Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Vincent C. Brenner


The individual tax compliance gap persists despite costly enforcement efforts that attempt to dissuade noncompliance by identifying noncompliant taxpayers and imposing sanctions on those identified. Enforcement strategies that reduce the psychological incentives of noncompliance may complement the dissuasive approach. The present study probes this psychological alternative by examining several factors that are potential determinants of how taxpayers perceive their tax return results, and accordingly, how much reporting risk they are willing to assume. Noncompliance is modelled as a two-tier process. The first relies on adaptation level theory that suggests taxpayers frame the filing of their current year tax returns with respect to the tax consequences of prior years' returns. The second relies on prospect theory that suggests taxpayers in loss frames are more risk prone (noncompliant) than those in gain frames. Sixty-four taxpayers participated in an experiment used to test the noncompliance model. Participant expectations were based on their individual prior tax return liabilities and refund status amounts. The task presented them with current year "estimates" that reflected increased or decreased liabilities with respect to prior amounts and increased or decreased refunds or additional amounts of tax owed with respect to these prior measures. Subjects rated their satisfaction with the estimates and their inclinations to modify the current returns to include ambiguous deductions. Results are supportive of both tiers of the model. Increases in tax liability, decreases in refund amounts, and increases in additional tax owed induce loss frames. Changes in the opposite directions induce gain frames. Taxpayers in loss frames are more inclined to include ambiguous deductions than those in gain frames. Cash position, the presumed framing determinant in previous studies, induces gain and loss framing, respectively, when there is a refund due and when additional tax is owed. The effects of frame on risk diminish over time although reported frame intensity does not appear to decline. Implications for formulating policy that may minimize changes and statuses that induce loss frames are discussed. Suggestions for further research regarding framing and timing are made.