The Conservatism Principle and Asymmetric Preferences Over Reporting Errors

At present, accounting conservatism is generally viewed from a measurement or reporting perspective. In contrast, we consider whether it relates to a moral rule of conduct. Conservatism has been described as deriving from a preference for reporting errors to be in the direction of understatement rather than overstatement. We experimentally pair Reporters who provide information with Users who rely on the information. We posit that under misaligned incentives that motivate aggressive reporting, Users view an aggressive report as reflecting Reporters’ exploitative intent and expect that a social norm prohibiting aggressive reporting applies. We predict that Users use noisy reporting errors to gauge Reporters’ norm compliance. Consistent with this we find that, ceteris paribus, Users prefer not to be paired with Reporters who produce overstatement errors that are likely to reflect aggressive reporting. This preference, revealed through Users’ incentivized actions, is both inconsistent with neoclassical economic models and cannot be explained by loss aversion. Alternatively, when Reporters’ motives are aligned with Users’, we find no such preference. While our evidence is indirect, it opens the possibility that conservatism emerged from a norm that enhances trust and cooperation among economic agents. We believe this insight can open new possibilities for conservatism research.