Now that Goldy isn’t the head honcho here any more, I think I’ll pick on him today. Over at Slog, he posts:
Following the success of last year’s local initiative outlawing red-light cameras in his hometown of Mukilteo, Eyman’s taking his latest for-profit/anti-government gimmick on the road. This year, he’s cosponsoring copycat measures in Bellingham, Monroe, Wenatchee, and Longview. But while Eyman provocatively characterizes the cameras as the “crack cocaine” of city budget writers and “taxation-by-citation, just another way for government to pick the pockets of taxpayers,” a definitive new study conducted by the Insurance Institute for Highway Safety (IIHS) finds that red-light cameras save lives.
Comparing crash statistics between 1992–1996 and 2004–2008 in the 99 US cities with populations above 200,000, researchers found a 35 percent reduction in red-light fatalities in cities that implemented red-light-camera programs, versus a 14 percent reduction in those that did not.
But the cameras’ benefits actually proved to be much bigger. When all crashes at signaled intersections were tallied, not just those due to red-light running, total fatalities dropped 14 percent in cities with cameras, while rising 2 percent in cities without.
This should be fairly obvious, but the evidence described in the third paragraph doesn’t exactly bolster Goldy’s assertion. It’s proof that there are a number of other factors causing declines in vehicle fatalities other than what’s happening at red light camera intersections. These could be related to safer car construction, fewer miles traveled, changes to traffic patterns, or something else. If there’s a reduction of 14 percent in red light crashes in cities that didn’t implement red light cameras, then there are another explanations for the decline. And that explanation could perhaps also explain why there was a gap in the overall statistics from city to city.
Here’s a page from the National Motorists Association that criticizes other aspects of the study, and another page from them that details out some studies which have shown that red light cameras increase accidents.
The National Motorists Association is an organization with a strong bias in this matter, and they often play up the increase in rear-end collisions that are seen with the implementation of red light cameras, while ignoring the decreases in side-angle crashes (which are more likely to cause fatalities) from the very same studies. In the end, I think there’s a case to be made that red light cameras provide some benefit, although I find this study to be completely unconvincing in the effort of making that case. In fact, this part at the end of their press release gives you an idea of how little their numbers are actually telling them and how they understand them even less:
Results in each of the 14 camera cities varied. The biggest drop in the rate of fatal red light running crashes came in Chandler, Ariz., where the decline was 79 percent. Two cities, Raleigh, NC, and Bakersfield, Calif., experienced an increase.
“We don’t know exactly why the data from Raleigh and Bakersfield didn’t line up with what we found elsewhere,” McCartt says. “Both cities have expanded geographically over the past two decades, and that probably has a lot to do with it.”
But Chandler has easily been one of the fastest growing cities in the United States over the past 20 years as well. Why did it experience such a dramatic decrease in vehicle fatalities while Bakersfield and Raleigh saw increases? There are certainly reasons for it, but it should be evident that red light cameras aren’t one of those reasons. When looking at red light cameras and trying to figure out whether or not they work, any study that isn’t looking at specific intersections and comparing data isn’t really worth much in this debate.