By James Baxter, NMA President
This past Tuesday the Insurance Institute for Highway Safety, IIHS, carpet bombed the media world with a press release claiming red light cameras saved 154 lives in 14 cities from 2004 through 2008.
Saying IIHS pulled this number out of thin blue air would be a tremendous compliment.
There is no meaningful evidence that red light cameras saved any lives in this or any other time period. The only credible evidence that does exist indicates that red light cameras have caused a lot of accidents.
The IIHS picked 14 cities that did not have red light cameras from 1992 to 1996, but did have them from 2004 through 2008. Then they compared the number of fatalities credited to red light violations for the same two time periods.
Here are just a few of the blunders in this IIHS “study”:
- They didn’t distinguish between the intersections with and without ticket cameras.
- They didn’t account for the long term trend in constantly declining fatality rates (fatalities per 100 million vehicle miles traveled) 1.7 in 1996 vs 1.3 in 2008, a 23 percent reduction, and the drop in actual fatalities of 11 percent.
- For a control, IIHS number crunchers looked at 48 cities that did not have red light cameras in either time frame; 1992 to 1996 or 2004 to 2008. In these cities too, the number of red light running fatalities dropped, but not at the same “rate” as the cities with red light cameras. But what rate are they referring to? It isn’t based on vehicle miles traveled, an actual measure of accident exposure. No, they used a rate based on fatalities per capita!
Demographics, infrastructure, and economies can change a great deal over 16 years, and radically so from city to city.
Per capita comparisons between Dallas, TX and Washington, DC could tell us a lot about the people who live in these cities. However, they also include a host of variables that make apples-to-apples comparisons impossible when measuring the utility of something like red light cameras.
That’s why we stick to using rates based on miles traveled.
The most telling research measuring the effect of red light ticket cameras involves exposing the same population of drivers to intersections with and without cameras, and logging the results.
When this is done (and other games aren’t played, like increasing the yellow light duration at ticket camera intersections and crediting the improvement to the cameras) invariably, the non-camera intersections continue the long term trend of reduced accidents and the ticket camera intersections experience an increase in accidents.
This totally contradicts the IIHS claims.
As a footnote, the true quality of the IIHS report is best reflected by their claim that there is broad public support for red light cameras, while omitting the fact that in the 15 times red light camera programs have been challenged through citizen referendums, the cameras came down every single time.
It turns out IIHS ticket camera statistics are as reliable as their measure of public opinion.
Editor’s Note: For more analysis of their ridiculous press release, check out TheNewspaper.com’s total deconstruction of their flawed numbers.