Association for Public Policy Analysis & Management

This study investigates the interaction between early-life shocks and human capital investments in children’s development. While the early-origins literature shows that early life trauma has long-term consequences, it does not demonstrate whether later human capital investments have the potential to mitigate these negative consequences (Almond and Currie, 2011). The reason this area remains largely unexplored is because of the challenge of identifying exogenous variation both in children’s early-life environments and in subsequent human capital investments (Almond and Mazumder, 2013). To analyze this question, the paper exploits two sources of exogenous variation to arguably provide causal evidence. First, we rely on variation in early-life conditions resulting from a child’s exposure to severe rainfall shocks. Second, we use variation in later human capital investments that result from the availability of conditional cash transfers (CCTs), which provide families with incentives to invest in children’s health and education. Using a census of low-income population and administrative records from Colombia’s CCT program (N~30 million individuals), this study combines a natural experiment and a regression discontinuity design (RDD) that exploits a poverty score used to determine the eligibility/assignment of households to the program.  The intuition behind the RDD is that, households just below and just above the CCT eligibility threshold are statistically comparable except for their program participation. Furthermore, children on both sides of the cut-off are equally exposed or unexposed to negative shocks in their early years. As a result, among children around the cut-off, the discontinuity in the participation to the program provides an exogenous variation in human capital investments both for children exposed to and unexposed to early-life shocks.  This variation is used to determine the interaction between early-life environments and subsequent human capital investments. The outcomes we study are secondary and high school completion, grade repetition, and achievement test scores. This paper contributes to the literature by: i) providing novel evidence on whether social programs can compensate for early-life shocks, evidence that is so far very narrow and mixed; ii) combining large-scale administrative data, which are becoming very important in measuring long-term outcomes and are helpful to apply demanding identification strategies; iii) informing about potentially heterogeneous effects of social programs on children who suffer early life shocks. While theory predicts that human capital accumulation is a dynamic process of skill formation (Cunha and Heckman, 2007) – i.e., the level of skills at a given developmental period influences the productivity of current and future human capital investments--, there is actually little empirical evidence on these “dynamic complementarities” (Aizer and Cunha, 2014).  From a public policy perspective, exploring the relationship between endowments and investments helps identify what interventions could be more effective to mitigate early-life disadvantage and reduce long-run poverty and inequality.