Macro Seminar - Sephorah Mangin (ANU)

Title:  The Effect of Search Frictions on Extreme Outcomes

Abstract:  Extreme value processes involving maxima are widespread in economics. We provide some general results regarding the asymptotic effect of search frictions on the outcomes of such processes. To do this, we allow the number of draws from the underlying distribution (e.g. of productivities, ideas, or utility shocks) to be given by a discrete probability distribution called the search technology. We derive a general expression for the extreme value distribution that arises in the limit as the expected number of draws becomes large. We find that the nature of the extreme value distribution depends on the search technology. For example, if the underlying distribution is Pareto, the extreme value distribution is Fr´echet if and only if the search technology is either Poisson or degenerate. In general, the extreme value distribution need not assume any of the three standard types. We also generalize a result from Gabaix et al. (2016) regarding extreme value outcomes and consider some applications of our results to aggregate productivity, markups, and large auctions. Some simple examples suggest that search frictions – or any other source of dispersion or heterogeneity in the number of draws – may have a quantitatively significant effect on extreme value outcomes.