Optimistic and Pessimistic Disagreement and the Cross Section of Stock Returns

Abstract

We decompose total disagreement about macro variables into the disagreement among optimists (i.e., forecasters whose forecast exceeds a certain threshold) and pessimists. Optimistic (pessimistic) forecasters tend to disagree more in good (bad) times. Pessimistic (optimistic) disagreement commands a negative and significant (positive, although often (insignificant) risk premium and total disagreement is often insignificant when included in the same regression. These results are robust across a variety of empirical specifications and sets of test assets. A theoretical model, in which the risk premia of optimistic and pessimistic disagreement depend in a non-trivial way on forecasters’ beliefs and on the joint impact of optimistic and pessimistic disagreement on the price of the assets used to speculate on individual beliefs, rationalizes the empirical findings.