UC Santa Barbara

In many work environments, production occurs in teams. The value of that team’s product, the productivities of workers’ teammates, and workers’ individual productivities help determine the value of workers to a firm, their wages, and the lengths of their contracts. This dissertation uses data from Major League Baseball to investigate these relationships. The first chapter estimates the relationship between team production and firm revenue and uses the estimated relationship to comment on the valuations of workers made previously in the literature. The second chapter analyzes wages and provides evidence that the young in the sample are willing to trade away wages to join more productive teams. The third and final chapter tests whether worker productivity uncertainty affects contract length and finds that length does increase with uncertainty. An abstract for each chapter is provided below.

Chapter 1: This study finds evidence that the absence of firm fixed effects from regressions of Scully’s (1974) firm revenue equation leads to overestimating a player’s marginal revenue product by at least 164%. This result is consistent across two sets of seasons, is robust to two different measures of firm revenue and the most commonly controlled revenue sources, and occurs even in commonly used variations of Scully’s (1974) revenue equation. This finding suggests that studies that have previously used the estimates of a baseball player’s MRP or assume a conclusion drawn from such a study may need to be reexamined.

Chapter 2: This study finds that an average free agent trades away wages to join a team expected to be more productive. More importantly, the young in the sample drive this result: an average, young free agent trades roughly 25% of his wages to join a team with an expected productivity one standard deviation higher. In contrast, the wages of older free agents are unaffected by expected team productivity. These results are robust to a variety of wage-determinant controls, remain consistent across a set of robustness checks, and suggest that better teams provide an important human capital investment opportunity. High-quality measures of both workers’ own past productivity and the expected productivity of a worker’s future team provide key advantages to identifying these effects. This study is the first to show that the expected productivity of the team a worker will join produces a significant and negative compensating wage differential and may offer an opportunity to invest in human capital.

Chapter 3: This study finds evidence supporting worker productivity uncertainty as a contract-length determinant. This result is robust to a variety of worker- and firm-specific controls, is consistent across two different measures of uncertainty for two different types of worker productivity, and supports Danziger’s (1988) efficient risk-sharing hypothesis. This study improves upon previous studies that analyze the relationship between real uncertainty and contract length by using worker- and firm-specific data for the first time. This key advantage allows this study to control, at the finest level, for contract-length determinants that could complicate analysis when using more aggregated data, a common problem acknowledged in the literature. Specifically, it allows Danziger’s (1988) hypothesis to be tested with real uncertainty measures derived from the productivity history of individual workers. Finally, the sensitivity of a third measure of uncertainty provides an initial look into a promising area of future research.