UC Berkeley

This dissertation contains three empirical studies on the impact of three distinct government policies, ranging from the provision of agricultural insurance in Mexico to improved pollution controls on school buses in California. All three papers take advantage of administrative data from the respective programs and use selection on unobservables designs to obtain causal estimates for the impacts of these programs. Chapter 1 estimates the impact of an index- based agricultural insurance offered to small-holder farmers by the Mexican government on their income and consumption following a negative weather shock, as well as investment decisions for the subsequent growing season. Chapter 2 analyzes how the nutritional quality of the meals provided through the National School Lunch Program (NSLP) affects students’ academic outcomes, as measured by standardized test scores. By studying the nutritional quality of these meals, as opposed to simply their availability, this analysis provides and important contribution to the body of literature on the educational benefits of the NSLP. In Chapter 3, I study the impact of the California Lower-Emission School Bus Program (LESBP), which seeks to lower the pollution emissions of school buses through replacements and retrofits, on students’ school attendance and standardized test scores.

The first chapter of this dissertation, co-authored with Alain de Janvry and Elisabeth Sadoulet, examines the ex-post, shock-coping impact of weather index insurance from a pioneering, large-scale insurance program in Mexico to cover smallholder farmers as a social safety net. Exploiting insurance thresholds as a source of plausibly exogenous variation in insurance payments, we find evidence that these payments allow farmers to cultivate a larger land area in the growing season following a weather shock. Households in municipalities receiving payments also have larger per capita expenditures and income in the subsequent year. These results suggest that the insurance payments can make smallholder farmers more resilient to shocks, although some of the full impact may be offset by reductions in remittances from abroad that act as informal insurance.

In the second chapter, coauthored with Michael Anderson and Justin Gallagher, we provide evidence on a topic of intense policy interest: improving the nutritional content of public school meals. Debate on this topic is frequently motivated by the health of school children, and, in particular, the rising childhood obesity rate. Medical and nutrition literature has long argued that a healthy diet can have a second important impact: improved cognitive function. We test whether offering healthier lunches affects student achievement as measured by test scores. We estimate difference-in-difference style regressions that take advantage of frequent changes in lunch-vendors California school districts and find that students served by healthy school-lunch vendor score higher on California state achievement tests. We do not find any evidence that healthier school lunches lead to a decrease in obesity rates. The test score gains, while modest in magnitude, come at very low cost, making this a cost-effective way to increase academic performance.

In the third and final chapter, I study the impact of a government program aimed at replacing old school-buses with inadequate pollution controls. Air pollution has been found to negatively impact children’s health, which in turn can affect their academic achievement by causing absences from school, among other mechanisms. School buses are important sources of exposure, both because older models lack adequate pollution controls, and because they are more prone to self-pollution than other vehicles. As a result, the state of California established the LESBP to provide funding to replace and retrofit buses of model year prior to 1986. I find that bus replacements increase attendance in the average school district, with some evidence of larger effects in areas that are out of compliance for PM 10 and PM 2.5. I find no effect of the program on standardized test scores.