An Equilibrium Analysis of Information Aggregation in Investment Markets with Discrete Decisions

This paper proposes a bridge between the Herding literature and the literature on Rational Expectations under asymmetric information. In particular we examine how the presence of discrete investment decisions affects the properties of a market equilibrium where information is costly to acquire. We choose to focus on the case where individual decisions are discrete since this appears to be the key element behind herding results. Our objective is to examine whether the equilibrium occurrence of herding type phenomena is likely to arise when actions are simultaneous (as opposed to sequential) and when prices can convey information. Our main result is that, as long as acquiring information is not too costly, the unique equilibrium outcome of our model is characterized by fluctuations in investment that resemble herding behavior. Specifically, equilibrium realizations of prices and investment may be high simply because uninformed investors are buying under the impression that the high price is a signal of good investment opportunities. Moreover, we find an interesting tradeoff between the size and the frequency of aggregate allocative errors, whereby as the cost of gathering information declines the size of allocative errors increases, even though there occurrence decreases. We believe these results provide new impetus for the view that herding type behavior may be relevant for understanding market fluctuations and even eventually business cycle phenomena.