A Dynamic Model of Investor Expectations and Financial Reporting

Abstract

This paper examines the implications of investor expectations for 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 accruals. Three investor expectations schemes with different levels of rationality are considered: rational expectations, adaptive learning, and misspecified Bayesian learning. In a stationary rational expectations equilibrium, whether the manager manipulates earnings depends on the real earnings process. Under adaptive learning and regime-shifting beliefs, earnings management exists to the extent that there is discrepancy between investor beliefs and the actual process of reported earnings. This model provides testable predictions for the joint impact of managerial opportunism and investor expectations on accounting-based return regularities. It also sheds light on the descriptive validity of different investor expectations schemes.