Essays on Applied Economic Theory

Abstract

This dissertation consists of three chapters (essays). The first two chapters share a common theme: the study of how the positive selection of employees -- identifying those who are more skilled and/or those who like their jobs more -- shapes the personnel practices of firms. The first chapter explores how firms' personnel policies -- retention and promotion -- shape employee careers with job moves both across and inside firms, whereas the second chapter focuses on the dynamic training investments of firms. The third chapter, instead, is an essay in "market design". In particular, I study the optimal design of a platform which hosts multiple auctions. The first chapter takes a step towards bridging the theory of external labor markets (ELM) and internal labor markets (ILM). A dynamic framework is constructed to understand how the personnel policy (e.g., retention and promotion) shapes the careers of employees with job moves both across and inside firms. The key insight of this chapter is that firms benefit from the positive selection of valuable employees in long-term employment relationships. In particular, an employee is valuable if she is highly skilled and/or likes her job. To begin with, the baseline model focuses on the retention policy in a labor market with firms competing for worker services. During working, each employee privately experiences a match quality, i.e., how much she likes a job, so she may change firms to find a better job. In equilibrium, firms are shown to adopt a dynamic screening mechanism -- wages are lowered in both the early careers and late careers of employees -- to positively select valuable employees, from whom a rent is captured. We then embed a job ladder in the baseline model. This extended model allows firms to also enhance the value of employment by promoting more able employees. The equilibrium of this integrative model characterizes the personnel policy (e.g., retention and promotion) of firms, which is often treated as a black box in macro-labor models. Most importantly, our model captures three sets of stylized facts in a single setting: (i) ELM results (e.g., findings concerning job moves across firms); (ii) ILM results (e.g., findings concerning job moves inside firms); and (iii) "interaction" results concerning ELM and ILM (e.g., findings concerning the relationships between job moves across firms and job moves inside firms). The second chapter explores the idea that firms make training investments on a dynamic basis. That is, in a multi-period setting, firms could provide training to workers during any period (stage) of their careers. By assuming that information concerning worker abilities is delayed and that abilities and training investments are complements, two novel results emerge in equilibrium. First, firms could delay training to concentrate training investments towards "positively selected" workers (in terms of abilities). Second, the level of overall training investments could be higher with delayed training than that without delayed training. Our model provides a potential explanation for common personnel practices at professional services firms. That is: (i) training is often delayed at professional service firms, e.g., career-development programs and sponsored MBA education; (ii) professional service firms offer more overall training than other firms. Finally, the third chapter is an essay in "market design". On many online and offline platforms, e.g., eBay and livestock auction houses, buyers often face the choice among several auctions which sell almost identical or indistinguishable items. In this chapter, I study the optimal design of such platforms. In particular, I consider a centralized platform with multi-unit competing auctions, i.e., a platform hosting several single-unit auctions that could differ in their reserve prices. With a full characterization of buyers' choices among auctions, I show that the optimal platform entails a uniform reserve price which coincides with the monopoly reserve price of Myerson (1981) for a revenue-maximizing platform owner. This suggests that on a decentralized platform where sellers each offer an auction and compete for buyers by undercutting reserve prices, a platform owner, who earns a commission fee proportional to sellers' revenues (e.g., 10% on eBay), should impose a floor on reserve prices to mitigate the price competition among sellers.