Do Conditional Cash Transfers Improve Education and Labour Market Outcomes in the Long Run?

This paper estimates the long-term impacts of PROGRESA, the Mexican flagship conditional cash transfer programme, on the educational attainment and the labour market performance of children aged 18 years or younger in 1997. Based on the household panel in 1997-2017, we utilise the initial experimental design where the programmes were allocated randomly at the village level in a phased manner. We estimate the intent to treat (ITT) effect by Propensity Score Matching (PSM) with (i) the correction for attrition bias and (ii) controlling for both unobservable and unobservable household characteristics utilising the panel data in 1997-2007. We also estimate the Unconditional Quantile Regressions (UQR) with the same corrections to examine the heterogeneous treatment effects. PSM results suggest that after 20 years, children of the poor eligible households in the early treated villages outperformed the matched children in the control villages. Early beneficiaries achieved better educational attainments in both durations and levels and were more likely to work and earn a higher salary in the regular and non-regular labour markets. We also find spillover effects on children of non-poor households in the treated villages for education and voluntary/non-salaried work. Disaggregation by age groups and gender shows that early-age nutritional support has a long-term positive effect on education and employment for females, while early educational support benefited males rather than females in terms of tertiary education, employment and wages in the regular market. However, UQR results show that the early exposure to PROGRESA increased overall education inequality, while it decreased inequality in education among women and in labour income for both men and women.