Do We Accept Accrual Profits at Our Peril?

The study described documented evidence that informed traders use their proprietary information on accrual quality to trade against average investors. The informed traders’ arbitrage strategy generated annualized abnormal returns adjusted by size and book-to-market value of 19.8 percent over the 1993–2002 period. The accrual profits were still significant after trading costs were subtracted. The findings suggest that (1) informed traders’ profits from accruals-based strategies derive mainly from their costly information on accrual quality and (2) the persistence of the “accrual anomaly” may be driven by nondiversifiable information risk. The study suggests a strategy for uninformed traders to overcome the information barrier by mimicking informed traders.The “accrual anomaly”—the evidence that companies with higher (lower) accruals earn lower (higher) abnormal returns—has been widely documented in the literature and is well understood by portfolio managers. Studies on the accrual anomaly have failed to explain the source of accrual profits, however, and the persistence of accrual mispricing. If accrual profits are significant in size, why do the more sophisticated investors not exploit this opportunity and quickly eliminate accrual mispricing?We used the return and stock trading information on NYSE and Amex stocks for the period 1993–2002 (from the CRSP, Compustat, and Trade and Quote databases) to examine whether informed traders can take advantage of accrual mispricing and develop profitable trading strategies based on information about accruals. We first used an algorithm developed to compute the “probability of information-based trading” (PIN) for a stock during a given time period. We then sorted the stocks by their PIN values. We identified evidence that the arbitrage strategy confined to the stocks with high PIN values—that is, the stocks where private information-based trading would be concentrated—generates an annualized abnormal return to a portfolio adjusted for company size and book-to-market value (B/M) of 19.8 percent. The evidence shows that informed traders actively use their proprietary information about accrual quality and trade against the average investor.We then applied three methods to compute trading costs and control for their impact on the profitability of the accruals-based trading strategy. We found that informed traders could make an abnormal return ranging from 6.5 percent to 17.5 percent after trading costs. These results are robust to testing methods, asset-pricing models used, and various ways of controlling for trading costs. The results suggest that the source of accrual profits and of their persistence may be nondiversifiable information risk rather than high trading costs. The accrual profits to the informed traders derive mainly from their access to costly information on accrual quality.The implications of our findings for portfolio managers are twofold. First, using the results reported here allows portfolio managers to better understand the source and persistence of accrual profits. This understanding may help them design effective arbitrage strategies around accrual information. Second, our analysis suggests a way for uninformed traders to overcome the information barrier and mimic informed traders’ behavior. Because of a high level of autocorrelation in the measure of the extent of informed trading (PIN), an accruals-based strategy can be based on PINs from the previous period, which is known to investors when forming their portfolios. In our tests, stocks with high previous-period PINs generated abnormal profits similar in size and significance to those informed traders could obtain. Thus, portfolio managers and even individual investors can harvest accrual profits by using data publicly available to compute the probability of informed trading.