Google
 

 Publications

Bio

New and extra publications

 

(A piece that is similar to this was published as Fueling Road Safety (PDF). adiNEWS, p. 20-21, October 2008)

Do increases in the cost of fuel reduce traffic fatalities?

Leonard Evans, 2008-08-04

I have been asked this question countless times since the 1973 oil embargo.  The frequency with which it has been asked has increased recently, especially since the publication of the Michael Sivak’s useful report Is The U.S. On The Path to The Lowest Motor Vehicle Fatalities in Decades?  Sivak’s report adds analytical substance to answering this recurring question.  He assembles estimates of gasoline sales, miles driven, and fatalities from May 2007 through April 2008, and finds declines in gasoline sales, miles driven, and fatalities, but the fatality decline is much larger than the others.  Sivak estimates that, if these trends continue, fatalities could drop below 40,000.

 I am in agreement with the thrust of Sivak’s report.  My purpose here is to offer some additional comments, including some that overlap with Sivak’s report, on why fatalities decrease far more than the decrease in miles driven, and to place any declines in US fatalities within the framework of the main points in the chapter The Dramatic Failure of US Safety Policy in my book Traffic Safety.

While there is uncertainty about the future of oil prices, there is little uncertainty about the most basic economic law.  When the cost of something increases, consumption decreases.  This law applies to fuel, and also to alcoholic beverages (see extensive discussion in Traffic Safety.).  For the case of the price of gasoline, there is no well established quantitative relationship saying how much travel will decline for a given increase in the price of fuel.  But when fuel goes up in cost, miles traveled go down.

 For a given drop in miles of travel, it might seem that fatalities would drop by the same percent.  This is not so because different types of driving are preferentially curtailed as gasoline prices increase, and different types of trips have different risks.

 Most travel is commuting to work.  This remains largely unaffected by fuel costs.  There are minor effects.  High fuel costs coincide with a sluggish economy producing some job loss, high fuel costs encourage more carpooling and use of public transport, and some organizations have encouraged or instituted 4 day work weeks.  Compared to the total miles traveled in commuting that remain unaffected, these are small effects.  The net result is that the total commuting miles remain relatively unaffected by fuel prices.

 The reduction in total miles traveled comes mainly from discretionary travel.  Discretionary travel poses far higher fatality risk than commuting travel because it involves higher average speeds.  Speed is the dominant factor in determining fatalities – fatality risk increases as the fourth power of travel speed.  Commuting travel speeds are kept low by congestion, while discretionary travel may involve high-speed intercity travel and travel at times of reduced congestion permitting higher speeds.  Discretionary travel is more likely during hours of darkness.

 As Sivak mentions, higher fuel costs lead to larger reductions in travel by drivers with limited funds, including young drivers and drivers in lower socioeconomic groups.  Such drivers have above-average crash risks, so reductions in their driving will more than proportionally reduce traffic deaths.

 One discretionary trip that is likely to be curtailed as fuel purchases leave less budget for entertainment is driving to bars and restaurants that serve alcoholic beverages.  This will reduce driving with elevated blood alcohol levels. 

Alcoholics and problem drinkers are responsible for about one third of traffic deaths.  Higher fuel costs could use so much of their money that they could no longer afford bar prices and be forced to seek a cheaper way to drink.  This is available by purchasing alcohol at supermarkets and taking it home.  After a night’s drinking the alcoholic or problem drinker rolls into bed instead of getting behind the wheel, with great benefits to road safety.

 Drivers choosing to reduce speed on freeways to conserve fuel will indeed save modest amounts of fuel but (as pointed out by Sivak) will sharply reduce their fatality risk.  Reducing speed from 70 mph to 65 mph reduces fuel use by about 8%, but fatality risk by 26%.  I don’t expect many drivers to slow down in order to save fuel.  However, even a few doing so will contribute to reducing fatalities because the fatality rate increases so steeply with increasing speed.  Substantial fatality reductions would result from enacting, and enforcing, lower speed limits.  The nationwide 55 mph speed limit introduced in 1974 was a major contributor to the largest ever one year drop in fatalities, from 54,052 in 1973 to 45,196 in 1974. 

 Increases in fuel prices lead to reductions in driving.  Reductions in driving lead to far greater than proportionate reductions in fatalities, so that a, say, 3% reduction in travel could lead to a 10% reduction in fatalities.  Would this not negate the conclusions in the chapter The Dramatic Failure of US Safety Policy that were based on comparing changes in US fatalities to those in other countries?  No -- while traffic deaths will now decline in the US, they will also decline in other countries because the increase in oil prices is a world-wide phenomenon.  Indeed, since the chapter was written the gap between the US and the other countries has already widened.  For example, in 2007 the UK recorded under 3,000 traffic deaths, their lowest total since records began in 1926, and a drop of 63% below their all time high.  For the US to achieve the same drop from our all time high we would need to record 20,100 annual deaths.  Even if our fatalities do drop below 40,000, we still have a very long way to go!