Ajps 4.t 1 POLICY RESEARCH WORKING PAPER 28 97 Gender Dimensions of Child Labor and Street Children in Brazil Enily Gtustafsson-Wlright Hi,i, H,1i-, Pyne The World Bank Latin America and the Caribbean RegionE Gender Sector Unit October 2002 Poi.icy RESEARCH WORKI.NG PAPFR 2897 Abstract Gustafssoni-Wriglit and Pyne review child labor and the the labor market. The earlier a child begins to wvork, the sitLuation of street childreln in Brazil from a gendcr lower his or hcr earniigs. And girls are more adversely perspectivc. Relying primarily on Brazil's national affected by early labor force entry th an boys, with the hoUsehold snrvev for 1996, the auithors examinie various gender differential increasing the earlier a child begins to dimenisionis of child labor by genidcr, incluiding Nwork. participation, intensity, and type of activities; the Taking poverty as the primary coontributor to child relationship between child labor, cducation, and futLire labor, governenilt programs to comlbat child labor are earninigs; and the risks of child labor to health and wvell- wvell desigiled in that they compensate familics for a beiig. Thev also summarize approaches to prevent and child's forcgone earnings and address family factors that eliminiiate child labor and street childreni in Brazil. lead to poverty. However, programils coLI]d be improved The authors find that more bovs thani girls wvork in by explicitly considering the gcnder dimenisionis of child Brazil especially in rural areas where boys arc labor. The authors point to the nted for analysis of the concentrated in the agricultLral sector, that maniv impact of child labor on health, and specifically to the clhildrein bothi wvork and attend sclhool, and that girls gender and sex-diffeciretiated impacts. They stiggest the attain highier lcvcls of educatiotn thani boys on average, need to address gendcr in intervention strategies for even when considering number of hours worked. --hc street childreni, as well as research on chiild labor in exception is the 11-14 category. Thev also find that an domestic service where girls are overreprcsented. individual's carninigs are correlated with age of entry inro This paper-a product of the Gender Sector Unit, Latin America and the Caribbean Regioni-is part of a larger effort in the region to identify and address gender issuLes relevanit to devclopment. Copies of the paper are available free from thc World Bank, 18 1 8 HStreetNW, Washiligtoin, DC 20433. I'lcase contact Maria Correia, room 18-1 15, telepholne 202-473- 9394, fax 202-676-0199, cmail address mcorreia(Cworldbalik.org. Policy Rescarch \Vorkiig P'apers are also posted oni thc Weh at http://econ.worldbank.org. The aLuthors may be contacted at egustafssolln( w\orldbanik.org or hpyne(iw\orldliank.org. October 2002. (30 pages) The l'ohicvI Reseairch Working Paper Series disseminates the lznzdimgs of ivork in prm,'ress to enCnerage' the, exchange of ideas abouzt developmzent issies. An obljective of the series is to gCt the findings o(ut quickly. even if tbe preSentat;I(Sn re Iess than fullv polished. The papers canr the IManies nfthe ant bors and shonil( he cited accordinglv. The findings, interpretations, an1d 61cinchisonis expressed in this paper are enitirelv those o/ the authors. They do inot nhecessaril/ represenit the viv w/ othe World Bank, itS Executive L)irectors. or the c(ountries thely represent. P'roduced bv the Rescarch Advisory Staff GENDER DIMENSIONS OF CHILD LABOR AND STREET CHILDREN IN BRAZIL Emily Gustafsson-Wright and Hnin Hnin Pyne I. INTRODUCTION 1. This paper provides an overview of child labor and the situation of street children in Brazil from a gender perspective. Understanding the manner and extent to which girls and boys work and are living on the streets has important implications for methods of intervention. This paper details various dimensions of child labor - such as participation rates and types of labor in which girls relative to boys are involved - and examines characteristics of children who work, the relationship between child labor and educational outcomes and future earnings, and the potential health hazards and risks. 2. The paper is divided into three parts. The first part provides a brief background of the legal context in which child labor exists in Brazil and discusses the definitions of child labor and street children as used in the paper. The second part examines three dimensions of child labor - incidence, intensity, and activities - and highlights the differences between boys and girls. Furthermore, it investigates the possible relationship between child labor, education and future earnings, and corollary risks to children's health and wellbeing. The third part summarizes key approaches undertaken to prevent and eliminate child labor in Brazil with a gender lens. Throughout, the paper brings to light some of the gaps in the existing literature and proposes topics for future research. II. BACKGROUND The Legal Context in Brazil 3. Throughout the world, the reality of child labor and children in the streets precedes recorded history. Child labor has generally shown a decline as technology has improved, education has become more accessible, and a legal framework protecting the rights of children has been established. In many parts of the world however, especially the developing world, child labor remains a serious problem. In recent years, a rise in the attention paid to child labor and street children at an intemational level has been demonstrated by such policy actions as the creation of the UN Convention on the Rights of the Child in 1989 and the International Program for the Elimination of Child Labor (IPEC) of the International Labour Organization (ILO) in 1992. 4. In Brazil, civil society organizations such as workers' unions, churches, and non governmental organizations (NGOs) have been instrumental in lobbying for governmental programs to address child labor and street children. As a result, the Brazilian Government has made children a policy priority. In 1988, the federal constitution of Brazil pronounced children and adolescents to be an "absolute priority": article 227 states that "it is the duty of the society and the state to ensure that children and adolescents are an absolute priority - and that they have the right to life, health, food, education, leisure, training, culture, dignity, respect, familial and community relations - as well as to ensure that they are not subject to any form of negligence, discrimination, exploitation, violence, cruelty or oppression." The Government also supported the prohibition of nocturnal, dangerous or unhealthy labor for children under 18 years of age and any type of labor for children under 14 except in a situation of learning or professional training. In 1990, Law 8.069 (the Statute of Children and Adolescents, or ECA) article 60 ratified the prohibition of labor for children, which prohibits children under age 14 to work because of the harm to their physical, psychological and moral development. It also establishes the guiding principles of apprentice work, access to basic education and mandatory attendance, activities compatible with the adolescent's development, and a reduced workday. 5. In 2000, the Government ratified both ILO Convention 138 (anuary); which addresses the minimum age for employment; and Convention 182 (September), which addresses the effective abolition of the worst forms of child labor. The Government has set out to eradicate the worst forms of child labor by the year 2006. 6. In its policy on the protection of children, the Government also vowed to support programs and projects that stimulate equality in conditions of access and return to schooling as well as in the supply of socio-educational activities, sports, and leisure before and after the normal school day. 7. Programs to eradicate harmful forms of child labor in Brazil have included the National Forum for the Eradication and Prevention of Child Labor (1994), which is linked to the Ministry of Labor and receives technical and financial support from the ILO and UNICEF and involves civil society. The Forum concentrates on the critical forms of child labor, such as sugar cane, sisal mate and charcoal production. The Federal Program for the Eradication of Child Labor (PETI), which was established through the Ministry of Social Security and Assistance as a pilot in 1996, provides cash stipends to low-income families who keep their children in school and out of work. PETI also 2 offers sports and cultural activities to children during after-school hours to keep them away from work. Other programs include: the Bolsa Escola programs, which are minimum income guarantee programs anchored in the school systems of metropolitan areas, the Program for the Assurance of a Minimum Income (FGRM), and programs specifically designed to work with street children. (Sedlacek and Gustafsson-Wright, forthcoming). 8. Recently, in an effort to scale up and integrate separate cash-grant initiatives, the FGRM joined with Bolsa Escola (2 million beneficiaries) and PETI (400,000 beneficiaries in 2000) under the umbrella Alvorada Program. The Alvorada Program subsequently established the Sentinela initiative to target children at-risk of working in prostitution. Defining Child Labor and Street Children 9. The World Bank lacks a universal definition of child labor, as the definitions of 'children' and 'labor' are highly dependent on the locality of these elements.' However, in Brazil child labor refers to any work, paid or unpaid, for at least one hour per week and that any type of labor is illegal for children under 14 years of age.2 This paper considers the phenomenon of street children as one element of child labor and, at the same time, recognizes the additional issues, concerns, and risks that working in the streets poses. UNICEF distinguishes two types of children in the streets (Barker and Knaul, 199?). The first group of children works in the streets and has a home to return to at night (maintaining regular contact with family and contributing to family income), whereas the second group works and live in the streets, lacking regular contact with famnily. Activities in the street fall largely into the following four main areas: (a) vending candy, gum, or newspapers; polishing shoes; and washing cars; (b) begging; (c) illicit behaviors such as drug trafficking or other crininal acts; and (c) prostitution. Worst Forms of Child Labor 10. Some argue that not all types of labor are harmful to children. It is true, in fact, that in some areas, child labor is associated with a nurturing working environment in the home with family members. Informal education and training are some of the benefits of this type of labor. In addition, because children can be the principal contributors to family income, their labor can lSee World Bank Child Labor Initiative. 3 substantially improve living conditions for the family when the alternative is even more extreme poverty. Nevertheless, for many children, the type of labor in which they are involved can pose substantial harm to their physical and mental health.3 11. In Brazil, key activities identified as the priority forms of child labor are grouped by urban or rural residence.4 In urban areas, children may be involved in illicit activities (drug trafficking, prostitution, etc.) or street vending (newspapers or other products). In rural areas, activities may include the collection or production of charcoal, agave, cotton, vegetable products, sugar cane, tobacco, horticultural products, citrus, salt, flour, fish, wood, textiles, tiles or ceramics and work related to the extraction of stones and gems - i.e. mining. 12. Given the overwhelning nature of the problem, the definition of priority forms of child labor in Brazil is narrow, as its focus is on the eradication of activities that have a negative impact on physical rather than psychological wellbeing, and on manual labor conducted in public spaces, that is, work that is more visible. Boys tend to participate in these forms of work. As the "worst form of activities" are expanded to consider the private sphere of work and the psychological harms of child labor, girls child labor will become more visible and as such policies and intervention sttategies will begin to take into account the gender differentiated forms of child labor. III. GENDER DIMENSIONS OF CHILD LABOR AND ITS IMPACTS 13. This section examines three dimensions of child labor (incidence, intensity, and activity) and the anecdotal evidence on the correlation between child labor and education and future earnings, as 2 In Brazil, previous to 1991, work was characterized as fifteen hours per week. 3 See Sedlacek and Gustafsson-Wright, forthcoming. 4The worst forrns of child labor convention of 1999 (Convention 182) states that all members must take immediate and effective measures to secure the prohibition and elimination of the worst forms of child labor as a matter of urgency. For the purposes of the convention, "child" refers to all persons under the age of 18 and "the worst forms of child labor" comprises: (a) all forms of slavery or practices similar to slavery, such as the sale and trafficking of children, debt bondage and serfdom and forced or compulsory labour, including forced or compulsory recruitment of children for use in armed conflict; (b) the use, procuring or offering of a child for prostitution, for the production of pornography or for pornographic performances; (c) the use, procuring or offering of a child for illicit activities, in particular for the production and trafficking of drugs as defined in the relevant international treaties; (d) work which, by its nature or the circumstances in which it is carried out, is likely to harm the health, safety or morals of children. 4 well as the exposures and risks to children's wellbeing. Using a gender lens, the paper focuses on differences between boys and girls and on how societal-based gender roles shape the differences and consequences. Incidence of Child Labor 14. The Pesquisa NaconalAmostra do Domicilio (PNAD), a national household survey conducted in 1996, estimates that in Brazil 2.6 million children between the ages of ten and fourteen, or 15 percent of the population in the age group, work at least an hour a day.5 This suggests a large decline from 1993 when 4.5 million children worked. Concerning the absolute number of children working in 1996, the number of children working in rural areas exceeded that of urban areas by only 292,000; however, participation rates (proportion of children between 10 and 14 year who are working) are higher in rural areas (36 percent) than in urban areas (9 percent). In both urban and rural areas, twice as many boys as girls work (see Table 1).6 Table 1: Incidence of Child Labor (number of children working) and Participation Rates (percentage of all children between the ages of 10-14) - - -- - . Urba. -- Rural - Total Girls 379,811 6% 450,314 23% 830,125 10% Boys 771,759 12% 994,042 48% 1,765,801 20% Total 1,151,570 9% 1,444,356 36% 2,595,926 15% Source: Calculations based on PNAD, 1996 15. School and work are not mutually exclusive activities; many working children in Brazil also attend school, as Table 2 demonstrates. Boys are more likely than girls to be engaged only in work as well as to combine work with school. Gender differences are particularly stark in the rural areas where 12 percent of boys compared to 5 percent of girls are working exclusively, and 36 percent of boys compared to 18 percent of girls are both working and attending school. 5For consistency reasons this paper examines 1996 data. It is noteworthy that more recent evidence demonstrates that child labor continues to decline. 6 The sample used for this paper consists of 37,401 observations of children ages 10-14. Of this population 75-85% lives in urban areas, 50% are girls, and each age group comprises roughly 20% of the children. The most highly represented *states are Para, Mato Grosso do Sul and Goias. 5 Table 2: Proportion of Children at Work and/or School, by Sex and Area 7-Gl -/) ' - -Boy %) ; Gd % B-y./ I Work Only 1 2 5 12 Work and School 5 10 18 36 School Only 89 83 66 45 Neither work nor attend school 5 5 11 7 100 100 100 100 Source: Calculations based on PNAD, 1996 Intensity of Work 16. Intensity of work is indicated by hours worked per week or per day. Although the survey found no child working over 20 hours a day, about 5 percent of boys and 6 percent of the girls worked exceedingly long hours, that is, between 10 and 20 hours (see Table 3). Differences between the intensity of work carried out by boys and girls were not dramatic. It is surprising to find, however, that although a greater number and a higher proportion of children work in rural areas, rural children are more likely to work fewer hours than their urban counterparts. Table 3: Proportion of Working Children by Hours Worked Per Day Hours of r ., Urbah --- Ruiiral--- Per day. - - Girl- /)- - Boys (4o) idiO) B.(Gis) Boy(°;o)- 0 20 0 0 0 0 0 0 100 100 100 100 100 100 Source: Calculadons based on PNAD, 1996 6 Sectors of Work 17. As is to be expected, the majority of working children in the rural areas, particularly boys (95 percent), are involved in agricultural work. Ten percent of rural girls work in the service sector, with the majority (81 percent) working in domestic work. In the urban areas, on the other hand, child labor can be found largely in service and commerce sectors, the proportions being 30 percent and 25 percent, respectively. Agricultural work follows close behind with 20 percent. 18. Gender differences in the urban areas are noteworthy (see Table 4). About half of girls work in services compared to 20 percent of boys but boys edge out the girls in both the agriculture and commerce sectors. Again, the service-oriented jobs for girls are primarily in domestic service (77 percent), whereas for boys, service jobs are primarily food- and drink-related services (42 percent) and auto repairs (32 percent). Table 4: Percentage of Working Children by Sector, Region and Sex -~ ~ ~~~~~~~~w P,S,tl Co.- >'F, $..4, Agriculture 23.7 12.2 19.9 94.6 82.1 90.7 Industry 13.3 8.7 11.8 1.9 4.5 2.7 Construction 6.4 0.9 4.6 0.4 0.1 0.3 Other Industry 0.5 0.3 0.4 0.3 0.0 0.2 Commerce 26.3 21.5 24.7 1.5 2.1 1.7 Services 20.2 48.7 29.6 0.8 10.3 3.8 Aux. Services 2.5 3.1 2.7 0.1 0.0 0.1 Transportation 2.5 0.4 1.8 0.2 0.3 0.2 Social 1.5 2.4 1.8 0.1 0.3 0.1 Public 1.3 1.1 1.3 0.1 0.3 0.2 Others 1.7 0.7 1.4 0.1 0.0 0.1 100.0 100.0 100.0 100.0 100.0 100.0 Source: Calculations based on PNAD, 1996 7 19. A recent study on domestic workers indicates that the incidence of girls ages 10-16 who work in this sector increased by 10 percent.7 This is striking considering that overall levels of child labor have declined substantially over time. Today, 20 percent of all Brazilian women who work are domestic employees. While most wotking girls are white, in the sector of domestic service, sixty percent of girls are non-white. Girls who work in domestic service receive on average 60 percent of minimum wage and some girls who do domestic work (4 percent) do so for no pay at all. Lastly, girls working in domestic service tend to drop out of school (30 percent). 20. Locality of work indicates the type of activities in which children are involved (see Table 5). As the majority of children are working in the agricultural sector, particularly in rural areas, more than half of children (54 percent) are working on a farm, ranch, poultry farm, or small rural property. Gender differences in work locality are particularly pronounced, again reflecting the sectors that differentially engage child labor. About 20 percent of working girls are in their employer's home, and 12 percent are at home carrying out domestic chores, compared to 1.4 percent and 2.7 percent respectively among the boys. In contrast, working boys are prominent in localities such as roadways or public areas. Table 5: Distribution of 10 to 14 Year Old Working Children by Gender and Work Locality W'ork Locahrv -Boy's (°`) -Girls.(O/%) TotaI o(%) Store, repair shop, factory, office, school 26.0 17.1 23.10 Farm, ranch, poultry farm, small rural 57.6 46.4 53.8 property Home 2.7 11.9 5.8 Employer's home 1.4 20.5 7.8 Site designated by employer 2.6 0.3 1.9 Automotive vehicle 0.3 0.1 0.2 Roadway or public area 7.9 3.0 6.2 Other 1.5 0.7 1.2 Total 100.0 100.0 100.0 Source: Calculations based on PNAD, 1995 7 Lavinas, Lena (2000). 8 Types of Activities 21. The Brazilian government has defined the worst forms of child labor as activities involving the collection or production of charcoal, sisel (/ute), cotton, vegetable products, sugar cane, tobacco, horticultural products, citrus, salt, flour, fish, wood, textiles, tiles or ceramics and activities related to the extraction of stones and gems. In addition, most harmful activities specific to urban areas include drug trafficking and prostitution, street vending (newspapers or other products), etc. 22. Boys' and girls' involvement in the worst forms of child labor varies by type of activities. As indicated in Table 6, rural girls comprise a higher proportion of working children in babacu and acai palm plantations, tobacco, and sisel (ute) farms, whereas rural boys are doniinantin sugar cane, cotton, fishing, and coal production. Table 6: Percentage of Working Children (10-14 years of age) in Rural Areas by Activity and Gender Agricultural Actvity Rural -Total Urban Total -Boys GLrls . Boys GLrls Worst formns Cotton 1.2 0.3 1.0 0.2 0.0 0.1 Agave/sisel (ute) 0.0 0.3 0.1 - - Sugar cane 2.1 0.3 1.6 5.0 1.2 4.3 Tobacco 2.7 4.7 33 0.6 0.0 0.5 Babacu (palm) 2.4 6.4 3.5 - - - Acai (paln) 16.7 20.6 17.8 9.2 9.8 9.3 Fishing 1.2 0.6 1.0 9.0 3.8 7.9 Coal 0.2 0.00 0.1 - - Less Harmful Corn 22.1 22.7 22.3 12.4 8.3 11.6 Aipim (sweet cassava) 11.0 11.1 11.0 6.3 8.4 6.8 Cattle 15.0 4.4 12.0 15.0 3.8 12.7 Poultry 2.2 10.5 4.5 11.9 36.5 16.9 Rice 8.1 4.2 7.0 7.0 3.8 6.4 Source: Calculations based on PNAD, 1996 9 Working on the Streets 23. Most street activities, both legal and illegal, are consideted the worst forms of child labor in Brazil. This paper treats street children as a subset of child labor while recognizing that children working in the street represent a different set of issues and concerns. The risks and exposures that street children confront daily and the very public nature of their work and their hardships have heightened concerns about this particular aspect of child labor, as well as having attracted local and international attention (Freeman, 1997, Scheper-Hughes and Hoffman, 1997, UNICEF, 1993). 24. According to PNAD 1995, about 161,000 children or 6.2 percent of all working children, work in roadways or public areas. Table 7 shows the number of children working in the streets in three Brazilian cities. In the city of Salvador, an estimated 16,000 children work in the streets; however, only 3 percent live and sleep in the street. In fact, in the three Brazilian cities the percentages of children living or sleeping on the streets are relatively low if compared to Central America where an estimated 25 percent of street children sleep or live on the streets (Takahashi and Cederlof, 2000). Table 7: Estimate of Children in the Street in Selected Brazilian Cities Ciies' in Braid .-' Working in the;street- '-`Nunb`er ' d.proportion sleeping.i. C " 'r.-..ihe street - - Sao Paulo 4, 520 895 (19.8 percent) Fortaleza 5, 962 184 (3.1 percent) Salvador 15, 743 468 (3.0 percent) Source: Rizzini, Barker, and Cassaniga, 1998 25. Though findings cannot be generalized, a study conducted in Brasilia on street children and adolescents under 18 years of age provides an insight into the population's gender composition. The 1996 -1997 survey indicated that although the absolute number of children dropped dramatically in the second year, the gender composition remained the same: 89 percent of the street children were boys and 11 percent were girls. Overrepresentation of boys among street children is a pattern that is also witnessed in other countries in Latin America. Boys comprised 72 percent of the working youth population in the streets of Mexico (UNICEF, 1993), and 70 percent of those in Honduras (Takahashi and Cederlof, 2000). However, the Brasilia study found no gender differences 10 in the number of hours worked, the time of day/night worked, or the likelihood of sleeping in the street (Araujo, 1998). The reasons children work 26. Children work for a variety of reasons though it is commonly believed that one of the main reasons is poverty. 8 Poor families are forced to send their children to work to contribute to family income (see Table 8). Many families do not see themselves as having an altemnative, as children's earnings may be necessary for family survival. The constraint to work is placed on children when the opportunity cost of receiving an education (not working) is extremely high. Other factors that drive child labor include organization of work activities, social norms that encourage children to work, inability to pay school fees supplies or uniforms, inability to access schools due to distance/transportation difficulties and poor quality of school and teachers.9 Table 8: Incidence of Child Labor, by Income Quintile, Area and Gender - (percentage of all children between the ages of 10-14) ~$iome L4 F-ov'~~4 iU o l-T--- Ok Bo~ idToa Bb, M~F WM ir : To JSt quintile 14 6 10 52 24 38 31 14 22 2nd quintile 14 6 10 40 22 31 19 9 14 3rd quintile 12 6 9 40 21 31 15 8 12 4thquintile 9 5 7 36 21 29 11 6 8 5th quintile 6 4 5 28 18 23 7 5 6 Source: Calculations based on PNAD, 1996 27. A parent's decision to put a child to work is also influenced by the organization of a family's productive activities, induding the nature of contracting and payment systems. Child labor becomes part of the famnily labor, which is particularly true in rural areas. Another critical factor is the social norms that uphold child labor, which is viewed as an integral and positive part of a child's development. A working child is perceived as receiving discipline and training for his or her future 8 See for example Graiter and Lerer (1998), Grootaert and Patrinos (1999), Jensen and Nielsen (1997), Psacharopoulos (1997) and Tzannotos (1998). entry into the labor market as an adult. Furthermore, working occupies time, which might otherwise be spent "getting into mischief" (Moura, 1999). 28. The question of why children work can often be answered by examining how the money they earn is spent. Employers frequently pay rural children's earnings directly to their parents, as part of the family labor. In the urban areas, however, the situation differs as the types of jobs that children engage in differ substantially. Some children who work (but do not live) in the streets bring the money they earn back to their parents if it contributes significantly to family income. Evidence is lacking, however, on how the money is spent by the parents once they receive it. Children that do not return to a home with parents or another adult, spend the money on their own subsistence, and some, as the problem is increasing, may move to the purchase of drugs. 29. Gender roles affect parents and families' decisions on who (boy or girl child) works and where they work. Girls, for example, may be forced to take on more of the household chores as their male siblings leave the home to earn income for the family. As demonstrated previously, girls are over-represented in domestic work and other service sector jobs, in both urban and rural areas. Boys, on the other hand, are over-represented in the streets. This division of labor in part reflects gender roles and expectations. Engagement of girls in domestic service sector is shaped by the role of girls/women positioned in the biological and social reproductive spheres (maintaining a household, whether it is her own or that of others), and the role of boys engaged in the productive realm. In addition, girls tend to primarily occupy the private sphere, while boys, the public. Other reasons for girls leaving their homes to work and live on the streets may differ from the reasons offered by boys. It is not uncommon for girls who face physical, sexual or verbal abuse in the home to leave this environment in search of a better one. The streets may be their only option. Child Labor, Education and Poverty 30. The growing effort to address child labor stems from the potentially negative social and economic implications of children working. In many cases, employers are able to take advantage of the poor bargaining power of children, since children cannot legally form unions and therefore employers are able to subject children to working conditions to which they could not otherwise subject adult workers. The negative social or developmental impacts on children are measured by 9 If parents find that there are larger gains from obtaining work experience than education. 12 the effects of actual working conditions (hours worked, exposure to physical or mental strain or repetitive action, exposure to hazardous substances) and other physical or psychological risks or abuse. Economic arguments against child labor are based on the inability of children to accumulate human capital (education), which prevents them from having higher future earnings. Stronger markets for child labor also induce increased incidence of child labor and lower school enrollment rates.'° Some evidence suggests that child labor leads directly to lower years of schooling (Psacharopolous, 1997) or at least to learning less per year of schooling (Akabuyashi and Psacharopolous, 1999). The low rates of productivity growth contribute to the cycle of poverty and affect economic growth." 31. The need for children to work does not mean that parents are not concerned about their children's welfare or that the consequences of child labor are negative. Child labor and child schooling may not be mutually exclusive. In fact, many working children are also in school as demonstrated in Table 2. In some cases child labor may even increase household income sufficiently to make school affordable by helping families pay for transportation to school, school materials or uniforms. Child labor can also enhance the lifetime earnings potential of children directly based on the theory of learning by doing. Education 32. As discussed above, child labor may be associated with poverty and educational attainment but whether the relationship is negative, positive or some combination is not clear. Anecdotal evidence indicates that in Brazil, the incidence of child labor in households that fall in the bottom 20 percent (lowest quintile) of income distribution is 22 percent, compared to only 6 percent in the highest quintile. About 90% of heads of poor households have not graduated from primary school, and 74% have not completed 4th grade. The consequences of child labor in the present, however, are only part of the story. How child labor relates to the future - in terms of education and future poverty - poses an even more interesting question. Poor households tend to have low levels of hurnan capital, which in turn, is often a result of early labor market entry. tO Rosenzweig and Evenson (1977) and Levy (1985) found that higher child wages led to increased child labor participation and decreased enrollments. King, Orazem and Patemo (1999) found that stronger local child labor markets increased the incidence of school dropout. Sedlacek and Gustafsson-Wright (forthcoming). 13 Enrollment 33. Most Brazilian children, on average 94 percent, are enrolled in school, with little difference between girls and boys (94 percent and 93 percent respectively). For working children, average enrollment is only slightly lower at 83 percent. The numbers do not, however, reflect the low levels of educational attainment of Brazilians owing to late entry into school, slow progression, and high drop out rates. The numbers of children who lag in grade as well as the levels of educational attainment of working and non-working boy and girls are more pertinent; and a correlation may exist between these indicators although not necessarily causality. Attainment byvAge 34. According to the PNAD household survey, the educational attainment of children in Brazil is only about 55 percent of what it should be for their age. Boys tend to lag behind more than girls, especially when they are working (60 percent compared to 51 percent for girls)."2 The greater number of boys involved in child labor may explain part of this gender disparity in education lags."3 However, as mentioned previously, these numbers may be distorted by measurement error because many girls work in domestic labor, which often goes uncounted. 35. Contemporaneous evidence of educational attainment between girls and boys who only work, only study, do both or neither in rural and urban areas suggests gender differences. In rural and urban areas, the difference between boys and girls who only study is relatively insignificant. However, for children in rural areas who only work, the evidence clearly demonstrates a substantially lower level of attainment for boys (1.7 years) than for girls (2.5 years). In urban areas, girls who only work or also both study and work, also exhibit higher attainment on average than boys. 36. Although it does not necessarily indicate a causal relationship, it is perhaps more telling to examine the relationship between the number of hours worked and educational attainment (see Table 9). In rural areas, girls who work 10 to 20 hours per day have a 35 percent higher attainment rate than their male counterparts who work the same number of hours, and a 20 percent higher rate than boys if both are working 5 to 10 hours. In urban areas, attainment for girls working 5 to 10 12 Note that this considers children who are currendy enrolled in school and currently working and does not describe overall attainment. 1 Sedlacek, Gustafsson-Wright and Ilahi (2000) 14 hours is 18 percent higher than boys working the same number of hours, but it is only 3.6 percent higher than boys' attainment when they work 10 to 20 hours per day. However, because the evidence is contemporaneous rather than retrospective, the effects of work on educational attainment over the lifeti;me are unclear. Econometric analysis would be necessary to measure the precise magnitude of these effects. Table 9: Level of Educational Attainment for Age by Hours Worked for Girls and Boys Hours . :1 : :Urban :Rural- Total Worked Per Day. -Boys :Gils Tot Boys Girls Total Boys Girls Total h=O 3.37 3.79 3.59 2.14 2.51 2.35 3.18 3.54 3.37 O