The District's red-light cameras have generated more than 500,000 violations and $32 million in fines over the past six years. City officials credit them with making busy roads safer.
But a Washington Post analysis of crash statistics shows that the number of accidents has gone up at intersections with the cameras. The increase is the same or worse than at traffic signals without the devices.
Three outside traffic specialists independently reviewed the data and said they were surprised by the results. Their conclusion: The cameras do not appear to be making any difference in preventing injuries or collisions.
"The data are very clear," said Dick Raub, a traffic consultant and a former senior researcher at Northwestern University's Center for Public Safety. "They are not performing any better than intersections without cameras."
The District started the camera program in 1999, and from the beginning, officials said they were aiming to curtail red-light running and accidents. At the time, Terrance W. Gainer, then the second-highest ranking D.C. police official, said the cameras would "get people to stop at red lights and avoid crashes. . . . Hopefully, we'll have a few less messes to clean up."
D.C. Police Chief Charles H. Ramsey said he remains convinced that the devices are worthwhile. Even if the number of crashes is not going down, he said, citations for red-light running have dropped by about 60 percent at intersections that have cameras.
Ramsey said the number of accidents would be even higher without the cameras, adding that he would like to install them at every traffic light in the city. He pointed to last year's steep decrease in traffic fatalities -- 45 people died compared with 69 in 2003 -- as evidence that the program is working.
"I'd rather have them than not have them," Ramsey said. "They make people slow down. They reduce the number of traffic violations, and that's a good thing."
City officials attribute the increase in accidents to higher traffic volume. But that does not explain why the presence of cameras has failed to slow the rate of accidents at those intersections, Raub and others said. The outside experts suggested that the cameras might be more useful at other locations, and D.C. officials said they are studying the issue.
The city has cameras at 45 intersections. They take photographs of cars running red lights, generating tickets that are processed by a private contractor. Police oversee the issuance of tickets, which carry $75 fines, and the money goes into the city's general fund -- nearly $5 million last year.
The Post obtained a D.C. database generated from accident reports filed by police. The data covered the entire city, including the 37 intersections where cameras were installed in 1999 and 2000.
The analysis shows that the number of crashes at locations with cameras more than doubled, from 365 collisions in 1998 to 755 last year. Injury and fatal crashes climbed 81 percent, from 144 such wrecks to 262. Broadside crashes, also known as right-angle or T-bone collisions, rose 30 percent, from 81 to 106 during that time frame. Traffic specialists say broadside collisions are especially dangerous because the sides are the most vulnerable areas of cars.
The number of crashes and injury collisions at intersections with cameras rose steadily through 2001, then dipped through 2003 before spiking again last year.
The results were similar or worse than figures at intersections that have traffic signals but no cameras. The number of overall crashes at those 1,520 locations increased 64 percent; injury and fatal crashes rose 54 percent; and broadside collisions rose 17 percent.
Overall, total crashes in the city rose 61 percent, from 11,333 in 1998 to 18,250 last year.
Lon Anderson, a spokesman for AAA Mid-Atlantic, said the data reinforce the motor club's view that the red-light effort is targeted more at generating revenue than at reducing crashes. "They are making a heck of a lot of money, and they are picking the motorists' pockets on the pretense of safety," he said.
Red-light cameras are used in 12 states, including Maryland, where they are deployed in Montgomery and Prince George's counties. In Virginia, the General Assembly eliminated red-light cameras this year partly because of concerns raised by some legislators about civil liberties. The action affected six Northern Virginia jurisdictions: Alexandria, Arlington County, Fairfax City, Fairfax County, Falls Church and Vienna.
The District installed its first batch of 26 cameras in 1999. City officials added 14 the next year. Some intersections have more than one camera to cover different approaches. All told, the cameras installed in 1999 and 2000 covered 38 intersections; a camera subsequently was removed from one of them.
Ramsey said city officials put the cameras where police noticed the most red-light running. At the start of the program, police officials said they also received advice on camera placement from residents and from the private contractor that operated the devices.
Nine more cameras were installed in July, boosting the number of monitored intersections to 45. Most of those drivers ticketed come from outside the city. In August, for example, less than one-fourth of the citations were issued to motorists from the District.
D.C. police also operate photo-radar devices that take pictures of speeding motorists. Because many of these cameras are mobile and used at varying times, they were not included in The Post's review.
Douglas Noble, the chief traffic engineer for the D.C. Department of Transportation, said his office was examining crash data and plans to review the red-light camera locations. The department collects the data from police reports and advises police about where to install the devices.
Noble said that no studies have been conducted on the District's red-light cameras in several years but that he "would not disagree" with The Post's analysis. "I don't necessarily have an explanation" for the trends, he said.
He added that he believes the severity of injury crashes has decreased at camera locations. The city crash database does not categorize the severity of crashes.
AAA and other critics have accused the city of installing cameras in high-volume locations where they could generate thousands of tickets, regardless of how many accidents happened there.
The analysis raised questions about where police installed the cameras. Nine intersections with cameras had two or fewer crashes annually in 1998 and 1999; seven reported no crashes that led to injuries or fatalities during that period. Officials installed cameras at six of the 20 most crash-prone intersections in 1998, data show.
Seventeen of the 45 intersections now covered by red-light cameras were ranked among the 50 most accident-prone locations in the District last year.
Individual results at intersections vary, the analysis shows.
The camera at New York Avenue and Fourth Street NW, for example -- on one of Washington's busiest commuter routes -- has generated the most tickets in the city: more than 150,000 since 1999. Although the number of monthly citations there has dropped 65 percent, crashes nearly doubled, from 12 in 1998 to 23 last year.
The number of crashes has decreased in recent years at another busy spot, Bladensburg Road and New York Avenue NE, where cameras have generated more than 73,000 tickets. The intersection had 35 crashes in 1998, 88 in 2001 and 71 last year.
The camera at Wisconsin Avenue and Brandywine Street NW has produced nearly 30,000 tickets, but its crash totals have hovered around two a year.
Advocates for the cameras point to research such as a recent national study by the Federal Highway Administration that showed the number of broadside crashes dipped 25 percent at sites with cameras. The study found that rear-end crashes rose 15 percent at camera locations. But because broadside crashes are more dangerous and cause greater damage, the study concluded that the cameras can help reduce the costs of traffic accidents.
Gang-Len Chang, a professor of civil engineering at the University of Maryland, said cameras can be useful in reducing serious crashes if deployed properly.
Chang and the other traffic specialists said the city should not abandon red-light cameras. Rather, they said, the mixed results indicate that D.C. officials should conduct a thorough review of camera sites.
"They definitely should look at the locations and find where the cameras would be much more effective," said Nicholas J. Garber, a professor of civil engineering at the University of Virginia who studied the use of red-light cameras in Fairfax County.