Accelerated Quality Discovery through Sponsored Search Advertising in Online Marketplaces (Job Market Paper)
Abstract: This paper analyzes the dynamic effect of sponsored search advertising on quality discovery on the Taobao.com retail platform. A stylized model spells out the dynamic from a theoretical perspective: A new product's boosted exposure from sponsored search ads could help the platform to infer how well the product converts exposure to sales (the quality measure). Because the platform awards top organic search ranks to products inferred of high quality, sellers with higher private quality signals would bid more aggressively for sponsored ads, which accelerates the platform's discovery of these high-quality products. An empirical specification, connecting search ranks to sales through exposure, is estimated using scraped panel data on sales of hundreds of products and their sponsored and organic search ranks under thousands of keywords on both PC and mobile interfaces. A platform-specific feature with individual organic ranks affected by asynchronous weekly cycles provides exogenous variation for identification. The estimation results echo the stylized model's characterization of the informational landscape and reveal a synergy between the platform's PC and mobile interfaces: PC consumers browse extensively and generate significant exposure for sponsored products. Observations of sponsored sales to PC consumers could improve the organic rankings in the mobile app, where consumers highly concentrate on top products.
Costs of Managerial Attention and Activity as a Source of Sticky Prices: Structural Estimates from an Online Market (with Sara Fisher Ellison and Christopher M. Snyder)
Abstract: We study price dynamics for computer components sold on a price-comparison website. Our fine-grained data—a year of hourly price data for scores of rival retailers—allow us to estimate a dynamic model of competition that identifies two stages of price setting: monitoring market conditions and changing to the new price. Using a framework inspired by Bajari, Benkard and Levin (2007), we back out structural estimates of managerial frictions associated with the two stages. The estimated frictions are substantial, concentrated in the act of monitoring market conditions rather than entering a new price online. We use our model to simulate the counterfactual gains from automated price setting and other managerial changes. Coupled with supporting reduced-form statistical evidence, our analysis provides a window into the process of managerial price setting and themicrofoundation of pricing inertia, issues of growing interest in industrial organization andmacroeconomics.
Keeping Good Quality as a Surprise
Abstract: This paper analyzes the revelation of hard information when two parties transact a good. The buyer's valuation of the good depends on both the buyer's type (taste), and the seller's type (quality). The seller can choose whether and when to credibly reveal quality information during communications. This paper first analyzes an alternating bargaining game allowing endogenously chosen delay between communications in the fashion of Admati and Perry(1987), and shows there exists an equilibrium involving delayed revelation of quality information as a way to extract surplus from the buyer type with a greater taste for quality. Under some parameters, this is the unique outcome. This result complements the unraveling prediction of Grossman and Hart (1980) and Milgrom and Roberts (1986), which state that the seller should fully reveal all hard information. The paper then analyzes an informed principal problem, in which the seller, knowing his own quality, can choose and commit to any mechanism. It is found that under some specification of the parameters, the unique equilibrium outcome is that the seller, regardless of quality, chooses a truth-telling mechanism that maximizes the revenue of the seller type with high quality.