Inside the accrual anomaly

Sloan (1996) presents evidence that a trading strategy based on publicly available accounting accruals of firms in extreme accrual deciles earns abnormal returns of approximately 10% in the year following its initiation. This empirical regularity has been named the 'accrual anomaly'.; In this dissertation, I examine Sloan's (1996) explanation that investors misinterpret accruals information and naively fixate on them. I label Sloan's conjecture the 'accrual-fixation' hypothesis. Under this hypothesis, accrual extremeness is likely to be transitory because of the tendency of accruals to reverse. Further, if aggressive accounting is related to accrual-fixation then, for a given firm, accruals are not likely to remain extreme in the following period because of disciplinary mechanisms such as auditor oversight.; The accrual-fixation hypothesis also argues that accrual mis-processing is symmetric in high and low accrual firms. In both cases, investors presumably do not appropriately adjust for the differential persistence of accruals and cash flows.; I test predictions arising from the characteristics of the accrual-fixation hypothesis. I examine the time-series pattern of extreme accrual firms to see whether they tend to leave the extreme deciles in the following period or whether they are habitual extremes. Inconsistent with the accrual-fixation hypothesis, I find that extreme accrual firms are habitual extremes (i.e. sticky). About 25% of firms in the extreme accrual deciles remain there in the following year. Further, I find that future abnormal returns are, to a large extent, driven by sticky firms.; Sticky high accrual firms are associated with extreme growth while sticky low accrual firms have a higher probability of bankruptcy. There is also a difference in future abnormal returns between sticky low and sticky high accruals. The symmetry in mis-processing under the accrual-fixation hypothesis does not predict such differences.; I make several additional contributions. I show that returns to the accrual anomaly partially overlap with returns following corporate events. The estimated upper bound of corporate events' effect on abnormal returns generated by an accrual strategy is at most 25%. My results also indicate that when book-to-market is added as a control for expected returns, returns to the accrual strategy decrease by approximately 20%.