Hirshleifer et al. (J. Account. Econom. 38 (2004)) and Taffler, Lu and Kausar (J. Account. Econom. 38 (2004)) document large and statistically significant abnormal returns from trading on balance sheet data and audit opinions. However, the statistical tests ignore high transactions costs, especially for selling short, that would likely make the trading strategies unprofitable. The accounting anomalies literature is adding little to what we know about how and why markets operate more or less efficiently. I identify some research questions and opportunities, highlighting those with accounting and auditing implications.

Prior research concludes that financial analysts do not process public information efficiently in generating their earnings forecasts. The ordinary least squares (OLS) regression-based tests used in prior studies assume implicitly that analysts face a quadratic loss function. In contrast, we argue that analysts likely face a linear loss function, and hence, try to minimize their absolute forecast errors. We conduct and compare rational expectations tests using these two alternative loss functions. We reproduce most prior findings of forecast inefficiency with OLS regressions, but find virtually no evidence of forecast inefficiency with least absolute deviation regressions, where we explicitly assume a linear loss function.