Essays on Urban Life and Labor Supply of Women

This dissertation focuses on two main research questions related to the effect of a factor in a local labor market. Both relate to a finding of Black et al. (2014) that married women are less likely to work in Metropolitan Statistical Areas (MSAs) with more traffic congestion. The first essay evaluates the appropriateness of an MSA as a geographical entity in estimating the effect of congestion on labor supply of married women. One concern with such aggregated data is that they do not take into account within-city variation in congestion. In order to address this issue, I replicate the works by Black et al. (2014) at smaller geographical levels. Once the coefficient on commute time at each level is estimated, I compare the coefficients from smaller geographical entities with one from the MSA to examine if they are statistically similar. I find that an MSA is a geographically proper unity when the effect of commute time on the LFP of married women is examined. Additionally, I explore whether commuting time has also a significant effect on other related to labor market issues. First, I find longer commuting time is associated with shorter weekly working time of high school educated women. Secondly, fewer married women are self-employed in the area with longer commuting time. The second essay begins with the possibility of correlation between congestion and the error term in the estimation equation by Black et al. (2014). The coefficient on congestion in their equation might be biased due to the endogeneity problem. I employ a structural approach with a multinomial logit in order to deal with the endogeneity issue. By examining the effect of congestion and its interaction term on city-specific fixed effects that can be estimated using a discrete-choice model of residential decision, I find that the negative relationship between congestion and labor supply of married women discovered by Black et al. (2014) is partially true. The effect of congestion is statistically uncertain or depends on the model specifications, with the exception of a finding that married women with children are especially responsive to the effect of traffic congestion on their willingness to work