The case for limited auditor liability : the effects of liability size on risk aversion and ambiguity aversion

Both the US and the EU consider limiting auditor liability in order to ensure the viability of the audit market, but fear its potentially negative impact on audit quality. Our paper discusses the existing empirical results on this topic in the auditing and behavioral economics literature, and provides new evidence based on a controlled laboratory experiment. Our experiment involves real losses and allows for direct inference of behaviour under limited and unlimited liability in situations of ambiguous liability risk. Our findings imply that limited liability can induce an efficient level of audit effort, while unlimited liability induces an inefficiently high level of audit effort. This paper contributes to the literature on auditor liability, as well behavioral economics research in general, by addressing recent controversial issues on behavior in the presence of ambiguity and real losses.