Ulric B. and Evelyn L. Bray Social Sciences Seminar
Abstract: Can we distinguish from choice data alone whether consumers are informed about the attributes of goods? Surprisingly, we frequently can. Suppose the data generating process is a search model with a hidden attribute observed to the econometrician; consumers search goods in order of "visible utility" (based on non-hidden attributes) and then choose the good that maximizes utility among searched goods. Canonical models will be biased: the value of the hidden attribute will be understated because consumers will be unresponsive to variation in the attribute for goods that they do not search. An alternative method of recovering preferences using second derivatives of choice probabilities succeeds regardless of the search protocol and is thus robust to what consumers know when they choose. The approach nests the standard full information case. It is also general, requiring few assumptions on utility beyond those required to identify utility parameters in the existing literature. Our approach can be used to forecast how consumers will respond to information as well as to recover consumer preferences when consumers are imperfectly informed.