Risk homeostasis theory and traffic accidents: propositions, deductions and discussion of dissension in recent reactions

Abstract

There is abundant evidence showing that technological innovations have led to major reductions in the accident loss per unit distance of mobility, in certain road sections, on certain roads as well as in the road network as a whole. However, the accident loss per time unit of road-user exposure and per head of population have not shown equally favourable downward trends. In order to explain this contrast, as well as many other findings regarding road-user behaviour, the theory of risk homeostasis (RHT) has been put forward. This posits that accident loss per capita and road-user behaviour are mutually related in a closed-loop regulation process, with the level of preferred risk as the controlling variable outsidethe closed loop. There is evidence also that the per capita traffic accidents can be reduced by motivational interventionsthat are effective in lowering road users' preferred level of accident risk. RHT has received support as well as opposition from other researchers. The purpose of the present paper is to identify what appear to be the major sources of disagreement. It will be argued that the opposition to RHT is largely due to misapprehension of its essential propositions and their derivations and that the allegedly contradictory empirical data drawn into the debate by some commentators are either inconclusive, compatible with, or in support of the theory in question.