Essays in labor economics and panel data analysis

Evan S Totty, Purdue University


This dissertation is composed of three independent chapters. The first chapter studies the impact of minimum wage hikes on employment of low-skill workers in the United States. The second chapter studies the lasting impact of attending a higher value-added high school on college performance. The third chapter studies the impact of stuttering on labor market outcomes. The first chapter resolves issues in the minimum wage-employment debate by using factor model econometric methods to address concerns related to unobserved heterogeneity. Recent work has shown that the negative effects of minimum wages on employment found using traditional methods are sensitive to the inclusion of controls for regional heterogeneity and selection of states that experience minimum wage hikes, leaving the two sides of the debate in disagreement about the appropriate approach. Factor model methods are an ideal solution for this disagreement, as they allow for the presence of multiple unobserved common factors, which can be correlated with the regressors. These methods provide a more flexible way of addressing concerns related to unobserved heterogeneity and are robust to critiques from either side of the debate. The factor model estimators produce minimum wage-employment elasticity estimates that are much smaller than the traditional OLS results and are not statistically different from zero. These results hold for many specifications and two datasets that have been used in the minimum wage-employment literature. A simulation shows that unobserved common factors can explain the different estimates seen across approaches in the literature. The second chapter studies whether there is any lasting benefit to attending a better "value-added" high school. Many states and school districts have tied teacher and school evaluation to value-added scores, which attempt to assess the impact that teachers and schools have on student test scores. Much work has been done to develop value-added models that can provide unbiased estimates of the impact that teachers and schools have on test scores. Much less work has been done to determine whether these higher value-added teachers and schools have a lasting impact on their students. This chapter address this issue by estimating value-added scores for high schools in Florida and North Carolina and then linking these value-added scores to a student-level dataset on college performance that includes the high school that the student attended. Results suggest that attending a higher value-added high school has a positive and statistically significant impact on college GPA, although the impact is small; a one standard deviation increase in high school value-added increases first semester GPA by 0.022 points and final GPA by 0.017 points. The effect on graduation is not statistically significant. Results are consistent across race and gender, although the effect is slightly larger for White students. The third chapter quantifies the difference in labor market outcomes between people who stutter and people who do not. Stuttering is a neurodevelopmental disorder of speech which is characterized by repetitions and prolongations of sounds, syllables, and words. It is estimated that 70 million people stutter worldwide and 3 million in the United States. Despite this prevalence, no study has adequately quantified the impact of stuttering on adult labor market outcomes; considering the ubiquitous nature of communication, it is likely that communication differences and difficulties would affect labor market outcomes. Results suggest that stuttering impacts employment status, labor force participation, receipt of public assistance, and hourly earnings. Additionally, stuttering seems to impact males and females differently.




Mumford, Purdue University.

Subject Area

Economics|Labor economics|Economic theory

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server