Abstract
This paper explores the notion that the firm's life cycle is an omitted variable in commonly used discretionary accrual models. I first document a mean positive (negative) bias in discretionary accruals for firms in the growth (decline) stage of their life cycle. Second, I re-examine prior studies' tests for income-increasing (decreasing) earnings management around IPOs (asset write-downs), and I show how inferences change after controlling for the firm's stage in its life cycle. Last, I show that incorporating controls for life cycle in accrual models leads to comparable, or lower, type I and type II error rates for random samples of discretionary accruals. Robustness tests indicate that these findings are not driven by earnings management incentives.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have