UC Berkeley

Markets are rich environments for study. Two features in particular make them attractive settings for the study of the roles of information. First, they are often characterized by a significant amount of uncertainty, an essential ingredient for information to be relevant. Second, strategic interactions abound.

Each chapter of this dissertation uses a different `lens' so as to focus on particular roles of information in strategic market settings. First, information is considered in its pure `aid to decision-making' role, in which agents use it to improve their decisions, not only based on the expected future conditions but also on the expected actions of the other agents which in turn depend on the information they are likely to hold. Second, information is considered in its `influencer' and `coordinator' roles. It can be used to influence the decisions of agents which will in turn affect the welfare of the original transmitters, while increasing overall market coordination. Finally, information is considered in its `homogenizer' role, in the sense that actions and preferences may become more similar after its repeated use and subsequent transmission. The purpose of this study is to propose, document and estimate mechanisms by which information fulfills these specific roles in actual market settings.

This dissertation is organized in the following way. Chapter 1 analyzes the use of information acquired by decision-makers in a strategic context. Chapter 2 introduces the layer of the information transmitter as a separate strategic entity, who may have different interests from those of the receivers. Finally, Chapter 3 considers the case where the same agents are simultaneously receivers, users and transmitters of information.

These scenarios are analyzed in well-defined market contexts. Chapter 1 looks at the role of information in the production and capacity decisions of DRAM producers. I estimate the precision of market information used by firms in the DRAM industry, implicit in the patterns and accuracy of their actions. The value of information is estimated as well as its impact on the decisions of the firms.

Chapter 2 investigates advertising between firms and consumers in a market for a search good. It is shown that even if firms have an interest in providing biased information through advertising, it is still possible for it to convey credible content that may be useful to consumers. In addition it is shown that the precision of information delivered is sparser as the quality of the product or service in question decreases.

Finally, Chapter 3 analyzes the context of online movie ratings and estimates how the opinions of consumers shape the opinions of subsequent consumers, ad infinitum. It is estimated that roughly half of the consumers' valuations for movies is a function of the valuations of other consumers who have had a chance to rate the movie earlier. A sensitivity analysis of how some extreme initial ratings influence consumer valuations is performed, revealing considerable effects.