2001年-世界发展银行全球_Childrens_Work_and_Schooling___Does_Gender_Matter__Evidence_from_the_Peru_LSMS_Panel_Data_44页_1mb

Children's Work and Schooling: Does Gender Matter? Summary

Core Content

This working paper investigates the determinants of time allocation among children in Peru, focusing on how gender influences their engagement in schooling, housework, and income-generating activities. Using panel data from the Peru LSMS (Living Standards Measurement Survey), the author, Nadeem Ilahi, explores whether changes in household welfare, adult female employment, sickness, and access to infrastructure affect the time use of boys and girls differently.

Main Findings

• Gender Differences in Time Use: Girls predominantly engage in housework, while boys are more likely to work outside the home. Housework is responsive to economic incentives and constraints.
• Impact of Household Welfare: Changes in household welfare have a greater effect on girls' schooling and labor than on boys'. Even though boys and girls have similar average educational attainment, girls' education is more sensitive to household welfare changes.
• Effect of Adult Female Employment: An increase in adult female employment leads to more time spent by children on housework, with a stronger impact on girls than boys. It has no significant effect on boys' outside labor.
• Sickness and Disease: Girls bear a greater time burden due to sickness and disease in the household, especially in rural areas. Sickness can lead to substitution effects where children are withdrawn from schooling to engage in housework or labor, with gender-specific impacts.
• Infrastructure Access: Lack of access to energy infrastructure negatively affects educational attainment for both boys and girls, but has little impact on their labor. Access to water and energy is crucial for time allocation decisions.
• Demographic and Life-Cycle Factors: Age and birth order significantly influence time use. Older children, especially girls, spend more time on housework. Birth order is associated with lower educational attainment and higher likelihood of working.
• Female Headship: Female-headed households are not significantly different from male-headed ones in terms of children's work patterns after controlling for age. However, female headship is used as an indicator of gender vulnerability and decision-making power.
• Ethnicity and Poverty: Indigenous children face higher poverty and lower educational attainment compared to non-indigenous children, even after controlling for other factors. This suggests that ethnicity plays a significant role in the determinants of time use and schooling outcomes.
• School Costs: Direct school costs (tuition, uniforms, transport) influence children's education and labor decisions. These costs are often endogenous and can lead to biased estimates if not properly accounted for. The paper aggregates school costs at the cluster level to mitigate this issue.

Key Variables and Models

• Time Allocation Equations: The paper models three main activities—schooling, housework, and labor—using reduced-form equations. These equations are estimated using a random effects model, which accounts for unobserved heterogeneity.
• Dummy Variables: The use of dummy variables for female headship, access to water, and energy infrastructure helps in capturing the impact of these factors on children's time use.
• Endogeneity and Instrumental Variables: The paper acknowledges the endogeneity of health and school costs and uses instrumental variables to address this, following the approach of Pitt and Rosenzweig (1990).

Policy Implications

• Gender-Sensitive Policies: The findings suggest that gender-sensitive policies are necessary to address the different impacts of household welfare and adult employment on girls and boys.
• Childcare and Safety Nets: Policies that provide childcare support and protect household incomes from shocks can help reduce the likelihood of girls being pulled out of school.
• Infrastructure Development: Improving access to basic services like water and energy can positively impact educational outcomes and reduce the burden on children, particularly girls.
• Education and Economic Incentives: The traditional approach to child labor and education often overlooks the role of housework, leading to an underestimation of girls' work and potential misguidance in policy formulation.

Methodology

• Panel Data Analysis: The use of panel data allows for the control of unobserved heterogeneity and provides insights into the dynamics of time use over time.
• Random Effects Model: This model is used to estimate the time allocation equations, assuming that individual-specific effects are random and uncorrelated with explanatory variables.
• Discrete Variables: The schooling and labor variables are treated as discrete, with values of 1 and 0 indicating full-time schooling or labor participation, respectively.

Conclusion

The paper concludes that while economic factors are important in determining children's time use, gender plays a significant role in shaping these patterns. Girls are more affected by household welfare, adult female employment, and sickness than boys. Therefore, policies aimed at reducing child labor and improving educational outcomes must consider the gender dimension, particularly in the context of housework and the broader socio-economic environment.