Uber vs. Taxi: A Driver's Eye View (with Joshua Angrist and Jonathan Hall)
Abstract: Ride-hailing drivers pay a proportion of their fares to the ride-hailing platform operator, a commission-based compensation model used by many internet-mediated service providers. To Uber drivers, this commission is known as the Uber fee. By contrast, traditional taxi drivers in most US cities make a fixed payment independent of their earnings, usually a weekly or daily medallion lease, but keep every fare dollar net of expenses. We assess these compensation models from a driver's point of view using an experiment that offered random samples of Boston Uber drivers opportunities to lease a virtual taxi medallion that eliminates the Uber fee. Some drivers were offered a negative fee. Drivers' labor supply response to our offers reveals a large intertemporal substitution elasticity, on the order of 1.2. At the same time, our virtual lease program was under-subscribed: many drivers who would have benefitted from buying an inexpensive lease chose to opt out. We use these results to compute the average compensation required to make drivers indifferent between ride-hailing and a traditional taxi compensation contract. The results suggest that ride-hailing drivers gain considerably from the opportunity to drive without leasing.
Research in Progress
The Reallocation of Workers Across the Business Cycle with Fane Groes, Nikolaj Harmon, and Daniel le Maire
Abstract: Many economic models assume that consumption and savings decisions today depend on beliefs about future income. In this project we combine administrative (tax, credit report) and survey (durable consumption) data with novel survey information on filers’ expectations about the size of their tax refunds. We first document that while many households in our sample of low-income filers have correct mean expectations about their refunds, there is substantial uncertainty, particularly among EITC-eligible tax filers. Furthermore, the accuracy of expectations closely tracks self-reported uncertainty, as measured through probabilistic survey questions. We then study how consumption and savings behavior responds to surprises in the size of tax refunds and to the resolution of uncertainty about tax refunds during tax filing. Revolving debt balances at the time of tax filing are on average roughly equal to refund size, and tax filers who expect larger refunds are more likely to use their refund to repay debt. Positive surprises about tax refunds affect self-reported savings rather than durable purchases. Households with more uncertainty resolved at the time of tax filing are on average equally likely to repay debt, save, or make durable purchases, suggesting little to no precautionary behavior.