UCLA

The added worker effect refers to the notion that married women increase their labor supply in response to an unemployment event by their husband. Previous literature has focused on developed economies and has generally found small or insignificant effects.

In chapter 1, I provide reduced form evidence of a large and significant added worker effect in the context of a developing country: Mexico. Wives whose husband lost their job are 11 percentage points more likely to enter the labor force, an entry rate that is 60% higher than wives whose husbands did not lose their jobs. Furthermore, the evidence suggests that entry is transitory as over 70% of wives exit the labor force within a year. Additionally, descriptive statistics suggest that wives who enter transitorily tend to choose low-skill, low-entry cost occupations such as domestic employees, street vending, etc. . . In this chapter I also develop and estimate a structural model of household time allocation decisions that captures the key determinants of household labor supply. The estimated model is then used to perform two counter-factual policies; one policy offers unemployment insurance (UI) at varying replacement rates, and highlights the important crowd out effects of UI on the added worker effect. The second policy studies the effects of shutting down access to low entry cost jobs for families with different levels of savings, and highlights the heterogeneous value of this sector for different families.

Chapter 2 provides a framework for understanding the decision of politicians to engage in corrupt activities. We developed and estimate a structural model of a politician's decisions to provide public goods and engage in corruption over the span of his political career. While our model is general, we develop and estimate it in the context of municipal governments in Brazil. Overall, we make two main contributions to the existing literature. First, the proposed framework captures many of the various mechanisms by which politicians choose to engage in corruption, which enables us to assess empirically the relative importance of these different mechanisms. This lies in contrast to the previous literature whose empirical evidence has been mostly based on simple correlations. As a second contribution, we use the estimated model to evaluate the effectiveness of anti-corruption policies that increase politicians' wages, induce a higher probability of being audited by a central authority, or increase term limits.