Essays on rational behavior in incomplete information

Abstract

This thesis examines the behavior of rational agents in market settings of incomplete information using tools of signal-jamming and signaling. I study the decision-making processes of agents who aggregate information by updating noisy information, knowing that their own actions may reveal private information to others, and knowing that others face the same problems. The first essay analyzes how the existence of noise traders causes a reporting and pricing bias in a “rational” financial market when managers may be signal-jamming by over-reporting earnings. I assume that traders differ in their available information on the possible reporting bias of a manager, and that the manager has private interests in manipulating the stock price. I conclude that the existence of noise traders causes systematic pricing bias and leads to an “inefficient” financial market. I also analyze the effects financial market structure and accounting systems regulation have on pricing bias and reporting bias. For example, I show that market pricing bias is lower if there are more people who are better informed about the manager’s private interest in manipulating the stock price, even though it leads directly to a higher reporting bias. The second essay theoretically investigates the factors influencing an established firm’s decision whether or not to introduce a new product under the same brand name, that is, to stretch the brand. I assume that a firm has better knowledge of the product’s quality than consumers do; that the characteristics of the established and the new products are related; and that consumers infer as much as possible from prices and brand name. The size of the original market relative to the new market and consumers’ ability to judge quality in both markets combine to determine the magnitude of backward or forward effects. I confirm Cabral’s (2000) conjecture that the greater the similarity of products (i.e., higher correlation of qualities), the less frequent brand stretching occurs when the original good is perceived as being high quality. However, in contrast to Cabral, as consumers gain higher accuracy in perceiving the quality of the goods, if the original good is perceived as being of low quality, then higher correlation between the products leads to more frequent brand stretching. The present analysis also sheds light on the importance of consumers’ ability to accurately detect quality—the more accurate consumers are, the less important is any strategic signaling by the firm. The last essay examines a combined idea of previous two essays in different context, which is whether or not a fund manager in the established market can signal his superior ability in a new financial market. The better informed traders in a stock market may have a chance to trade in other financial sectors, e.g. the bond market. His ability in the new market may be different from the previous ability and is a private information. Now the decision of the manager is whether or not to introduce an additional performance fee considering both backward and forward effects. As in the second essay, the manger with the ability of high performance in the new market declares his type by charging the additional fee.