Prospect theory is often used to predict individuals’ risky tax decisions. For example, individuals who are in a tax due (refund) position are predicted to engage in more (less) tax noncompliance. This is known as the “withholding phenomenon”. However, tax noncompliance is not just risky, it is also unethical. We hypothesize that because there is an ethical component inherent in this risky decision, the feelings described by prospect theory are insufficient to cause increased risk-seeking. In a series of experiments, we show that moral disengagement is the primary theoretical mechanism that explains noncompliant tax behavior. The feelings evoked from being in a loss domain provide motivation for individuals to morally disengage, while moral disengagement is directly related to noncompliant tax behavior. We also develop an intervention that deters individuals from morally disengaging, specifically when they are in a tax due position. These results have important practical and theoretical implications.

The Organization for Economic Co-operation and Development (OECD) reports that a male favoring pay gap exists in every one of its member countries. To reduce the gender pay gap, governments and investors are demanding that companies disclose gender pay information. While companies seem to resist these demands, there is little evidence about how investors might react to the disclosure of gender pay information. We draw on theories of fairness and the instrumental perspective of corporate social responsibility activities to predict how the disclosure of gender pay information influences investor judgments. Our experimental findings indicate investors are more willing to invest in a company that discloses gender pay equity compared to either a company disclosing a gender pay gap or one disclosing no gender pay information. Further, our mediation analyses results show a sequential mediation process whereby the gender pay disclosure affects perceptions of fairness (economic consequences) directly (indirectly through fairness), with only perceptions of the economic consequences resulting from the gender pay disclosure directly influencing willingness to invest. In addition, despite the presence of the same sequential mediation relationship, we find limited evidence investors are less willing to invest in a company that discloses a typical gender pay gap compared to a company disclosing no gender pay information. Two additional experiments indicate it is the information about gender pay, not its disclosure by the company, that influences investors; and the effect is intentional. Our results are consistent with investors anticipating real economic consequences from the disclosure of gender pay information.

Prior research has established that tax liability increases lead to decreased compliance. However, tax liability changes do not happen in a vacuum. Notably, prior research has also identified a withholding phenomenon: individuals in a tax due position are less compliant than those in a refund position. Additionally, tax law changes are often enacted in politically polarized environments. We examine how three factors—tax liability changes, withholding position, and group affiliation—combine to influence individuals' tax compliance decisions. Our experimental results show that a tax increase is universally experienced as a loss, even when coupled with a tax refund and enacted by an ingroup, leading to decreased compliance. However, a tax decrease coupled with a tax due position is viewed neutrally and leads to less compliance than a tax decrease coupled with a tax refund. Further, group affiliation influences compliance in some situations. Individuals in a tax due position are less compliant when an outgroup, versus an ingroup, is responsible for the tax change. This study contributes to the mental accounting literature by examining how individuals react to mixed gain/loss situations when the gains and losses are of different types. We also integrate the previous separate research streams on the withholding phenomenon and tax liability changes. Practically, our results contribute to tax policy by showing how individuals react when tax law changes are enacted. Importantly, even when a tax change results in a decrease in tax liability, tax compliance may be affected by individuals' withholding position and group affiliation.

ABSTRACT Based on prospect theory's value function, we predict how reference points adapt to influence individuals' tax evasion choices during and after experiencing temporary tax changes. Results from a multi-round experiment indicate reactions to temporary changes depend jointly on the direction of the change and expectations. Specifically, individuals experiencing a tax increase evade more while the increase is in effect. More interestingly, knowing, versus not knowing, a tax decrease is temporary prevents an increase in evasion after the temporary change expires, and may lead individuals to reduce evasion during the change. In a supplemental condition, we induce uncertainty by repeatedly extending a tax decrease. We find when uncertainty is introduced, both benefits of knowing the temporal nature of the decrease are lost. Overall results are consistent with individuals failing to adapt to a loss state and adapting quickly to a gain state unless they are certain the gain state is temporary.