'Meet Me Halfway': The Costs and Benefits of Bargaining

Abstract

Bargaining is an important pricing mechanism, prevalent in both online and offline markets. However, there is little empirical work documenting the costs and benefits of bargaining, primarily due to the lack of real-world bargaining data. We leverage rich, transaction-level bargaining data from a major online platform and supplement it with primary data to quantify the costs and benefits of bargaining for sellers, buyers, and the platform. We do this by building a structural model of buyer demand and seller pricing decisions while allowing for the existence of bargaining initiation cost, loss-of-face cost, and price discrimination. Using our results, we perform three policy simulations to quantify the importance of not distinguishing between no-bargain and failed-bargain transactions, ignoring the loss-of-face cost, and not allowing bargaining. These simulations provide rich details on how the various costs of bargaining impact our understanding of buyer and seller behavior and transaction outcomes. Banning bargaining, in particular, benefits the buyer and the platform greatly, but only has a modest benefit for sellers. Finally, we show that our results are robust to our assumptions and replicate in another product category.

Keywords: Bargaining, Pricing, Platforms, Digital Markets, Structural Models, Alibaba, China

JEL Classification: D40, L10, L11, M30, M31