This study comprises three essays exploring labor market discrimination using new data, a new application of an equilibrium search model, and a new game theoretic model of the dynamics of economic discrimination. In the chapter "Testing Standard Theories of Economic Discrimination: Productivity, Prejudice, and Lost Profits During Baseball's Integration" evidence from the integration of white professional baseball is used to explore the empirical dynamics of integration, and in so doing reveal the nature of the discrimination present in that market. An important finding is that owner discrimination appears to be the only traditional model of discrimination present in the market. Estimates of the profits forgone by owners are also generated, and are both statistically significant and substantial. In "Opportunities Denied, Wages Diminished: Using Search Theory to Translate Audit Pair Study Findings Into Wage Differentials," a new application of a search model of discrimination is used to estimate the extent to which documented levels of hiring disparity affect the economic outcomes of job seekers. A key finding is that even seemingly small differences in hiring rates can lead to substantially different realized wages. Perhaps even more important than the findings is the use of a theoretical tool to translate findings from audit studies of the labor market into more relevant metrics. In the third essay "Do Wages Approach Value When Productivity Signals Are Private?" a game theoretic model where only tenure and wages are publicly observable is posited. It is found that wages should converge to productivity even in this market of limited information. The model's predictions are also consistent with the stylized fact that a black-white wage gap persists at the high-skill end of the distribution, yet no "reverse gap" exists at the low-skill end. Additional empirical evidence is offered that is consistent with the dynamics proposed by the model. In combination, these three essays improve upon our understanding of economic discrimination by empirically testing various models of discrimination, translating audit study findings into more relevant metrics, and positing a model of employer learning that incorporates private signals.