Attention, Demographics, and the Stock Market

University Of California-Berkeley, Berkeley CA

Investigators

Abstract

According to the theory of efficient financial markets, stock prices should reflect all available information. However, evidence on post-earnings announcement drift and momentum effects suggests that stock prices do not fully adjust to new information. This project explores a novel model of investor attention and some preliminary evidence on a test of the model. In addition, it suggests further empirical tests. The intellectual merit of this proposal is that it outlines a straightforward model of the allocation of attention and suggests a test of its implications for demographic variables. The hypothesis is that individuals neglect information about long-term outcomes, even when these outcomes can be forecast. For example, investors may neglect long-term determinants of earnings. This project considers a prominent example of a variable that can be forecast years in advance, and therefore may be partially neglected - demographics. In a model with attentive investors, forecastable fluctuations in cohort size should not generate predictability of stock returns, because stock prices react immediately to the demographic information. Under the alternative hypothesis of inattention, current demographic information should forecast future stock returns. Preliminary evidence from the US stock market over the period 1936-2002 supports the hypothesis of inattention. It also suggests two additional tests of the hypothesis of inattention to long-term demographic information. The first additional test is a cross-country study of the response of industry stock returns to demographic variables. The analysis of markets of countries such as France, Japan, and United Kingdom would add substantially to the available evidence, which is currently limited to the United States. The second empirical test focuses on prominent media articles on demographic trends. On days when such articles appear, inattentive investors respond by correcting their previous under-reaction to information regarding demographic variables. Lastly, the proposal describes a third project on the corporate response to anticipated demand shifts induced by demographics. The decisions studied include entry, exit, and investment. Of particular interest is whether companies react in advance of the forecastable demand shifts. The broader impacts of this proposal are twofold. First, this proposal improves the understanding of the economic consequences of demographic changes. Given the large demographic changes looming for American and European societies, a better understanding of the impact of demographic variables is a first-order task. Second, this proposal adds to the interdisciplinary work in behavioral finance and in behavioral economics. This project takes a simple insight about attention allocation - long-term events may be neglected - and examines its relevance in markets.