主 题：Quantifying the Economic Value of Location Preferences and the Gravity Effect in Online C2C Platform Shopping
Locations of sellers on online consumer-to-consumer (C2C) platforms can affect purchases directly via the gravity effect and indirectly via a buyer’s location preferences. Using detailed transaction data (from Alibaba Group’s Taobao.com) between buyers and sellers who are located in provinces all over China, we quantify the economic value of seller location for online purchases. We take a hierarchical Bayesian approach to estimate a destination choice model that allows for asymmetric pairwise origin-destination preferences, while accounting for distance, shipping costs, seller characteristics, product quality guarantees, conventional and novel marketing tools, the selling environment, etc. Additionally, as part of the Bayesian hierarchy, the model characterizes pairwise location preferences in terms of socioeconomic, cultural and commercial factors at and between origins and destinations. An endemic issue with studying choice behavior in online C2C platforms is that the large number of potential sellers of a product in a given location creates the problem of choice set explosion when formulating and estimating the model. To address this issue, we use a matching method to first identify the most compatible seller from each province for each choice occasion. Our empirical results reveal that even for identical products, provinces in which sellers are located are not viewed as being equal by buyers in different origin provinces; there is substantial variation and considerable asymmetry in preferences across any pair of buyer-seller locations. Preferences explain 21.8% of the variation in destinations’ origin-specific market shares. Buyers show stronger preferences for their home provinces, for provinces that are perceived to be more trustworthy or have better commercial credit environments; buyers seem to use a seller’s location as a heuristic for transaction security and risk reduction. Distance remains a barrier for online shopping, but the gravity effect is moderated by seller reputation, and seller-reputation and product quality guarantee are the most important choice promoting factors. We then show how our model can be used to quantify the benefits of “location-based” pricing so sellers in more disadvantaged provinces can use price concessions to overcome their locational disadvantage.
关键词: Online C2C platform, Location preference, Gravity effect, Home bias, Trust, Big Data, Emerging Market, China
Professor Chu is interested in empirically modeling consumer behavior and the strategic interactions among firms. Her research falls into the New Empirical Industrial Organization (NEIO) paradigm. She employs both the classical approach and Bayesian approach to study two-sided markets, network effects, e-commerce, distribution channels, retail competition, high-tech markets, social interactions and networking. She is also interested in emerging markets in general and China’s economy and market in particular.