In the US motor vehicle fuel economy standards were imposed in the late 1970s, in response to the oil crises of that decade. Since then, efforts to increase the standards have not occurred, one reason being the argument that smaller vehicles (which are generally more efficient) are considered less safe. Recent analyses (Energy J.( 2004)) suggests that variance in vehicle weights may be more important than the absolute weights of vehicles in making the highway network less safe. In Europe and other countries, which generally have smaller more efficient vehicle fleets, due to relatively high gasoline taxes, this debate has not occurred. In particular, countries such as Great Britain and Sweden have far safer road transport systems than the US but also have much more efficient vehicle fleets. This suggests that either vehicle weight and size are unimportant or if they have an effect it is small compared to other factors. This paper uses international data to build econometric models that examine whether average vehicle fuel economy has any association with road traffic fatalities, while controlling for other factors that are associated with fatalities. The effect on pedestrian fatalities is also analyzed. Cross-sectional time-series data on traffic fatalities from OECD countries is used and negative binomial regression models are developed using panel data to determine whether any associations are present. Results find that changes in vehicle efficiency are not associated with changes in traffic fatalities, suggesting either that size and weight changes over time have not had a strong effect or are not associated with fuel economy improvements. (c) 2004 Elsevier Ltd. All rights reserved.