Doctoral Dissertations

Date of Award

5-2005

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Economics

Major Professor

William F. Fox

Committee Members

Kenneth Anderson, Donald Bruce, Michael McKee, Matthew Murray

Abstract

Provided herein is policy relevant empirical evidence on the effects of tax policy on three different areas of behavior: business tax planning, entrepreneurship, and personal income tax compliance. A study of business tax planning is meritorious in light of the recent erosion of state corporate income tax bases and the corresponding search for solutions. The effect of tax policy on entrepreneurship is especially important given the attention that policymakers devote to development policies and the potential social benefits that entrepreneurship can generate. A consideration of income tax compliance behavior among self-employed individuals is warranted due to the costly distortions that could arise from asymmetric tax compliance patterns across groups.

Essay 1

The focus of the first essay is on the extent to which business tax planning in response to variations in state tax policy has affected state corporate income tax bases and revenues. Tax planning is defined as a broad set of actions undertaken by firms to reduce their tax liability. Financial or accounting tax planning is contrasted with what is referred to herein as locational distortions, in which firms move physical operations to avoid higher tax liabilities. Results from a fixed effects instrumental variables regression model using a 1985-2001 panel of state-level data provide highly suggestive evidence that tax planning activity significantly diminishes taxable corporate profits in high tax states. Specifically, results indicate that state corporate income tax bases decline by around seven percent following a one-percentage-point increase in top marginal corporate income tax rate, controlling for locational distortions. Results also suggest that throwback rules are usually ineffective in restoring corporate income tax bases while combined reporting requirements are often effective. Further analysis indicates that tax planning has not diminished the locational distortions of tax policy.

Essay 2

In Essay 2, a 1989-2001 panel of state-level data is used to examine the influence of state tax policies on entrepreneurial activity. Entrepreneurship is defined specifically as self-employment. Fixed effects regression results indicate that marginal tax rates for corporate income, personal income and sales taxes do not have statistically significant impacts on state entrepreneurship rates or state shares of the national entrepreneurial stock. However, other key aspects of state tax policies such as combined reporting requirements, LLC provisions, and personal income tax progressivity are found to be important determinants of the observed level of entrepreneurial activity in many cases.

Essay 3

In the final essay, experimental methods are used to better understand personal income tax compliance when a portion of an individual’s income is relatively difficult to detect upon audit by the tax authority. This is motivated by the likelihood that individuals whose income is not reported by a third party, the self-employed for example, may be less likely to be found evading taxes relative to scenarios where there is third party reporting. Experimental methods provide several advantages for examining this issue. Most importantly, an appropriately designed experiment will allow for a better isolation and control of the fundamental influences arising from variations in income matching policy across modes of employment, relative to naturally occurring data. Results indicate that tax compliance rates decline as individuals earn larger shares of income that is not perfectly detectable. Results also indicate that tax compliance rates decline with higher income, lower audit rates, and higher tax rates.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Included in

Business Commons

Share

COinS