Increasing Returns to Scale in Carpool Matching: Evidence from Scoop

Systems that depend on matching often exhibit scale economies, whereby increased participation leads to improved performance for all users. This paper examines the presence of such increasing returns to scale in carpool matching. Data from Scoop, a carpooling app, is used to demonstrate this phenomenon across various markets using regression. As the number of requests to carpool in a certain market rises, the share of proposed matches that users accept rises, while the extra distance traveled to accommodate these carpools declines. These relationships hold in four specifications of the regression model, and they suggest there are increasing returns to scale in matching.