The Speed of Information Revelation in a Financial Market Mechanism

Suppose that information about the value of a risky asset is dispersed among many agents in the economy. The paper studies the rate at which successive price quotations from competitive market makers, which reflect the desired (notional) trades of risk- averse informed agents, reveal the value of the asset. The situation considered is akin to the real-time dissemination of theoretical prices in the opening batch auction of some continuous stock trading systems. The issue is studied in the context of an information t tonnement process in which informed agents submit market orders to market makers who quote prices efficiently. Informed agents in turn revise their estimates of the value of the asset and resubmit orders. The equilibrium of the t tonnement is fully characterized and it is found that price quotations converge to the underlying value of the asset at a rate of n-1/2, where n is the number of rounds of the t tonnement, and have an asymptotic precision negatively related to the degree of risk aversion, the noisiness of private signals and the amount of noise in the system. The analysis makes clear the role of competitive market makers: by increasing the depth of the market as the number of rounds increase they induce insiders to respond more to their information and speed up convergence. In fact, in markets in which depth is exogenously fixed, convergence is slow. The approach used allows also the study of the comparative dynamic properties of equilibria, such as the