Trust Information and Compliance

The principal-agent problem is one of the most important and extensively studied issues in economics. A large literature has examined the role of trust in principal-agent problems both theoretically and empirically. Most previous studies have treated trust as a signaling device (i.e., part of the strategy) to supplement contracts. In the current study, we experimentally test the impact of trust information when it is part of the institutional arrangement. That is, we randomly determine whether the trust/trustworthiness of the agent/principal is revealed to the counterparty. This study design allows us to distinguish between the impact of trust per se and the strategic use of trust in a principal-agent game. To test the possible impact of trust/trustworthiness, we first elicit the agents’ and principals’ trust/trustworthiness in a trust game. We then adopt a tax audit framework in which a taxpayer (agent) chooses the amount of income to report to the tax authority (principal). The tax authority then determines whether to audit the taxpayer at a fixed cost. Although the tax audit game features a unique Nash Equilibrium, our results show that revealing trust/trustworthiness information affects taxpayers’ and tax authorities’ behaviors. We find that (1) when taxpayers’ trust in the tax authority or tax authority’s trustworthiness is revealed, the audit rate significantly decreases while compliance does not improve; (2) when the tax authority’s trust in taxpayer or taxpayer’s trustworthiness is revealed, audit rate significantly decreases with significantly improved compliance. Although we use a tax audit framework to analyze our experimental results, the conclusion applies more generally to other principal-agent settings, especially those involving possible audit and compliance decisions.