The Effects of Industry Specialization on Auditors' Inherent Risk Assessments and Confidence Judgements*

This study experimentally examines how industry specialization affects auditors' inherent risk assessments and their confidence in those risk assessments. Two groups of participants ‐ experienced banking specialist auditors and equally experienced nonbanking auditors ‐ provided inherent risk assessments for a hypothetical banking client for two financial statement accounts. They assessed inherent risk for an industry‐specific account (loans receivable) and for a nonindustry‐specific account (property and equipment). The results indicate that nonbanking auditors assessed inherent risk significantly higher than industry specialists for all but the valuation assertion for the loans receivable account. However, the difference between the nonbanking auditors' and banking specialists' inherent risk assessments was not as great for the property and equipment account. Further, nonspecialists were less confident about the appropriateness of their inherent risk assessments compared with industry specialists. Potential implications for research and practice are discussed in light of the study's findings.     

Borrower Misreporting and Loan Performance

Borrower misreporting is associated with seriously adverse loan outcomes. Significantly more residential mortgage borrowers reported personal assets just above round number thresholds than just below. Borrowers who reported above‐threshold assets were almost 25 percentage points more likely to become delinquent (mean delinquency was 20%). For applicants with unverified assets, the increase in delinquency was greater than 40 percentage points. Misreporting was most frequent in areas with low financial literacy or social capital. Incorporating behavioral cues such as threshold effects into a risk assessment model improves its ability to uncover delinquencies, though at a cost of mischaracterizing some safe loans.     

The Limits of Investor Behavior

Many models use noise trader risk and corresponding violations of the Law of One Price to explain pricing anomalies, but include a storage technology in perfectly elastic supply or unlimited asset liability. Storage allows aggregate consumption risk to differ from exogenous fundamental risk, but using aggregate consumption as a factor for asset returns can make noise trader risk superfluous. Using (i) limited asset liability and limited storage withdrawals, or (ii) an endogenous locally riskless interest rate eliminates violations of the Law of One Price. Our main results use only budget equations and market clearing, and require virtually no assumptions about behavior.     

The Controllability of Financial Outcomes

Accounting performance measures that capture results not controlled by the person being evaluated may be dysfunctional. The aim of this paper is to provide a framework for identifying situations in which outcomes are captured by such measures as are more (less) likely to be controllable. Results of a questionnaire survey support the claim that as task uncertainty increases, individuals perceive less control over financial outcomes.     

Catastrophic Risk and Credit Markets

We provide a model of the effects of catastrophic risk on real estate financing and prices and demonstrate that insurance market imperfections can restrict the supply of credit for catastrophe-susceptible properties. Using unique micro-level data, we find that earthquake risk decreased commercial real estate bank loan provision by 22% in California properties in the 1990s, with more severe effects in African–American neighborhoods. We show that the 1994 Northridge earthquake had only a short-term disruptive effect. Our basic findings are confirmed for hurricane risk, and our model and empirical work have implications for terrorism and political perils.