E-scooters and traffic accidents : evidence from staggered roll-out in Swedish municipalities

The rapid rise of e-scooters (electric scooters) in cities around the world, boosted by the introduction of shared e-scooter services has visibly reshaped the way people move around cities, sparking both excitement and controversy. With the increase in popularity of these vehicles, concerns regarding their impact on traffic safety and accidents have become a rising public concern. In this paper, we investigate the frequency of traffic accidents involving escooters following the introduction of shared e-scooter services in Swedish municipalities during the period 2019-2022. We use a staggered difference-in-difference regression to identify the causal effect of shared e-scooters on various types of traffic accidents using municipalities without e-scooters as a control group. We present three main findings. First, overall accidents increase by approximately one standard deviation in the first quarter following the introduction of shared e-scooters, but the overall effect decreases (0.5-1 standard deviation) over five quarters and vanishes over nine quarters. Second, the increase in accidents involving e-scooters is not associated with an increase in pedestrian or bicycle accidents. Instead, e-scooters are predominantly involved in accidents with cars. Third, the observed increase in accidents is largely attributable to large metropolitan areas, where urban traffic is usually more complex and intensive.