Three Essays in Public Finance and Labor Economics.

Abstract

The first paper of this dissertation utilizes unique data on teachers' union election certifications in Iowa, Indiana, and Minnesota to analyze the effect of teachers' unions on education production. I find teachers' unions have no impact on teacher pay, class sizes, or per-student expenditures, but they increase teacher employment by between 5 and 10 percent. Further, I estimate education production functions using high school dropout rates; my results are consistent with unions causing an increase in the returns to lower class sizes and higher teacher pay. These findings are in conflict with much of the past literature on teachers' union impacts, which I argue is due to measurement error in the union measure utilized in previous work.

The second chapter uses micro-data on cigarette consumption from the CPS Tobacco Supplement to estimate demand models that incorporate cross-state purchasing behavior. I find demand elasticities with respect to the home state price are indistinguishable from zero on average and vary significantly with the distance individuals live to lower-price borders. However, when smuggling incentives are eradicated, the price elasticity is negative, though still inelastic. I also estimate cross-border sales cause a modest increase in consumption, and 13-25 percent of consumers purchase cigarettes in border localities in the CPS sample. These findings imply traditional models of cigarette demand have not yielded a complete picture of the impact of cigarette taxes on smoking behavior and tax revenues.

The final chapter analyzes the increased time to baccalaureate degree and decrease in college completion rates in the United States over the last three decades. Using data from the NLS72 and NELS:88 longitudinal surveys, the increase in time to degree is found to be localized among those starting college at non-selective public and two-year institutions. There is no evidence changes in student characteristics explain the observed trends. The results indicate the increases in time to degree tend to be concentrated in states that experienced rapid growth in the size of their college-age cohort. There is also evidence of increased hours of employment among students, which is consistent with students working more to meet rising college costs.