Pay-as-you-drive, driving behavior and risk of accident: evidence from Norway

Abstract

Through collaboration with a telemetry company and an insurance company we analyse detailed driving data from 550 vehicles equipped with in-vehicle data recorders (IVDRs) as a part of an ongoing pay-as-you-drive (PAYD) pilot study conducted in Norway. By combining a naturalistic driving dataset with data on insurance claims, we seek to understand the relationship between accident involvement and driving behaviour. We estimate several logistic regression models in order to investigate the effect of various exposure variables on accident risk and find that accident-involved vehicles generally accumulate more mileage, speed more and drive more during rush hours, compared to accident-free vehicles.