||Rules are for the obedience of the foolish and the guidance of the wise
||LTSA & Police Myth|
|Higher speed limits reduced crash rates and injuries without increasing fatalities.
What caused an accident? Driver, vehicle, road, atmospheric, surface or traffic conditions?
Interpreting statistics of accident data where these factors can neither be measured nor controlled is hazardous. Identifying trends and causes of those trends in data subject to so many variables is even more difficult. Untrained and vested interests often rush to claim certainty where none exists. Statistical studies of accident data must attempt to control as many of those variables as possible.
Many of these studies examine the impact of major speed limit changes.
In 1995, as a result of pressure from the National Drivers Association (see www.motorists.org) the US Congress removed the national open road speed limit of 55 mph which had been imposed as a fuel saving measure. Many States subsequently raised their speed limits and various studies have been done of the consequences. The classic analysis was the Cato Institute's May 1999 Speed Doesn't Kill report which concluded that raising the speed limits reduced injuries significantly, increased fatalities insignificantly and had a considerable net cost benefit to the public.
Various other studies have come to varying conclusions since then, including one by LTSA which claimed increased U.S. speed limits had led to substantially increased injuries and fatalities.
However, the Cato conclusions are supported by the most recent and most thorough academic study. re-examined the impact of March 1996 increased open road speed limits from 55mph to 65mph (88kph to 104kph) in Washington State of the USA. This study classified over 6000 road segments averaging 700 feet long to ensure that the before and after comparison was based on equivalent road and traffic conditions. Otherwise, for example, selective assignment of speed limits by authorities to particular conditions may bias the results.
The number of fatalities was virtually unchanged when all other factors were the same. Drivers had 12% fewer accidents at the higher speeds. But drivers who were going to have a serious crash, were very slightly more likely to kill themselves when they did. Level sections of road with no median barriers and little congestion had the highest fatal crash rates.
One of the authors recently summarised their conclusions:
Responding to a hypothetical 10 mph (16 kph) speed limit increase, the final models predict just a 0.07 percent increase in fatalities, a 1.62 percent increase in fatal crashes, an 11.7 percent decrease in injuries, a 12.3 percent decrease in injury crashes, a 9.2 percent decrease in Property Damage Only crashes, and an 8.4 percent decrease in total crashes. Using Blincoe et al.’s 2002 findings, a 10 mph speed limit increase on high-speed roadways in the State of Washington is estimated to offer benefits worth $2.9 million (in 1996 dollars) to society.
This result is consistent with the intuition that serious accidents are caused mostly by a few drunk, drugged, reckless, inattentive, over-tired, suicidal, or incompetent drivers and occasionally by rare combinations of unforeseeable conditions. At general open road speeds, it is unlikely these factors are sensitive to moderate speed changes.
The following chart shows that after the oil shocks of 1974 and 1979 which increased costs, introduced carless days and reduced travel, both fatalities and injuries were trending up until the open road speed limit was increased from 80 kph to 100 kph in 1985. That increase together with freeing up the imports of cheap better quality cars ushered in a sustained period of declining fatalities and injuries.
Spatially Disaggregate Panel Models of Crash and Injury Counts by Kockelmann & Kweon, Univ of Texas, presented at Transportation Research Board annual meeting in Washington, DC, January 2004.
Y-J Kweon, University of Texas - personal communication, March 2004.