|Abstract||Some people consider that small cars necessarily have low secondary safety, i.e., a relatively high probability of serious injury to their occupants. In collisions with other vehicles, this is true, as there is a strong effect of mass ratio. Beyond that, this paper will argue there is probably little or no further effect. Three lines of evidence are presented to support this.
• Theoretical analysis of the movement of occupants in collisions, and how injury occurs.
• There is no compelling evidence for an effect of car mass in data from the 1970’s and 1980’s. The most frequent reasons for rejecting claims based on that data are as follows. (1) Lack of relevance (e.g., the occupants were largely unrestrained). (2) Damage-only crashes were included in the denominator number of crashes (thus giving misleading results if the degree of under-reporting of damage-only crashes depends on size of car).
• We have recently analysed South Australian data on single-car crashes. Mass of car was found to have no effect. Allowance was made for some covariates: characteristics of driver (age, sex), geography (speed limit, in vs. outside the metro area), and time (night hours vs. daytime).
Naturally, if small cars were encouraged (explicitly, or by a substantial increase in fuel price), attention would need to be given to their safety. But at present, there is little evidence to support a conclusion that society should shy away from reducing the average mass of cars in its fleet.