Seller-Pricing Frictions and Platform Remedies

Abstract

Many platform sellers struggle to set prices when facing complex market conditions. These difficulties motivate the platform to offer (centralized) algorithmic pricing as a remedy, yet at the expense of sellers’ private information. This paper studies the extent of seller-pricing frictions and explores underlying mechanisms. I use granular data on Airbnb sellers’ demand and pricing strategies to pin down the extent of pricing frictions. Leveraging a price-setting interface change, I disentangle two different mechanisms –sellers’ price-adjustment costs and their use of suboptimal heuristics– and show both are essential drivers of the frictions, but with distinct implications on the optimal platform remedy. I develop a scalable model to characterize the competitive market equilibrium, separately quantify sellers’ price-adjustment costs and use of heuristics, and simulate alternative platform remedies. My results show that a centralized pricing algorithm will distort market outcomes if it cannot fully account for sellers’ private information, suggesting that the platform should let sellers set prices even if their decisions are imperfect. An alternative remedy is to improve the seller-side interface, and I show that reducing sellers’ price-adjustment costs alone can alleviate up to 43% of the total profit loss from frictions.