Essays on labor economics

Abstract

[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI AT AUTHOR'S REQUEST.] This dissertation consists of two chapters on labor economics. In the first chapter, we study redistributions via the United States Social Security retirement system for cohorts of men born during the second half of the 20th century. Our focus is on redistributions across race and education groups. The cohorts we study are younger than cohorts studied in previous, similar research, and thus more exposed to recent increases in earnings inequality. All else equal, this should increase the degree of progressivity of Social Security redistributions due to the structure of the benefit formula, but we find that Social Security redistributions exhibit little progressivity for individuals born as late as 1980. Differential mortality rates across race and education groups are the primary explanation. While black-white mortality gaps have narrowed some in recent years, they remain large and dull progressivity. Mortality gaps by education level are also large and unlike the race gaps are widening, which puts additional regressive pressure on Social Security redistributions. In the second chapter, I investigate the technique of imputing top-coded income data in longitudinal surveys. The incomes of top earners are typically top-coded in survey data to protect individuals' identities. Common imputation methods used to recover top-coded income values are limited in several ways when applied to longitudinal data. I show that the accuracy of imputed income values for top earners in longitudinal surveys can be improved significantly by incorporating information from multiple time periods into the imputation process in a simple way. Moreover, I introduce an innovative, nonparametric empirical Bayes imputation method that further improves imputation quality. With a sample of individuals for whom incomes are pseudo top-coded (i.e., in which the exact income figures are accessible but temporarily expunged), I show that the Bayesian imputation method reduces the root mean squared error of imputed income values by 19-46% relative to standard approaches in the literature. After documenting this improvement in performance, I illustrate the benefits of the Bayesian method for investigating multi-year income inequality.