Previous analyses of market structures characterized by gradual information dissemination presume the equilibrium price existing after all market participants are informed is independent of the order of information dissemination. In these papers, final market clearing price, given the investors' posterior beliefs, is known a priori and is assumed to equal the price that would exist if data were disseminated simultaneously. We demonstrate that final equilibrium price is dependent, in general, on the order of information dissemination. This implies that, if the dissemination sequence is stochastic, price is unknown prior to the complete dissemination of information, even if the investors' posterior beliefs given the information event are known. We derive a necessary and sufficient condition for equilibrium price to be independent of the dissemination sequence in our economy. Our analysis highlights the importance of the wealth redistribution dynamics inherent in the information dissemination process.

We examine a model of market equilibrium in which there is less information available about some of the securities in the market than about others. We consider the model as a potential explanation of the well-known small firm anomaly. Using period of listing as a proxy for quantity of information, we find an association between period of listing and security returns that cannot be accounted for by firm size and which is not diminished by an elimination of January returns data from our sample. Thus, we observe a new empirical regularity in the data and refer to the regularity as the ‘period of listing’ effect.

Stephen J. Brown, Christopher B. Barry, Anomalies in Security Returns and the Specification of the Market Model, The Journal of Finance, Vol. 39, No. 3, Papers and Proceedings, Forty-Second Annual Meeting, American Finance Association, San Francisco, CA, December 28-30, 1983 (Jul., 1984), pp. 807-815

We examine the hypothesis originally advanced by Roll [12] that observed anomalies in excess returns can be explained by misspecification of the market model used to estimate systematic risk. We find substantial misspecifications in the model systematically related to size and period of listing of the securities in question. There is some evidence that these misspecifications are associated with systemic biases in measured betas used to construct excess returns.

This paper addresses the investor's decision to employ multiple managers for the management of investment funds. Under conditions such that specialization of managers and diversification among managers are not motives for the use of multiple managers, the paper shows that risk sharing considerations may be sufficient. A model is developed in which the decision to use multiple managers is explicitly treated, and conditions are studied such that an increase or decrease in the number of managers would be desirable. Under some conditions, a multiple manager solution is preferred over a single manager solution.

This paper addresses the investor's decision to employ multiple managers for the management of investment funds. Under conditions such that specialization of managers and diversification among managers are not motives for the use of multiple managers, the paper shows that risk sharing considerations may be sufficient. A model is developed in which the decision to use multiple managers is explicitly treated, and conditions are studied such that an increase or decrease in the number of managers would be desirable. Under some conditions, a multiple manager solution is preferred over a single manager solution.