Removing accident liability through a no-fault system may only have minor effects on driver care, yet existing studies assessing whether no-fault auto insurance requirements increase accidents use fatal accidents as the primary outcome. In this paper we estimate the effect of no-fault on overall accident rates using detailed data from police accident reports in New Jersey, North Carolina, and Utah. We focus on accidents involving out-of-state drivers and account for differences in driving prevalence using accidents beyond the control of the driver. Controlling for a wide variety of accident, driver, and vehicle characteristics, we find little evidence that drivers under no-fault cause more accidents. Results on accident severity are mixed.