Essays on Firm Behavior

Abstract

This dissertation studies three questions in which economic behavior at the firm level plays an important role. Chapter 2 studies how individual owner-managers shape firm conduct in the labor market. I use survey and administrative microdata from the U.S. Census to link firm and worker outcomes to the past local unemployment rate exposure (URE) of owner-managers. Using a difference in differences approach centered on changes in firm ownership, I find that firms acquired by high URE owner-managers increase worker earnings on average while displaying no differential trends in firm employment. These results also hold in worker-level analysis, which reveals that firm-level differences are driven in part by immediate pay increases for older and more educated workers, resulting in greater retention among these cohorts. These results are further validated among firms that do not experience ownership changes. Using an instrumental variables design, I find that URE is associated with greater rent-sharing at the firm level. Together, these results demonstrate that owner-managers have substantial scope to determine pay and hiring policy at their firms. Chapter 3, coauthored with Mark Curtis, Daniel Garrett, Eric Ohrn, and Juan Carlos Suarez Serrato, studies plant-level responses to a large federal tax incentive, known as bonus depreciation, that lowered the cost of capital investment. Difference-in-differences estimates using confidential Census Data on manufacturing establishments show that tax policies increased both investment and employment, but did not stimulate wage or productivity growth. Using a structural model, we find that the primary effect of the policy was to increase the use of all inputs by lowering costs of production and that capital and production workers are complementary inputs in modern manufacturing. Our results show that tax policies that incentivize capital investment do not lead manufacturing plants to replace workers with machines. The fourth chapter assesses how state and local taxes influence firm entry decisions in the video gambling industry in Illinois, which comprises almost 7,000 establishments each operating up to five slot machine-like gambling terminals. Using variation in local gambling ordinances and an event-study framework, I estimate that gambling legalization leads on average to a 3.0% increase in local tax revenue and a noisy 1.6% increase in local spending. I then develop and estimate an equilibrium model of entry and exit to explore the effects of counterfactual tax increases. Simulations reveal that uniform increases in the marginal tax rate increase tax revenue while reducing the extent to which gambling establishments select into low-income neighborhoods. Taken together, these results suggest that taxation can effectively offset the regressivity of gambling activity if revenue is effectively targeted to local governments.