Market Reaction to an Earnings Shock : A Test of the Conservatism Effect

In this paper, I empirically test the conservatism effect of Barberis, Shleifer and Vishny (1998). Conditioning on a shock to quarterly earnings, firms ranking in the top (bottom) earnings shock quintile exhibit substantial price momentum over the next three-month periods following the initial earnings shock. In the subsequent quarter, firms reporting earnings performance that maintain their ranking positions in the highest (lowest) earnings quintile exhibit a marginal incremental price run above that of the initial earnings signal. However, firms that fail to keep (succeed at moving out of) their ranking positions in the highest (lowest) earnings quintile experience a strong price reversal. These findings are robust to the four-factor regression (the Fama-French three-factor model extended by the momentum factor) and various robustness tests. Evidence reported in this paper is not consistent with the view that investors underreact to a recent earnings change. Rather, the evidence points to a market that systematically overreacts to extreme earnings news