I motivate and implement a methodology that decomposes a firm’s discretionary accruals into a firm-specific and an industry-specific component. I find that the “accruals anomaly” (Sloan 1996) – the finding that firms with high discretionary accruals subsequently earn negative abnormal returns – is driven by firm-specific discretionary accruals. More importantly, although industry-specific discretionary accruals do not contribute directly towards the anomaly, I find that it is precisely when industry-specific discretionary accruals are high that firms with large firm-specific discretionary accruals subsequently earn negative abnormal returns. The results suggest that industry-wide use of high discretionary accruals adversely affects investors’ tendency to fairly price high firm-specific discretionary accruals, possibly due to increased search costs associated with detecting earnings manipulation during such times. The results are also consistent with the notion that firms use high (firm-specific) discretionary accruals to convey value-relevant information to investors when most firms in the industry have relatively low discretionary accruals. The findings potentially have an important bearing on the earnings management literature that has used high discretionary accruals as a proxy for earnings manipulation.