Learning by Consuming: Optimal Pricing for a Divisible Good

Abstract

We study the revenue-maximizing mechanism when a buyer's value evolves because of learning-by-consuming. The buyer chooses the initial consumption based on his rough valuation. Consuming more induces a finer valuation estimate, after which he determines the final consumption. The seller faces the tradeoff that selling more initially makes selling the rest more profitable but on a smaller base. The optimum is a try-and-decide contract. In equilibrium, a higher first-stage valuation buyer chooses more initial consumption and enjoys a lower second-stage per-unit price. Methodologically, we address the difficulty that without the single-crossing condition, monotonicity plus envelope condition is insufficient for incentive compatibility. Our results help to understand contracts with learning features, e.g., course packages with included sessions and leasing agreements for experience goods.

Keywords: Dynamic mechanism design, Learning by consuming, Endogenous type distribution, Quantity effect, Experience good.