Modeling the operational and labor costs of autonomous robotaxi services

Autonomous vehicle (AV) companies are rolling out driverless taxi-type services in cities around the world, promising to enhance road safety, promote equity, and foster environmental sustainability while simultaneously threatening the jobs of current taxi and ride-hailing drivers. The potential impacts of these “robotaxi” services hinge upon not only whether AV technology becomes sufficiently performant, but also whether AV services become economically competitive. In this study, we develop ground-up cost models for a traditional taxi and a robotaxi service to compare their relative competitiveness. We draw on direct observations of commercially available robotaxi services, semi-structured interviews with AV operational and regulatory experts (N = 27), and archival documents to include the most detailed accounting of frontline robotaxi labor roles to date. We find that labor remains a significant cost for existing robotaxi services but that robotaxi operating costs are still lower than those of traditional taxi services. Ultimately, utilization rates and annual mileage will serve as the most influential factors for robotaxi competitiveness. Finally, if jobs shift from traditional taxi to robotaxi services, the total number of frontline jobs could decrease by between 57% and 76%, but the distribution of worker wages would shift higher. Transportation planners, researchers, and policymakers should continue accounting for labor costs in an AV future, as these costs will influence where AV deployment and job losses are (and are not) likely to occur, and should proactively investigate alternative career paths for potentially displaced frontline workers. Robotaxi costs are highly sensitive to lower labor ratios (i.e., fewer vehicles per worker), but firms reach economies of scale relatively quickly. Robotaxi firms should weigh marginal labor cost reductions against impacts to passengers and workers.