33753 TheWorld Bank A u g u s t PREMnotes 2 0 0 5 n u m b e r 1 0 0 Gender Gender issues in child labor Developing effective policies against child labor requires attention to gender differences among working children. This note reviews the available evidence. Worldwide, an estimated 352 million children higher total work participation rates at all ages. age 5­17 are engaged in economic activities Because household chores account for a grow- (paid and unpaid market production; ILO ing share of girls' work as they age, the under- 2002). This figure would undoubtedly be estimation of their total work resulting from higher if it was possible to estimate children's the exclusion of those chores rises accordingly. involvement in household chores. Work performed by girls may also be under- Working long hours in economic activities estimated because more girls than boys are or household chores can adversely affect chil- reported as being "idle" in household sur- dren's health and education (ILO 2004). veys--indicating that they are not in school Developing effective policies to combat child and are not working. In rural Morocco in labor requires understanding gender differ- 1998, for example, 38 percent of girls and ences among working children, because: 19 percent of boys were reportedly idle. "Idle" · Standard definitions of child labor tend to children might really be doing nothing, but underestimate girls' work. it is also possible that parents do not consider · Economic activities of boys and girls differ the tasks their children undertake to be "real" by country and industry. work (Cigno, Rosati, and Tzannatos 2002). · Determinants of child labor may differ Domestic chores, in particular, may be by gender. ignored, thus explaining why more girls than · Consequences of child labor may differ by boys are reported to be idle. gender. Standard definitions of child labor tend to underestimate girls' work Standard definitions of child labor focus on economic activities and on the "worst" forms of child labor (defined by the International Labour Organization as children's involvement in hazardous work, armed conflict, pornog- raphy, and the like). These definitions typically exclude household chores, and so underesti- mate children's work. This is particularly an issue for girls in developing countries, as they tend to perform more household chores than do boys. In rural India, for example, boys usually have higher participation rates in economic activi- ties than do girls (figure 1). However, when household chores are included, girls have from the development economics vice presidency and povert y reduction and economic management network Including time use modules in household time that children--especially girls--spend surveys provides information on the house- on household chores. hold chores performed by children and can help avoid underestimation of the total work Household welfare they perform. But most household surveys Much empirical evidence shows that children's do not include such modules. economic activity decreases as household wel- fare increases (Dar and others 2002). How- Economic activities of boys and ever, a more nuanced story is emerging from girls differ by country and recent research. Using land size as an indi- industry cator of household welfare for rural families, Global estimates of children's involvement Bhalotra and Heady (2003) find that girls' in economic activities hide substantial farm labor actually increases with household national differences. Boys are more likely welfare in Pakistan and Ghana. A similar effect than girls to be economically active in Latin is not found for boys. Ray (2000) shows that American countries such as Bolivia and while changes in household welfare influence Colombia, while girls are more likely to be the composition of girls' work (economic activ- economically active in African countries such ities relative to household chores), it has no as Côte d'Ivoire and Ghana (Grootaert and effect on the total hours they work. Patrinos 1999; Blunch and Verner 2000). Girls and boys also tend to be concen- Water infrastructure trated in different types of economic activ- In many countries, collecting water is largely ity. In an international overview, Ashagrie the responsibility of women and girls. Thus (1998) finds that boys are more concen- improvements in water infrastructure can trated in manufacturing, trade, restau- play a crucial role in reducing children's, rants, hotels, and transport, while girls are especially girls', work hours and increasing more concentrated in agriculture and per- their school attendance. A connection to a sonal services. The variation in girls' and public water network is expected to affect boys' economic activity by country and parents' decisions about their children's time industry points to the need for data col- allocations in two ways. First, it reduces the lection and analysis at the national and cost of purchasing water through private local levels. It also highlights the impor- schemes, freeing up household income for tance of identifying the needs of both boys school fees. Second, it reduces the time and girls when designing child labor pre- required to collect water, reducing the vention policies. opportunity cost of children's time. In urban Yemen, connection to a public Determinants of child labor may water network reduced the likelihood that differ by gender girls were reported to be idle by 13 percent- The differences in the amount of household age points, reduced the likelihood that they chores and economic activities performed were engaged in economic activity by 2 per- by boys and girls suggest that the determi- centage points, and increased the likelihood nants of child labor may also differ by gen- of school attendance by 16 percentage points. der. We examine the differential impact on The effects on boys were much smaller. These boys and girls of household welfare (as mea- results suggest that the largest portion of the sured, for example, by household spending increase in attendance comes from report- per adult equivalent), water infrastructure, ill- edly idle girls--indicating that many of these ness in the household, and adult employment. The girls had an important role in water collec- evidence suggests that increases in house- tion (Guarcello and Lyon 2003). hold welfare may not be sufficient to reduce the total work burden of children, and indi- Illness in the household cates that investments in water infrastructure, Idiosyncratic shocks to the household may health, and child care may help reduce the affect the time allocations of household mem- PREMnote 100 August 2005 bers differently. The limited research on this These results may reflect boys' ability to topic suggests that the time allocations of gain marketable work experience that they girls may be particularly sensitive to illness can use in their later careers, while girls mostly in the household. Peruvian data shows that gain experience as domestic workers that is illness among young children increases not transferable to other occupations. Where household work for girls but not for boys early economic activity is unavoidable, girls (Ilahi 2001). Using Indonesian data, Pitt and would benefit from gaining more transferable Rosenzweig (1990) find that a 29 percent rise work experience while staying in school. in the incidence of child illness decreases an older sister's likelihood of labor force par- Policy implications ticipation by 25 percent, decreases her like- A number of policy implications follow from lihood of going to school by 15 percent, and the evidence presented above. First, includ- Policies for increases the likelihood of her participation ing time use modules in household surveys in home care by 53 percent. would capture unpaid household chores per- reducing child formed by children, thereby providing more Adult employment accurate estimates of total work time. Cross- labor must When both parents are engaged in economic checking answers with several household mem- activity, children, particularly girls, often sub- bers (including children) further increases recognize that its stitute for mothers in performing unpaid the accuracy of such data and can reduce the household work. Ilahi (2001) finds that in prevalence of apparent idleness. The diversity causes may differ urban Peru, both girls and boys spend more of child labor practices across countries and time on household work when adult female industries further necessitates research and by gender employment rises, but the effect is larger for data collection at local levels, and highlights girls. Research on the effect of changes in the importance of considering the needs of the cost of child care on girls' school enroll- both girls and boys when formulating policy. ment offers further evidence, albeit indirect, Second, interventions aimed at reducing of this type of substitution. Glinskaya, Gar- child labor should address its specific causes, cia, and Lokshin (2000) find that in Kenya, and should recognize that these causes may a 10 percent increase in child care costs differ by gender. The determinants of child reduces older girls' school enrollment rate labor should be examined by running sepa- by 3 percent, while there is no significant rate regressions for boys and girls, or by inter- effect for boys. This evidence suggests that acting the gender dummy with the main access to low-cost child care could reduce explanatory variables. Doing so would improve girls' household chores and have positive understanding of the nature and size of dif- schooling effects. ferential effects, and enhance the targeting and effectiveness of policy interventions. Consequences of child labor Third, investing in water infrastructure, may differ by gender providing low-cost child care, and increas- Because males and females engage in dif- ing access to health care facilities can sig- ferent types of work, both early and later in nificantly reduce the time that girls spend life, the impact of child labor on future labor on household chores, thereby increasing market outcomes may also differ. Research their school attendance. Improving oppor- on Mexico and Brazil finds that the future tunities for income generation by fathers earnings penalties for engaging in early eco- and mothers can reduce the need for eco- nomic activity (with or without concurrent nomic activity by children, but care should schooling) are more severe for girls than boys be taken to prevent an increased burden (Knaul 2001; Gustafsson-Wright and Pyne of unpaid household work for older girls 2002). In Mexico, the returns to early work when their mothers go to work. experience can compensate males for the Finally, interventions aimed directly at associated earnings penalties, but this is not increasing children's schooling--such as pro- the case for females. viding subsidies for school fees, reducing dis- PREMnote 100 August 2005 tance to schools, and improving school qual- Gustafsson-Wright, E., and H. H. Pyne. 2002. ity--are also likely to reduce the prevalence "Gender Dimensions of Child Labor and of child labor. As with policies aimed at reduc- Street Children in Brazil." Policy ing child labor directly, these interventions Research Working Paper 2897. World may have gender-differentiated effects. Bank, Washington, D.C. Ilahi, N. 2001. "Children's Work and School- Further reading ing: Does Gender Matter? Evidence from Ashagrie, K. 1998. "Statistics on Working Chil- the Peru LSMS Panel Data." Policy dren and Hazardous Child Labour in Research Working Paper 2745. World Brief." International Labour Organization, Bank, Washington, D.C. Geneva. ILO (International Labour Organization). Bhalotra, S., and C. Heady. 2003. "Child 2002. "Every Child Counts--New Global Farm Labor: The Wealth Paradox." The Estimates on Child Labour." Geneva. World Bank Economic Review 17: 197­227. ------. 2004. "Helping Hands or Shackled Blunch, N. H., and D. Verner. 2000. "Revis- Lives? Understanding Child Domestic iting the Link between Poverty and Child Labour and Responses to It." Geneva. Labor: The Ghanaian Experience." Pol- Knaul, F. M. 2001. "The Impact of Child icy Research Working Paper 2488. World Labor and School Dropout on Human Bank, Washington, D.C. Capital: Gender Differences in Mexico." Cigno, A., F. Rosati, and Z. Tzannatos. 2002. In E. Katz and M. Correia, eds., The Eco- "Child Labor Handbook." Social Pro- nomics of Gender in Mexico--Work, Family, tection Discussion Paper 0206. World State and Market. Washington, D.C.: Bank, Washington, D.C. World Bank. Dar, A., N. Blunch, B. Kim, and M. Sasaki. Pitt, M., and M. Rosenzweig. 1990. "Esti- 2002. "Participation of Children in mating the Intrahousehold Incidence of Schooling and Labor Activities: A Review Illness: Child Health and Gender of Empirical Studies." Social Protection Inequality in the Allocation of Time." Discussion Paper 0221. World Bank, International Economic Review 31: 4. Washington, D.C. Ray, R. 2000. "Child Labor, Child School- Glinskaya, E., M. Garcia, and M. Lokshin. ing, and Their Interaction with Adult 2000. "The Effect of Early Childhood Labor: Empirical Evidence for Peru and Development Programs on Women's Pakistan." The World Bank Economic Review Labor Force Participation and Older 14 (2): 347­67. Children's Schooling in Kenya." Policy Research Working Paper 2376. World This note was written by Alexandra van Selm Bank, Washington, D.C. (Consultant, Gender Division, PREM Network) Grootaert, C., and H. A. Patrinos, eds. 1999. under the supervision of Nayantara Mukerji. The Policy Analysis of Child Labor: A Com- The author is grateful to Jean Fares, Anne Kiel- parative Study. New York: St. Martin's Press. land, Estela Rivero-Fuentes, Karen Mason, and Guarcello, L., and S. Lyon. 2003. "Children's Cecilia Valdivieso for useful comments. Work and Water Access in Yemen." A If you are interested in similar topics, consider report from Understanding Children's joining the Gender and Development Thematic Work--An Inter-Agency Research Coop- Group. Contact Waafas Ofosu-Amaah (x85872) eration Project. Florence. or click on Thematic Groups on PREMnet. This note series is intended to summarize good practices and key policy findings on PREM-related topics. The views expressed in the notes are those of the authors and do not necessarily reflect those of the World Bank. PREMnotes are widely dis- tributed to Bank staff and are also available on the PREM Website (http://prem). If you are interested in writing a PREMnote, email your idea to Madjiguene Seck. For additional copies of this PREMnote please contact the PREM Advisory Service at x87736. PREMnotes are edited by Paul Holtz and laid out by Suzanne Luft. Prepared for World Bank staff