Drive for the season, year-round
Fri 19 Feb 2010 by Infometrics Ltd.

One of the grimmer aspects of the holiday season is the prominent focus given to the holiday road toll.   Although it will be small comfort to the friends and relatives of those who died on the roads this summer, 2009’s holiday road toll of 13 was a substantial improvement on the 25 who died in 2008.   After adjusting for the number of vehicles on the road, 2009 would be the third lowest holiday road-toll since records began in 1958.

There is often a temptation to divine an explanation for the year-to-year swings in the holiday road toll (or the yearly road toll).   The reality is that over such a small sample period (the holiday period is normally 11.6 days), the normal variation inherent in such figures swamps any ability to attribute changes to causes like public safety campaigns or changes in police visibility.   We just can’t say that the drop from 25 to 13 reflects any underlying change in behaviour.

The good news is that over a longer time period, there is a remarkably clear and significant downward trend in the overall road toll.   The raw road toll statistics need to be adjusted for the increasing number of drivers on the road.   Ideally this adjustment would take into account the actual number of kilometres driven, or time spent in cars, but lacking that data the common adjustment is to compare road accidents to the number of vehicles in New Zealand.

Using that as our metric, annual road fatalities have fallen from around six per 10,000 vehicles in the early 1950s to about one now modern drivers have just 20% the annual mortality risk of drivers two generations earlier.   Over that period, the holiday road-toll has declined inlock-step with the overall road-toll.   The record road toll of 37 in 1972, if scaled up to modern vehicle fleet size, would have been 82.

As with most significant changes, there are a number of contributing factors to the improvement in road safety.   Vehicle safety standards have steadily climbed; road standards are higher; seat-belt usage has improved; and rates of drink-driving and speeding have both declined.  Naturally these tend to be slow-moving processes, which is another reason why we’d not expect to see statistically significant changes in year-to-year road tolls.

The interesting fact about the holiday road toll is that, despite the large amount of media attention given to it, the holiday period is not strikingly a more dangerous time to drive.   On average fatalities are 20% higher per day in the holiday period than on the other days of the year (put in context, it was as dangerous driving in the 2009 holiday period as it was on a normal day in 2004).   Injuries are only barely (5%) higher.

The graph below demonstrates the relationship between the road toll period and the rest of the year comparing the ratio of fatalities per day during the road toll period to fatalities per day in the rest of the year.

And clearly we are not comparing apples with apples here.    If your holidays bear any resemblance to mine, much of it involves the inside of a vehicle.   Vehicle miles are probably higher in the road toll period, and a much greater percentage of that driving is open-road (i.e. at higher speeds).  Once an allowance is made for this fact it is quite plausible that the drive to your holiday destination is no riskier than any other open-road drive you would undertake during the year.   It may even be safer!

This outcome should not necessarily surprise us more cars on the road means more cars to hit, but also slower average speeds.   Drivers may be tired from driving long spells, but they also benefit from more day-time driving.

Now it is possible that the holiday period would be a particularly dangerous period if not for the fact that we drive under the perception that driving is unusually dangerous.   Driver behaviour is often strange like that studies have shown, for instance, that mandatory seat belt laws led to people driving faster (as they believed themselves to be safer).

Overwhelmingly, most road deaths are still attributable twosome form of human error (as opposed to weather, health problems, or vehicle malfunction).   So the lesson to draw from these statistics is not to relax your driving standards on your next summer vacation, but rather to apply the same degree of healthy caution to your driving in the other 353 days of the year.

Related Articles