So another Easter has come bringing with it the traditional Easter road toll and press hyperbole…
Out come the state premiers and police commissioners to solemnly pronounce another crackdown on speeding and drink driving across the Easter holiday, and we are all put on notice that our collective reckless behavior will not be tolerated, this time.
But are we all really behaving more recklessly?
Let’s strip away the rhetoric and think about the subject cooly and rationally. We know for example that road accidents go down in times of economic hardship (Thoresen et al., 1992). Quite simply people drive less and therefore with fewer kilometers driven there are less accidents (1).
So there is a demonstrated proportional relationship between road usage kilometers and accident rate. One obvious reason is that people are prone to err (a recognized major causal factor in road accidents) so the longer people do a task (or the more of them do it) the higher the cumulative likelihood of error (2).
Given this relation, we would reasonably expect to see higher accident rates in those months where more people spend more time traveling, that is the Easter and Christmas breaks.
In fact we’d expect to see an increase regardless of whether people behave more or less riskily across the Easter or Christmas break.
This relationship could also be non-linear because accidents happen to both single vehicles and between vehicles. The more vehicles proportionally the greater the number of potential interactions. If so then even a small increase in road usage could have an appreciable affect upon accident rates.
Normalising the data
The problem is that our perception of risk as a society is informed by un-normalized data. We perceive in absolute terms the accident rate rather than the rate per trip or kilometers travelled. So when we make decisions based on this skewed perception they may often make little sense from the perspective of managing an individual risks.
For example the increased police presence on the roads over Easter comes at a cost in terms of police overtime and in the immediate period after the ‘blitz’ a reduced level of enforcement.
Now if enforcement really does work in reducing accidents then the normalized per kilometer accident rate may well go up with a reduced police presence, thereby increasing an individual road users risk.
A simple proposal
All of the discussion above would be of mild academic interest if it weren’t directly relating to the health and safety of the members of our society, including ourselves.
Perhaps now is the time to step back and ask whether these traditional holiday safety blitzes are really cost effective and whether our safety dollar might be money better spent elsewhere.
1. There may also be an associated affect where those road users more likely to have an accident, primarily young males in lower socioeconomic groups, are the first to leave the road users pool in times of economic hardship.
2. This does ignore the effect of performance shaping factors such as fatigue, circadian disruption and environment (such as night and unfamiliar terrain). But in my opinion as a simplifying assumption, proportionality is still valid.
1. Thoresen, T., Fry, T., Heiman L. & Cameron, M. Linking economic activity, road safety countermeasures and other factors with the Victorian road toll, Monash University Accident Research Centre, Report No. 29, 1992.