Seminar

Broad Matching and the Market for Search Platforms

Ran Spiegler (University of Tel Aviv - University College London)

September 30, 2014, 11:00–12:30

Toulouse

Room MS 001

Economic Theory Seminar

Abstract

We construct a two-sided market model, in which consumers can only provide a noisy signal of the type of product they want. Each signal functions as a platform, to which (multi-homing) firms get access if they pay its competitive market price. A "broad match" function, designed ex-ante by the search intermediary, links signals to one another: a firm that attaches itself to one signal (by paying its market price) can get access to the search pool of consumers who provide another signal. We ask the following question: Is there a broad match function that induces an efficient market equilibrium, given the underlying search technology? In the case of random sequential search, we provide a necessary and sufficient condition, in terms of the underlying joint distribution over consumers' tastes and signals - specifically, a simple inequality that involves the relative fractions of consumers who like different products, and the Bhattacharyya/Hellinger distance between their conditional signal distributions. The same inequality turns out to be the condition for joint implementability of efficiency and full surplus extraction under a general anonymous mechanism. The role that Bhattacharyya distance plays in our analysis links our paper to the machine-learning literature on recommender systems.