Repairing the Accruals Anomaly

Abstract

We document how the effectiveness of an accruals-based trading strategy changes systematically with the financial health of the sample firms or with the benchmark used to identify an extreme accrual. Our refinements significantly improve the strategy's annual hedge return, and do so mostly because they improve the return earned on the long position in low accrual stocks. These results are important because recent evidence has shown that, absent these "repairs," the accrual strategy does not yield a significantly positive return in the long portion of the hedge portfolio. We also find that our new measure of accruals is not dependent on the presence or absence of special items and it identifies misvalued stocks just as well for loss firms as for gain firms, in contrast to the traditional accruals measure. Finally, we show that our repairs succeed where the traditional measure of accruals fails because they more effectively select firms where the difference between sophisticated and naïve forecasts are the most extreme. As such, our results are consistent with Sloan's earnings fixation hypothesis and are inconsistent with some alternative explanations for the accrual anomaly.

Keywords: accruals anomaly, market inefficiency, trading strategy

JEL Classification: M41, M43, G14