementing chaniges in the research agenda resulted. development projects for * The project's extension agents, who were attached to the Ministry of women. Agriculture and tend to be technicians, resisted working with women. * Conducting research on Anecdotal information indicates that male extensiornists preferred to wait to women's access to PRONADER provide services on the weekend when men were likely to be present as services. opposed to working with women during the week when men were away * Providing technical assistance on working. women in development, * On land titling, thefjthas de investigad6n,which register the history of the including financing an property and by law are required to register the names of both spouses, international consultant. tend to note only the name of the male partner (common law marriages are * Financing contests to promote work in recognized by law). the visibility of women's productive * Due to lack of experience, women's rural groups require more intensive agricultural and other support in developing and strengthening their organizations and productive activities. events projects than men's groups. * Carrying out learning * Mixed organizations tend to be more culturally acceptable than segregated including sponsoring workshops women's and men's groups. on gender and rural and * Women's productive activities often fail because of the time women must development and gender dedicate to reproductive and domestic roles, such as fuelwood and water project formulation. collection and childcare. * Producing the document * Women's participation in productive activities implies a redistribution of "Strategies, Processes and domestic work between men and women. Guidelines for Incorporating Women in the Design of PRONADER Projects". * Promoting miixed men's and women's groups (initially the project favored women's groups). EducationI The project promotes the SEM model,which creates"clusters" or "networks"of * The project establisheda pilot Given the innovationassociated with the schools for administration,teacher trainingand school-communityrelations. trainingprograms for teachers teacher trainingon stereotypes,examine the T'hemain genderissue identifiedin the project was that of messagesand gender on gender stereotypesin impact of the programin terms of changing stereotypestransmitted through textbooksand teachingmethods, which can teachingpractices. teachers'attitudes and practicesand affectboys and girlsnegatively. In the case of girls,these stereotypescan * The project took measuresto students' performance. underminetheir careerpotential and earningpotential. Maiestereotypes and identifyand eliminategender messagesmay contributeto machistabehavior, including violence and risk taking. stereotypesin competitivebids for textbooks. EducationII The project willinclude: (a) institutionalstrengthening; and (b) improvementsto * Project under preparation. * Designand implementschool the qualityand coverageof primaryand secondaryeducation. Possiblegender retention programs aimed to address issues include: the specificconstraints of both boys * Lowerenrollment levels for girlsin the Andeanregion; girls' educational and girls(e.g. PROGRESAMexico attainmentlower in rural areas. and Programade la Ninia,Guatemala). * Boyshave higher repetitionand drop out rates than girls.Demand (e.g., * Strengthenthe bilingualeducation cost of opportunity)and supply(e.g., boys' dysfunctionalbehavior in programs for both students and school)contribute to this situation. mothers (e.g. Mother to Mother * Reasonsfor not going to school are differentfor boys and girls.For Program,Panama). instance,girls tend to leave schoolto do home-basedwork and boys drop * Expand the gender stereotypes out to enter the paid labor market. teachers'training program,induding * Gender differencesin boys' and girls' academicperformance on examiningparents' attitudesabout mathematicsand language. their sons' and daughters'schooLing. * Higher illiteracyrates and lower educationattainment rates of mothers, c Explore options for career orientation particularlyindigenous and those livingin rural areas,may limittheir role as programs for students. school'seducaton partners. * Genderroles and stereotypesinfluence boys' and girls' choicesin career paths. Education The project willrestructure and strengthenthe Ministryof Educationand Culture. * Project under preparation. * Insufficientinformation to make Management The project may offer opportunitiesto address the impact of genderroles in recommendations. Transformation educationalaccess and performance. TeacherDevelopment The project willstrengthen the frameworkfor pre-serviceand in-serviceteacher * Project under preparation. * Examineteachers salaries vis-a-vis training. Gender issuesin this sectorarc labor issues. Teacherstend to be female. similarprofessions requiringthe same Teachers'pay is low and workingconditions are difficult. Project also offersan level of educationand experience. opportunityto addressgender stereotypesin educationthrough teachers'training. * Explore the option of includinga . module on gender roles in teacher trainingprogram. Health & Nutrition Project objectives include: (i) expanding and strengthening basic health services * Project targets poor women and Project IS near completon, project (FASBASE) (including nutrition and water and sanitation) in peri-urban and rural areas; (ii) childrcn in rural and pcri-urban evaluation should examninegender- promoting community involvement in the organization and financing of health areas across the country. differentiated programs and impacts and services delivery and rural water and sanitation programs; and (iii) institutional identify implementation lessons. strengthening of public institutions involved in the delivery of health services. Possible gender issues include: * Rural and urban poor women have lirnited access to basic reproductive health care and family planning. * Teenage and early pregnancy is high. * Culture and ethnicity influence women's decisions in seeking formal health care for birth delivery, which may lead to maternal morbidity and mortality. * Access to contraceptives cited as a problem among the poor in the Amazon as well as among the mestizos in the Costa. * The use of contraceptive methods by men is extremely low due to the lack of integration of men into reproductive health programs, and cultural values and stereotypes. hlealth Sector Reform 'l'he project aims to decentralize health systems, modernize hospitals and improve * Project will give priority to * Explore options for using community policies, human resources management, and consensus building. Possible gender maternal and child care services. organizations and NGOs to provide issues include: basic services, information and training _ Same as above. in family planning and maternal health. * Gender roles influence the degree, type of exposure and impact of certain * Ensure reproductive health programs health risks. This is the case for HIV/AIDS, hazardous work, mental health target men (particularly male youth) as problems and violence-related accidents. well as women. * Ethnicity combined with gender influence men's and women's access to * Include a strategy on Hf-n/AIDS health care as well as demand for services. prevention in reproductive health programs. Projeets Possible Gender Issues Current Gezder-Reited Recommendartions

Eicuador With the uiltimateaimn of strengtheninggovernance, addressing the growingfiscal * Sinicestudy focuseson macro Monitor resultsof the study to deternune if Deccntralimation.Study and politicaldemands of the diffcrent regionsin the country,and improvingthe institutionalissues, opportunities entry points for issues such as gender exist. provisionof social services,the study will:(a) evaluatethe pros and cons of to raise gender issuesare limited. deconcentrationand decentralizationlaws, as well as some of initer-government revenueand expenditurepractices anid proposals for FEcuador'sdeccntralization; and (b) recomnmendcriteria and strategiesfor decentralizating.Short-term recommendationswill focus on core decentralizationprinciples and entrypoints to produce immediateresults, i.e., pilot projects(e.g., agreements for intergovernmentalcoordination, incremental sector certificationof provincesand municipalities,etc.,.); small-scale administrative measures (e.g., decentralization or deconcentrauonof proceduresthat add transactioncosts when centralized);and consolidationand replicationof existinggood practices. Long-term recommendationswill focus on processesfor incremental,widening consensus buildingaround flexiblescenarios, some of which willrequire legal and other fundamnentalchanges to the roles and responsibilitiesof governments,as wellas intergovernmentalarrangements. Gender issuesrelated to decentralization include: * Men predominatein localgovernance structures; women's voicetends to be more muted than men's when it comes to identifyinglocal level needs. * Gender socializationbegins in the home and the communitythus changing socializationprocesses requires local level initiatives. * Geniderdifferences in preferencesfor socialservices and assistancein productiveactivities. * Gender roles influenceaccess to servicesand opportunitiesto benefit from services,for example,social services tend to bc more accessibleto women and productiveactivities tend to be more accessibleto men. Recommendations Pro.ects Possible Gender Issues Curreat Gender-Related ..- . __._ __._ _ _-__ _ _lterventirnis ._ _ _ .______-_ __._ __._ * In the Municipal Development I Given that the Municipal Development I MTunicipal The Municipal Development I Project finances institutional development and Project, women werc involved in Project is near completion, the project Development I and II mifrastnicture. The Municipal Development Tl Project, which is currently under processes related to should consider comparing municipalities preparation, will support: (a) strengthening institutlional capacity of medium-sized consultation in four pilot with and without gender-based consultative cities in line with the government's decentralization pohcies; (b) financinig high- service provision processes to see if gender affects types of priority infrastructure investments in eligible medium-sized cities; and (c) municipalities. investments selected as well as their impact. privatization. Possible gender issues include: The new project should consider: * Gender differences in preferences of municipal services. 0 introducing gender in municipal * Gender roles influence access to services and opportunities to benefit from planning processes and methodologies. services, for example, social services tend to be more accessible to women * Integrating gender in surveys to and productive activities tend to be more accessible to men. determine who uses and is willing to * Privatization will potentially affect men and women differendy givcn that pay for municipal services. the labor market is segmented by gender. * Examnining options for decentralizing services to the municipal and community level, such as reproductive health for women and men. * Introduce participation mechanisms with a gender focus with a view to making services more efficient, responsive and appropriate. l Source: Review results APPENDIX 2: PRELIMINARY LESSONS FROM THE NATIONAL RURAL DEVELOPMENT PROJECT (PRONADER)

One of the main objectives of the Ecuador National Rural Development Project is to assist rural women. In the project's agricultural technology component, 35 percent of extension recipients have been women to date. But project experience indicates that without dedicated gender staff, this level of involvement can not be achieved or sustained. Project staff also point to a number of other important lessons such as: the need to establish concrete monitorable gender targets and indicators; the importance of providing continuous technical assistance to extension workers in addition to training; the need to address women's reproductive and domestic time constraints in conjunction with supporting their productive activities; and the importance of providing additional support to women to overcome their limnitedorganizational experience. Project staff also questioned a number of the project's strategies, such as its ideological womenin developmentapproach over a true gender perspective, and its strategy to organize segregated over mixed groups.

I. Background:

1. Agriculture remains an important sector in Ecuador's economy. Despite high rural- urban migration, about half of the economically active population works in agriculture. The proportion increases to 60 percent if direct and in-direct employment in agro- industries is included. Although Ecuador's agricultural potential is high, agricultural production is low. Only about 50 percent of its land base suitable for agricultural production and 62 percent of its pasture land is being exploited. With rural incomes substantially below urban levels and high underemployment (about 50 percent), improved agricultural productivity and production are necessary to expand employment and raise income of the rural poor. The Ecuador National Rural Development Project (PRONADER) was designed to address these problems.

2. PRONADER was conceived as a first phase of a longer-term effort to reduce poverty and promote socio-economic development in rural areas. The project aims to: (a) increase and diversify output and improve agricultural productivity in the rural sector; (b) increase employment and incomes of small farmers and improve their standard of living; (c) improve rural infrastructure (roads, irrigation, marketing) and help reduce the risk of flooding; (d) promote grassroots involvement of rural communities and beneficiary groups, giving special emphasis to rural women; (e) strengthen beneficiary organizations and local non-governmental organizations; (f) promote a better use and conservation of natural resources; and (g) enhance institutional capacity for planning and implementing rural development activities.

3. Project components involving beneficiary groups include: (a) adaptive research and extension services to validate and transfer technology to small farmers; (b) land tenure regularization; (c) a community development fund to help finance small local projects; and (d) training for beneficiaries and beneficiary organizations. The project would directly benefit 23,100 families through its productive components and an additional 10,000 through processing, marketing, water supply and community development activities. Although the Ministry of Social Welfare (NISW) has overall responsibility for

- 76 - implementing, coordinating and managing the project, a number of othet agencies also participate in project implementation. The Ministry of Agriculture and the National Institute for Agricultural Research (INIAP) participate in the agricultural extension and research component of the project. The project contracted the Inter-American Institute of Cooperation for Agriculture (IICA) to provide technical assistance on the project.

4. During project design, actions directed towards women were justified on efficiency grounds. First, high levels of male migration from rural areas leave women largely responsible for both productive and reproductive roles. Second, economic hardship has increased the role of women in productive activities. And third, lessons from other projects have demonstrated that women's productive projects are viable, and moreover, that income in the hands of women leads to higher increases in family welfare than when income is controlled by men. Thus PRONADER includes a number of specific objectives related to women. For example, according to the staff appraisal report: (a) a minimum of 20 percent of professional posts should be held by women; (b) the project should establish a team of social promoters to promote women's participation; (c) the project should include staff training to reduce institutional constraints related to providing services to women; (d) the project should include special lines of credit for women; and (e) the project should fund studies on women's roles and the constraints they face in the agriculture sector.

II. Activities and Actions Related to Women:

5. Project activities actually carried out to date include the following:

u Recruitment of Specialized Staff The MSW formed a team of 12 social promoters whose main role on the project was to organize community members prior to agriculture technicians and other technical staff arriving in the community. Promoting women's involvement in the project was among social promoters' responsibilities. In addition, from 1993-96 IICA retained a Gender Specialist in its project team, whose main role was to develop a self-teaching manual on how to work on women's projects. In 1998, IICA recruited a new Gender Specialist, who currently is responsible for overseeing women's activities.

O Training Manual on Women and Rural Development. During the early stages of the project, the IICA gender specialist developed a self-training manual for project staff, which provided guidelines on planning and implementing rural development projects for women.

o Studies on Women in Agriculture. The project supported research to learn about women's access to PRONADER services. The studies aimed to identify mechanisms for increasing female beneficiaries and reducing constraints to their participation. The project also carried out a study in the highlands and coastal areas of Ecuador on how to adapt agricultural and livestock technologies to meet the needs of women. The latter, however, did not lead to policy changes or alterations in agricultural extension and research practices.

- 77 - o Training Events. The project supported a number of training events on rural women's productive and domestic roles, women's constraints in accessing and controlling resources, women's self esteem, participation of women in leadership positions and decision making, and women's microenterprise activities. u Technical Assistance. The project included technical assistance on women in development, including an international consultant who visited Ecuador over the project cycle.

O Contests to Promote Women's Visibilityin the Agriculture Sector. The project included a number of contests to promote the visibility of women's work in agricultural and other productive activities. These contests led to a better understanding of women's work and resulted in strategies for improving women's welfare, independence and autonomy. o Workshop on "Gender: A Path to Human Development'" In 1998, the project sponsored a workshop which brought together project staff from the 12 geographical areas of PRONADER. In addition to sessions on gender concepts and analytical tools, the workshop centered on institutional questions, that is, on how to change staff attitudes and practices so that they work more effectively with women producers.

0 Document on Strategies, Processes and Guidelines for Incorporating Women in the Design of PRONADER Projects. The project developed a document that outlines strategies on integrating women in project activities, provides instruments for carrying out a participatory diagnosis with a gender perspective, and describes tools for identifying gender roles during the pre-identification phase of PRONADER projects.

• Workshops on ProjectFormulation. The project financed a number of workshops for farmers and technicians (male and female) on gender concepts, rural women's roles, needs and demands in the Ecuadorian context, and participatory planning with a gender perspective. About 180 staff and beneficiaries have participated in the workshops.

0 Promoting Women's Groups. The project's community fund component included a special category to finance women's activities. Initially the project favored women's groups over mixed groups, although the strategy changed as time went on.

L Documenting Project Experiences Related to Women. IICA project staff are currently in the process of documenting experiences on reaching women over the last seven years of the project. The exercise aims to have project staff and beneficiaries: (i) reflect on processes and results of the project; (ii) identify constraints and opportunities for reaching women; and (iii) identify, lessons to improve the effectiveness of activities directed toward women.

- 78 - III. Results of Gender Actions:

Table 1: Clients Reached by the Community Fund (1991-1998)

YEAR MALE FEMALE TOTAL % CLIENTS CLIENTS CLIENTS MALE FEMALE 1991 2,489 889 3,378 74 26 1993 1,480 768 2,248 66 34 1994 2,017 990 3,007 67 33 1995 2,892 1,565 4,457 65 35 1996 564 121 685 82 18 1997 1,514 529 2,043 74 26 1998 3,992 1,296 5,288 75 25 TOTAL 14,948 6,158 21,106 71 29 Source: PRONADER

6. Between 1993 and 1995, the program provided support to 165 mixed groups, of which 1,637 beneficiaries were women (gender-disaggregated statistics unavailable) and 40 groups of women only(836 beneficiaries). Specific benefits by gender included: (a) 35 percent of land titles were given to women; (b) 35 percent of farmers who received agricultural extension were women; and (c) 33 percent of producers who received credit were women. According to available statistics, female participation was dependent to a certain extent on IICA technical assistance and involvement of MSW social workers. The number of social workers attached to the project declined as MSW resources became tight. The dip in women's participation in 1996 also coincides with the period that IICA did not have a gender specialist on staff.

V. Lessons:

o Presence of Gender Staff. As suggested by the statistics above, the presence of gender staff was an important factor contributing to women's participation. According to the current gender specialist, MSW social workers, who were initially attached to the project, were also a central force in organizing women's groups and promoting their involvement.

0 Extension Staff. The project's extension agents, who were attached to the Ministry of Agriculture, were unaccustomed to working with women. According to the current gender specialist, early on in the project, they often preferred to visit communities on weekends when migrant husbands were home rather than working with women during the week.

L Research Agenda. Over the life of the project, INIAP has not generated and/or validated agriculture or livestock technology packages corresponding to women's

- 79 - needs. While the project financed a study on adapting technologies for women's crops, no changes in the research agenda resulted. o Gender Approach. According to project staff, the gender approach promoted by the project was perceived as a feminist orientation, which led to rejection and resistance. The approach to gender was also criticized for being too ideological-gender staff were unable to translate concepts into concrete actions. Experience from other projects demonstrates that a more convincing argument is to show how applying gender analysis contributes to project goals. o Lack of Continuous Gender Support to Staff. One of the problems of the training program designed for the project was the lack of continuity. Another constraint is the staff rotation, which was high on the project, thereby reducing effectiveness of gender training that did take place. o Composition of Women's Groups. Project experience indicates that in some cases segregated women's groups are the best strategy for promoting female participation. In other cases, communities rejected separate women's group. Women's activities, when segregated, are also often marginalized or treated as secondary. Hence, the project is now tending to promote more mixed organizations.

• Lack of Organizational Capacity on the Part of Women's Groups. Due to lack of experience, women's groups require more intensive support in developing and strengthening their organizations than men's groups. PRONADER experience shows that without special attention to develop organizational skills, women's groups are not sustainable.

• Support to Women's Reproductive Activities: As with so many other projects, women's productive activities often fail because of the time women must dedicate to reproductive and domestic roles. Thus to be effective, productive endeavors for women must be accompanied by measures to help them decrease their domestic workload, such as improved technologies to reduce time spent on fuelwood and water collection and childcare. Early on, the project entered into an arrangement with the Ministry of Education and Culture to provide childcare services in rural communities.

O Limitations of a Wozmenin DevelopmentApproach. Project experience demonstrates the importance of a true gender approach as opposed to one that treats women as a segregated group. For example, staff indicated that women's projects were often rejected because community members did not understand the rationale for organizing women on their own. Also, changes in women's roles imply that men's roles must change too, thus the need to work with men alongside women. For example, the involvement of men and other community members needs to be established at the outset of an intervention prior to forming women's groups.

- 80 - j Lack of Gender-Based Targets: According to project staff, lack of gender targets was another problem encountered during project implementation. As such, goals to improve women's participation were too nebulous and thus too difficult to monitor.

- 81 -

Distributors of World Bank Group Publications

Pricesand credit terms vary from CZECHREPUBLIC INDIA EulyooPublishing Co, Ltd. PERU SWEDEN cosnty tocountry. Consult your USIS,NIS Prodejna AlliedPublishers Ltd. 46-1, Susong-Dong EditorialDesarrollo SA WennergrenWilliams AR localdistributor before placing an ta0e0kagae2 751MountRoad Jongro-Go ApartIrdo3824. lca 242 OF106 P.0. Bot305 order 13000 Prsgae3 Madras-600 002 Seoul Lima1 S-t17125 Solna Tel:(420 2) 24231486 Tel:(91 44) 852-3938 Tel (822) 734-3515 Tel.(51 14) 285380 Tel (46 8 705-97-50 ARGENTINA Fax:(420 224231114 Fax:(91 44) 852-0649 Fax (822)7329154 Fax:(51 141)286628 FaxP(46 8) 27-00-71 WorldPublications SA URL:http:/Mww.nis.cz/ INDONESIA LEBANON PHILIPPINES E-mail:marl@wwi 06 112Cirdoad deBono7 Are DENMARK Pt. tndiraLimited Librairiedo Liban InternationalBnoksosrce Center toe SWITZERLAND 1 120CiudaddeBuenosAires SamrundsLltteratur JalanBorObudur20 P. Boo11-9232 1127-A Antipolo St, Barangay, LibrairiePayot Service Institutinnel Tel (541) 4815-815 RosenoernsAlIt It P.OBoo 181 Beirut Venezuela C(tm)tes-de-Montbenon30 Fax:(541 1) 4815-8156 DK-1970Frededksberg C Jakarta10320 Tel:(961 91217 944 MakatiCity 1 02 Lausanne E-mail:[email protected] Tel:(45 35 351942 Tel:(62 21) 390-4290 Fax:(961 ) 217434 Tel:(63 2) 8966501: 6505: 6507 Tel.(41 21) 341-3229 AUSTRALIA,FIJI, PAPUA NEW Fax:(4535 357822 Fax:(62 21) 390-4289 E-mail:hsayegh@librairie-du- Fax:(63 2) 896 1741 Fax:(41 21) 341-3235 GUINEA,SOLOMON ISLANDS, URL:hrtp:/1www.sl.cbs.dk iRAN lIlbancomlb POLAND ADECOVan Diermen VANUATU,AND SAMOA ECUADOR Ket~~~~IAbNaaC.Pbihr URL:http:/Avwwwlibrairin-du- InternationalPublinhing Servece EditionsTechniques D.A,Intormation Servics Libri Mundi KhaledEsiamboli Ave., 6th Street LAcmb t. Piekna3 648Whitehorse Road LbeilteninaDaarnAlsNo8MAYSA00-677 Waroawa CH1807Blonay MTchlm361392,0 Vicori RDO.Boo 17-01-3029 RDO.Beox 1574973 Unioersityot Malaya Cooperative Tel:(46 2 6-08 Tel:(61) 392107777Joan LeonMera 851 Tehran151 17 Bookshop,Limited Fax:(48 ) 621-7255 Fax(41 21) 9433605 Fax:(6() 3 92107788 Quito Tel:(9821) 8717819:8716104 PO.Box 1127 E-mail:books%[email protected] THAILAND E-mail:[email protected] Tel:(593 2) 521-606:(593 2( 544- Fan:(98 21) 8712479 JulanPantai Baru URL: CentralBooks Distribution URL http:1/www.dadirect.com.au 185 [email protected] 59780 KualaLumpur http://www.ipscg.wawplAps/export 306Silom Road AUSTRIA Fan:(593 2) 504-209 - KowkahPublishers Tel:(60 3(756-5000 PORTUGAL Bangkok10588 Geroldand Co. E-mail:[email protected] P.O.Box 19575-511 Fax:(603)755-4424 LivrariaPortugal 2 30 Welbuormgasse26 E-mail:[email protected] Tehran E-mail:umkoopOtm.net.my Apartado2601, Rua Do Carm Fax.(66) 237-8321 A-11 len COOEU Tel:(96 21) 258-3723 MEXICO o 70-74 TRINIDAD& TOBAGO TAe(41(1-4- Ruizde Castilla763, Edit Expocolor Fax:(98 21) 258-3723 INFOTEC 1200L ANDTHE CARRlBBEAN Fa:(431(512-47-31-29 Primerpiso, Of.1#2 RENDAs. SanFernando No. 37 Tel-(1) 347-4082 URL:hItp://www gerold.co/at.online QudLo D Col TorielloGuerra Fax:(() 347-0264 ynamAugicnStudies Ltd. QuitoGovemment Supplies Agency 14050 Mexico.D.F ~~~~~~~~~~~~~St.AngustineShopping Center BANGLADESH Tel/Fax:(593 2 507-383;253-091 Oefegan tSotathaie 14050Maxico, 0.F EasternMain Road, St. Augustine MicroIndustries Development E-mal coa@impsaLnmtec 4rt5HancourtRoair Tel:(525) 824-2800 Trinidad& Tobago,Went Indies AssistanceSoeietyi (MIDAS) EGYPT,ARAB REPUBLIC DF Dubli 2acor d Fax:52 5)~~624-2822 ROMANIA uursiSA Tel:(8681 645-8466 Dhi R E tecrInnet.mx CompaniDe LibrariiBcesi Fax:(86 645-8467 HouseS,Rood 16 AlmaotiAgencyAl AhamDithbuion Aency Tel:(35315681-3111 E-maL.itp/nneat mxS.Lisain 2,eco3 miltDernddne DhanmondiFt/Area AlGalaa Street Fan:(353) (475-2670 URBu:/t.etm I.Lpcharine 2,setr3smtl oetrndr,e Dhraka1289 BodcPenaMeior..aere 64 GtD Tnt:(880 2) 326427 Cairn ISRAEL MTeilre aM40c S1)e V.UGND Fant(080 2()3821814128T8 Tel: 20 2 57-86083 YozmotLiterature Ltd. C/RioPanuso, 141-Colo eia Fax:(40 t) 3124000 GustroLtd. Fax.811188(20 (880 2) 2) 578-6823 P.D.Boo 56055 Cuauhtemoc PDBoo 9997, Madhoani Building BELGIUM TheMiddle Cant Observer 3 rohaoaaHanandlar Street 0e:65205eico 533505 RUSSANe Kampala As. dir Ral 202 CioTl(92 58-37Fan. (52T(514-6799 9n. KolpachniyPerealok Tl 264 5 6 1060Brussels aTel:(202(393-9732 Far:(972 3)5285-397 NEPAL Moscow101831 Fax (2564 )251 468 Fax:(32 2) 538-0841 Fax:(20 2) 393-9732 R.O.YInternational InternationalInterntional Service (7 09) 917 49 E-mal:gus@swiftuganda Com FINLAND POBox 13056 JR Ltd.43 [email protected] UNITEDKINGDOM Poblicac005TecnicasAloateeminenKir(okaoppa tat TeldomainAvis 61130ox5443Mierrinno Ltdd PubcacZes TecnicasInternacionais As DteemineaKitjakama TelAvivS1130 SINGAPORE;TAIWAN, CHINA RD.Box 3, Omega Park, Alton. Ltda. I Boo128 Tel:(972 3 6499469 Tel:(9771) 416026 MYAMAR;BRUNEIU FIN-00101Helsinki Fao:(972 264860389 Fa:x(977 1 224431 HomiSpherePa blication Services EHngasirnerGU34l2P 010a aontoPelo 209 (35805-Fan:(121-4435 EURl: royinatvisionnet.il 41 KallangPodding Rood #04-03 Tel (44P142006848 Tel: (55llPa259l664 GERMANaYl-ai: oURLnt 5iFre ttp:/Mww.royint.co.il NErHERLANDS GoldenWheel Building Fun.(44o142009889 Ten.(5512596o644 UaL: pakatilausstkann.m PalespeianAirthority/Middle East De Lindeboom/Tnternationale Singapare349316 E45-415w ail: postm 9sterpiwaolebrlaom Iden lniormationServices Pabliradeshovk - Tel:(65) 741-5166 E-maiL:wtba/[email protected] URL:hrtlp:/Avnwwmuot.br FRANCE PROB,19502 Jerusalem PRD.Boo 202, 7480AE Huaksbergan Fan:(65( 742-93586R:ttp/wwirit.Ou CEdibonsEska; OB Tel (9722( 6271219 Tel:(31 53) 574-0084 E-mail:asegiRte@asianconnectAcom Toe Stationery Office RaxNA:u 48, rue GayLussac Fax:(9722) 8271834 Fan:WllE(31,53(572-9296 NineElms Lane lindnboo@worldonlinen, Gospodarskivstnik Publishing Telond SW8573a8m08 539ChnaBcotk5369Canotek Road Rmpood Centre TAIf7580533o-1nsook22BParis -0o8p ITALYL Servi E-mail:URL:http./lwwwwT c-n4/23 FTel 1267 ) 87382427 Ottawa,Ontario KIJ , Pan:(33-1)54329-91- LicosaCommiGonaria Sansoni SPA URL hnternatiskalA i G cewstai Tel:(613) 745-2665 Fax (3364-254 Via DucaDi Calabria,1/1 batandRL- OFwax(271kucost8e24 745-7660 GERMANY CaseloaPostale 552 N D1000 LjubljanuaLoffics,ow.the-Ftk63an819- E-mail:Fair-(613 UNO-Verlag 50125Firanz 2C 692 Ld .,Z Fax:(386 1) 1330983: 1 ENEZUELA [email protected] PoppeosdorferAllee 55 Tel:(39 55) 645-415 PdivateMolt Bag 99914 Fn 388(3 03 EEUL URL:http://awwwrenouthooks.com 53115 Bonn Fan.(3955) 641-257 Ne aktE-mail: repunsekr(@gvestniksi Tecnri-CienciaLibros, SR.A CNINA ~~~~~~~~Tel:(49228) 649020 B-marl:licosaCItbcc.it Aucklund SOLUTHAFRICA, BOTSWANA GeraCiaGmrcuTmno ChinaFinancial & Economic Pan:( Department URL: s lcir Tel.(64( 524-81519 Far singlletitlan: Nivel: 2,9Car PublishingHouse URL:http://ww-w.uoa-vedag.de JAMAICA Fax,(64 9( 524-0067 DoxfordUniversity Press Southern Fax5035.590016 0. Da FoSi bongJie E-mail:[email protected] IanRandle Publishers Ltd. Africa Beaijing GHANA 206Old Hope Roirad Kinogton 6 Oasi Official VascoBoulevard. Goodwooc ZAMBIA Tel: 06 ral 6401-7355 EppBooks Services Tel 876-927-2065 PD Boo3627 RD Boa12119, Ni City7453 UniversityBookshrop, Uniersity ot Fai: B061b) 64010-7365 P.r. Bo 44 Fan.076-977-0243 Wellington GapeTown Zambia 462 15 31 Tel:(27 21( 5954460 GreatEust Road Gampus ChinaBook import Centre AeH-t3eTUG 0 E-mail:irplclis coiTel: Faxn(64 944) 4()4991972 Pus:(27 21( 5954430 RD.Boo 32379 Chin Ronk Tnx:2321) 7700 JAPAN [email protected] E-mail:oxord@oup o.za Lusaka PDBooin285eai: 22321 770099 EasternBook Service FrsbcilinodT Fan:p/wwoaibosc.z/e (2601) 253 9762 j 285-2798Tel.(509)239260 Nairobi~~~~~~~~~~~~~NE ZEahore400 Tel:(38691)4136834700 32123 Tel (s$t Riaxi:neuonomi.mtv 3-13Hongo 3-chame, Bunky5-ku U w iF (2601(252576 GhineseCorporation tar Promotion GREECE Tokyo113-International SubscriptionService ot Humanities Ppsrro A.Tl(83;38-61NIGERIA RDO.Boo 41095 ZIMBABWE Academicand Baobab Rooks (For I CHINADIVDIRE:ON49KNG,52.You Fang Hu Tong. 35pasotirourarS.ArC21A74 MAAO UFa:Fan:(61 htp254 a)2)3818-0864 wftc irs ThivrsinFa:(2azr7831acastiblwwwmundipenacmCrowns Ruidingterich Graighali N330C22 Xuan5ei Da Jie 10650 AthernaaSt E-ai:(l- 3rdr818re0864 p hr MaronBulding509ch Johannesburg2024 Ltd. Rei(in Te.(01)364-1820s URmaLores tes P PrivateMiBa505Tel: (2711(0880-1440 4Conald Road, Granitesiga Tel-(26617(66072 494 Fan:(30 1( 364-18254 otptR wbL kam.r.- ltbadarn Pan (27110880-6248 RD Ron567 Fan-(06 10166072 494 Fx 3 )3485hfpwwbkom0Ip-V Tel:(234 22( 41-1356 E- sc.aHarare COLOMBIA ~~~~HAITI ebs Fair:(234 22) 41-2056 SP-INa:ins23 scou03 CnoLOMBaceLd CultureDiffusion KENYA SPAINos Tel:Fa263 4 7550135 Carrera6 tNo,51-21 5, RueCatrois AtticaBook Service (E.A.( Ltd PAKISTAN Mondi-FrensairsSAFn23711 ApartaoaAereo 34270 C.P.257 Quaranhouse, Mfangano Street MimeaRook Agency Castello37 SantaldRonalddeD.C ~Port-an-PrincePO.Boo 45245 65, Shahrah-e-Quaid-e-Azum 28001Madrid -a ntaf deBgt0CTel. (589)23 9260 Nairobi Lahore54050 Tel:(34 91( 4363700 Tel285-2798Fan: (57 1 (509(23 4858 Tel:(254 21223 641 Tel-(92 421 735 3601 Fair:(34 91( 5753990 Fax,2HNG8KNG5CHIN;7MCAD (57 1) Fan:(254 24 330272 Fan:(92 424 576 3714 E-mail:[email protected] COTED'IVUIRE HOGKN,CIAOAA RL http-//wwwo,mundiprnnsucoml Centerd'Edition et do Ditfusion Asia2000 Ltd. LegacyBeaks OxfordUniversity Press Mandi-PrensaBarcelona Atricaines(GEDA) Soles& CircslationDepartment LoitaHouse 5 BangaloreTown Conseildo Gent,391 04 P51302 SeabirdHouse Mnzzanine1 SharaeFaisal 08009Barcelona AbiPan541 22-28Wyndham Street, Central RD Boo68077 PDBoa 13833 Tl 3 )4839 Tbld(22n24650246511 Hog ons,CiaNroiKac-730Fan: (34 3) 487-7659 Te:(852)5530-1409 Tel.(254, 2-330853, 221426 Te921 44rc35076-7 TEmal:(34 3) 488-3492c a a.e Fax:25 0567Fan: (225) (852)2526-1107 Fan:(25 (2-330854,5818654 Tel:(92 21) 4-ar:bac47a@64i0ose CYPRUS E-mail:[email protected] E-mail,Legacygform-neL.com F-anl:(922 ThOfc.nt SILA" H MLIE Centertor AppliedReerh UL tp/wwai2O.o.k KOREA,REPUBLIC OF LakeHouse Bookshop CyprusCollege HUNGARY DayaogBooks Trading Co. PakBook Corporation 100,Sir Chittampalam Gardioer 6, PiogenresStreat. Engomi EuroInto Service InternationalDivision AzimChambers 21, Queen's Road Mawalha PD Boo2006 MaroitsogedEurnpa Han 783-20,Pangba Bon-Dong, Lahore Calombo2 Nicosia H-1138oBudapest Socho-ku Tel:(92 42(6363222: 638 0885 Tnl-(94 1)32105 Fax:~~~ (357 Tel:(36 1 35088824,350 80 25 Seoul Fan:(92 42(8636 2328 Pan:(6 r432104 66-2051 F~~ax:(38 r'(35890 32 Tel:(82 2) 538-9555 E-mail:[email protected] E-mail: [email protected] E-mail:[email protected] Fan:(82 2) 536-8025 E-mail:[email protected]

THE WORLD BANK

1818 H Street, N.W. Washington, D.C. 20433 U.S.A. Telephone: 202-477-1234 Facsimile: 202-477-6391

Telex: MCI 64145 WORLDBANK MCI 248423 WORLDBANK Internet: wwwvworldbank.org E-mail: [email protected]

ISBN0-8213-4778-0