Report No. 23442-BR
Brazil Gender RZeview
Issues and Recommendations
January 23, 2002
Brazil Country Management Unit
Poverty Reduction and Economic Management Unit
Latin America and the Caribbean Region
Document of the World Bank



Currency and Equivalents
Currency Unit: The Real (R$)
R$2.34 = US$1 (March 2002)
Weights and Measures
The Metric System is used throughout the report
Fiscal Year
January 1 - December 31
Vice President:   David de Ferranti
Country Director:  Vinod Thomas
.Sector Director:  Emesto May
Sector Leader:    Joachin von Amsberg
Task Manager:     Maria Correia



LIST OF' ABBREVIATIONS AND ACRONYMS
AIDS                Acquired Immune Deficiency Syndrome
AIDS I and II       First and Second AIDS and STD Control Project
BEMFAM              Societal Civil do Bem-Estar Familiar no Brasil
(Brazilian Civil Association for Family Welfare)
CCR                 Comi3sio de Cidadania e Reprodu,co
(Citizenship and Reproduction Commission)
CENEPI              Centro Nacional de Epidemiologia
(National Epidemiology Center)
CNDM                Conselho Nacional dos Direitos da Mulher
(National Council on Women's Rights)
CONTAG              Confederac,o Nacional dos Trabalhadores na Agricultura
(National Confederation of Agricultural Workers)
ECCE                Early Child Care and Education
ECLAC               Economic Commission for Latin America and the Caribbean
ESW                 Economic Sector Work
ENAP                Escola Nacional de Administra,co Publica
(National School of Public Administration)
FLACSO              Facultad Latinoamericana de Ciencias Sociales
FUNASA              Funda,co Nacional de Saude
(National Ilealth Foundation)
HIV                 Human Immunodeficiency Virus
IBGE                Institato Brasileiro de Geografia e Estatistica
(Brazilian Census Bureau)
IBISS              Institato Brasileiro de Inova,co em Saude Social
(Brazilian Institute for Innovation in Social Health)
IDAC                Institato de A,co Cultural
(Cultural Action Institute)
IDRC                International Development Research Center
ILO                 International Labor Organization
IPEA                Institato de Pesquisa Econ6mica Aplicada
(Institute of Applied Economic Research)
MED                 Ministerio da Educagco e do Desporto
(Ministry of Education and Sports)
MNMMR               Movimento Nacional de Meninos e Meninas da Rua
(National Movement for Street Boys and Girls)
MS                  Ministerio da Saude
(Ministry of Health)
NGO                 Non-governmental Organization
PACS                Programa Agente Comunitaria de Saude
(Corrmmunity Health Program)
PSF                 Programna Saude da Familia
(Family Health Program)
PAISM               Programa de Saude Integral da Mulher
ii



(Program on Comprehensive Health Care for Women)
PETI                Programa para Erradicagco do Trabalho Infantil
(Project to Erradicate Child Labor)
PNAD               Pesquisa Nacional por Amostra de Domicflios
(National Household Survey and Research)
PNDS               Pesquisa Nacional sobre Demografia e Sauide
(National Survey on Demography and Health)
PROSAD             Programa de Saude do Adolescente
(Adolescent Health Program)
SAEB               Sistema Nacional de Avaliacio da Educa,co Basica
(National System of Evaluation of Primary Education)
STD                Sexually Transmitted Disease
SUS                Sistema Unico de Saude
(Integrated Health System)
UNDP               United Nations Development Programme
UNICEF             United Nations Children's Fund
WHO                World Health Organization
MNMMR              Movimento Nacional de Meninos e Meninas da Rua
(National Movement for Street Boys and Girls)
MS                 Ministerio da Saude
(Ministry of Health)
NGO                Non-governmental Organization
PACS               Programa Agente Comunitaria de Saude
(Community Health Program)
PSF                Programa Saude da Familia
(Family Health Program)
PAISM              Programa de Saude Integral da Mulher
(Program on Comprehensive Health Care for Women)
PETI                Programa para Erradicagco do Trabalho Infantil
PETI                Programa para Erradicacio do Trabalho Infantil
(Project to Erradicate Child Labor)
PNAD               Pesquisa Nacional por Amnostra de Domicilios
(National Household Survey and Research)
PNDS               Pesquisa Nacional sobre Demografia e Saude
(National Survey on Demography and Health)
PROSAD             Programa de Saude do Adolescente
(Adolescent Health Program)
SAEB               Sistema Nacional de Avahiago da Educagco Basica
(National System of Evaluation of Primary Education)
STD                Sexually Transmitted Disease
SUS                Sistema Unico de Saude
(Integrated Health System)
UNDP               United Nations Development Programme
UNICEF             United Nations Children's Fund
WHO                World Health Organization
iii



Acknowledgments
This report was written by Maria Val6ria Junho Pena and Maria Correia and subsequently revised
by Bernice van Bronkhorst. It is based on a background report prepared by Lourdes Beneria and
Fulvia Rosemberg (consultants) and research work carried out by Angela Umbelino de Souza
Albernaz (consultant). John Garrison, Wendy Cunningham, Chris Parel, Joachim vonI Amsberg
and Geoffrey Chalmers provided comments on earlier drafts. Dan Gross acted as peer reviewer.
Selpha Nyairo and Chris Humphrey worked on the editing and formatting of the report.
While this report has been discussed with institutions and individuals of the Brazilian
Government, the views expressed in this report are exclusively those of the World Bank.
iv



TABLE OF CONTENTS
List of Abbreviations and Acronyms
Executive Summary _                                                        vii
I.    Introduction                                                            1
II.   Overview of Gender Issues and Trends                                   3
(a) Demographics                                                      3
(b) Reproductive Health                                               6
(c) General Health                                                   10
(d) Indigenous Peoples' Health                                       14
(e) Violence and Private Relations                                   15
(f)  Education_                                                      18
(g) Labor                                                            22
(h) Poverty, Structural Adjustment and Social Protection             31
(i)  Governmental/Non Governmental Organizations Dedicated to Gender  34
III.   Summary and Implications of Findings                                  37
IV.   Strategy and Recommendations                                           41
(a)  Overall Strategy                                                41
(b) Health                                                           42
(c) Violence                                                         43
(d) Education                                                        44
(e) Labor                                                            44
(f)  Poverty, Economic Fluctuations and Social Safety Nets           46
V. References                                                                47
Tables
Table l a: Demographic Characteristics of the Population: Brazil and Regions, 2000  3
Table lb: Demographic Characteristics of the Population: Brazil and Regions, 2000  4
Table lc: Child Mortality (Children less than five years): Brazil and Regions, 1996  4
Table ld: Child Mortality (Children less than five years) according to years of Mother's Study
and Region                                                                  4
Table 2a: Use of Contraception by Women Age 15-49 Years who live with a Partner _ 5
Table 2b: Fertility Rate for Women Aged 15-49 Years According to years of Study and
Region 1999                                                                 5
Table 3: Pregnancies At Risk, Space Between Pregnancies and Health Care During
Pregnancy and Delivery, Brazil and Regions, 1996                            7
Table 4: Number of Reported Cases of AIDS, Brazil 1991-98                  12
Table 5: Homicide Rates by Sex and Region per 100,000, 1998                15
Table 6: Violent Homicides and Violent Body Injuries Registered by the Civil Police, Rio de
Janeiro (1991-97)                                                          16
Table 7: Victims of Homicides Registered by the Civil Police by Sex, Rio de Janeiro (1991-
97)                                                                        17
V



Table 8: Illiteracy Rates for Population 15 years and Older by Sex and Region  _  19
Table 9: Years of Schooling for 10 Year Olds and Older, by Sex, 1999, Brazil  19
Table 10: Median Years of Schooling for Individuals 10 Years and Over by Sex and
Region                                                                      20
Table 11: Reasons for Abandoning School Among Females Aged 15-24 Years, According to
Place of Residence (%) Brazil, 1996                                         21
Table 12: Labor Force Participation Rates by Sex and Age, 1985-1999         23
Table 13: Individuals in the Labor Market by Sex and Age, as Percent of the Labor Force,
1996, Brazil and Regions -_                                                 23
Table 14: Unemployment rate, by Gender, Population Aged 15-65, 1996         23
Table 15: Labor Force Participation Rates by Gender and Years of Schooling, 1995  24
Table 16: Employment by Sex and Sector (%), Brazil, 1985-1999               24
Table 17: Median Monthly Wage, Individuals Age 10 and Over, by Sex          24
Table 18: Average Hourly Earnings by Sex and Schooling in R$                25
Table 19: Income by Gender, Ethnicity and Education, Brazil 1996            28
Boxes
Box 1: The Scope of Gender                                                   1
Box 2: The Family Planning Law                                               6
Box 3: The Programa de Assijt6nda Integral d Sau.de da Mulher                8
Box 4: Rural Women's Views on Their Health                                  13
Box 5: Health indicators among Indigenous Children in Rio Grande Do Su_     14
Box 6: Child Labor Conditions                                               30
Box 7: The Lipstick Lobby_                                                  34
Box 8: Integrating Gender in Fundac,o Abrinq's Innovative Programs          35
Graphs
Graph 1: Women who had Children in the Previous Five Years with < 24 Months between
the Last and the Previous to Last Pregnancy(/o), Brazil and Regions, 1996    7
Graph 2: Women who had Prenatal Care in the First Three Months of Pregnancy and During
Delivery (%), Brazil and Regions, 1996                                       9
Graph 3: Percent of Women with High Risk Pregnancies (1991-96)              10
Figures
Figure 1: Work and Age_                                                     26
vi



EXECUTIVE SUMMARY
INTRODUCTION.
This report documents findings of a review of gender issues in Brazil carried out by
the World Bank in 1999 and updated in 2001. It emerged as a response to the growing
recognition - both in the World Bank and in Brazil - of gender as an important issue in
increasing economic and social well being as well as reducing poverty. The review examnines
gender in terms of demographic trends, health indicators, the effects and causes of violence,
education indicators, labor market trends and social protection. Gender relevance and the
composition of the Bank's portfolio were the main criteria for selecting sectors to be
analyzed. The review identifies gender issues across sectors with a view to improving the
Bank's efficiency and effectiveness and reducing gender inequities and inequalities that affect
men's and women's well being.
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 assigned gender roles. However, because the report relies primarily on
secondary sources, and because previous gender work has tended to focus on women,
discussion on male gender issues is limited.
MAIN FINDINGS.
Overall, Brazil has progressed significantly in addressing gender issues and reducing
gender gaps. Three major advances are worth noting. First, women's access to and use of
contraceptives has increased enormously, resulting in a sharp drop in the fertility rate and the
size of households. Second, education levels for women have increased to the point that, on
average, women now have more schooling than men do. And third, although men still
predominate in the labor market, women's participation has steadily increased over the last
two decades, and at the same time the wage gap between men and women has decreased.
These changes can be attributed in part to a prolonged economic crisis in Brazil, which
deeply altered the allocation of men's and women's labor outside the household. Brazil's
rapid democratization, which spurred a demand for equal rights and treatment in the family,
in the labor market, under the law and in civil life, also contributed to greater gender
equality.
However, while Brazil has witnessed many advances related to gender, a number of
issues affecting men's and women's welfare remain. For example:
Mortality rates linked to external factors, such as traffic accidents, homicide and
suicide, differ greatly by gender. During 1977-93, for example, mortality from
external factors increased by 45 percent among men and by 13 percent among
women. For Brazil, the life expectancy gender gap - 12 percent greater for women
than for men - is higher than that of other countries in the region (e.g. Mexico) and
of more industrialized nations (for example, Canada and Greece).
vii



* In terms of health services, pre-natal care for pregnant women continues to be
inadequate. Almost half of Brazilian women who gave birth in the five years
previous to 1996 were at-risk; the proportion of at-risk pregnancies increased to two-
thirds in the Northeast. Maternal mortality is on the rise in Brazil, and the problem
is particularly acute among indigenous women. On the other hand, contraceptive
use among males is minimal - only six percent according to the 1996 Sociedade Ciil do
Bem-Estar Familiar na Brasil (BEMFAM) survey - thereby posing a significant risk in
the spread of acquired immune deficiency syndrome (AIDS) and other sexually
transmitted diseases. Reproductive health programs, however, tend to target women
and exclude men, who are important decision-makers and are affected by family
planning choices.
* Quantitative and qualitative information from non-governmental organizations
(NGOs) and the government, including police reports, indicates that violence among
young adult males, male violence against women and sexual violence against boys
and girls continues l:o be high. Men are the primnary victims of homicides, whereas
both men and women are victims of violent aggressions. In 1998, according to
police registries 93 percent of victims of homicides were men; an equal nurnber of
men and women were victims of violent aggressions. The high incidence of violence
in Brazilian society is taxing on the health care system, and it also affects work
productivity and quality of life. Violence against women also reduces wornen's
ability to negotiate <afe sex and avoid sexually transmnitted diseases and human
immunodeficiency virus (HIV)/AIDS.
*  Studies in Brazil harve concluded that textbooks and teaching methods tencl to
reinforce gender segregation and stereotypes. Men are linked to public life and all its
dimensions of wort, leisure, politics, wealth and power, while women are relegated
to the private sphere of the household. Poverty and inadequate schools have led
boys in particular to enter the labor market at a premature age, with negative impacts
on their education.
* The availability of early childhood care and education programs remains poor.
Because household labor continues to be assigned to women, low coverage of
childcare facilities disproportionately affects women's labor market opportunities,
particularly those of poor women.
*  Although it has narrowed, Brazil continues to register one of the widest gender wage
gaps in Latin America and the Caribbean - women eam 66 percent of what men
earn. Black women are the most disadvantaged of all groups. The gap prevails even
when education and hours worked are taken into account, and indeed increases with
more years of education. Researchers suggest that factors outside the labor market,
such as marriage, children or work interruptions, may explain part of this gap. Given
'that women have higher levels of education, their lower salaries vis-a-vis mnen means
that women's returns to education are lower than that of their male counterparts.
* Analysis of data from the Northeast shows that male- and female-headed households
differ only marginally in the extent to which they are likely to be poor, 39 percent
viii



and 41 percent respectively. Furthermore, male-headed households experienced a
slower reduction in poverty compared to female-headed households. However, when
controlling for education (women have more years of schooling than men) and other
individual characteristics, female-headed households have a much larger likelihood
(46 percent) of being poor than do male-headed households. In addition, households
with young children are more vulnerable and more likely to be poor than households
with no children under the age of five.
STRATEGY AND RECOMMENDATIONS.
The report puts forth a three-pronged strategy. Firs4 it recommends that, having
made significant advances in reforming the legal and policy framework, Brazil now turn to
changing societal gender roles and expectations so that women and men take advantage of
the opportunities provided to them by the law and government policies. Different channels
need to be used if gender-based societal expectations are to change, induding the education
system, the media, the family, peer groups, the community and cultural practices. These all
play a part in socializing men and women and affect the choices they make over their
lifetime. Second, the report recommends acting on gender issues through conmmunity and
local level organizations. Many problems identified in this report, such as substance abuse,
early child-hood education, maternal health care, teenage pregnancy, sexual abuse of
children, reproductive and sexual health care services, would best be addressed by local and
civil society groups. The strategy of working through the local level has several advantages:
(a) local organizations are the closest to the target groups and have a better understanding of
local contexts; (b) working through local organizations strengthens their capacity and
contributes to Brazil's democratization processes; (c) working through local groups takes
advantage of existing capacity and therefore increases efficiency; and (d) the public sector
faces serious budget constraints and is unable to provide the kinds of services and attention
being proposed. And third the report recommends that gender work target men as well as
women, for two reasons: male issues such as violence, alcoholism, under performance in
school unemployment, etc. are key social issues, which can be in part attributed to men's
narrowly defined gender roles; and effective women's programs often require men's implicit
or explicit cooperation and involvement. Specific sector-specific recommendations follow.
HEALTH
Recommendations for the health sector include: reducing Brazil's relatively high
maternal mortality rate; improving access to family planning for the poor; and targeting men
as partners in reproductive and sexual health programs. Increasing tetanus immunizations
for women to reduce the risk of infections during and after pregnancy would be a relatively
low cost measure with high payoffs. And including fathers in programs such as childcare
and nutrition would promote their roles as husbands/partners, parents and care providers.
In terms of family planning, sterlization as a method of contraception should be
discouraged given that it represents a high expense to the health care system, it may lead to a
reduction of condom use and it is not an option for younger women who plan to have
children later in life. Male forms of contraception should also be heavily promoted, given
the low incidence of condom use and male sterilization. In areas such as the Northeast,
actions should focus on reducing the incidence of teenage pregnancy. In terms of the best
ix



vehicles to deliver services, existing government programs, including the PACS (Programa
Agente Comunitaria de Saude) and PSF (Programa Saude da Familia), play an important role. In
addition, local and community-based organizations and NGOs can be mobilized to provide
information and basic servic:es during different stages of pregnancy, particularly during the
first trimester when failure to seek services places women in a high risk category, and to
provide post natal care sucb as information on child nutrition.
Two other recommendations include: designing control and prevention programs for
violence, depression, suicide and risky behavior among men by carrying out epidemiiological
studies on the incidence and gender-related risk factors associated with these problems; and
paying attention to the alaning health problems of indigenous groups which have links to
gender, for example high male mortality rates and maternal mortality/morbidity.
VIOLENCE
Given its incidence and gendered nature, violence should be a priority area for
gender work in Brazil. But addressing violence from a gender perspective should go beyond
dealing with the effects of domestic violence on women, on which Brazil has made
important advances. Over the long term, gendet-related work should turn to violence
prevention by examining how gender affects male violence and identifying steps to counter
the effects of gender roles :nd socialization on violence. The report identifies the education
system, community prograi-ns and the media as vehicles for preventing gender-related
violence. In terms of education, possible interventions include retraining teachers and
eliminating gender stereotypes in textbooks, and developing special programs to teach
children nonviolent conflict 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 provide social
services for victims of violence. Peer group programs, such as the Big Brothers and Big Sisters
Program in North America, have also been found to be important mechanisms for reaching
and reforming at-risk youth.
EDUCATION
Given that Brazil has progressed significantly in terms of increasing girls' education,
attention should now turn to ensuring that boys do not fall behind, as well as improving the
quality of schooling as a whole. On keeping boys in school the report supports e fforts that
are currently underway to examine the gender-differentiated impact of the Bolsa Escola and
Programapara Erradicafdo do Trabalho Infantil (PETI) programs in improving educational
attainment and reducing child labor. In terms of school quality, key measures would be to
reduce gender stereotypes transmitted through education by developing teacher-training
modules on gender and removing stereotypic images and messages in textbooks and other
classroom materials. Such measures would go much beyond affecting the career choices of
men and women; they would contribute to changing socialization processes that lead to
docility and passivity among girls and aggression among boys. Partnerships between the
Ministry of Education and Sports and civil society organizations have been a cost-effective
way of addressing gender stereotypes and socialization processes. Brazil could also learn
from the successful experiences of neighboring Argentina, which in the late 1980s and early
1990s advance significantly in removing sexist language and depictions in textbooks.
x



Because early childhood education represents a future investment in the human capital of
the country, this should also be a priority over the long term. Providing training and
regulation for existing mdes creches (nursery schools) could be one strategy for the short term.
LABOR
Brazil needs to address its persistent gender wage gap, which is larger than that of
many of its less developed neighbors. Ensuring better adherence to labor discrimination
laws by examining current enforcement mechanisms and institutional support available to
female workers is one important strategy; making available information on labor rights to
female employees and employers is another. A further priority would be to counter the
possible negative effects of Brazil's generous maternity leave, which may contribute to
discrimination against women.
Discrimination, however, is only one factor contributing to the wage gap.
Redressing the gender imbalance in the division of household work, so that men and women
share parenting, care giving and domestic chores, thus freeing women to participate in the
workforce and advance on the job, is another precondition to reducing gender-related labor
inequalities. Greater gender equality in the workplace will also require change in the way
work is currently organized by, for example, allowing for more flexible arrangements such as
job sharing. Projects and programs that work with families could also begin to promote
changes to gender roles in the household, following the model of the new 'family capacity
building' program being proposed for Argentina.
Two other recommendations are: having civil society and community-based groups
act as a clearinghouse for jobs as well as provide training on grooming and preparing oneself
for a job, which would benefit poor women in particular; and building up social and human
capital in the poorest communities to address the issue of street children.
POVERTY, ECONOMICFLUCTUATIONSAND SOCL4L SAFETYNETS
Lastly, findings emphasize the central importance of poverty reduction strategies that
reduce barriers for and address the needs of poor women. These include providing access to
childcare and family planning, continuing improvements in education and reducing barriers
to women's participation in the labor market.
Given the permanent state of volatility and risk associated with globalization and the
liberalization of markets, having a better understanding of how households are affected by
economic fluctuations would help to better design social safety nets. The report makes a
number of recommendations related to gender, particularly regarding information analysis.
First, intrahousehold data should be collected and analyzed in a way that better captures the
heterogeneity of household structures rather than categorizing households in terms of the
gender of the household head, which can be simplistic and misleading. And second,
intrahousehold data needs to better capture household dynamics, decision making, coping
strategies and responses to incentives, by household composition and gender roles of
household members. Data collected in Mexico and World Bank research carried out using
that data provide a good precedent that Brazil could follow. Other recommendations
xi



include examining gender-related demand and supply factors associated with any emergency
employment initiative that Brazil considers adopting.
xii



I.      INTRODUCTION
1.      This report documents findings of a review of gender issues in Brazil carried out by
the World Bank during 1999 and updated in 2001. It emerged as a response to the growing
recognition, both in the Bank and in Brazil, of gender as an important variable in reducing
poverty and increasing economic and social well-being. The review examines gender in
terms of demographic trends, health indicators, the effects and causes of violence, education
indicators, trends in labor market participation and social protection. Gender relevance and
the composition of the Bank's portfolio were criteria for selecting sectors to be reviewed.
2.      The review identifies gender
.,  .                     ~~~~~~~Box 1: The Scope of Gender
issues across sectors with a view to
imnproving the Bank's development                    Although grounded biologically, gender is a relational
effectiveness in three aspects: the          category that points out socially constructed roles and relations
efficiency of its operations; the            between men and women. In the words of Simone De
reduction of gender inequities and           Beauvoir, "one is not born woman, one becomes woman."
inequalities; and the improvement of         Becoming women, but also becoming men, are learning
processes, born out of established social patterns; they are
social well being. It also discusses the     enforced through norms, but also through coercion, and they
gender dimensions of poverty and social      are modified over time to reflect changes in the power and
exclusion in Brazil.                         normative structure of social systems.
3.      Gender in this report pertains to        Thus gender in this report refers to aspects of social life
both men and women and refers to the         and poverty which are experienced differently because men
different experiences, preferences, needs,   and women have different ascribed roles. These result in:
opportunities and constraints they face      a   men and women manifesting different preferences,
because of their socially ascribed gender        interests and priorities;
roles and expectations (see Box 1).          *   inequalities and inequities based on whether one is male
However, because the report relies               or female;
*   men and women facing different opportunities,
primarily on secondary sources and               cosrit.    n  hlegs n
constraints and challenges; and
previous gender work - which have                men and women being affected differently by and
tended to focus on women - the                   contributing in different ways to social and economic
discussion on men's gender issues is             development.
limited.                                   I__
4.      DATA SOURCES. The report draws principally from secondary informnation. It is
based on: (a) a literature review of published and unpublished materials on gender in Brazil;
(b) a desk review of Bank-financed projects and economic sector work (ESW) in the areas of
labor, education, health, rural poverty and social protection in Brazil; (c) official statistical
information produced by the Instituto Brasileim de Geo,grafia e Estatistica (IBGE), specifically the
censuses and Pesquisa NacionalporAmostra de Domidlios (PNAD'), the 1996 Pesquisa Nacional
sobre Demografia e Saude (PNDS), and the Sociedade Civil do Bem-Estar no Brasil (BEMFAM); (d)
interviews with governmental officials, academics, representatives of civil society and
women's organizations in Washington, Brasilia, Sao Paulo, Rio de Janeiro and Joao Pessoa,
and World Bank staff based in Washington and Brasflia; and (e) field visits to World Bank
1 PNAD is the Brazilian Annual Household Survey carried out by the Instituto Brasileir de Geografia e Estatistica.
Its sample is statistically representative of all the Brazilian territory, with the exception of the Rural North
Region, which was covered during census years only.



financed rural development projects in Campina Grande, Galante, Cha dos Pereira and
Juarez Tavora.
5.     ORGANIZATION. This report is organized into three parts. The first part provides
the cross-sectoral overview of gender issues and trends in Brazil. The second summarizes
and discusses implications oJ main findings. The final section provides a strategy to move
forward on gender as well as makes sector recommendations related to gender.
2



II.    OVERVIEW OF GENDER ISSUES AND TRENDS
(a) Demographics
6.     POPULATION AND MORTALITY RATES. According to the 2000 census, women
outnumber men in Brazil. From 1980, the female/male ratio increased from 100:98.7 to
100:96.9. Brazil's population is concentrated in urban areas: almost 80 percent of Brazilian
citizens are urban-based. Over two-fifths of Brazilians are black orpardo (mixed race). The
distribution of the population is pyramidal, although its bottom has narrowed and its top has
widened in recent years as a result of both declining mortality rates (a 14-year increase in life
expectancy between 1960 and 1991) and fertility rates (a 60 percent reduction over the past
two decades). Infant mortality is 60.7/1000 live births for children less than five years old,
and (as shown in table Id) increases significantly with fewer years of schooling of the
mother. A child born in Brazil in 2000 had a life expectancy of 68.4 years; the corresponding
figure was 64.6 for boys and 72.3 for girls. The life expectancy gender gap for Brazil is about
12 percent, compared to about five percent for Canada, seven percent for Greece and 10
percent for Mexico.2 In Brazil, income does not influence mortality rates.
7.     The Brazilian population has aged. The 2000 census suggests that the proportion of
the population over 65 years will increase more than the entire population. Between 1940-
1960, the elderly male population increased at a higher rate than the elderly female
population; after 1960, this trend changed as a result of a lower female mortality rate. The
elderly, particularly elderly females, tend to be concentrated in the urban areas (Beltrao et
Camarano, 1997). Tables la, lb and lc show the main demographic characteristics of the
Brazilian population.
8.     As indicated in Table lc, mortality rates are higher for boys than for girls in all
regions. Higher male mortality is more significant for the 15-49 year age group as a result of
external causes such as traffic accidents, homicides and suicides, and it is linked to the higher
exposure of males to risk situations outside the home (Laurenti et al, 1998). Race also
influences mortality trends, with rates being lower for white children.
Table la. Demographic Characteristics of the Population: Brazil and Regions, 2000
Region                Population        Rate of         Rate of    Men/Women
population    urbanization
growth
Brazil                 169,590,693        1.63          81.23         96.87
North                  12,893,561         2.86          69.83         102.43
Nordieast              47,693,253         1.30          69.04         96.13
Southeast              72,297,351         1.60          90.52         95.75
South                  25,089,783         1.42          80.94          97.6
Midwest                11,616,745         2.37          86.73         99.38
Source: IBGE, Demographic Census- 2000
2 Both men and women live longer in Mexico than in Brazil; the life expectancy in Mexico is 69.2 years for men
and 75.9 years for women.
3



Table lb. Demographic Characteristics of the Population: Brazil and Regions, 2000
Region               Fertility rate           L ife expectancy
total men women
Brazil                   2.3                  68.4 64.6 72.3
North                    3.1                  68.2 65.3 71.4
Northeast                2.6                  65.5 62.4 68.5
Southeast                2.1                  69.4 64.9 74.1
South                    2.2.                 70.8 67.1 74.8
Midwest                  2.2                  69.2 66.0 72.7
Source: IBGE, Synthesis of Soial Ind icators 2000
Table lc. Child Mortality (Children Less Than Five Years): Brazil and Regions, 1996
Region                  Male         Female          White          Black
Brazil                   65.5          56.0           45.7           76.1
Urban North               -              -             -              -
Northeast                105.7         86.1          82.8           102.1
Southeast                41.4          32.0          30.9            52.7
South                    36.2          29.6          34.8            47.7
Midwest                  46.1          34.9          31.1            51.4
Source: PNAD, 1996
Table Id. Child Mortality (Children Less Than Five Years) According to Years of Mother's Study
and Region
Region              Less than four years   Four to seven years     Eight years and more
Brazil                    93.0                    47.4                   29.7
North                     50.2                    55.1                   37.4
Northeast                124.7                    69.5                   45.0
Southeast                 62.3                    35.4                   24.4
South                     74.9                    36.2                   18.5
Midwest                   68.1                    36.6                   24.3
Source: IBGE, Synthesis of Social Indicators 2000
9.     FERTILITY RATES. 'Since the 1960s, the historical trend of a stable but high fertility
rate was reversed in Brazil. Total fertility declined from 6.3 to 4.3 from 1960 to 1980 and to
2.4 in 1999. The change was due to the introduction and frequent use of birth control,
particularly among females (although a narrow range of choices in methods was available)
(Carranza, 1994). However, Total Desired Fertility Rate is reported to be 1.9 for Brazil as a
whole (BEMFAM, 1998). This would indicate that there is still a substantial unmet demand
for high quality and reliable family planning services, information and resources. In 1986,
according to PNAD, with the exception of few states, more than 50 percent of women 15-
49 years old and living with a partner were using some form of contraception. A decade
later, the proportion increased to 77 percent. In some regions, such as the Midwest, it
reached 85 percent. As indicated in Table 2a, although male sterilization is a simpler, safer
and less costly intervention, female sterilization had become the most frequent form of
contraception used by Braz:ilian women. However, injectables have rapidly gained in
popularity in recent years.
4



Table 2a. Use of Contraception by Women Age 15-49 Years who Live with a Partner
Region           Any method      Female         Male             Pill        None
sterilization  sterilization
Brazil               76.7            40.1           2.4           20.7        23.3
North                72.3            51.3           0.0           11.1        27.7
Northeast           78.2             43.9           0.4           12.7        31.8
Midwest              77.8            38.8           2.6           21.8        22.2
Midwest              88.3            29.0           3.5           34.1        19.7
South               84.5             59.5            1.8          16.1        15.5
Rio deJaneiro       83.0             46.3            1.0          22.5        17.0
Sao Paulo           78.8             33.6           5.3           21.4        21.2
Source: BEMFAM - PNDS, 1996
Table 2b. Fertility Rate for Women Aged 15-49 Years According to Years of Study and Region 1999
Region             Less than four years    Four to seven years     Eight years and more
Brazil                      3.1                     2.6                      1.6
North                       3.4                     2.9                      1.9
Northeast                   3.4                     2.7                      1.7
Southeast                   2.8                     2.5                      1.6
South                       3.4                     2.8                      1.9
Midwest                     2.3                     2.3                      1.9
Source: PNAD, 1999
10.    TEENAGE PREGNANCY. A review of available statistics shows that fertility among
teenagers 15-19 years old increased from 1970-90 but finally stabilized during 1990-95. In
1995, 13 percent of teenagers between 15-19 years of age had already given birth to a child
according to the 1995 PNAD. Teenager fertility is highest, and actually increasing, in the
North and Northeast, with 24 percent of adolescent females having experienced at least one
pregnancy. Lower school attendance has been associated with pregnancy among youth.
BEMFAM statistics indicate that 54 percent of female adolescents with one year of
schooling are mothers, in contrast to only four percent among those with 9-11 years of
schooling.
11.    FAMILY PLANNING. The public sector did not officially provide access to
contraceptives until 1985. By 1995, however, 71 percent of sterilizations were either
performed in public hospitals or paid by the government through the Sistema Unico de SaAude
(SUS) (BEMFAM, 1996).3 In 1996, 88 percent of women using oral contraception
purchased them in pharmacies (ibid.).
3 The Brazilian health care system is unique in Latin America and comparable to the systems in Canada and
several European countries. As defined in the Constitution and in the SUS law: (a) the system is single-payer;
(b) coverage is universal; (c) health benefits are all-inclusive and free at the time of use; (d) financing comes
mostly from general taxation; (e) funds are allocated to states and municipalities on the basis of negotiated
budgets; (f) most care is delivered by private providers who compete for patients; (g) providers are reimbursed
on the basis of services; and (h) ownership and administration of public services are partially decentralized to
municipalities and states. The principles of the SUS include social participation through consultations with
communities and membership in state and municipal health councils. Private health insurance plays a
significant complementary role for about 38 million people (about a quarter of the population).
5



12.    Until 1996, sterilization was
prohibited by both the penal and medical               Box 2. The Family Planning Law
ethics codes. As a consequence, to be
sterilized, the patient and her doctor
planned a cesarean section -- frequently    *   The right to surgical sterilization (male and female) for
paid by the public health system - after        those older than 25 years and for those younger than 21
which the sterilization procedure took          years old and with two living children.
place. Doctors sometimes received an        *   Access to birth control and counseling by a
additional fee for this service (Berquo,        multidisciplinary team with the aim of discouraging
precocious sterilization.
1996). In 1998, when for die first time     *   Prohibition of surgical sterilization of women during
statistics were made public, the high           birth, immediately after birth, or after an abortion.
number of sterilizations led some to            Exceptions are granted in cases of proven need, for
suspect that a non-explicit racist policy       example for health reasons and earlier successive
was under way, particularly in those            cesarean sections.
wtates wnder way,i proprticu in thofblakse      The obligation for men and women to obtain their
states with a high proporticin of blacks.       partner's consent to carry out sterilization procedures.
The Brazilian Congress thus created a        *  Compulsory notification on the part of the SUS of all
coninission to examine the allegation,          surgical sterilization perforned.
but it was unable to establis;h a link.     *   Prohibition of induced or instigated sterilization.
However, the controversy lied to the        *   Registration, enforcement and control by the Muiistry
regulation of family planning (Law              of Health of institutions and services that perform
family planning procedures and research to ensure
9263/96, summarized in Box 2). The              institutions offer all contraception options, and that
law provides women and men greater              reversible contraception and surgical sterilization are
access to accurate informalion on family        authorized.
planning as well as a broader array of       *  Penalties for doctors who disobey the law.
choices. It also prohibits dloctors from
carrying out female sterilization in conjunction with child birth unless under exceptional
circumstances. The extent to which the law is enforced is not known.
(b)     Reproductive Health
13.    AT-RISK PREGNA ICIES. Almost half of Brazilian women who gave birth in the five
years previous to 1996 were at-risk (see Table 3). In all regions, at least one-third of
pregnancies was at-risk. In the Northeast the proportion was two-thirds. As shown in Table
3 and in Graph 1, spacing between children was only 24 months for about one-third of
mothers. The proportion of women who had not received health care during the first three
months of pregnancy was also about one-third. A 1993 study of 307 deliveries in the Clinica
Obst6trica do Hospital Univer.titdio da Universidade de Sdo Paulo confirms that at-risk pregnancies
are related to mothers' low weight, age (over 35 years old), spacing between pregnancies and
quality of pre-natal care (Spalliacci, 1997). In Brazil, studies have shown that closely-spaced
births negatively affect child growth. Also, as the number of siblings increases, sc, does the
proportion of severely stunted children.
6



Table 3. Pregnancies At-risk4, Space Between Pregnancies and Health Care During Pregnancy and
Delivery, Brazil and Regions, 1996
Area/Region         At-risk   Pregnancies with <24  Women with health care  Women with health
pregnancies  months between the last during the first three months  care during delivery
(%)     and previous to the last  of pregnancy (o%)      ()
pregnancy (%/6)
Brazil               45.5           29.2                 66.0                87.7
Urban                39.8           26.9                 72.7                92.3
Rural                59.1           33.3                 45.7                73.3
Rio deJaneiro        38.7           20.5                 79.4                96.2
Sao Paulo            34.1           24.7                 75.9                96.5
South                37.8           17.4                 79.7                93.1
Minas Gerais and    46.9            30.6                 67.3                94.8
Espirito Santo
Northeast            54.9           37.7                 51.9                76.3
Urban North         49.1            33.3                 55.7                75.0
Midwest              34.4           22.4                 71.7                96.4
Source: BEMFAM, 1996.
Graph 1. Women who had Children in the Previous Five Years with < 24 Months
Between the Last and the Previous to Last Pregnancy (%), Brazil and Regions, 1996
0       10      20       30      40
BRAZIL
Urban Brazil
Rural Brazil
Rio de Janeiro
Sao Paulo
S outhl                     _     li
Central-East (1)
Northeast
Urban North
Central-West
Source: BEMFAM, Pesquisa Nationa/ sobre Demografia e Sa,ide, 1996
14.    USE OF MATERNAL HEALTH SERVICES. Among women who had had children in the five
years prior to the BEMFAM inquiry, 87 percent had used some form of pre-natal health care
services. In 81 percent of cases, a doctor most frequently provided that care. Mothers' level of
schooling, place of residence and age, as well as the birth order of children affect the use of pre-natal
care. Women age 20-34 years old and those experiencing a first pregnancy were more likely to seek
services. The study also showed that almost 50 percent of women visited a doctor more than seven
times. Women in Rio de Janeiro made more frequent visits to health care services (66 percent of
them had at least seven visits in contrast with only five percent who had made no visit at all). In the
4 At-risk was defined as pregnant women less than 18 years and more than 35 years; mothers with more than
four pregnancies; and mothers whose previous two pregnancies were spaced less than 24 months apart.
7



Northeast only 27 percent of women had made seven visits or more, and 26 percent had made
none. Failure to seek maternal health services is strongly associated with maternal mortality.
15.    MATERNAL MORTAI.TY. While no clear consensus exists on the maternal mortality
rate in Brazil different authors           Box 3. The Programa de Assist&ncia Integral a Saude da
concur that the maternal death rate is                           Mulher
too high for Brazil's level of
economic and social development;                The present configuration of women's health services in
that maternal mortality is Ott the rise;  Brazil is a result of the integration of the Programa de sfisllna
andthat maternal mortality can be   Inte,gral a Saside da Mulher (PAISM), created in 1984, and the Sistema
and that maternal moJtaltty can be       Unico de Saude (SUS), which came into effect with the 1988
prevented (Carranza, 1994;               Constitution. Consolidating these two institutions has been
Ministerio da Sauide, 1994; Berquo,      complex and arduous. The PAISM is a top-down Ministry of
1996). According to official 1996       Health program, in contrast to the SUS, which is a decentralized
Brazilian statistics, maternal deaths    system to the municipal level and has a horizontal implementation
were 55.8 per 100,000 live births.       style. The PAISM was designed to offer comprehensive health
care for all women's needs at all stage of life. Over the last years,
The World Health Organization            however, its scope has been reduced. The program now responds
(WHO) and the United Nalions            mainly to women's needs during their fertility years, and
Development Programme (UNDP),           emphasizes reproductive heath. The Ministry of 1-Iealth's present
however, estimate Brazil's mnaternal     priorities on women's health and maternal and infant health are:
mortalit rate t be consderably(a) to improve the quality of care during pre- and post-natal period
mortality rate to be considerably        with a view to reducing maternal and neo-natal mortahty; (b) to
larger, that is, 220 per 100,000 live    overcome prevailing distortions in access to contraceptives,
births. This compares to a rate of 12    including family planning education; (c) to provide cervical and
per 100,000 live births in th e United   cancer screening; and (d) to better integrate programs, particularly
States, 110 per 100,000 live births for  the HIV-AIDS and the Adolescent Health Program (PROSAD).
Mexico nd 63  er 100,00 hveA recent study which examined state and municipal government
Mexico and 63 per 100,000 live           health offices (but covered only capital cities) concluded that (a)
births for Chile. Variations in rates    coverage of health services pertaining to the PAISM was less than
are due to methodological                40 percent in 81 percent of the municipal health offices in the
differences employed and to              capitals, and in 67 percent of the state heath offices; (b) some 38
correction indices applied to control    percent of the municipal health offices and 44 percent of the state
health offices were providing prenatal care for less than 20 percent
under-reporting (Tanaka, 1994), due      of the pregnant women in their respective areas; (c) coverage for
for example to the lack of r egular      gynecological care is less than 10 percent in roughly 25 percent of
transmission of death certiicates by    the institutions studied; and (d) control of cervical cancer covers
the Health Ministry. It its worth       less than 10 percent of the female population in 44 percent of the
noting that in 1995, the Ministry        municipal health offices in the capital and 36 percent of the state
created the "National Plan to Reduce     health offices.
Maternal Mortality" to mouitor           Source: Costa, A.M. Uma Politica de Assistinaa Integral Sasde da
maternal mortality rates.                Mulbera Ser Revatada. (11999)
16.    Maternal mortality is caused by, in decreasing order, hypertension specific to
pregnancy, hemorrhages, puerperal infections and/or abortion (Tanaka,1994). The high rate
of maternal death resulting from infections and hemorrhages in a country where 95 percent
of births take place in hospitals suggests serious problems in the quality of health care.
Another probable cause of infections is that women have not had tetanus shots; however,
insufficient information exists to confirm this speculation. Some authors also associate the
high rate of matemal mortality with the high percentage of cesarean deliveries.
8



Graph 2. Women who had Prenatal Care in the First Three
Months of Pregnancy and During Delivery (%), Brazil and Regions, 1996
20       40        60       80        100
BRAZIL
Urban Brazil
Rural Brazil
Rio de Janeiro
Sao Paulo
South
Central-East (1)
Northeast
Urban North
Central-West
F medical assistance G pre-natal
Source: BEMFAM, Pesquisa Nacional Sobre Demografia e Sa,ide, 1996
17.    The high proportion of cesareans performed in Brazil places pregnant women at-risk
and adds to health costs. It is estimated that more than half of deliveries in Brazil are
performed through cesareans.5 Fundac,o SEADE data indicate that in three regions of Sao
Paulo, cesareans are performed in more than two-thirds of deliveries. In 12 of the regions,
cesareans are carried out in three-fifths of deliveries, and in the other 28 regions, they are
performed in over 70 percent of deliveries (Berqu6, E. 1999). A recent study in the British
Journal of Medicine found that of women interviewed a month before their due-date in
Porto Alegre, Sao Paulo, Natal and Belo Horizante, 80 percent would prefer to have a
normal (vaginal) delivery. This indicates a real discrepancy between what women want and
what actually happens in the delivery room.6
18.    ABORTION. Given that abortion is illegal in Brazil (with the exception of rape cases
and when the mother's health is at risk), it is difficult to establish the extent to which the
practice is performed. According to statistics colected by the Alan Guttmacher Institute in
six selected countries in Latin America and the Caribbean, an estimated 1,443,350 abortions
were performed in 1991 (Facultad Latinoamenrcana de Ciencias Sociales, FLACSO, 1995).
This was the equivalent of 44 per every 100 live births, or an annual rate of 3.7 per 100
women in the 15-49 year age group (ibid.). Based on these data, about one-third of all
pregnancies in Brazil are interrupted voluntarily, which is similar to rates for Chile and Peru,
but greater than in Colombia (one-fourth of all pregnancies) and the Dominican Republic
and Mexico (one-sixth of all pregnancies) (ibid.). In recent years however, the number of
abortions declined markedly. The Alan Guttmacher Institute estimates a decline from 37
per 1000 women in 1991 to 27 per 1(000 women in 1996 in Brazil. The wider availability of
family planning methods and the free distribution of the 'morning after pill' by the public
health service and the easy availability of the illegal abortion pill are thought to be behind
5 The World Health Organization (WHO) recommends that the C-section rate should be no higher than 15
percent.
6 GaZeta Mercantil, December 7, 2001.
9



this decline.7 Indeed the number of hospitalizations due to abortion complications fell from
350,000 per year in the early 1990s to 238,000 in 2000. The numbers of deaths recorded due
to abortion complications also fell from 80 to 27 in this period.8
Graph 3. Petcent of Women with High Risk Pregnancies (1991-96)
20      30       40      50       60
BFlAZL
Urban 13razl-
Rural 13razil
Rio de J- neiro
Sao Paulo
South
Central-East (1)
Northeast
Urban North
Centra- West
Source: BEMFAM, Pesquisa Nacional sobre Demografia e Sagde, 1996
19.    MALE ROLES IN REMRODUCTIVE HEALTH. Family-health and family-planning
programs have been almost exclusively directed at women. Only recendy have reproductive
health programs for men become the focus of attention in Brazil, particularly among
academic and activist groups. Thus it is no surprise that changes in contraceptive use on the
part of men have been slow, even though their use is on the increase. In 1996, despite the
fact that almost all men surveyed claimed to be aware of contraception alternatives, only six
percent used condoms and four percent had had a vasectomy (1996 BEMFAM). Also, many
men are unacquainted with basic reproductive functions and systems, which affects the
contraceptive choices they mnake. According to a study based on 200 in-depth interviews
with low income men and women in the city of Porto Alegre in the southern state of Rio
Grande do Sul, 50 percent of men and women believed that women were fertile during
menstruation while only 15 percent of the sample, the majority women, were familiar with
the current scientific knowledge on fertility (Leal 1998).
(c)    General Health
20.    MORTALrrY. The primary cause of mortality among men and women is circulatory
disease. The second, which has risen over the last two decades, is cancer. According to 1993
data, lung cancer (at a rate of 17 per 100,000), stomach cancer (13.7 per 100,000) and
prostate cancer (9.6 per 10C0,000) ranked among the top causes of death among men. Breast
7 While the pill has been important in helping to reduce abortion-related hospitalizations and deaths, it often
results in partial abortions and other complications when taken without adequate medical supervision.
8Ministry of Health, as cited in IIEJA, November 28, 2001.
10



cancer (12.6 per 100,000) and cancer of the uterus (11.0 per 100,000) are the most frequent
types of cancer among women, and both have increased in the last decade (Laurenti et al
1998). In 1998, the Health Ministry launched a campaign to screen four million women aged
35-49 years as part of a program to prevent cancer. No parallel efforts have been made,
however, in the case of rnale types of cancer. Box 3 provides an overview of the Health
Ministry's strategy on women's health; the Ministry has no comparable plan for men's health.
21.    Gender and age affect causes and costs of hospital admissions in the SUS. In 1994,
psychiatric treatment was the main reason for admitting 25-54 year old men (Nunes and
Piola, 1998). For men in the 25-34 year age group, psychiatric admissions absorbed US$33
million at an average cost per admission of US$347. This cost was only surpassed by AIDS,
which corresponded to a cost of US$600 per patient for this same age group. Among
women aged 15-44 years old, the most frequent cause of admission was obstetric care, with a
high rate of caesarian deliveries and post-natal reproductive system surgeries. In 1994, the
SUS admitted 1.9 million women for normal delivery and 811,000 women for cesarean
delivery at an average cost of US$125.30 and US$191.50 per admission respectively (Nunes
and Piola, 1998).
22.    AIDS. AIDS first appeared in Brazil in 1980, about two years after emerging in
developed countries. The HIV/AIDS epidemic falls into the category of a 'concentrated'
epidemic, i.e. HIV prevalence has surpassed five percent in one or more sub-populations
presumed to practice high-risk behavior, but prevalence among women attending urban
prenatal clinics (presumably a group with low risk of contracting the disease) is still less than
five percent. As of August 1997, Brazil ranked fourth in the world in terms of number of
cases (approximately 116,000). Ranked according to relative incidence, however, Brazil
ranks 40'h with 730 cases per million inhabitants. The number of reported cases of AIDS
rose from 550 in 1985 to 8,201 in 1990, appears to have peaked with nearly 21,000 cases in
1997, but slowed in the subsequent three years. As of 1996, an estimated 338,000 to 488,000
individuals (15-49 years of age) were infected with HIV.
23.   The first wave of the AIDS epidemic was concentrated among homosexual men,
who accounted for three-quarters of those inflicted in 1984. This was followed a few years
later by an epidemic among intravenous drug users, most of whom were also men. Cases
attributed to sharing of contaminated needles rose from three percent in 1984 to 20 percent
in 1995. Later, the disease spread to sex workers, female partners of bisexual men and
intravenous drug users, which caused the proportion of heterosexual cases to rise from two
percent in 1984 to 28 percent in 1995. Of reported cases in 1995, about 55 percent were
attributed to sexual transmission, 20 percent to intravenous drug use, four percent to
contaminated transfusions, three percent to mother-child transmission and the remainder
was unknown. While two-thirds of all cases are still found in the Southeast region, the
epidemic has spread to other areas of Brazil. Every state in the country has been affected,
and nearly half of all municipalities have reported at least one case of AIDS. The average
age of patients has also gotten younger over time. Although data are not available on
income levels, evidence suggests a link between AIDS and education, with lower-income
people being affected.9
9 Information provided by the World Bank-financed AIDS II Project.
11



24.   The incidence of AIDS by gender has changed over the years. Initially, nearly all
AIDS cases were male, but now about 50 percent of new cases are female (see Table 4.)
Women have a greater chance of contracting AIDS for physical reasons. Gender-based
differences in positions of power also make women more vulnerable to AIDS and other
sexually transmitted diseases. Early on, AIDS programs were run mainly by gay men's
groups and targeted mostly homosexual men. However, the focus of programs has changed
over the years as information on risk groups and AIDS trends became available. For
example, the World Bank-financed AIDS II Project develops targeted interventions to sub-
groups at greater risk of contracting HIV, such as intravenous drug-users, sex workers,
adolescents, prisoners, truck drivers and others, in addition to homosexual men. Moreover,
preventive activities are oriented to both women and men as well as low-income
populations.
Table 4. Number of Reporte<l Cases of AIDS, Brazil, 1991-98
Year           Men               Women
1991          9.616               1,992
1992          11,815              2,890
1993          12,905              3,623
1994         13,771               4,179
1995         14,455               4,944
1996          15,029              5,976
1997          14,452              6,473
1998          12,662              6,112
Source: IBGE, 1999
25.   ALCOHOLISM AND SUJBSTANCE ABUSE. Although statistical information is
notoriously inadequate, development specialists highlight that Latin America and the
Caribbean has one of the world's highest incidences of alcohol consumption10, with a rate
three times greater than the r-est of the world (Londono, 1996). Within Latin America, Brazil
has experienced one of the highest increases in alcohol consumption (World Bank, 1997a).
From 1970 to 1989, alcohol use increased 242 percent in Brazil, compared to 72 percent in
Colombia, 45 percent in Costa Rica and 21 percent in Mexico. In contrast, Argentina,
Venezuela and Chile experienced drops in consumption (ibid.). Annually, alcohol leads to
two million deaths worldwide. According to Cercone (1993) (as cited in World Bank,
1997a), alcohol consumption is expected to continue to rise due in part to aggressive
marketing efforts on the part of beverage companies, which have lost markets in more
industrialized countries.
26.   According to a Valladolid University Hospital study carried out between 1980 and
1984, alcohol abusers and alcohol dependent inpatients tend to be poor men. The research
conducted with records frora 150 cases of the hospital's Psychiatry Department indicated
that patients are typically unskilled blue collar workers who tend to have low levels of
education (primary education only), be in their mid forties, be married, live in urban areas
and have low incomes (Conde Lopez, 1990). NGOs confirm that alcohol abuse is
reportedly much greater among low income, poorly educated men. While reliable data is not
available, there is some evidence that drug consumption is considerable and on the rise,
particularly among young people. The use of crack cocaine and injectable cocaine appears to
10 Consumption measures were based on sales.
12



have become common amongst marginalized groups such as prostitutes and those living on
the streets. Cases have been reported of children as young as seven and eight years old using
crack and injectable cocaine.
27.    Health Ministry statistics indicate that costs associated with patients admitted with
illness directly provoked by substance abuse, including alcohol, drugs and tobacco,
represented one-third of the total costs of mental health admissions in 1995, amounting to
US$ 113 million. Additional costs related to substance abuse include compensation paid for
lost days or work under the format of auxilio doenfa (sick pay).
28.    INTRA GENDER DIFFERENCES: HEALTH RISKS AMONG RURAL WOMEN. Rural
workers experience specific health problems and risks relative to their urban counterparts.
An International Development Research Center (IDRC) funded study that sampled rural
women in Brazil's Northeast
Region showed that women                 Box 4. Rural Women's Views on Their Health
work more hours than men
and, as such, are more          *  'Women get ill more than men. If men did the same work as
women, they would get iU too." (Woman sugarcane worker)
vuhierable to health risks.     a  "Looking after beans is more a woman's job than a man's. We
Headaches, which women             work with sacks on our backs and plant beans as we go. My
ranked as their most frequent      shoulder sweUs up and my spine is kiling me because, in addition
ailment, can be caused by          to the weight of the sack, I have to work hunched over the whole
fatigue, emotional stress,         time to pick beans." (Woman bean worker)
*   "If I had a fuil-length skirt, a long-sleeve shirt, gloves and boots,
infections, excessive              for the thoms, I would not have health problems. The landlord
exposure to sun and toxic          should provide all of this but he does not." (Woman rice worker)
substances (Fischer and         *  "I drink the warm water from the channel. I can't bring water from
Albuquerque, 1997). This is        home because I live far away. I think the water is bad for you.
frequently the case for those      There are lots of snails in the channel and they carry disease. Warm
working in irrigation areas,       water and spraying brings a lot of sickness to women." (Woman
tomato worker)
independent of their sex. In    *  "My grandmother's sister (a tomato worker) died because of the
the case of women, their           poison. It gave her a urinary infection. The doctor said she got the
domestic chores add to             disease from planting. He said it was a disease of the uterus. She
fatigue and other symptoms.        died of cancer." (Woman tomato worker)
For example, 60 percent of      Source: Fischer and Albuquerque, 1997
women interviewed indicated
that they suffer from frequent headaches. Harvesting and planting - work activities that
most require laborers to bend down - are typically performed by women. These cause
chronic back pain and spinal cord problems. Gynecological problems suffered by women in
irrigation areas are also common. Lack of clean water or the presence of water
contaminated by agrochemicals in the workplace provoke headaches, nausea, loss of
appetite, parasites and itching among women. Schistosomiasis and schistosomiasis-related
infections are also a problem in these irrigation areas. Women are more exposed than men
to the risks of schistosomiasis infection because they are responsible for clothes and dish
washing as well as for fetching water, which places them in greater contact with the source
of infection. Planting and transplanting rice in Rio Sao Francisco, another female task,
require women to stay in the water while working. Box 4 illustrates how rural women in the
Northeast relate work to their health.
13



(d)    Indigenous Peoples,'Health
29.    INDIGENOUS MORTALITY
AND FERTILITY RATES.                 BOX 5. Health Indicators among Indigenous Children in Rio
Information and statistics on                            Grande do Sul
indigenous health is clearly
inadequate. Available evidence             Results from research carried out by the Coordenafdo Nacional
-aeut.A in,  do Indio/Fundapio Nacional de SaAide point to serious problems in the
however, indicates that the         health status of indigenous children in the Guarani and Kaingang
mortality rate for the Amazo nian   territory (reserva) of Rio Grande do Sul, which is one of the most
indigenous population is strilingly  developed states in Brazil. Out of the 905 children examined (440
higher than the Brazilian national  male, 465 female), mothers of 53 percent of children had not
average. Moreover, important        received pre-natal care and only 16 percent had pre-natal care during
the three first months of pregnancy; 28 percent of mothers had not
gender differences exist among      any type of help during delivery; and two-thirds of mothers have not
the indigenous. Tiroy6 women's      been immunized against tetanus, which is associated with neonatal
life expectancy, for example, is    mortality. A deficit height/age (stunting) was prevalent among 47
almost 50 percent higher thana for  percent of children, two times more than in the Northeast, the
males. Mortality rates among        poorest region of Brazil; and the deficit weight/age was prevalent
among 16 percent of chldren, one and a half times higher than in the
indigenous groups are highet than   Northeast.
the rates for the North of Brazil
as a whole, where the majority of   Source: Fundaftio Nacional de SaAide (1999) and Mapa da Fome Eantrr as
the indigenous people live. AIDS    PopulafJes Incgenas no Brasil (1995)
has become a serious health
concern, and suicide is an increasing health risk. The fertility rate is also very high. Fertility
and family size are closely related to poverty and malnutrition. Despite the diverse cultures
among indigenous groups in the Amazon, ethnographers concur on factors associated with
high fertility rates: girls begin sexual activities at quite young ages; reproductive activities
among women are extended over the course of almost their entire fertile cycle; positive value
is placed on large families; marriage is almost universal for both men and women; and
because of high infant mortality rates, families have more children to compensate for the
dead (Coimbra Jr. and Santos, 1994).
30.    AIDS AMONG THE INDIGENOUS. AIDS is an emerging issue among indigenous
peoples in Brazil. The epidejic is thought to have spread to Brazil from the Yanomnani,
who circulate freely between Venezuela and Brazil and who respond to epidemics by
migrating (in 1990, the WHC) reported 1,061 AIDS cases in Venezuela). Miners who travel
between southern Venezuela and Brazil are believed to have introduced AIDS in areas close
to the Venezuela border. AIDS is reported to have become common among miners of
Mato Grosso, who infected the local population through prostitution. Indigenous women
of certain ethnic groups are at the same, or greater risk as the spouses of miners, because of
the incidence of rape and prostitution. Many different social and economic arrangements
exist between indigenous women and migrants to their areas. Having temporary wives is
one such arrangement. Among some indigenous peoples, such as the Yanomani, women
who engage in sexual intercourse with outsiders face little social reprobation, and as such,
AIDS transmission has increased.
31.    MEDICAL COVERAG -." While the Brazilian health system as a whole has been the
subject of major reforms in recent years, specific efforts have been made to increase
culturally appropriate medical care for indigenous groups. Special Indian Sanitary Districts
14



(Distritos Sanitdrios Espedais Indigenas) provide this care; they are controlled by District
Councils (Conselhos Distritais), and are organized with representatives from state and
municipal governments, NGOs and indigenous communities. However, access to medical
care remains very inadequate, both in the indigenous communities themselves and in the
wider community.
(e)    Violence and Private Relations
32.    Both men and women face violence in Brazil. Data indicates that prevailing types of
conflict vary by gender, resulting in a bipolar public/domestic pattern. Men are more
involved in violence related to labor conflict and crime. Women are more likely to be
involved in conflict related to conjugal break-ups and other types of domestic conflict. The
same patterns of public (male) victimization and private (female) victimization emerge once
the gender of the victims and aggressors are taken into account. For example, aggression
against men tends to involve acquaintances as well as strangers, but rarely family members.
On the other hand, women are attacked equally by acquaintances, strangers and relatives.
(PNAD, 1988). As shown in Tables 5, 6, and 7, violence in general and specifically
homicides involving men are more frequent than those involving women; however, women
and men account for about the same proportion of victims of violent acts causing bodily
harm. In addition to the economic costs (e.g. foregone earnings) and the more obvious costs
to the judicial and health systems, fear and intimidation may deter women and men from
engaging in productive activities.
33.    MEN's EXPOSURE TO VIOLENCE. Men's exposure to violence begins at an early age.
Boys in the 0-9 age group face violent situations more often than girls, and do so in a variety
of settings, including the home, school street, public transportation and sports centers. This
suggests that gender socialization makes violent behavior more acceptable when practiced by
males. According to Jorge (1998), mortality rates linked to external factors (traffic accidents,
homicide and suicide) are very different according to gender: 114 and 25 per 100,000 men
and women respectively. This pattern of male/female mortality in Brazil is similar to that of
other countries (Laurenti et al 1998), with a higher incidence of homicide among young men
(161 per 100,000 for the 15-49 year age groups).
Table 5. Homicide Rates by Sex and Region per 100,000, 1998
Region                          Males                   Females
Brazil                           48.1                     4.3
Urban North                      35.1                    3.8
Northeast                        35.0                     2.6
Southeast                        67.2                     5.6
South                            26.1                    3.5
Midwest                          45.9                     5.4
Source: Health Ministry/Fundafao Nacional de SaAde/ Centn National de Epidemiologia
34.    SUS hospitals confirm the frequency of violence among men, and particularly among
male youth. For exarnple, violence is the main cause of admnission to SUS facilities among
male 15-24 year olds. In 1995, close to 108,000 of male admissions to hospitals (seven
percent of the total) were due to poisoning and other physical aggressions, while the
15



equivalent number for women was 36,000 (Laurenti et al, 1998). The prison population in
Brazil is also predominantly male - 95 percent of 170,000 (Censo Penitenciario, 1997).
Table 6. Violent Homicides and Violent Body Injuries Registered by the
Civil Police, Rio de Janeiro (1991-97)
Absolute number         Percent
Violent Homicides
Total victims                           50,729              100.0
Men                                     47,065               92.8
Women                                    3,664               7.2
Violent Body Injunies
Total victims                          250,197              100.0
Men                                    125,197               50.0
Women                                  125,139               50.0
Source: Pokiia Civil/Reisftros de Oaryncia in Garotinho (1998, p.115)
35.    DoMEsI1c VIOLENC:E. Within the family, violence is related to the subordination of
women by men on the one hand, and on the other, of children by adults."1 Domesti.c
violence is widespread in Brazil, as in other Latin American countries and indeed the
developed world.12 Barker's 1998 study of adolescent men living in poor areas of Rio de
Janeiro provides important insights into men's perceptions of violence against women and
the relationship between male violence and demonstrating manhood. The study of 58
adolescents (15-19 years) and 32 young men (19-30 years) living in low income
neighborhoods andfavelas ccncluded that: (a) being a man is related to working harcl, being
responsible, being the financial provider and being sexually active; (b) becoming a man is a
public act, something that has to be shown to others: "To be a manyou have to be conplete,you
can't stray or slp - like not being a man, beinggay or something. You can't have afeminine side and be a
man"J (c) the basic tone of male-female relationships is one of mistrusting women; and (d)
violence against women reptesents cowardice, but in some circumstances is understandable
and acceptable. One young tnan interviewed explained an unwritten rule about cohabitation:
if men are financial providers, they can expect certain things from the woman in return, e.g.
fidelity, childcare and housekeeping. A man is perceived to have the right to use violence
against her if she does not comply with these unwritten norms. According to the
interviewee, '"a man is allowed to beat a woman if she doesn 'tfeed the children, when she gossips all the
time and when she doesn't clean t,5e house".
36.    Brazil has taken several initiatives to address violence against women, including: (a)
creating 200 police stations for women, staffed and directed by women (120 of these stations
are located in the state of Sio Paulo); (b) training individuals on physical and sexual violence
The problem is not limited to mten; women are also violent with their children.
12 According to WHO data, in Switzerland, 20 percent of women reported having been physically assaulted
(based on a sample of 1,500 women aged 20-60) and in the United States, 16 percent of women reported
having being physically abused by spouses according to a nationally representative sample. In Latin America,
26 percent of Chilean women repcrted at least one episode of violence by a partner in a representative sample
of women in Santiago, age 22-50. In Mexico, 30 percent of women in a representative sample of 650 women
with husbands or partners from Metropolitan Guadalajara reported being physically abused by their partner
within the last year of the survey ( VHO,1999).
16



against women in sectors such as the police, the judicial system and health; (c) carrying out
awareness raising campaigns through media (while television, and particularly Brazilian soap
operas, or telenovelas, reportedly continue to glamorize violence against women); (d) working
to change the law to distinguish between sexual violence against a person and sexual
behavior that violates customary norns, such as exhibitionism; (e) creating shelters to
support victims of violence and their children; (f) establishing a 20 percent quota in a
Prefeitura de Porto Alegre housing project for female victims of domestic violence; and (A)
providing psychological support to aggressors in an effort to prevent the repetition of
violent acts.
37.    Sexual violence has been on the agenda of the women's movement in Brazil since the
late 1970s. It is presently a key priority of NGOs, the National Council for Women's Rights
(CNDM) and other women's councils operating at the state and municipal levels. The state
of Rio de Janeiro has developed an integrated violence prevention plan which has involved
creating a Council of Public Insurance, investing in investigative intelligence, reforming civil
and military police systems, establishing a working program and other alternatives for
prisoners, reducing violence against women, searching for alternatives to incorporate
perpetrators into civil life (targeting youth in particular).
Table 7. Victims of Homicides Registered by the Civil Police by Sex, Rio de Janeiro (1991-97)
Year                  Men                        Women                   Total
Number         Percent      Number         Percent
1991           6,691         92.8           521           7.2            7,212
1992           6,750         93.6           461           6.4            7,211
1993           6,691         93.4           469           6.6            7,160
1994           7,342         93.0           553           7.0            7,895
1995           7,303         93.1           539           6.9            7,842
1996           6,308         92.0           546           8.0            6,854
1997           5,980         91.2           575           8.8            6,555
Source: Policia GOtllRegistros de Ocorrincia in Garotinho (1998, p.1 15)
38.    VIOLENCE AGAINST CHILDREN AND CHILD SEX ABUSE. As in some other Latin
American countries, child prostitution is conspicuous in tourist areas of Brazilian coastal
cities. In cities such as Fortaleza and Salvador, sex trade among children has become blatant.
According to the United Nations Children Fund (UNICEF), poor black girls suffer the
worst abuse. INTERPOL also reported that in 48 cases of missing children who were
traced by the agency, 23 were from Brazil and 16 of these had been raped before being
killed. In recognition of the problem, the Brazilian Government launched a strong
campaign against sexual abuse of children, with airports and hotels displaying signs stating:
'Exploitation of minorsfor sexual tourism: Beware, Bra#il is watching,you"
39.    An evaluation of sex abuse among children in several municipalities of Mato Grosso
conducted by the Instituto Brasileiro de Inovafdo em Sauide Social (IBISS), made the following
recommendations with regard to children's sexual abuse and child prostitution: (a) create a
data base about sexual abuse among children and teens; (b) support police investigative
processes; (c) establish a network to assist victims; (d) support Conselhos Tutelares (uridical
institutions responsible for applying laws related to children); (e) put in place awareness
campaigns about the magnitude and seriousness of the problem; and (f) provide training to
education, health, and social welfare professionals to enable them to identify and treat sexual
17



abuse (IBISS, 1998). Recomnendations were presented at the UNICEF-supported Forum
to Fight Sexual Abuse of Children and Teenagers, which took place in Campo Grande, Mato
Grosso in 1996.
40.   ALCOHOLISM AND VIOLENCE. The effect of alcoholism on violence has received
very little attention in the literature; however, available information suggests a strong
correlation between the two. For example, a Mexico study found that almost 50 percent of
those convicted of homicide admitted to having consumed alcoholic beverages prior to
committing the crime (World Bank, 1997a). Special units for domestic violence in the state
of Rio de Janeiro have also found alcohol to be strongly linked with incidences of violence.
As previously stated, Brazil has experienced one of the highest increases in consumption
rates in Latin America and the Caribbean, and alcohol abuse is significantly higher among
men than women.
(6)   Education
41.   TRENDS IN EDUCATION. As in other parts of Latin America and the Caribbean, and
in contrast to developing nations outside the region, women's education levels are higher
than men's in Brazil. As Table 8 shows, female illiteracy is the same as male illiteracy for the
population over age 14. Table 9 indicates that in 1999, just 14 percent of men age 10 and
older did not have any schooling, also according to PNAD. The proportion for women was
lower (13 percent). However, 18 percent of men had more than 10 years of schooling,
compared to 20 percent of women. In 1990, the probability of enrolling in school was 85
percent for a white child, and only 65 percent for a black child; the probability of a white
child finishing the first grade was 57 percent and dropped to 36 percent for a black child
(Sant'Anna e Paixao, 1999).
42.   The median years of schooling for individuals aged ten and over is slightly higher for
women (5.7) than for men (5.5). As Table 10 illustrates, there are regional differences in
schooling, with the Northeast lagging behind other states. Also, differences between male
and female schooling is greai:est in the Northeast and disappear almost entirely in the
Southeast. In 1995, the combined primary, secondary and tertiary enrollment ratio For
women was 72 percent, which was just slightly above that of men (69 percent). The gender
gap is smaller than in the United States (98 percent for women and 93 percent for nmen),
Argentina (80 percent for women and 69 percent for men), Uruguay (80 percent for women
and 65 percent for men) and Colombia (71 percent for women and 63 percent for men).
But the gap is larger than that of Canada (100 percent for both) and Norway (93 percent for
women and 92 percent for men).
18



Table 8. Illiteracy Rates for Population 15 Years and Older by Sex and Region
Region                        Total                 Men              Women
Brazil                         13.3                  13.3              13.3
North                          11.6                 11.7               11.5
Northeast                     26.6                  28.7               24.6
Southeast                       7.8                  6.8               8.7
South                          7.8                   7.1               8.4
Midwest                        10.8                 10.5               11.0
Source: IBGE, Syntbeis of Soal Indicators, 2000
Table 9. Years of Schooling for 10 Year Olds and Older, by Sex, 1999, Brazil
Year                              Total         Men          Women
Without education/less than one year  13.4      13.6           13.2
1 - 3 years                        18.3         19.4          17.2
4 - 7 years                        34.2         34.5          34.0
8 - 10 years                       14.6         14.5          14.7
11 years or more                   19.0         17.5          20.4
Source: PNAD, 1999
43.    BoyS' EDUCATIONAL PERFORMANCE. A number of factors might explain why boys
have under-performed girls in school in comparative terms. Many of these are related to
gender roles and expectations. For example:
*  Culturally, girls are expected to need more protection than boys, and schools are
seen as more protective than the street or labor market.
*  Traditional 'female' behavior is more compatible with the school environment.
Principals of Brazilian public schools are centered on the firm role of teachers
who demand that students be passive and reward obedient behavior, order and
cleanliness; girls tend to be socialized to act in this way and hence feel more
comfortable in school settings.
*  Boys feel greater pressure to work outside the home, particularly if they do not
perform well in school.
*  Women may feel compelled to excel in school to compensate for discriminatory
practices in the gender-segregated labor force; women see schooling as
instrumental in upgrading their skills.
*  In both rural and urban areas, domestic work (for which girls tend to be
responsible) is more compatible with school schedules than outside work (which
boys tend to carry out). To compensate, boys often enroll in evening courses,
particularly boys living in urban areas. Evening courses tend to be of poor quality
and to have high failure and attrition rates (Franco and Zibas, 1997). Women try
to avoid night classes for safety reasons.
19



Table 10. Median Years of Schooling for Individuals 10 Years and Over
by Sex and Region
Regions                    Males           Females
Brazil                      5.5              5.7
Urban North                5.3              5.6
Northeast                  3.8              4.5
Southeast                  6.3              6.4
South                      6.0              6.1
Midwest                    5.6              6.0
Source: PNAD, 1996
44.   GiRLs' EDUCATIONAL PERFORMANCE. In terms of girls' performance, available data
suggest girls are not in school for a number of reasons (see Table 11). The most important
of these are related to poverty, such as lack of resources to pay tuition or girls being pulled
out of school to work or help their families in other ways. A second is related to
characteristics of the school system, for example, the inaccessibility of schools or the
inability to attract female students. Accessibility is a problem particularly in rural areas, and
is the single most important factor explaining differences in enrollment rates between urban
and rural young women. The proportion of those who abandoned school because they need
to help their families is smaD! in both rural and urban areas, but nonetheless it is twice as high
in rural areas. Finally, the proportion of girls who dropped out because of work is about the
same in both urban and rural areas, as is the proportion of those who dropped out because
of lack of interest.
45.   GENDER ROLES TRANSMITTED THROUGH EDUCATION. Some studies have
concluded that Brazilian text books tend to reinforce gender segregation and 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 and Amado, 1989). Some
studies also show gender stereotypes to be intertwined with racial ones (Telles, 1987; Pinto,
1989). This issue has been taken up by some national and municipal institutions, and recently
by the CNDM. These entities have called for the elimination of sexism in books that are
bought and distributed through the Ministry of Education and Sports (MED).
46.   GENDER DIFFERENCES IN ACADEMIC PATHS. Despite increases in schooling for
women, they continue to follow different scholarly paths than men. Women tend to take
courses in the humanities, leading to low wage professions typically considered 'feminine'
such as teaching. Men, on the other hand, are rarely found in professions related to
humanities. Likewise, school tests indicate that, from fourth grade on, boys do better in
math and girls excel more in Portuguese, according to a 1995 report by the Sistema lacional
de Avaliafdo da Educajdo Bdsica (SAEB). However, more research on this subject is needed
given that girls' advantage in Portuguese seems to disappear during high school years (SAEB,
1995).
47.   Conversely, technical night courses distant from homes reinforce the tendency for
women not to accumulate technical skills (Barroso and Mello, 1975). The low participation
of women in technical courses has not changed much over time. A study of metropolitan
Sao Paulo shows that in 1991, women represented just over 10 percent of the students
registered in technical training, up from eight percent in 1980. When women do follow
20



more specific training, they tend to concentrate in traditional feminine sectors such as
garment, shoe and food processing (Matesco and Lavinas, 1994).
Table 11. Reasons for Abandoning School Among Females Aged 15-24 Years,
According to Place of Residence (%) Brazil, 1996
Reasons                 Rural         Urban         Total
Attending school                      34.6           50.4         47.7
Became pregnant                        3.5            5.3          4.9
Got married                            7.4            5.7          6.0
Taking care of their own child         1.7            2.4          2.2
Needed to help family                  4.8            2.1          2.5
Couldn't pay tuition                   1.0            3.2          2.8
Needed to work                         9.1           10.1          9.9
Graduated, schooling                   2.0            3.1          2.9
Bad grades                             0.5            0.9          0.8
Didn't like school                     9.6            7.8          8.1
Access to school difficult            21.6            3.3          6.6
Medical reasons                        1.2            1.1          1.1
Other                                  2.8            3.9          3.7
Don't know / no response               0.2            0.6          0.6
Total (/o)                           100.0          100.0         100.0
Number                                765           3,517         4,282
Source: BEMFAM (1996:31)
48.    EARLY CHILDHOOD CARE AND EDUCATION. A very important issue in Brazil, both
in terms of education and female labor market participation, is the extent and quality of child
care and pre-school education, both of which fall under the rubric of early childhood care
and education (ECCE). The 1988 Constitution introduced the concept of shared
responsibility between society and parents for the care and education of pre-school age
children (age 0-6). A recently promulgated law, the Lei de Diretri.es e Bases da Educafdo
Nacional, operationalized this constitutional principle and incorporated ECCE into the
conventional educational system. This event was the result of intense mobilization on the
part of civil society, which began in the 1970s and included governmental sectors concerned
about the quality and lack of regulations of services for poor children.
49.    The recent increase in female labor force participation has resulted in an expansion
of ECCE centers, but quality has remained poor. In practical terms, any person can set up
an ECCE center without governmental control. With the focus on primary levels of
education, federal and state governments have been withdrawing funding for ECCE centers.
As a result, tension exists between the need to improve the quality and regulation of the
system on one hand, and the scarcity of allocated resources on the other. Quality is
particularly bad in poor areas given the limited bargaining power of parents, and since
higher-income families use private services and/or nannies.
50.    The low-cost expansion of early childhood care has relied on the 'natural' abilities of
women to extend their family care taking roles. These women do not have the appropriate
qualifications and are poorly paid, however, thus undermining the effectiveness of the care.
The system fails children even before they enter primary school affecting poor and black
21



children in particular. For example, the 1995 PNAD indicates that preschool education
centers often have children enrolled who are in the 7-11 age group.
(g)    Labor
51.    GENERAL TRENDS. Over the past two decades, Brazilian society has witnessed
profound changes in women'; work and social roles as a result of demographic, socio-
economic, political and cultutal transformations. This process has its roots in the 1950s. A
major drop in fertility brought the national fertility rate from 6.3 in the 1950s to 2.4 in 1999
(PNAD, 1999). Although registered particularly in urban centers and the more developed
regions, the drop in fertility reduced the amount of work associated with reproductive and
domestic tasks, thus facilitatirg the incorporation of women in the paid labor force.
52.    Women's greater access to all levels of education also contributed to the
transformation of gender roles and the gender division of labor, which in turn altered
women's labor force participation and career paths. As Table 12 shows, during the 1990-99
period, while participation rates for men remained fairly constant, they increased for women
from a national average of 39 percent in 1990 to 47 percent in 1993 and 49 percent in 1999
(with variations according to age).13 In addition to educational attainment and demographic
changes, high rates of growth and the increase in the degree of urbanization and
industrialization generated labor demand that absorbed new workers, and women in
particular, into the labor force. At the same time, changing consumption patterns and the
growing availability of new products in the market generated a demand for complementary
family income. Changes in household survey methodology also resulted in a more
statistically accurate description of participation rates, particularly in agriculture. Finally, the
women s movement and changing views on women's place in society intensified these
trends.
53.    While the number of economically active women has increased, the overall
proportion of women in the labor force compared to men remains inferior in all regions,
even for women over 18 year:s old, as shown in Table 13. In 1996, the proportion was 41
percent for women and 60 percent for men. For women, this represented an increase from
33 percent in 1981 and 39 peacent in 1989. The unemployment rate is also consistently
higher for women than for men. As indicated in Table 14, in all regions of Brazil, more
women were unemployed than men, particularly in the urban North and in the Midwest,
where the gap was substantial (PNAD, 1996).
13 The corresponding figure for 1981 was 33 percent. ECLAC estimates of labor force participation rates in
Latin America, based on household surveys for each country, report a female rate for Brazil as high as 51 per
cent in 1994 (Arriagada, 1998).
22



Table 12: Labor Force Participation Rates by Sex and Age, 1985 - 1999
Age                  1990                 1993               1995            1999
Men      Women        Men      Women     Men    Women   Men    Women
10-14           24.3      10.6        28.1       14.9    26.4    14.4    21.7   11.4
15 - 19         71.8      41.4        72.2       45.4    68.8    44.1    62.8   42.0
20 - 24         92.1       52.9       91.1       59.6    90.5    60.9    89.3   63.3
25 - 29         26.2       52.7       95.8       61.0    95.2    62.7    94.7   65.7
30 - 39         96.9       54.7       96.5       63.7    96.3    55.4    95.9   68.4
40 - 49         94.5       49.5       94.7       61.0    94.5    63.5    94.6    64.8
50 - 59         82.3       34.5       82.3       46.0    83.6    48.0    81.7   49.7
60 and up       46.0       11.5       50.5       21.4    49.4    20.4    48.0    19.7
Total           75.3       39.2       76.0       47.0    75.3    48.1    73.8    49.0
Source: PNADs 1990, 1993, 1995, 1999.
Table 13: Individuals in the Labor Market by Sex and Age, as Percent of the Labor Force,
1996, Brazil and Regions (*)
Sex and Age               Brazil    North     Northeast   Southeast   South    Midwest
Men 10-14                  2.3        2.1        3.7         1.4        1.9      2.1
Men 15-17                  3.7        3.5        4.4         3.3        3.5      3.6
Men 18+                   55.5       56.4       53.6        57.0       54.3     57.4
Women 10-14                1.0        1.0        1.5         0.7        1.2      0.9
Women 15-17                2.0        2.1        2.0         1.9        2.1      2.1
Women 18+                 35.4       34.8       34.7        35.7       37.0      33.7
Total                    137,560     9,290      41,124      47,073    25,562    14,511
Source: PNAD, 1996
(*) Excluding the ural north
54.    IMPACT OF EDUCATION ON LABOR              Table 14. Unemployment Rate, by
FORCE PARTICIPATION. As in other countries, an    Gender, Population aged 15-65, 1996
association exists between schooling and labor                       Male       Female
force participation in Brazil. In 1995, whereas 16  Brazil            5.7         8.8
percent of men in the labor force had less than    Urban North        6.0        10.2
one year of schooling, 17 percent had more than    Northeast          5.2         7.8
eleven years of schooling (see Table 15). The      Southeast          6.2         9.8
same trend is valid for women. Of those women      South              4.5         6.6
in the labor force, 13 percent had less than one   Midwest            6.2        10.5
year of schooling and 25 percent had more than    Source: PNAD, 1996
11 years of schooling. The 1996 PNAD also
shows that the majority of working men and women has 4-7 years of schooling. Overall, as
shown in Table 15, women in the labor market had higher schooling than men, and the
association between years of schooling and labor force participation is stronger for women
than for men.
55.    OCCUPATIONAL SEGREGATION. Despite changes in recent years, gender
socialization at home and throughout the educational system still tends to channel men and
women into gendered workplaces and segregated occupations. As Table 16 indicates, male
workers are concentrated in agriculture and industry, while the largest proportion of women
works in the service sector, followed by agriculture. But occupational segregation in itself
23



would not be an issue if men's and women's average wages were equivalent when average
respective amounts of human capital (both education and experience) were taken into
account. Also, female or male dominance in one occupation or another per se may not be
important as long as they are spread across occupations at large. Thus greater analysis is
required before making substantive conclusions on the occupational segregation issue.
Table 15. Labor Force Participation Rates by Gender and Years of Schooling, 1995
Schooling      Participation r ates %       Distribution by years of schooling
Men           Women              Men               Women
Less than 1  73.5           40.2              16.3               13.2
1-3         65.6            39.0              19.0               16.5
4-7         73.9            44.0              34.1               31.9
8-10        82.5            52.8              13.1               13.6
11-14       88.6            69.0              12.2               18.1
15 or more  90.6            82.3               4.8                6.5
Total %     75.3            48.1             100.0              100.0
( Milions )   --             --              [44.21              [301
Source: Bruschini, 1998
Table 16. Employment by Sex and Sector (%), Brazil, 1985 - 1999
Sector                 1985              1990              1995             1999
Men    Women     Men    Women     Men     Women     Men    Women
Agriculture        33.6     18.4    28.1     14.0     28.4     22.5    26.8    20.5
Industry           27.0     12.4    29.1     13.7     26.4     9.3     25.8     9.5
Trade              11.2     10.4    12.3     12.2     13.3     12.8    13.4    13.5
Services            8.9     32.1    10.4     30.9     12.0     29.8    12.5    29.4
Other Services      2.9      2.3     3.3      3.0      3.5     2.9      4.3     3.3
Transport and       5.0      0.8     5.2      1.0      5.6     0.8      5.9     0.1
communications
Social Activities   3.2     17.1     3.3     18.6      3.6     16.3     4.1    17.4
Administration      5.0      3.3     5.1      4.0      5.1     3.9      5.1     3.9
Other               3.4      3.1     3.0      2.7      2.1     1.7      2.0     1.7
Total             100.0    100.0   100.0    100.0    100.0   100.0    100.0   100.0
Source: PNADs 1985,1990,1995,1999.
56.    WAGE GAPS. Brazil has
one of the highest wage gaps in  Table 17. Median Monthly Wage, Individuals Age
Latin America. In 1990, urbmn    10 and Over, by Sex (in # of minimum wages)
women's wages averaged 66              Male         Year             Female
percent of men's (FILACSO,              349         1988              2.1
1995). Only Bolivia, Paraguay          437          1989              2.5
and Uruguay registered larger           3.7         1990              2.2
wage gaps than Brazil's in 1990         3.1         1992               1.6
(ibid.). While having one of the        3.4         1993              1.7
widest wage gaps in the region,         4.4         1995              2.4
women have gained relative to           4.5         1996              2.7
men in the 1987-97 period, as           4.5         1997              2.6
Table 17 indicates. In 1987,     Source: PNAD, 1987-1997
monthly wages for women were
53 percent of men's wages; by 1997, it had increased to 58 percent Table 18 shows that
24



even after controlUing for schooling, a significant gap remains in the average hourly wages
earned, which in fact increases to almost to almost 50 percent with 11 years of schooling and
above. While the tendency has been for gender gaps to decrease, this has been accompanied
by a trend toward greater intra-gender inequality. The disaggregation of female earnings
according to educational levels shows that the male/female wage convergence is due to
higher wages for women with higher education (Lavinas, 1996). But the income distribution
is even higher among men than for women. For example, the Gini coefficient for men is
0.58 compared to 0.56 for women.
57.   GENDER ROLES AND WORK.               Table 18. Average Hourly Eamings by
Institutions that reinforce gender roles can  Sex and Schooling in R$
contribute to women's lower wages. For       Years of Study  Men       Women
example, labor legislation which provides   < 1             0.93         0.96
women with four months of paid maternal leave  1 - 3         1.54        1.02
and allows them to retire five years earlier than  4 - 7     2.82        1.79
men, makes female workers more costly to    8 -10            3.38        2.51
potential employers than men. For its part, the  11         6.21         3.78
workers movement in Brazil has reinforced the
role of men as the primary breadwinners and has
worked to protect women's maternal role, thus  Source: IBGE, 1999
contributing to men's predominance in the labor market.
58.   Men and women accept and reinforce these socially prescribed roles through their
actions and relations. Through in-depth interviews with poor male and female workers in
Sao Paulo, Martins Rodrigues (1991) shows the importance of gender roles and the pressures
they place on men. According to female respondents, men are the primary income earners
and if families are poor, fathers and husbands are at fault for not bringing in enough
resources. Because men also consider their main role to be that of provider, they blamed
themselves for being poor just as women did.
59.   A study of the auto industry in Sio Paulo also points to the key nature of gender-
differentiated roles and abilities in determining labor market outcomes. The research found
that perceptions of men's and women's work roles and abilities lead to different training and
promotion opportunities for men and women, as well as expectations. For example,
whereas supervisors asked men to be trained in technical skills, women were encouraged to
take behavioral training. The supervisors' explanation for lower female wages suggests a
'catch 22' situation: women did not eam the same wages as men because they did not have
the technical skills, but they also did not 'require' the same technical skills as men did.
Women were also considered to be more delicate than men, which affected some but not all
of the tasks assigned to them (Posthuma, 1998).
60.   JOB ADVANCEMENT. Women find it difficult to climb to positions of responsibility
due to a variety of factors that range from discriminatory promotional practices to women's
own reluctance to pursue such positions. Obstacles to women's advancement to higher
levels result from different factors at the household level and the workplace. Reproductive
roles, which women tend to carry out, reduces women's flexibility and ability to participate in
the paid labor force under the same conditions as men. Despite women's entry into the
labor force, a parallel redistribution of housework between women and men has not taken
25



place. According to a numbet of authors in Brazil the greater wage disparity among tnarried
women compared to single women can also be attributed to factors outside the labor
market, such as housework, child rearing and work interruption (Steliner, Smith, Breslaw and
Monette, no date).
61.   THE EFFECT OF CHILD                      Fgure 1. Work and Age
BEARING ON WOMEN'S                m-.             _ _                  1
EMPLOYMENT. In urban areas,
women's activity rates decline                      ...
during child bearing years
(Bruschini, 1994); however, the
tendency has been for the
participation curve to even out as               ......--- .--. -...........
more women with children remain
in the labor force. As Figure 1
illustrates, the mild M-shape of the
1985 age participation curve had                                        I
smoothed out by 1995. Despite      ''
women's concentration in domestic I
work, important differences among  °      ,0. , 3'*3' 2'*31
them exist according to socio-
economic levels. While middle
class professional women can make use of domestic help to release them from household
chores, working class and poor women have to absorb the double burden of domestic and
paid work. In all cases, however, women's choices regarding paid work are influenced by
household work.
62.   UNPAID WORK. Incomplete information is available for Brazil on unremunerated
activities by gender such as domestic and conmmunity work, which women tend to perform.
Over the past two decades and at the international level an important effort has been
underway to conceptualize and measure this invisible part of the economy, which tends to
be excluded from labor force and national income statistics (Beneria, 1992). UNDP cross-
country estimates for the early 1990s, for example, indicate that about two-thirds of women's
work was unpaid and unaccoumted for in national statistics, both in developing and high
income countries; for men, the corresponding figures were 24 percent and 34 percent
respectively (UNDP, 1995). TIhe measurement is an important approximation of men's and
women's contribution to hufrian welfare, and demonstrates the invisibility of a large
proportion of women's, and to a lesser extent men's, work.
63.   LABOR LEGISLATION. In 1932, female labor was for the first time comprehensively
regulated through the Decree 21.417. Brazil was a founding member of the International
Labor Organization (ILO); by 1932 it had signed all resolutions on women's labor and three
years later had ratified them. Since then, some of the protective measures have been
strengthened and others have been eliminated, including those considered to protect men's
labor market rights. For exarnple, women continue to have special privileges for retirement.
Businesses employing a certain number of mothers - but not fathers - are required to
maintain child care facilities. In the 1988 Constitution, maternal leave was expanded from
three to four months, a short paternal leave was created and job stability for mothers was
26



guaranteed up to five months after delivery. Finally, domestic workers were provided with
the same rights as formally employed workers, including paid vacations, social security and
paid maternal leave (Linhare, 1996).
64.   ETHNICITY AND EARNINGS. As previously mentioned, educational attainment is
positively associated with income. Yet, this association varies widely according to gender and
ethnicity. Table 19 shows the variation of income according to the last level of education
attained by black females, black males, white females and white males. Although this
variation may be explained by a number of factors such as women's preference for part-time
jobs, the data suggests the following:
*  Overall, income is more sensitive to gender than to ethnicity. With the exception of
those with graduate school training, white and black males consistently earn higher
incomes than white and black females for all the other levels of education.
*  In terms of income level, white females perform better than black females when the level
of education is constant. The same trend is valid for white males when compared to
black males.
*  Income is not blind to either gender or ethnicity. When ethnicity is combined with
gender, white males are clearly privileged in terms of their personal income, and black
females are unmistakably disadvantaged.
*  About two-thirds of black females and two-thirds of white females who have completed
only the first grade of education have a monthly income of less than one minimum wage.
Among black males with the same level of education, the proportion of those making
less than one minimum wage is about one-third, and among white males it is about one-
fourth.
*  Among those who have completed second grade, white males have the highest income
levels, with the majority making between three and ten monthly minimum wages. The
corresponding figure is between one and five minimum wages for black men and
between zero and two minimum wages monthly for the majority of black and white
women. At this same level of education, about six percent of white males make more
than 20 minimum wages monthly, double that of black males, five times more than the
proportion of white females and six times more than the proportion of black females.
*  About two-fifths of white males with college education earn more than 20 minimum
wages monthly. The corresponding figure is about one-fifth of black males, one-sixth of
white females and one-tenth of black females.
* Having completed graduate school, the difference in income between black males and
white females tends to disappear. Yet, white males continue to hold the most privileged
position in terms of income, with more than 70 percent eaming 20 minimum wages
monthly. This is double the proportion of white females and black males. Again, black
females with graduate training lag behind, with less than one-fourth earning more than
20 minimum wages.
27



Table 19. Income by Gender, Ethnicity and Education, Brazil 1996 (in # of Minimum Wages)14
Education          Less than 1 1 and 2  3 to 5   6 to 10  11 to 20  More than 20  Total
Black Female
lst.grade        68.7      26.8      3.0        1.3       0.1         0       16,978
2nd grade        39.5      36.3     13.1        8.4       2.1         0.6      6,140
College          12.5      13.3     15.2       29.3      19.8         9.8      1.164
Graduate school   8.3       2.8      8.3       16.7      38.9        25.0         36
White Female
1st grade        64.6      26.9      5.4        2.3       0.6         0.2     15,796
2nd grade        38.5      28.2     15.2       12.2       4.7         1.2      9,392
College          14.8       8.1     12.4       27.6      22.9        14.1      5,280
Graduate school   5.3       3.9       3.4      22.8      26.7        37.8     - 206
Black Male
1St grade        34.5      44.6     13.2        6.4      1.4          0.3     17,945
2nd grade        14.4      30.3     27.7       21.6      9.5          2.6      4,713
College           4.4       7.1      10.9      22.8      31.2        23.7      1.051
Graduate school    0        2.3       2.3      15.9      40.9        38.6         44
White Male
1St grade        24.2      39.3     18.7       13.1       3.7         1.1     15,360
2nd grade        11.0      19.2     20.7       27.3      15.7         6.4      7,683
College           4.0       3.3       5.3      19.5      28.4        39.5      4,652
Graduate school   1.3       0.7       2.2       4.3      21.2        70.3        232
Source: PNAD, 1996
65.    THE INFORMAL SECFOR. Women predominate in the informal sector. According to
the 1990 PNAD, 50 percent of working women and 15 percent of working men work in this
sector, most of them on a part-time basis (Lavinas, 1996). The informal sector includes a
high proportion of non-agricultural activities, including domestic service. Among those
engaged in paid work at horme, 82 percent are women. As Lavinas points out, the data
suggests that women are constrained to work in areas that are compatible with domestic
work, particularly in the case of poor households that have to function without basic
infrastructure. Race also de termines infornal sector employment. In 1990, whites
represented 59 percent of the work force as a whole, compared to 41 percent for non-
whites; the corresponding proportion for the informal sector was 44 and 57 percent for
whites and non-whites respectively (Republica Federativa do Brasil, 1994).
66.    THE PUBLIC SECTOIL. During the 1980s, women's employment in the public sector
increased significantly. Women currently represent over 44 percent of federal employees, a
figure well above the national average. Among working men, nine percent work in formal
public sector jobs, but this figure rises to 15 percent if the military is included (1996 PNAD).
The proportion of women varies according to sector (comparable data are unavailable for
men). For example, women. comprise about 15 percent of employees in the Ministry of
Justice, 63 percent of employees in the Ministry of Social Welfare and 51 percent of
employees in the Ministry of Planning, according to a 1998 study by the Escola Naconal de
14 In 1996, one monthly minimwn wage was US$100; the definition of black used in the table also includes
'pandos" (mulattos).
28



Administrafdo Puiblica (ENAP). At this point, the public sector provides one of the more
important sources of employment for women for three reasons: (a) the public sector
includes many jobs in the teaching and nursing professions, which have traditionally
attracted women; (b) decreasing relative salaries have made the social sector of the public
service a less attractive employment option for men; and (c) the government tends to abide
by labor laws and be less discriminatory (ENAP, 1998). Their predominance in the public
sector, however, makes women more vulnerable to downsizing of staff during public sector
modernization processes.
67.    DOMESTIC WORK. Domestic work continues to be a major sources of female
employment.15 In 1995, just under five million women worked in domestic jobs, compared
to 250,000 men. Domestic work is characterized by low wages: in 1995, 67 percent of
domestic workers made less than one minimum monthly wage. Men who work in domestic
jobs earn higher salaries than women even though female domestic workers have higher
educational levels than their male counterparts. Domestic workers also tend to be young -
about one quarter of domestic workers are in the 10-17 age category. Although Brazilian law
mandates the use of contracts for domestic workers and the payment of social security
benefits, the law is often not enforced. 16
68.    Domestic work allows poor woman to enter the paid labor force, such as migrant
women with low education and without previous work experience. Domestic jobs also
provide a stepping stone for other jobs and represent flexible employment. As well, by
being able to hire domestic help, middle and upper class women have been able to more
easily enter the work force. But on the downside, domestic workers mostly work in isolation
and without institutional ties with their co-workers, hence they are less able to gain human
capital on the job. As well, the ease of hiring domestic help means that educated women feel
less pressed to have their male partners share domestic chores. In this sense, domestic
service reinforces traditional gender roles.
69.    THE TEACHING PROFESSION. Teaching is another traditionally female occupation.
Information from the teachers' union, the Confederafao Nacional dos Trabalhadores de Ensino,
indicated that men comprised only three percent of pre-school to fourth grade teachers, 19
percent of fifth to eighth grade teachers, and 39 percent of middle school teachers.
Nineteen percent of principals were men. Gender stereotypes prevail among female
teachers themselves, according to interviews conducted by Carvalho (1998). Excerpts from
interviews included:
We see men as more radical, as teachers here, students there. As teachers I can't see that
they can develop a close relationship with students". According to another teacher,
"Women live the problems in the classrooms. Perhaps men can't reach the students. The
male sex does not have this, let's say, maternal side. The malefigure in itself is more
imposing, more linked to discipline, right? Because women are more motherly, they are
15 The proportion of employed women in domestic service is between 16 and 20 percent. A study by the
Instuto de Pesquisa Econmica Ap/icada (IPEA) estimates that over the past decade 19 percent of employed
women were domestic workers (Pereira de Melo, 1998).
16 There are currently some efforts to organize trade unions among domestic workers and to press for the
implementation of existing legislation.
29



moneflexible, they provide more support, onegets more relaxed." A third teacher
commented: "Eveyone is equal in the teaching profession. However, I think women
are more patient than men, don'tyou think? Principally because we are mothers and if
you are a mother, you ar? more patient, understand children better. "
70.     RURAL FEMALE LABOR FORCE PARTICIPATION. Women's labor force participation
in rural areas of Brazil has traditionally been undercounted. A 1991 revision of the concept
of "work" in the national census, however, resulted in an increase in women's agricultural
work from 14 percent in 1990 to 23 percent in 1995.17 For men, the rates were 28 percent
for both years (Bruschini, 1998).
71.    The number of wage-earning women in rural areas has been increasing, particularly
in agricultural production such as food processing. Rural industrial employment for women
is also increasing in some areas, such as the Northeast, where capital has been relocated from
the South and new industries (like shoe and garment production) have taken root.
However, much concern remains about the difficulty for rural women to gain access to
technical and professional training on an equal basis with men. During the 1997 and 1998
meetings on Gender and Famnily Agriculture, criticisms were voiced about the marginal
participation of women in seminars, courses and technical meetings organized by institutions
providing agricultural technical
assistance (Silipandri, 1998).18                 Box 6. Child Labor Conditions
72.    Creating women's                       Despite efforts on the part of Government and
commercial cooperatives of            NGOs, the presence of street children in Salvador, Bahia,
traditional feminine crafts has been  continues to cry for attention. In contrast to what was
expected, public lighting increased rather than decreasing the
one approach applied to help poor     number of hours children work. Children now are seen late
rural women, but these have           at night washing cars and seling fruits and candies. A.D., a
experienced limited success. On       twelve year old boy, spends that entire day on the streets
the whole, traditional women's        seDing coffee and cigarettes. "I work to help at home," he
says. He also says he is enrolled in a school, though he is
crafts hav  limiteddemandandunable to tell his name or address. Violence, drug abuse and
must compete with product& from       alcoholism are part of the street children's life.
lower income countries with lower
average wages. Encouraging            Source: A Tarde, August 1998.
women's crafts production i, also
frowned upon because it is seen to reinforce women in traditional female occupations.
73.     CHILD LABOR. In 1997, according to the PNAD, out of the 17 million children in
the 10-14 age group, 2.8 milion children participated in the labor force, of which 67 percent
were boys and 33 percent were girls. In the South, the proportion of girls reached about 40
percent but represented abouat 30 percent in the Northeast (where female enrollment in
17 The change took place as a resudt of a national campaign organized by rural women workers around the 1991
"Censo Econ6mico" arguing that they were "rural producers" rather than "unpaid family members" (Lavinas,
1996).
18 The Ministry of Labor has developed a flexible training program (PLANFOR) imnplemented jointly with
states, which has involved a good proportion of women (30 percent in rural areas according to 1997 data). Its
objective is to increase employabi)ity for a variety of workers.
30



school is also the greatest). According to the 1997 living standard measurement survey
(Pesquisa de Qualidade de Vida), about 50 percent of black male children aged 14 had worked at
least once in their lives, in contrast with one third of white male children. For females, the
proportion of child labor, although lower than for males, continues to be higher among
black children than among white children: 23 percent of white girls and 26 percent of black
girls had worked at least once before age 14. Child labor is both a result and a cause of
poverty, to the extent that work reduces the time that children spend in school and affects
their ability to acquire human capital.
74.    The 1985 PNAD Supplement on Children indicated that, in Sao Paulo, children
from families headed by single mothers had a 10 percent higher risk of working than
children from conjugally-headed households. Black children had an eight percent higher risk
of working than white children (Barros e Mendonca, 1990). Poverty is also transmitted
through the generations, with children of poor parents much more likely to be subjected to
poverty than those of less poor families (Barros, 1991).
75.    All over Brazil, the Bolsa Escola and PETI (Programapara Erradicafdo do Trabalho
Infantie programs are rapidly expanding. These are programs designed to increase school
attendance and attainment as well as reduce child labor through targeted financial transfers
as part of a package of various measures. Children of school going age (7-14 years old) from
poor households are the target group. The programs are widely seen to be successful, both
in terms of improving educational outcomes as well as in improving family welfare of the
recipient fanmilies. Nevertheless, serious evaluation of these programs would serve to further
enhance their effectiveness. A gender analysis of the Bolsa Escola and PETI programs would
also be useful to gain a better understanding of the factors that lead to school abandonment
by boys and girls and would help to design even better programs to keep both boys and girls
in school.
76.   Another effect of the Bolsa Escola and PETI programs has been on mothers receiving
the transfers. The programs specifically give the money to the mothers of participating
students. It was thought that by doing so, a larger amount would go to the improvement of
the welfare of the family as a whole and of the children in particular. For many of these
women this is the first time that they have received and have been entrusted this kind of
financial responsibility, in many cases opening a bank account for the first time. This 'official
recognition' has also led to a greater recognition and valuation of women's role in the family
- including by their partners - and in many cases increased self-confidence. In field visits,
men and women repeatedly mention the money received by mothers through the Bolsa Escola
and PETI programs as one way in which gender relations and the roles of men and women
are slowly changing.
(b)   Poverty, StructuralAdjustnent and Social Protection
77.   POVERTY. Recent analysis carried out by the World Bank of IBGE data for the
Northeast, in particular the states of Ceara and Rio Grande do Norte, finds various factors
related to poverty including: (a) ethnicity (the white population is at considerable smaller risk
of poverty than the black or mixed race population); (b) age of household head (the older
the household head, the lower the incidence of poverty); (c) level of education; and (d) labor
31



status and occupational sector (informal and agricultural workers display higher incidence of
poverty) (Fiess and Verner, 2001). The analysis also showed that male- and female-headed
households differ only marginally in the extent to which they are likely to be poor, 39
percent and 41 percent in Rio Grande do Norte respectively. Furthermore, male-headed
households experienced a slower reduction in poverty compared to female-headed
households. However, when controlling for education (women have more years of schooling
than men) and other individual characteristics, female-headed households have a much larger
likelihood (46 percent) of being poor than do male-headed households. Overall, the most
important factor contributing to the likelihood of a household being poor is the completed
level of education by the household head.
78.   In addition, the study found that households with young children are more
vulnerable and more likely to be poor than households with no children under the age of
five. Furthermore, the probability of experiencing poverty for households with small
children seems to have increased over the past decade. Households with members between
the ages of five and 15 also have a large probability of being poor, although slightly less than
those with children under the age of five. The numerical predominance of families with
small children located below the poverty line as well as the findings on increased
vulnerabilities of female-headed households imply that policy interventions should not
ignore these groups. In fact, targeting these groups would go a long way in reducing poverty
and might be at the center of a poverty reduction strategy.
79.   A recent study by Ferreira and Leite (2001) simulating the effects of an expansion in
education on income distribution and poverty in the state of Ceara, shows further evidence
to this effect. The study found that household dynamics played a crucial role in the impact
of educational expansion on income distribution and poverty. It also found that, as in other
places where educational levels rose rapidly, there appears to be a 'reserve army' awaiting
conditions to enter paid or sclf-employment, largely composed of women. As women
acquire education and enter ihe labor force their fertility rate declines, reducing the number
of children in the famnily. In fact, labor force participation and demographic changes arising
from educational expansion account for around half of its overall poverty reduction impact.
A large inflow of women into the labor market, however, may generate downward wage
pressure or enhance job competition. The gains of a more educated labor force depend, to a
large extent, on how effectively "a level playing field for its women" is ensured (Ferreira and
Leite, 2001).
80.   This study again emphasizes the central irnportance for poverty reduction strategies
that reduce barriers and address the needs of poor women. These include access to childcare
and family planning, continuing improvements in education and reducing barriers to
women's participation in the labor market. It also includes long term goals that require
changing gender roles and socializing men and women differently such as reducing levels of
violence against women and redressing the gender imbalance in the division of household
work. When men and wome n share parenting, care-giving and domestic chores more
equally, women are freer to participate in the workforce and advance on the job, itself
another precondition to reducing gender-related labor inequalities. Such policies will all have
important effects on the overall welfare of poor families.
32



81.   STRUCTURAL ADJUSTMENT AND SOCIAL PROTECTION. Women and men are likely to
be affected by macroeconomic shocks in different ways. For example, downsizing will have
a differentiated impact given the gender-segregated nature of the labor market. Deficit
reduction leading to job losses in the public sector, where women's presence is above the
national average, is likely to disproportionately affect women. But cutbacks, layoffs and
closures in heavy industries subject to increased globalization and competition mean greater
unemployment among men. Male workers in the Sao Paulo and Rio de Janeiro areas face
the greatest risk.
82.   While some women will likely lose their jobs, others will enter the labor market
during economic downturns. Women represent a disproportionate part of the inactive
population and, as men in the household become unemployed and/or are unable to sustain
the household financially, women will be forced into the labor market. Among the poor,
women often take jobs that offer them easy entry and flexible working conditions, albeit
with lower earnings. Historically, economic crisis has been an important trigger for women
to entry into the labor force in Brazil. Various authors suggest that while part of the increase
in women's labor force participation rates during the past decade might be due to higher
educational levels among women, the economic crisis has been an important factor in
women's search for employment during the 1980s and 1990s (Rep6blica Federativa do
Brasil, 1994).
83.   Without changes in the gender division of labor at the household level, an increase im
women's participation in paid work implies new pressures on their time. In other countries,
survival strategies have resulted in an intensification of women's domestic work (Floro,
1994). A World Bank study, however, suggests that while women's workload does indeed
increase, time dedicated to domestic tasks decreases overall during economic downturns due
to women's increased participation in the paid labor force, with corollary negative
consequences on the upkeep of the family (Cunningham, mnimeo). Also, in Ecuador, Moser
(1997) found that pressures on parents to work results in spending less time supervising their
children. This leads to their sons, in particular, to spend more time on the streets.
84.   For men of all ages, unemployment threatens their role as family providers and often
creates problems of self-esteem, depression and suicide. Although more information is
needed in the case of Brazil, other countries have observed that unemployment can result in
male alcoholism and violent behavior, with both men and women as victims. Among young
men, it often results in destructive and criminal actions, with high costs to their families and
communities as well as to themselves. Violence increases health costs for families as well as
society at large. An increase in child labor among boys is another possible effect of
economic downturns, which in turn causes them to interrupt or drop out of school.
33



(i)     Governmental/Non Governmental Organizations Dedicated to Gender
85.    THE ROLE OF CIVIL SOCIETY.
As in many other Latin Ameican                        Box 7. The Lipstick Lobby
countries, civil society groups have             In 1987, civil society organizations, NGOs and
played an important role in putting      the Conselbo Nacional dos Direitos da Mulher (National
gender on the public agenda. 19 In        Council of Women's Rights, CNDM) banned together to
1987, civil society organizations,        advocate for gender equitable civil rights, which became
NGOs and the CNDM barned                  known as the "lipstick lobby". Their actions led to the
following legal rights being instated:
together to advocate for gender    0   Women's entitlement to make decisions in the
equitable civil rights, which became             household, with the court arbitrating in cases of
known as the "lipstick lobby" (see               conflict. Men's supremacy in family affairs was
Box 6 for a description of what the              eliminated.
lobby achieved). The civil society            *  Married women's right to separate tax returns
movement also worked closely with                and to claim children as dependents for tax
deduction purposes.
the national Congress to improve the             ThIle emination of the male right to irnpede their
gender balance in political                      wives from working under any circumstances.
participation and representalion. As a        *  The same tights for children bom out of wedlock
result, in 1997 Congress passed a law            as for those of married couples.
requiring all political parties to include    *  The same rights of common law partners as
those of formnal marriages.
at least 25 percent female candidates         *  The right to take paid paternity leave.
(WEDO, 1998). Following the                  a   Sexual violence as a human rights crime as
passing of the law, women's                      opposed to a morality crime, thus implying
representation increased from eight to           harsher penalties to sexual offenders.
11 percent (Htun, 1998). In addition          m  Labor and welfare rights expanded to domestic
,  . .  .  .         ~~~~servants.
to increasing women's participation in
politics, such a measure is also          Source: Junho Pena, 1991.
expected to encourage women to
exercise their rights to vote. 2( In 2001, after many years of debate and amendments, having
already been guaranteed in Brazil's 1988 constitution, Congress finally passed a law
guaranteeing equal rights for men and women. Among other measures, the new law
abolishes the traditional concept of 'paternal power', which gives fathers the unrestricted
rights to make decisions on behalf of their families. Under the new legislation, husbands and
wives will share that authority, and single women will be regarded as heads of households.
86.     Civil society organizations have also been active in developing programs and services
with gender specific objectives (see Box 7). A few examples of organizations carrying out
important gender work include: (a) CEPIA in Rio de Janeiro, which provides training on
gender issues for police officers and for medical doctors and nurses in the health sector; (b)
19 Civil society organizations have proliferated over the last few decades. According to an Instituto Brasileiro de
Estados Rek4giosos (IESER)/Johns Hopkins University study, NGOs and civil society organizations employ about
1.1 million people. Another 330,000 work as volunteers for NGOs. More than half of the volunteers are
linked to religious organizations.
20 Also according to Htun (1998), Brazilian citizens elect individual candidates as opposed to a party list, as is
the case in neighboring Argentina and Costa Rica. Hence in Brazil women must compete for publicity and
support within parties, which according to Brazilian potential female candidates has inhibited the achievement
of quotas.
34



Instituto de Afdo Cultural (IDAC), which manages a successful Inter-American Development
Bank-funded training program for female leaders covering areas such as politics, unions, and
press; (c) the SOS Corpo, in Recife, which has a long tradition of working in reproductive
health and other health issues with a gender perspective; (d) IBISS in Campo Grande, which
addresses the sensitive issue of domestic violence and sexual abuse against children with a
gender perspective; (e) Ecos, in Sao Paulo, which works with men on sexuality and
reproduction issues; (f) Cunha, a feminist group in the Northeast city of Joao Pessoa, which
raises public awareness on gender through street theater, musical shows and videos; (g)
CEMINA in Rio de Janeiro, which produces weekly radio talk shows transmitted in the
Amazonia to discuss gender related to other social issues such as the environment, health
and violence; and (h) REDEH in Rio de Janeiro which works with MED to provide gender
education to adult literacy teachers. While gender work in Brazil carried out by civil society
has traditionally focused on women's issues, the above examples clearly show a shift in
emphasis to integrating male gender issues, particularly in the area of sexual and reproductive
health.
Box 8. Integrating Gender in Funda4ao Abrinq's
87.    THE PUBLIC SECTOR RESPONSE.                            Innovative Programs
In 1985, the Govemnment created the               Funda,co Abrinq in Sao Paulo is an example of a
CNDM as a consultative body to promote        Brazilian NGO dealing with complex gender issues in the
gender equity and eliminate discrimination    course of working widh children, which is its main target
against women.21 The CNDM is located in       group. Funda,co Abrinq, which was founded widh resources
the Ministry of Justice. Although the         from the private sector, uses a unique approach in its work.
President appoints the head of the Council,   Instead of funding and implementing programs directly,
Abrinq invested in developing a stamp which distinguishes
civil society group and NGOs run the          products and businesses that foUow child-friendly practices.
Council. It is legally mandated to            The foundation also awards municipalities for innovative and
coordinate actions across the federal         successful social programs directed at poor families and
government in sectors such as education,      children. The types of programs awarded address issues such
labor, justice, health and human tights.      as:
,                   education, including early childhood and preschool
education, youdh and adult education, education for
88.    According to Htun (1998), after                street children, programs to reduce dropout and
undergoing a decline in power and clout in           repetition and reeducation of child criminal
the late 1980s, the CNDM experienced a               offenders;
resurgence under the current                         health, including programs that support pregnant
administration of Fernando Henrique                  women, provide incentives for breast feeding,
education in family health care and family planning
Cardoso. The Council currently maimtains              services, address pediatric AIDS, drug addiction and
a small staff, but has a large consultative          at-risk pregnancy, and prevent uterine and breast
council. Since 1997, it has been successful          cancer;
in fulfilling its mandate to coordinate           *  violence, including programs to reduce violence
across sectors and government agencies,               against children and attend to victims of domestic
hacrossand other forms of violence; and
having established agreements with four           *  children, including summer camps for low income
ministrLies to implement actions related to           children and programs that address eradication of
gender.                                               child labor and provide shelters for street children.
89.    In Brazil's federalist structure, state
agencies have played a critical role in addressing gender issues. For example, the Sao Paulo
State Council for the Condition of Women was instrumental in developing policy measures
21 CNDM was created through Law 7.453.
35



on violence and health. Sao F'aulo has the most extensive network of women's police
stations in Latin America. Sao Paulo's public hospitals have established innovative
procedures for treating rape and domestic violence victims.
36



III. SUMMARY AND IMPLICATIONS OF FINDINGS
90.   Brazil has progressed significantly in addressing gender issues and reducing gender
inequalities. Four major advances are worth noting:
(a)   Women's access to and use of contraception has increased enormously,
which has resulted in a sharp drop in the fertility rate and size of households
in all regions of the country. Female sterilization has become the most
frequent form of contraception.
(b)   Since 1988, women's rights have expanded within the household, in the work
place and in terms of land rights and personal safety. The 1988 Constitution
also created short-termn paternal leave.
(c)   Female education has increased to the point that women now have more
schooling on average than men.
(d)   Although men still predominate in the labor market, women's participation
has steadily increased over the two last decades. The wage gap between men
and women has also narrowed.
91.   While Brazil has witnessed important improvements in regard to gender, a number
of issues affecting men's and women's well-being remain. For example:
(a)   DEMOGRAPHICS. Fertility, infant mortality and mortality rates remain very
high among indigenous groups. Teenage pregnancy is also a continuing
problem. Although the national rate is 15 percent, teenage pregnancy is 24
percent and increasing in the Northeast. Mortality rates linked to external
factors such as traffic accidents, homnicide and suicide differ greatly by
gender.
(b)   WOMEN'S HEALTH. Inadequate pre-natal care persists and the proportion of
at-risk pregnancies remains high. Maternal mortality is also on the rise,
which suggests serious problems with the public health care system. Women
are contracting AIDS at a greater rate than men, although in absolute terms
more men are infected with AIDS than women. AIDS prevention programs,
which in the past tended to focus on homosexual men, have recently been
broadened to include female risk groups.
(c)   MEN'S HEALTH. Health programs, including sexual and reproductive as well
as general health programs, have been directed mostly at women. For
example, prevention programs have been established for breast and cervical
cancer but no parallel efforts have been made to address prostrate cancer
among men. Similarly, reproductive health programs have tended to exclude
men. As a result, contraceptive use among males is minimal in Brazil. Of
particular concern is that an insignificant proportion of men report using
condoms, thus putting themselves and their partners at-risk of contracting
AIDS and other STDs.
37



(d)   VIOLENCE. The incidence of violence among men, male violence against
women and :iexual violence against children of both sexes continues to be
high. Violence is a male gender issue in the sense that the way men are
socialized and the expectations placed on men by society contribute to male
violence.
(e)   EDUCATION    On average, boys now have fewer years of schooling than
girls. And boys tend to have higher school drop-out rates than girls because
they are forced to enter the labor market prematurely. For both sexes, gender
socialization results from girls and boys being treated differently in the
Brazilian school system. For exarnple, rebellion, which is punished in girls, is
accepted for boys. And violence among boys is considered a side-effect of
being male. As for girls, gender stereotypes and socialization are believed to
hold back their educational and career aspirations.
(f)   EARLY CHIIDHOOD CARE. While remaining a government priority, the
quality of early childhood care and education programs remains poor.
Because of the division of labor within the household, low quality services
disproportionately affect women's opportunities to enter into and advance in
the labor fojrce. Poor women are the most affected group.
(g)   LABOR. Women's labor force participation increased to 49 percent in 1999.
But as in other countries, the labor force is highly segregated by sector. For
example, women continue to be concentrated in low-skill occupations such
as domestic work and teaching. Education reinforces this tendency, as do
household role models, thus leading to inter-generational transfers of
occupational segregation. More importantly, Brazil continues to register one
of the largest gender wage gaps in Latin America and the Caribbean. While
the gap has decreased overall, the disparity between women with higher and
lower educational levels has increased. However, the gender wage gap is
narrower between men and women with lower levels of schooling, and it is
wider between married men and women. Race is as important as gender in
determining wages.
(h)   POVERTY. Data from the Northeast shows that male- and female-headed
households only differ marginally in the extent to which they are likely to be
poor, 39 percent and 41 percent respectively. Male-headed households
experienced a slower reduction in poverty compared to female-headed
households. However, when controlling for education (women have more
years of schooling than men) and other individual characteristics, female-
headed households have a much larger likelihood (46 percent) of being poor
than do male-headed households. In addition, households with young
children are more likely to be poor than households with no children under
the age of five.
92.   Gender differences such as the ones identified in this report have economic and
human development consequences. For example:
38



*  poor health affects economic productivity and is a drain on a country's resources; therefore,
policy makers need to be aware of men's and women's different health risks;
*  the exclusion of men in reproductive health initiatives means programs are potentially less
effective because men are also involved in decisions on contraception and reproduction;
*  an increasing risk of contracting AIDS and other sexual transrmitted diseases is associated
with the lack of condom use among men; but reproductive health programs continue to
target women for the most part;
*  the high incidence of violence in Brazilian society is taxing on the health care system and
affects work productivity; as reported by the WHO, violence against women has a direct
negative impact on several important health issues including safe prevention of sexually
transmitted diseases and HIV/AIDS;
*  low levels of educational attainment - due to early entry into the labor force in the case of
boys and household work and other issues in the case of girls - affect human capital and
economic potential; but lack of attention to the particular needs of girls and boys affects the
viability of education initiatives designed to keep children in school; and
*  given higher female levels of education, women's lower salaries in the labor force vis-a-vis
men's means that their returns to education are lower than their male counterparts.
93.    The report's findings point to the importance of gender roles in shaping
opportunities, in affecting access to resources, in influencing choices, and, ultimately, in
determining well-being. While economic opportunities for women have broadened, the
primary role of many women continues to be that of mother and caregiver, which means
they are likely to be more economically dependent. Men's roles in contrast, have remained
fairly rigid, the expectation being that men be the main breadwinners in charge of the
family's economic well-being. Narrowly defined gender roles affect men and women over
their lifetime. Beginning with school age children, girls are often forced to leave school or
combine domestic work with their education because of gender roles, while boys drop out or
repeat grades to participate in income-generating activities. These choices affect human
capital later in life. As adults, women earn lower salaries than men, due in part to women
opting for lower income jobs that offer greater flexibility, and that allow them to juggle paid
work and household responsibilities.
94.    Differences in gender and sex roles can also lead to discrimination. For example,
Brazil's generous maternal leave policies can act as a double-edged sword for women by
raising their labor costs. And while no evidence of significant gender gaps exists in terms of
primary- and secondary-level education, gender differences in the returns to human capital in
the labor market suggest possible forms of discrimination or inequality. Gender biases can
also result from lack of information, gender stereotypes or false perceptions on the relative
attributes of men and women.
95.    The heterogeneity of men and women is also noteworthy. For example, gender gaps
in wages are less among lower skilled workers than among the better educated. And while
the gender wage gap in Brazil has been declining, the gap between different groups of
39



women had widened. Ethnic differences must also be considered in exarnining gendler
issues in Brazil: it is one thing to be categorized as a white woman in Brazil and quite
another to be categorized as a black female.
40



IV. STRATEGY AND RECOMMENDATIONS
(a)    Overall Strategy
96.   As in many other countries, gender work in Brazil has focused on broadening
opportunities for women, reducing discrimination against women and increasing women's
rights. Laws and policies on family planning, women's health, child care, rights for domestic
workers, women's political participation, domestic violence and protection of female
workers have been the main instrument for expanding women's opportunities and rights.
Brazil's advances in the legal arena are indeed impressive. While efforts to strengthen the
legal framework and enforcement mechanisms should continue, equality will not be achieved
if societal expectations of what men and women should and should not do, and how they
should and should not behave, do not change. Overall, women have more education than
men yet labor market disparities continue. In particular, wage gaps are large compared to
other countries in Latin America and the Caribbean, and researchers suggest that only part
of this gap can be attributed to discrimination. Household decisions regarding the allocation
of men's and women's labor and the choices made by women themselves play a big part in
explaining gender differences in occupations, salaries and opportunities for advancement.
97.   FOCUSING ON GENDER SOCIALIZATION PROCESSES. Attention should turn,
therefore, to changing societal expectations and socialization processes so that women as
well as men make different choices. In contrast to enacting laws and establishing policies,
which can have a relatively short time frame and can require limited technical capacity and
resources, changing socialization processes will take several generations and will require
wide-ranging efforts through many different channels. Socialization takes place through
education, the media, the family, peer groups, the community, and cultural practices, so
efforts on each of these fronts need to be made if far-reaching change is to take place.
98.   ADDRESSING MALE GENDER ISSUES. As many Brazilian groups and development
practitioners have detected through experience, gender can not continue to focus solely on
women if gender barriers are to break down. Both men and women are socialized according
to gender norms and expectations, and this socialization leads to problems for both groups.
While women's issues are well-known and documented, men's issues such as violence and
aggression, alcoholism and substance abuse, risky behavior, the effects of unemployment
and aging and absent fathers are relatively unresearched. Experiences in Latin America and
the Caribbean as well as in industrialized nations such as the U.K and the U.S.A. suggest
that parallel gender efforts need to be made to reach men if real change is to take hold.
99.   WORKING AT THE LOCAL LEVEL. As an overall strategy this report recommends
working through local level and community-based programs to change socialization
processes as well as addressing other issues raised in this report (substance abuse, early child-
hood education, maternal health care, teenage pregnancy, sexual abuse of children,
reproductive and sexual health care services, etc.). Working through civil society and local
level organizations offers several advantages, the most important of which is that these
groups exert powerful peer pressure and influence by operating close to and interacting
directly with target groups. They also have a better understanding of local contexts and can
adapt programs according to local conditions and practices. Employing local organizations
41



also utilizes existing capacity, increases efficiency and puts less of a strain on limited public
sector budgets. Finally, working through local organizations strengthens their capacity and
contributes to democratization processes. One interesting example of such a local program
is the PAPAI Program (Daddy) in Pernambuco, an NGO that conducts research and runs
projects in the areas of teenage fatherhood, STD and AIDS prevention, health and the
media. Using the education system, public campaigns and other media are two alternate
strategies for changing socialization processes. Sector-specific recommendations follow.
(<a)   Health
100.   Reproductive health priorities as they relate to gender should include reducing
Brazil's relatively high mater-nal mortality rate, improving access to family planning for the
poor and targeting men as partners in reproductive and sexual health programs. With
respect to reducing maternal mortality, it is important to note that the problem is part of a
larger institutional issue which affects all health care services in Brazil, such as poor
regulation of the public-private sector 'conveniados.' As an immediate action, however, local
and community-based organiizations and NGOs can be mobilized to provide information
and basic services during diFferent stages of pregnancy, particularly during the first trimester
when failure to seek services places women in a high risk category. Post-natal care
including the provision of information on child nutrition could be part of those
community-based services. Another basic service could include increasing tetanus
immunizations for women :o reduce the risk of infections during and after pregnancy.
Finally, including fathers in programs such as child care and nutrition would serve to
promote their roles as husbands/partners, parents and care providers.
101.   REPRODUCTIVE AND SEXUAL HEALTH. Efforts to improve reproductive and sexual
health services should continue. But sterilization as a method of contraception should be
discouraged given that it tmay lead to a reduction of condom use, it is not an optioii for
younger women who plan to have children later in life, and it represents a high expense to
the health care system. Also, reproductive and sexual health programs should target both
women and men. In particalar, it would be important to promote male forms of
contraception, given the low incidence of condom use and male sterilization.' While
Brazilian men claim to be aware of choices on contraception, information available suggests
that men do not receive accarate information on sexual and reproductive health. Efforts to
reach men would go a long way to addressing Brazil's AIDS epidemic. A good example of
work being done in Brazil i, the PROMUNDO Institute in Rio de Janeiro, which develops
workbooks for health providers on the reproductive and sexual health needs of boys,
organizes regional meetings to exchange ideas on working with boys and produce health
educational materials for boys produced by adolescent boys. Reproductive health programs
should also be region-specific. For example, in areas such as the Northeast, actions should
focus on reducing the incidence of teenage pregnancy.
102. VIOLENCE. DEPRESSION AND RISKY BEHAVIOR. With a view to better designing
control and prevention programs, epidemiologists should study the incidence and gender-
related risk factors associated with violence, depression, suicide and risky behavior among
22 Experiences from the U.S.A. suggest that comfortable 'male friendly' settings are key to attracting men
and encouraging open discussion on reproductive and sexual health issues.
42



men. Little has been done to examine these social issues as gender issues and specifically to
examine how gender socialization processes and gender roles and expectations contribute to
their incidence.
103.  INDIGENOUS PEOPLE'S HEALTH. More attention needs to be paid to the alarming
health problems of indigenous groups that have links to gender, such as high male mortality
rates and maternal mortality/morbidity. It has been argued that high opportunity costs are
associated with indigenous people's programs given their low population coupled with the
overwhelming economic and social problems facing Brazil today. But investments for the
indigenous are justified for reasons of poverty reduction and cultural heritage.
(c)    Violence
104.   Given its incidence and its gendered nature, violence should be a priority area for
gender work in Brazil. But addressing violence from a gender perspective should go beyond
dealing with the effects of domestic violence on women, in which Brazil has already made
some important advances. Over the long term, attention should turn to violence prevention;
that is, examining how gender affects male violence and identifying steps to counter the
effects of gender roles and socialization processes on violence.
105.   The education system, community programs and the media are all important vehicles
for preventing violence related to gender, according to Morrison and Biehl (1999). The
education system, for example, can influence cultural values that promote aggressive
behavior among boys and docile behavior among girls. Possible interventions include
retraining teachers and eliminating gender stereotypes in textbooks and developing special
programs to teach children nonviolent resolution skills. Community level programs can be
used to provide informal education programs, teach citizens about legal sanctions against
violence, establish violence prevention strategies, and provide social services for victims of
violence. Furthermore, media can be used to transmit educational programs and telenovelas
(soap operas) to produce and disseminate more positive images of interpersonal interactions
(male-female and adult-child) as well as illustrate nonviolent conflict resolution. And lastly,
peer group programs, such as the Big Brothers and Big Sisters Program in North America, have
been successful in reaching and reforming at-risk youth.
106.  In terms of sanctioning violence, it would be important to continue to strengthen
programs that make men and women aware of the illegality of domestic violence, provide
women with access to legal services and improve judiciary processes. Also, given that
women's police stations in Brazil have been used as a model in Latin America and the
Caribbean for helping the female victims of violence, learning more about the effectiveness
and sustainability of these police stations is another priority. Lastly, programs that address
the needs of street boys and girls in general and those engaged in prostitution in particular
are important. An example of a successful program working in this area is the MNMMR
(Momimento Nacional de Meninos e Meninas da Rua, or National Movement for Street Boys and
Girls), which has been instrumental in putting the issue of street children on the national
agenda. Another example is the Casa de Passagem in Recife, which provides alternatives and
support for girls and young women living on the streets, many of whom are prostitutes.
43



(d)   Education
107. FORMAL EDUCATION. Given that Brazil has progressed significantly in terms of
increasing female education, aLttention should now turn to ensuring that boys do not fall
further behind and to improving the quality of schooling in general by reducing gender
stereotypes conveyed through education. These can work to undermine girls' career choices,
to promote docile behavior among girls, to hinder boys' performance in school and to
contribute to aggressive behavior among boys. Specific actions include developing teacher
training modules on gender si:ereotypes conveyed through education; and eliminating gender
stereotyping in textbooks and other classroom materials.
108. Partnerships between the Ministry of Education and organizations such as RIDEH
have been a cost-effective way of addressing the problems associated with gender
stereotypes and socialization; hence, these partnerships should be strengthened. Brazil could
also learn from the successful experiences of neighboring Argentina, which in the late 1980s
and early 1990's made important advances to remove sexist language and depictions in
textbooks. Other recommendations related to formal education include: (a) examining the
gender-differentiated impact of the Bolsa Escola and PETI programs in reducing child labor
and retaining boys and girls in school; and (b) getting a better understanding of determinants
of education by gender, including how place of residence, parents' education and presence of
mothers or fathers affects girls and boys' educational attainment.
109.  DAY CARE/EARLY CJHILDHOOD EDUCATION. Early childhood education represents
a future investment in the human capital of the country and should be a priority over the
long term. Studies have shown that children who have attended pre-school perform better
academically than those children who do not. Providing childcare is also critical for
mothers, particularly those whao are forced to work. Publicly funded care may not be a
viable nor necessarily desirable option given the potential quality problems. Thus community
level organizations can be used as the vehicle for strengthening and expanding early
childhood education and care. In the short-term, improving programs that already exist
could be a viable option given that parents are willing to pay for good services. For example,
providing training and better regulating existing maes creches (nursery schools) would be a first
step.
(e)   Labor
110.  Brazil needs to address its persistent gender wage gap, which is larger than those of
many of its less developed neighbors. One priority would be to ensure better adherence to
labor discrimination laws by (a) examining current enforcement mechanisms and
institutional support available to female workers; and (b) making available information on
labor rights to female employees and employers. A second priority would be to leam how
Brazil's generous maternity leave contributes to discrimination against women, and how to
counter the possible negative effects of this regulation.
111.  As mentioned throug hout this report, discrimination is only one factor contributing
to the wage gap. Redressing the gender imbalance in the division of household worlk - so
that men and women share parenting, care giving and domestic chores, thus freeing women
44



to participate in the workforce and advance on the job - is another precondition to reducing
gender-related labor inequalities. But because changing gender roles requires socializing men
and women differently, this will be a long-term process.
112.   Attaining greater gender equality in the labor market will also require changing the
way work is currently organized, according to well-known femrinist and former head of the
CNDM, Rosiska Darcy de Oliveira. She suggests, for example, that more flexible
arrangements such as job-sharing need to be made available to men and women. Recent
research carried out in the United States supports the notion that work systems need to
change if greater equality in the household and in the labor market is to be achieved.23
113.   In the short-term, projects and programs that work with families could begin to
promote changes to gender roles in the household. For example, a new 'family capacity
building' program is being proposed for Argentina, which would provide unemployed men
with new skills in fathering and care giving and make it easier for their female partners to
seek work opportunities and excel on the job. Recommendations related to other gender
and labor issues follow.
*  JOB MATCHING. NGOs and other civil society groups that are well known in the
community could act as clearinghouses for jobs, as well as provide training on
grooming and preparing for a job. Women in particular would likely benefit from
these services because they have difficulties leaving the favelas (slums) to search for
jobs (due to household constraints), presenting themselves to potential employers
and getting access to information on jobs.
*   STREET CHILDREN/CHILD LABOR. According to a World Bank report, reducing the
number of street children requires investments in human and social capital, such as
integrated early childhood development programs. Others concur that the best
approach is to build up social and human capital in the poorest communities, which
has the added benefit of providing basic services in sanitation, health and nutrition
(Moran and Moura Castro, 1997).24 Lastly, the rapidly expanding Bolsa Escola and
PETI programs deserve more gender-sensitive investigation.
23 See Deutsch, Francine M., 1999. Halving it All How Equaly SharrdParentinA Works. Harvard Universit
Press.
24 The IDB study by Moran and Castro (1997) suggests three approaches to addressing the problem of street
children. First, the shelter (abnigo) approach provides a boarding school environment for poor children,
including education, training, health care, lodging, food, sports and leisure activities. FUNDEM in Rio de
Janeiro is using this approach. Pitfalls have been the high unit costs, difficulties in keeping children enough
time to make a difference and limited success in obtaining jobs for youth who reach the age limit. Second, the
"if the mountain won't come to Mohammed, then Mohammed will go to the mountain" approach takes food
and services to children in the streets. Problems with this approach include the inability to attract children and
opposition from pedestrians and shopkeepers. A third 'mainstreaming' approach seeks to reunite children with
families. Constraints to this strategy include the resources needed to provide families with a more nurturing
environment for children and the frequent 'beyond repair state' of most street children and their families.
45



(f) Poverty, Economic Fluctuations and Social Safety Nets
114.  The numerical predominance of families with small children that are below the
poverty line as well as the hndings on increased vulnerabilities of female-headed households
imply that targeting these groups would go a long way in reducing poverty and might be at
the center of a poverty reduction strategy. Specific recommendations include: increasing
poor women's access to chilcdcare and family planning, continuing improvements in
education, reducing barriers to poor women's participation in the labor market, reducing
violence against poor women, and redressing the gender imbalance in the division of
household work so that men and women share parenting, care giving and domestic chores.
115.  The highest incidence of poverty and vulnerability is amongst families with children
under the age five, particularly those headed by women. Given the demonstrated effects on a
reduction in poverty of increased economic participation by women, a program of financial
transfers (along the lines of Biolsa Escola and PETI) targeting these households and linked to
early childhood development: and education centers that include the provision of daycare
could be particularly effective in reducing poverty.
116. Lastly, given the permnanent state of volatility and risk associated with globalization
and the liberalization of marliets, having a better understanding of how households are
affected by economic fluctuations would also help to better design social safety nets.
Specifically, data collected and analyzed in Brazil needs to better reflect the heterogeneity of
households. As in other countries, households are typically categorized as either male- or
female-headed; social programs are then targeted to male- or female-headed households
depending on their poverty conditions. While this classification represents the easiest form
of analysis, it is simplistic in that it does not take into account the multitude of family
arrangements that have come to exist in Brazil. Moreover, it can lead to misleading and
erroneous conclusions. For example, the concept assumes a hierarchical relationship
between household members that may or may not be present. It also implies that the head is
the most important person, tchat the head is present in the household, that the head has
overriding authority in household decisions and that the head provides consistent and
central economic support. T'hese commonly held assumptions have proven to be inaccurate
in describing typical households in Latin America. Thus data needs to be collected and
analyzed in a way that better captures the heterogeneity of household structures, as well as
describes household dynamnics, decision making, coping strategies and responses to
incentives, by household composition and gender roles of household members. Data
collected in Mexico and Wo:rld Bank research carried out using that data provide a good
precedent, which Brazil could follow.25
2 See Cunningham, 1999.
46



V. REFERENCES
Arriagada, Irma, 1998. "The Urban Female Labor Market in Latin America: The Myth and the
Reality." ECLAC: Women and Development Unit Series #21.
Beltrio, K.I. et Camarano, A. "Caracteristicas S6cio-Demograficas da Populacao Brasileira." Revista
EstudosFeministas, Vol.5, No. 1: 106-119.
Barker, Gary, 1998. Boys in the Hood, Boys in the Bairro: Exploratory Research on Masculini'y, Fatherhood and
Attitudes Towards Women Among Low Income Young Men in Chicago and Rio de Janeiro Bra#l. Paper
Prepared for the IUSSP/CENEP Seminar on Men, Family Formation and Reproduction,
13-15 May 1998, Buenos Aires, Argentina.
BEMFAM, 1996. Pesquisa Nadonal sobre Demografia e Saude (PNDS). Rio de Janeiro: BEMFAM.
Beneria, L., 1992. "Accounting for Women's Work: The Progress of Two Decades." World
Development, Vol. 20, No. 11: 1547-1560.
Berquo, E., 1996. A SaAude Reprodutiva no ContextoAtual. Campinas: UNICAMP/NEPO.
Bruschini, C., 1998. "Trabalho Feminino No Brasil: Novas Conquistas ou Persistencia da
Discriminacio?" XXI International Congress, Latin American Studies Association, Chicago:
24-26 September.
Deutsch, Francine M., 1999. Halving itAll. How Equally Shared Parenting Works. Boston: Harvard
University Press.
Carranza, M., 1994. "Sauide Reprodutiva da Mulher Brasileira," in MulherBrasileira eAssim. Edited by
H. Saffioti and M. Munoz-Vargas. Brasilia and Rio de Janeiro: Rosa dos Tempos,
UNICEF/NIPAS 1994: 95-150.
Carvalho, M.P, 1998, "Vozes Masculinas Numa Profissao Feminina." Estudos Feministas, Vol.6, No.
2: 406-422.
Coimbra Jr.. C. and R. Santos, 1994. Epidemiologic Profile ofAma.Zonian Ameridiansfrom Brazil.
Washington D.C.: The World Bank (unpublished manuscript).
Conde Lopez, V., 1990. " Descripci6n de los pacientes diagn6sticados de 'Sindrome de la
Dependencia de Alcohol' y 'Abuso del Alcohol,' segiin criterios C.I.E.-9a de la O.M.S., en
un servicio de hospitalizaci6n psiquiatrica." Actas-Luso-Esp-Neurol-Psiquiatr-Cienc-Afines. 1990
May-Jun; 18 (3): 193-204.
Correia, M.C., 1998. Gender Pornfolio Review: Argentina, Chile, Unrguay Program. Washington D.C.: The
World Bank (unpublished manuscript).
47



Costa, A.M., 1999. Uma Po/itica de Assistencia Integral d Saude da Mulher a Ser Resgatada. Sao
Paulo: Commissao cle Cidadania e Reproduc,ao.
Cunningham, W. "Whose Double Shift? Household Allocation of Domestic Labor in
Macroeconomic Fluctuations." Washington D.C.: World Bank (mimeo).
----------------1999. "Breadwinner versus Caregiver: Labor Force Participation and Sectoral
Choice over the Mexican Business Cycle." In The Economics of Gender in Mexico. Edited
by E. Katz and M. Correia. Washington D.C.: World Bank, 2001: 85-132.
De Beauvoir, S., 1953. The Second Sex. H.M. Parshley (Iranslator)
ENAP, 1998. Diagnostico da 6ituaqfo daMulhernaAdministrafdo Puibica Federal. Relat6rio de Pesquisa,
Brasilia, April.
Fiess, Norbert M. and Vemer, Dorte, 2001. The Dynamics of Povery and its Determinant.s:
The Case of the Northeast of Brazjl and its States. Washington D.C.: World Bank (draft).
Ferreira, Francisco H.G. and Leite, Philippe G., 2001. The Effects of Expanding
Education on the Distriution of Income in Ceard: A Micro-simulation, PUC, Rio de Janeiro.
Fischer, I.R. and L. Albuquerque, 1997. Women andAgriculturalModernigation in Northeast Bra#j/.
Ottawa: International Development Research Center.
(source: http://wwwidrc.ca/socdev/document/Women/intro.html).
FLACSO, 1995. Latin American Women Compared Figures. Santiago de Chile: FLACSO.
Floro, M., 1994. "Work Intensity and Women's Time Use." In Color, Class and Country: Experiences of
Gender. Edited by G. Young and B. Dickerson. London and New York: Zed Press: 162-181.
Gateta Mercantil, 'Comodidade, a falsa armadilha da cesariana', Sao Paulo, December, 7, 2001.
Htun, M. N., 1998. Women's Political Participation, Representation and Leadership in Latin America Issue
Brief. Washington D.C.: International Center for Research on Women, November 1998.
Instituto Brasileiro de InovaS6es em Sauide Social (IBISS), 1998. Exploraf,co Sexual Infanto-Juvenil e o
Turismo em Mato Grosso do Sul.
Jorge, M. H. P. de M., 1998. "Mortalidade por Viokncia." In Seminirio Nacional sobre Emprego e
Violencia. Brasilia: Comissao Nacional de Populacao e Desenvolvimento.
Junho Pena, M.V., 1991. A Situafdo da Mulher Brasileira na Decada de 80. Report to the Inter-
American Development Bank.
Junho Pena, M.V., 1981. Presenfa Feminina na Constituifdo do Sistema Fabril Rio de Janeiro: Editora
Paz e Terra.
48



Lavinas, L., 1996. "Aumentando a Competitividade das Mulheres no Mercado de Trabalho." Estudos
Feministas, Vol. 4, No.1: 171-182.
Leal, Ondina, 1998. "Reproductive Culture and Sexuality:" Estudos Feministas, Vol, 6 No 2: 425-437.
Linhares, L., 1998. "Genero e Violencia." In Seminario Nacional sobre Emprego e Violencia.
Brasilia: Conselho Nacional de Popula,io e Desenvolvimento: 103-108.
-----------------1996. " Genero, Trabalho e Legisla,co Trabalhista no Brasil. Estudos Feministas, Vol 4,
No 2: 447-463.
Morrison, Andrew R. y Maria Loreto Biehl (Editors), 1999. Too Close to Home. Domestic
Violence in theAmericas. Washington, D.C.: Inter-American Development Bank.
Moser, C., 1997. Household Responses to Poverty and Vulnerabiliy, Volume 1, Confronting Crisis in Cisne
Dos, Guayaquil, Ecuador. Washington D.C.: World Bank Urban Management Program.
Moran, R. and Mouras Castro, 1997. Street-children and Inter-American Development Bank - Lessonsfrom
Brajl. Washington D.C.: Inter-American Development Bank Social Programs Division.
Nunes, A., Piola, S. F., 1998. "Internaq5es Hospitalares do SUS: Caracteriza,co por Sexo e Grupos
de Idade." In Politicas, Mercado, Etica. Edited by M. Arilha, M. T. Citeli. Sao Paulo: Editora 34:
111-135.
Pereira de Melo, 1998. 0 Seruifo Domistico Remunerado No Brasil: De Criadas A Trabalhadoras. Rio de
Janeiro: IPEA.
Posthuma, A.C., 1998. "Mercado de Trabalho e Exclusao Social da For,a de Trabalho Feniinina."
In Ginero e Trabalho na Sociedade Latino-Americana. Edited by Abramo, L. and A. Paiva Abreu.
Rio de Janeiro: Associa,co Latino Americana de Sociologia do Trabalho.
Republica Federativa do Brasil, 1994. Relatdrio Geral Sobre a Mulher na Sociedade Brasileira. IV
Conferincia Mundial sobre a Mulher Afdo para Igualdade, Desenvoluimento e Pa!, Pequim 1995.
Brasflia: December.
Rosemberg, F. A., 1997. Expansdo da Educaf,o Infantil no Brasil e Processos de Exclusdo. Sao Paulo:
FAPESP (Research Report).
Sant'Anna, W. and Paixao, M., 1999. "Muito Alem da Senzala." Unpublished.
Silipandri, E., 1998. Profissionali:afdo para a Mulher Rural e a Diviseo Sexual do Trabalho. Brasilia:
CONTAG, Oficina Nacional de Forma,co Profissional para Mulheres.
Spalliacci, L, 1997. Estudos de Algumas Variaveis Maternas Reladonadas d Prematuridade. Sao Paulo: M.A.
Dissertation for the Universidade de Sao Paulo.
Steliner, Morton, Smith, Breslaw and Monette "Labor Force Behavior and Earnings of Brazil
Women and Men." (Unpublished).
49



UNDP, 1995. The Human Development Report. New York: Oxford University Press.
V/EJA, 'Aborto: uma pratica em queda livre', Sao Paulo, November 28, 2001.
Women's Environment and Development Organization (WEDO), 1998. Mapping Progress: Assessing
Implementation of the Beying Platform 1998. New York.
World Bank, 1997a. Alcohol Polig as a Means of Controlling Crime and Violence: A Retiew) of the Research.
Washington D.C.: The World Bank (unpublished manuscript).
World Health Organization, I999. Violence Against Women Information Pack: a Priori¸y Health Issue.
50















IMAGING
Report No.:        23442  BR
Type:     SR