Cross Sectional Variation in Cash Flow Asymmetric Timeliness and Its Effect on the Earnings-Based Measure of Conditional Conservatism

Abstract

Earnings asymmetric timeliness in recognizing losses versus gains is the most commonly used measure of conditional accounting conservatism. However, this measure captures both accrual asymmetric timeliness and operating cash flow (CFO) asymmetric timeliness. Because cash flow asymmetry is not related to differential verification thresholds in recognizing gains versus losses, we argue that it adds noise or bias to tests of conservatism. Importantly, we show that CFO asymmetry varies cross-sectionally with variables associated with firms’ life-cycle stage -- size, age, growth, and capital expenditures. We attribute this result to the differential weight that the market places on assets in place versus growth options in good news versus bad news environments during various stages of a firm’s life cycle. Using three empirical settings, we show that CFO asymmetric timeliness confounds results in studies where the sample is over-represented by early life-cycle stage firms or the partitioning variable deemed to give rise to conservative reporting is correlated with the firms’ life-cycle stage. Going forward, we recommend that researchers either use an accrual-based asymmetric timeliness measure or control for firm characteristics associated with life-cycle stage when testing for conditional conservatism.

Keywords: Cash Flow Asymmetry, Conservatism, Asymmetric Timeliness

JEL Classification: M40