The information quality cost of earnings management: evidence from the CDS market
Subject
Accounting
Publishing details
Social Sciences Research Network
Authors / Editors
Ertan A; Karolyi S; Kolev K S
Biographies
Publication Year
2016
Abstract
The capital markets’ reward to meeting or beating earnings benchmarks through earnings management has been explained through signaling models: earnings management carries cash costs, which render it a credible signal. Using flatness, a theoretically-inspired measure of information quality based on the ratio of short-term to long-term credit default swap spreads, we provide evidence of another cost: deterioration of the quality of the accounting signal. First, we validate the measure by documenting that flatness predicts future information events. Then, consistent with the conjecture that accounting-based earnings management garbles the reporting signal, we find that flatness is statistically and economically larger when firms narrowly beat the analysts’ earnings consensus forecast via accruals. In a falsification test, we document that flatness is not sensitive to benchmark beating through cash-flow-based earnings management practices. Finally, we find that the documented signal-garbling effect is concentrated among highly-levered and small firms, and does not carry over to the level of credit default swap spreads or equity returns.
Keywords
Analyst Forecasts; Credit Default Swaps; Earnings Management; Information Quality
Series
Social Sciences Research Network