Tax compliance: Are enforcement efforts working

In the late 1970's, the IRS audited 2.15% of all individual tax returns filed. This percentage dropped to 0.49% in 2001. In an attempt to keep compliance levels stable while reducing the number of audits performed, the IRS has continued to pile on penalties at an increasing rate (Graetz, Dubin and Wilde, 1990). However, despite the increased penalties, delinquent and uncollectible taxes have continued to increase (Hamilton, 2002). The purpose of this dissertation is to expand the deterrence research and provide additional information regarding the impact of enforcement efforts on tax compliance.; This study examines three issues concerning enforcement efforts. The first issue concerns the impact of prior audit activity on future compliance. The second issue is whether or not taxpayers consider the risk of audit and the risk of detection differently and if their reporting behavior differs depending on which type of risk is higher. The final issue relates to compliance behavior when taxpayers know they may receive or actually incur a reduction in assessed penalties.; This dissertation uses experimental economics to investigate the proposed issues. Evening and graduate students participate in the experiment in order to extend the generalizability of the results. These students are more likely to have tax filing experience.; Regarding prior audit activity, the results indicate that audits do increase compliant reporting in the period following an audit. These results suggest that there is a future economic benefit to performing an audit in addition to revenues generated during the examination.; The study shows no significant difference in reporting behavior when detection is presented as a single probability than when it is presented as a compound two-stage probability. Also, there is no significant difference in reporting behavior when audit risk is high relative to discovery risk than when discovery risk is high relative to audit risk.; The final issue relates to the assessment of penalties. The results suggest that participants report higher income after receiving moderate, but stable, penalties than when they are assessed high penalties that are subsequently reduced.