Essays on Financial Fraud and Tax Evasion

This dissertation focuses on the problem of misreporting in the corporate setting, where managers may commit accounting fraud, and in the public sector, where taxpayers may not truthfully report their income. Both accounting fraud and income misreporting have contributed to unprecedented financial losses to shareholders and governments respectively. As a result, policy-makers and shareholders are focused on one goal, that is, to mitigate the occurrence of accounting fraud and income misreporting. The process of achieving this goal starts with understanding how compensation contracts and tax schemes influence an agent’s willingness to misreport. This dissertation pursues these objectives using a blend of theory, experimental techniques, and exhaustive empirical analyses. Chapter 1 has a theoretical focus; this chapter evaluates the incentive effects of various contracts within the class of stock option contracts. In this chapter, we develop a principal-agent model of managerial fraud to determine whether there exists a contract that ‘dominates’ another contract by generating relatively greater effort while minimizing fraud. While there exists an infinity of stockoption contracts that induce a given level of effort, we show that within the class of stock option contracts, any two contracts that induce the same effort must necessarily induce the same level of fraud. We also characterize the schedule of implementable effort-fraud pairs. Chapters 2 and 3 have an experimental focus; in Chapter 2, we implement the theoretical model in Chapter 1 and test whether contracts that are predicted to induce the same level of effort and fraud are behaviorally equivalent. The experiment produced strong results in support of our hypothesis. The predicted equivalent class of stock option contracts induced the same level of effort and the same level of fraud. In a behavioral sense, stock option contracts are the same as simple equity contracts. Chapter 3 focuses on tax compliance behavior under the progressive and the regressive tax systems in an experimental setting. This chapter contributes to the growing literature on tax compliance by experimentally testing whether tax compliance behavior of taxpayers is sensitive to either the progressive or the regressive tax system. All else constant, experimental results showed no difference in average tax compliance between the progressive and the regressive tax systems. However, fairness, risk-aversion, inequality aversion, and gender played an important role in explaining variations in tax compliance behavior