| Empirical evidence suggests that the returns-earnings relation has a few particular properties. These include a nonlinear returns-earnings relation, an underreaction to earnings announcements, and a stronger reaction to 'bad news' than 'good news.' In this study, I propose and test a behavioral theory for these characteristics of the returns-earnings relation in the laboratory. The primary contributions of this study are twofold. The first is a theoretical explanation of nonlinearity, underreaction, and an asymmetric response to earnings announcements. The second contribution is a test of whether prospect theory holds in an investment context, where valuation includes future earnings and uncertainty. The results should be useful to both capital markets and judgment and decision making researchers. The results are somewhat mixed. As predicted, the response per unit of surprise decreases as the absolute value of unexpected earnings increase (nonlinear returns-earnings relation). However, the subjects overreacted to earnings surprises, and the reaction to gains was more than the reaction to losses. |