| The social learning literature studies an economy in which individuals who face a decision problem under uncertainty, make a once-in-a-lifetime decision, one after the other. The individuals differ from each other with respect to the 'private' information they have about the uncertainty. The typical conclusion of the literature is that despite the asymmetry of information, eventually every individual imitates his predecessor, even though he might have chosen a different action on the basis of his own information alone. In this sense, individuals rationally 'ignore' their own information and 'follow the herd.' Moreover, despite the available information, a great number of individuals can make the same incorrect choice.; A central assumption of nearly all social learning models is perfect information: everyone is assumed to be informed about the entire history of actions that have already been taken. An individual is thus comparing his information with the information deduced from the actions of a large number of other individuals. In reality, individuals have imperfect information. If each individual observes the actions of only a small number of other individuals, it is not clear that 'herd behavior' will arise. This is the central question that my dissertation scrutinizes in depth both in theoretical and experimental grounds.; To this end, the first chapter studies a model under both perfect and imperfect information. We observe that the behavior can be radically different under the two information regimes. Precisely, while under perfect information we replicate the results of the literature, under imperfect information we observe longer and longer periods of uniform behavior, punctuated by increasingly rare switches.; To better understand the social and economic impact of learning, the remaining chapters substantiate the theoretical findings with a series of experimental studies. These chapters test the models' predictions, evaluate experimental methods and study the effects of variables about which the existing theory has little to say. Chapter 2 and 3 report the results of experiments under perfect and imperfect information respectively. Finally, chapter four attempts to capture the effects of a more realistic information transmission, namely when individuals learn by the advice of others rather than their actions. |
