UC San Diego

This dissertation is a collection of three essays on development economics. In the first essay, we randomly assigned beneficiaries of a conditional cash transfer program in Peru to attend a 3 hour training session designed to build their trust in financial institutions. We find that the intervention: (a) significantly increased trust in banks, but had no effect on financial literacy; (b) significantly increased savings over a ten month period, and (c) had no effect of the use of accounts for transactions.

The second essay estimates crime avoidance costs in the aftermath of homicides that occur near employees’ workplaces. I combine incident-level data on homicides with a matched employer-employee dataset for Sao Paulo City, Brazil, and estimate causal effects by exploiting timing and hyper-local variation in how close employees work to a homicide. Exposed employees experience a significant and persistent reduction in labor earnings due to a decrease in the hourly wage rather than a reduction in hours worked. In terms of incidence, I do not find evidence of firm labor market responses to homicides. On the contrary, I find that the effects are driven by employees switching to establishments that typically pay lower wages and are located in other municipalities. In addition, workers move to establishments located farther from the crime scene and in municipalities with lower murder rates, consistent with avoiding future crime.

In the third essay, I employ a close elections regression discontinuity design to study how political alignment affects the income and assets police officers disclose. Police officers in aligned municipalities report to have 5% more total income and 52% more net assets. The effects of political alignment are greater for nonadministrative police officers, those with higher tenure and those who work in higher crime areas. Taken together, these results are consistent with a corruption based explanation, either by an increase in extracted rents or a decrease in corrupt bureaucrats’ misreporting (i.e., through an effect in the financial disclosure law’s enforcement).