Ad-valorem platform fees and efficient price discrimination

Posted by D. Daniel Sokol

Zhu Wang (Federal Reserve Bank Richmond) and Julian Wright (National University Singapore) address Ad-valorem platform fees and efficient price discrimination.

ABSTRACT: This paper investigates a puzzle and possible policy concern: Why do platforms such as eBay and Visa that enable the trade of goods of different unobserved costs and values rely predominantly on linear ad-valorem fees, that is, fees that increase in proportion to the sale price of the trades that they enable? Under a broad class of demand functions, we show that a linear ad-valorem fee schedule enables a platform to maximize its profit as if it could actually observe the costs and values of the goods traded and set a different optimal fee for each good. Surprisingly, we find for this class of demands, allowing the platform to set ad-valorem fees (i.e. price discriminate) increases social welfare, both when the platform is regulated to recover costs and when the platform is unregulated.