Nonparametric Analysis of Two-Sided Markets

Abstract

This paper considers an empirical semiparametric model for two-sided markets. Contrary to existing empirical literature on two-sided markets, we do not rely on linear network effects. Instead, network effects and probability distribution functions of net benefits of two sides are specified nonparametrically. The demand functions and the network effect functions of readers and advertisers are estimated by nonparametric IV estimation using a data set from German magazine industry. The ill-posed inverse problem faced during the estimation is solved by Tikhonov Regularization. We show that semiparametric specification is supported by the data and the network effects on readers' side are neither linear nor monotonic. With a numerical illustration we demonstrate that the mark-up of the magazine on readers' side is 27% higher with the nonlinearly specified network effects than in the case with linear network effects.