In this ongoing debate, a new study finds that as red-light cameras go dark, red light running goes up — immediately.
Hundreds of American cities use red-light cameras to police their intersections, but a few major ones have recently had second thoughts about the devices. Last year Houston and Los Angeles turned off their cameras, and earlier this year a New Jersey state senator introduced legislation to ban them throughout the entire state. Sen. Michael Doherty believes the cameras are less about safety than city revenue:
"It's clear that towns have little interest in actually making intersections safer. They want violations to occur so they can continue collecting fines to prop up government spending. It’s an easy way for elected officials to justify bloated local budgets."
The red-light camera question often turns on the safety question, partly because research on the topic has been inconclusive to date. Some studies have found a solid connection between red-light cameras and a reduction in car crashes; last year the Texas Transportation Institute reported an 11 percent drop. Others, like a 2005 report from the Department of Transportation, have found a potential increase in rear-end collisions.
Add in the tendency for cities to sign contracts with camera distributors that encourage more tickets, and you've got the makings of a major debate. Resolving the matter is critical: there's a big difference between cities doing something to improve safety that just happens to bring in money, and doing something to bring in money that just might improve safety.
Recently a team of psychologists from Old Dominion University, led by Bryan E. Porter, had a rare chance to study how drivers adjust their behavior in the presence (or absence) of red-light cameras. In 2005, the Virginia legislature allowed the state law permitting the cameras to expire. (The law has since been reinstated.) Porter and colleagues, who had been studying red-light cameras in the state anyway, designed a new set of observations at intersections where the cameras were set to go dark.
The researchers focused on eight intersections in southeast Virginia: four in Virginia Beach with red-light cameras that would expire with the law, and four others (two in that city and two in Newport News) where cameras didn't exist. They recorded the light status of the final car to cross through an intersection — documenting only those cars that went straight — with a particular focus on the period just before and after the cameras turned off.
In nearly 2,800 light cycles, about a quarter of all last cars to enter the intersection went through on green, and 63 percent on yellow. The remaining 12 percent crossed on red — but when the cameras were still on, that rate was only 3 percent. (At intersections that never had cameras, the last-driver-through crossed on red 14 to 15 percent of the time.)
That finding alone wasn't terribly surprising: when punishment for a behavior goes from nearly certain to random at best, you expect the behavior to increase. What intrigued (and unsettled) the researchers was how quickly drivers reverted to red-light running form. In the immediately aftermath of the law's expiration, the risk of someone running a red light at an intersection was three times higher than it had been when cameras were on.
A year later it was four times higher, with all risk reductions having been erased, the researchers report in an upcoming issue of the journal Accident Analysis & Prevention. Their chart (the solid line shows the before-and-after camera intersections):
That should be enough to get the attention of those public servants truly interested in road safety. According to recent figures from the Federal Highway Administration, running red lights led to 676 fatalities in 2009, and has been implicated in as many as a thousand driver deaths since 2000. If cameras are what it takes to keep light-running rates low, then cameras might be worth the complaints over tickets. Porter and colleagues conclude:
Our findings suggest red light running reductions are likely to recidivate quickly, and certainly within a year, once cameras go dark. While more research is important, our findings should give pause to those wishing to remove the technology without significant scientific reasoning for doing so.