Back to Publications

Designing a Marketplace for Distributed Data

Abstract:
In this paper, we propose a model of a production economy of data markets. Such markets typically exhibit a complex production structure when the databases produced by different data providers can be joined to produce answers for buyers’ queries. We focus on three specific challenges:(1) different providers have the capability to produce different sets of databases;(2) data providers have high fixed costs for producing a database; and (3) buyers have combinatorial values over which databases are produced and thereby become available in the marketplace. To illustrate our model, we provide a possible market design solution for this domain. The key idea of our solution is to use a reverse auction for the sellers, a posted-price mechanism for the buyers, and a fixed-point iteration algorithm for finding an outcome that balances the two sides of the market. To achieve this, we illustrate how to infer values of buyers for particular databases from the values for their queries. Via simulations, we show how our market distributes the surplus between buyers and sellers. In particular, we demonstrate that our design rewards providers of ‘‘unique” data much more than providers of ‘‘common data.”