Abstract: This paper examines how consumers learn from their market experience in a homogeneous goods market. Using consumers' six-month purchase history data in a unique empirical setting in one of China's largest online marketplaces, this study finds that consumers buy from cheaper sellers as they gain market experience. To investigate how market experience improves the outcomes of price search, it incorporates price learning into a flexible structural search model, in which consumers have Dirichlet priors and update their price beliefs based on their past purchase prices in a Bayesian fashion. The results suggest that consumers have an upward bias in their prior price beliefs and are increasingly more price sensitive as they gain market experience. Early in a consumer's purchase history, the memory of sellers and prices from previous purchases accounts for a large portion of the price improvements, whereas an increasing price sensitivity plays a larger role in explaining the price advantage later on.
Research in Progress
Does the Emergence of Online Reviews Affect Quality? Evidence from Yelp.com
Abstract: I investigate whether the emergence of online reviews – a popular new form of word-of-mouth communication – has any impact on restaurant quality with a differences-in-differences strategy. Specifically, based on the survey findings that tourists rely much more heavily on online reviews for local restaurant choices, I exploit the exogenous variation in the impact of online reviews on chain restaurants with different numbers of locations worldwide, small vs. large chains, as well as the within-chain variation at different locations, local vs. tourist areas. This paper uses Yelp review data for all restaurant chains with at least two locations in Las Vegas, one of the world’s most popular tourist destinations. The empirical strategy is to treat large well-known chains as the control group, small chains as the treatment group, and the distance of a restaurant to the most dense tourist area in Las Vegas, the Strip, as the amount of treatment. Controlling for chain and zip-code fixed effects, I find that (1) within a small chain, no relation appears between the average rating of a restaurant and how far away the location is from the Strip, but (2) a significant difference in average ratings exists at different locations for larger chains: the closer to the Strip, the worse the rating. Moreover, results on the rating trends over time show that (3) small chains' locations closer to the Strip improve significantly more on Yelp ratings, compared to their locations farther out into the suburbs; in contrast, (4) large chains experience a stronger decline in ratings at their locations closer to the Strip. These results together suggest that online reviews play a more significant role in motivating small chains than large chains to maintain high quality at their dense tourist locations, where restaurants have little incentive to establish a good local reputation.
Obfuscation and Price: Evidence from E-Commerce
Abstract: How do merchants who sell the same product compete on an e-commerce platform with a price-comparison feature? This paper presents empirical evidence that some sellers in a marketplace intentionally make their prices on the price-comparison page lower than the actual prices by taking advantages of a loophole. Given the fact that the marketplace advertises only a seller’s lowest price, some sellers create several prices for one listing by offering a quantity discount, with one of the prices being very cheap but not applicable to any consumer due to the enormous amount of products required to purchase. In stark contrast to their advertised price, the actual applicable prices of sellers implementing this type of obfuscation techniques are significantly more expensive than others’. In addition, experienced sellers are more likely to use this obfuscation strategy to successfully mislead inexperienced consumers.