Second-Degree Price Discrimination on Two-Sided Markets

Abstract

The paper provides an analysis of the second-degree price discrimination problem on a monopolistic two-sided market. In a framework with two distinct types of agents on either side of the market, we show that under incomplete information the extent of platform access for high-demand agents is strictly lower than the benchmark level with complete information. In addition, we find that it is possible in the monopoly optimum that the contract for low-demand agents is more expensive than the one for high-demand agents if the extent of interaction with agents from the opposite market side is contract-specific.

Price Discrimination by a Two-Sided Platform: With Applications to Advertising and Privacy Design

“…5 10 then they are the points above the dashed curve in Figure 1-(b); if r = − 9. 5 10 , then they are the points above the thin curve in Figure 1-(b). In the case of negative sorting (i.e.…”

“…For instance, if r = − 8. 5 10 then (14) fails to hold for the points to the right of the dashed curve in Figure 1-(a); if r = − 9. 5 10 , then (14) fails to hold for the points to the right of the thin curve in Figure 1-(a).…”

“…We denote with F the set of (q A H , q A L ) satisfying (I A ), given (q B H , q B L ). 5 In order to describe F , we let • M (from "monotonic") denote the set of (q A H , q A L ) such that q A H > q A L ≥ 0;…”

Second-Degree Price Discrimination by a Two-Sided Monopoly Platform

“…They study how this parameter crucially affects the welfare under non-neutral networks where price discrimination is allowed and the welfare under neutral networks where price discrimination is banned. Böhme (2016) considers a model in which there are two types on each side and each agent's utility consists of two components: a type-specific intrinsic utility from access to the platform and an additional type-specific indirect network effect from interacting with agents on the opposite side of the market. We show that the model of Choi et al (2015) can be obtained as a special case of congruence between the two sides in our baseline model (see Remark 1).…”

“…We show that the model of Choi et al (2015) can be obtained as a special case of congruence between the two sides in our baseline model (see Remark 1). In Böhme (2016), the externality generated by an agent to an agent on the other side depends only on the latter's type, such that a high type enjoys a larger externality than a low type does. For this reason, the model of Böhme (2016) belongs to a limit case in which there is neither congruence nor conflict (in a strict sense) between the two sides and there is no conflict within a side (see Remark 2).…”