Investor Expectations, Earnings Management, and Asset Prices

Abstract

This paper examines the implications of investor expectations for the joint determination of earnings management and asset prices. I propose a dynamic stochastic model of financial reporting, in which a firm manager opportunistically shifts earnings in order to manipulate investor expectations, subject to a constraint on accumulated discretionary accruals. Four alternative investor expectation schemes are studied: rational expectations, constant-gain learning, regime-shifting beliefs, and accounting-information-system beliefs. The main findings are threefold. First, there does not exist a stationary rational expectations equilibrium; under the other three expectation schemes, optimal reporting decisions are characterized by threshold rules based on investor beliefs. Second, regression analysis using simulated data indicates that the relationships among earnings, discretionary accruals, and stock returns critically depend on investor expectations. Third, structural estimation uncovers considerable heterogeneity in investors' belief-formation processes, with accounting-information-system beliefs providing the best model fit for the majority of S&P 500 firms.