Trading across Borders in Online Auctions
利用在线编程支持市场数据，研究服务贸易中买卖双方如何匹配，发现卖家按质量和成本内生选择项目，解释了跨境交易的地理集中现象。
We invoke the insights from the auction literature to study trade in services using data from an online market for programming support. We find that the observed clustering of trade between countries can be rationalized through a model featuring endogenous sorting of sellers who are heterogeneous in both quality and costs across projects offered by buyers who differ in outside option and willingness to pay for quality. To accommodate the possibility of such an outcome we extend a single-auction entry model to a setting where sellers choose among multiple projects. This feature plays an important role in explaining the data and understanding the effects of various trade policies.