The Informational Content of Consumer Choice

How do mergers influence prices and quantities in differentiated product markets? How does advertising change purchasing decisions of consumers? Which products are perceived as good substitutes? Without a realistic model of consumer decision making in differentiated goods markets, these questions cannot be answered in an empirically robust way. Given the flood of products in today`s markets, it does not seem plausible that consumers consider all available products. The subset of products that are actually considered is referred to in the literature as the "consideration set." This project focuses on developing novel empirical techniques for identifying these subsets or "consideration sets." Example product categories where consumers have limited-information include beer, cars, or coffee, where many variants are offered. Despite large choices, previous research has shown that consumers only consider between 4 (coffee) and 8 (cars) different products. More recent empirical evidence also shows that consumers do not always make fully informed purchase decisions and choose from a reduced number of alternatives. The alternatives considered then do not include the entire product choice. However, most of the existing economics literature on differentiated products demand assumes that consumers are aware of and consider all products available in a market. If this assumption is violated, the result is a bias in demand estimates, which in turn leads to misleading economic policy conclusions. Despite its importance there is, however, a lack of empirical knowledge about how consumers shape their consideration process and which products they are more likely to include in their consideration set. Our goal is to estimate the extent to which different consumers consider the available products. To obtain credible estimates of consideration sets, our modeling approach will allow consumers to differ in the extent to which they consider sets. We apply our modeling approach to a large number of categories from the U.S. consumer goods industry, and investigate the differences between model predictions based on the assumption of fully informed purchase decisions and the alternative assumption of limited informed purchase decisions. To this end, we analyze scenarios in which prices change due to events like firm mergers or cost shocks.