Essays on Tax Policy and Firm Behavior

Abstract

This dissertation consists of three chapters studying corporate tax policy and firm behavior. Chapter 1 (coauthored with Lucas Goodman, Adam Isen, and Matthew Smith) studies the impacts of limiting interest deductions on firms’ investment and financing choices using U.S. tax data. The 2017 law known as the Tax Cuts and Jobs Act (TCJA) implemented an interest limitation for big, high-interest firms. Using an event study design comparing big and small high-interest firms, we rule out economically significant impacts of the interest limitation on investment and leverage, and find evidence that the interest limitation led firms to increase their equity issuance. Complementary triple difference and regression discontinuity designs using different comparison groups yield similar estimates. Our results indicate many firms do not use debt as their marginal source of financing and provide evidence consistent with capital structure models with fixed leverage adjustment costs. Chapter 2 studies how firms respond to book income alternative minimum taxes (AMTs) by examining the AMT book income adjustment in 1987. Using Compustat data and an event study approach, I find no evidence that firms avoid the tax, and no evidence of significant real production or investment responses. Firm tax base responses imply an elasticity of book income of -0.03 [-0.63,0.56], smaller than previous estimates because I correct for mean reversion. The null results indicate that firms face strong, non-tax incentives to report high book incomes. Corporate tax incentives are one critical tool that governments use to influence economic activity, but not all firms benefit equally from tax incentives. Some firms receive larger and more immediate benefits, while financial frictions imply the value of additional cash varies across firms. Chapter 3 (coauthored with Eduard Boehm) develops a novel dynamic investment model featuring financial frictions, and a realistic tax loss system that changes the timing of tax incentives. The model flexibly replicates patterns in the data. We use the model to quantify the impacts of allowing tax losses to target investment incentives, and to consider the efficiency of alternative policies aimed at stimulating investment.