How effective are red light cameras?

MSNBC posted an article about red light cameras, describing how many cities are taking them down. Why? Because the cameras are too effective: people learn where the cameras are and stop running the red lights. According to the article, the city of Dallas recently turned off a quarter of its red light cameras because it couldn’t justify the cost of running them. Other cities have had similar experiences.

Not surprisingly, the cameras’ effectiveness at reducing accidents is inconclusive. According to a study released in 2005 by the Federal Highway Administration, intersections with red light cameras get about 15% more rear-end crashes than they would have without the cameras. It seems as though drivers know that the cameras are there and will brake aggressively to avoid running the light. However, the study also reports a 25% decrease in T-bone crashes, with about 16% fewer injuries. Overall, the study reports no appreciable difference in the number of accidents, and an almost 5% decrease in injuries.

All told, the FHA study concluded that red light cameras give, at best, a “modest aggregate crash-cost benefit.”

And then there are opponents’ claims that, even if the cameras provided a huge safety benefit, that still doesn’t justify the systematic violation of drivers’ constitutional rights. I tend to lean towards this belief myself.

However, it’s rather amusing that cities are finding it increasingly difficult to justify the cost of red light cameras because drivers are running fewer red lights. One wonders if, were the cameras to provide a demonstrable safety benefit, whether cities would continue operating them at a loss. It’s probably a good thing for city public relations that the negligible safety benefit means that they can turn the cameras off without a loud public outcry.