Microeconomic essays on market entry, optimal education, and measured experience

Abstract

This dissertation consists of three essays in applied microeconomics. The first essay investigates the effects of generic entry on post-patent price competition in the prescription drug market using NDC Health data on 18 oral solids that lost their patent sometime between February 1998 and 2002. I am able to characterize the impact of endogenous generic entry on branded and generic prices, conditional on payment type (i.e., cash, Medicaid, third party). Based on the findings in this paper, the overall, long-term impacts of the 1984 Drug Price Competition and Patent Term Restoration Act (Waxman-Hatch Act) are yet to be determined. The second essay develops a theoretical model of earnings where human capital is the central explanatory variable. The analysis and estimation strategy stems from the Mincerian simple schooling model. Human capital investments (i.e., schooling) are incorporated into a model based on individual wealth maximization. We utilize the conventional economic models of supply and demand to derive an optimal level of schooling function. Using the NLSY79, we stratify our sample into one-year work experience intervals for 1985-1989 to identify the "overtaking" cohort (i.e., the years of work experience at which an individual's observed earnings approximately equal what they would have been based on schooling and ability alone). We employ the AFQT score as an ability proxy and consider its possible endogeneity for several estimation strategies. The third essay attempts to address the bias inherent in the use of potential, as opposed to actual, work experience measures in human capital models. While such a proxy is often deemed reasonable for males, problems still exist---specifically, unemployment spells manifesting themselves as active job searches or withdrawal from the labor market. Presumably, such activities have different effects on one's work experience. Potential work experience measures also abstract away from employment status, over-time work, moonlighting, and multiple-job holding. We employ actual work experience data from the NLSY79 and the PSID and extend our findings to a data set in which actual measures of work experience are not available---specifically, the IPUMS, with the creation of predicted work experience measures.