An Analytical and Empirical Measure of the Degree of Conditional Conservatism

Abstract

The Vuolteenaho (2002) return decomposition is linear because it assumes that the market's return expectations are obtained solely from accounting information. By restricting accounting recognition rules to specific (and primarily) negative future cash flow shocks, conservative accounting drives a wedge between the market's return expectations that are based upon all positive and negative cash flow shocks and return expectations that are based solely on accounting numbers. This insight allow us to derive analytically a nonlinear relation between revisions to returns and earnings news for conservative firms, of which the Basu relation is a special case. This nonlinear relation is shown to be mathematically equivalent to two linear relations conditioned on the firm's degree of conservatism. From these relations, we derive a model-based measure of the degree of conservatism at the firm-year level which is a function of the determinants of conditional conservatism. To account for the endogeneity of the firm's degree of conservatism and mitigate potential sample selection bias (Dietrich, Muller and Reidl 2007), the model is implemented empirically using a switching regression approach in which the switch point, namely, the degree of conservatism, is both unobservable and endogenously determined. Consistent estimates of the parameters of the switching regression, including the endogenous determinants of conservatism posited by Watts, are obtained by simultaneous maximum likelihood estimation. The results indicate that the degree of conservatism is a positive function of contractual information asymmetry and litigation risk but a negative function of taxes.