Investor sentiment and stock returns

Abstract

Since Keynes coined the term animal spirits economists have been debating what the real impact human psychology is on economic variables. The major challenge in identifying these effects is the close ties between negative (positive) emotions and poor (good) future real outlook. I exploit a historical weighting anomaly in a widely cited US stock index to examine the impact of psychology on stock returns. I first argue this is a plausibly exogenous shock, and compare this measure to other measures found in the literature. I find that the measure doesn't seem to relate to previous proxies for investor sentiment, however, when I examine survey measures of interest rates and consumer confidence we find a relationship. I then examine how sentiment affects the cross section of stock returns, consistent with predictions I find that small stocks earn low subsequent returns when sentiment is low, and high returns when sentiment is high.