According to a study completed last month and released to The Lens this week, 59 camera sites had 21 percent fewer crashes than would have been expected without them.
The analysis by the Office of Performance and Accountability began in March, at the tail end of The Lens’ probe into the effectiveness of the cameras. We reported that the city had never studied how the cameras affect crashes, injuries or deaths.
The new study was done with a tool, provided by the nonprofit group DataKind, “to rigorously evaluate the impact of any traffic safety interventions,” Erin Burns, Mayor Mitch Landrieu’s spokeswoman, wrote in an email.
The study is the first of “an expanding effort to take a more data-driven approach to traffic safety policy,” she wrote.
Crashes increased citywide during the study period, 2005 to 2015. But 76 percent of the sites with cameras had a smaller increase in the number of crashes than comparison sites without them.
And 54 percent of camera locations saw a decrease in crash rates, compared to 9 percent of the comparison sites.
Traffic cameras have generated about $16 million annually for the city, with about 30 percent going to the company that runs them, American Traffic Solutions. Some critics have called the program a money-grab.
The cameras have become an issue in this year’s mayoral campaign, with one candidate, City Councilwoman LaToya Cantrell, promising to stop their use until a comprehensive study shows they make roads safer.
We sent the study to Cantrell to see if it satisfies her concerns. Not quite.
“While we need to study the analysis further, I do have some concerns about the data,” Cantrell said in a written statement provided by spokesman David Winkler-Schmit.
She said she wants more information on individual cameras. “That way we can determine which cameras, if any, can stay and which ones should be removed,” Cantrell said.
The document provided by the city includes a single-page chart of results by location, but it doesn’t provide further details on each one.
Assertions of safety, but no evidence
Last year Landrieu pushed the city council to add 55 new cameras, including 10 units mounted in cars.City officials repeatedly asserted the cameras make New Orleans roads safer. But before the new study, the city had not measured the program’s effect on crashes.
In March, as city employees began work on the new cameras, The Lens reported that the city could provide no evidence that the program, which went online in 2008, had reduced crashes, injuries or deaths.
The city’s camera contractor did not perform crash studies at proposed locations before installing them, even though its contract requires them.
The lack of data didn’t get in the way of the rhetoric.
“We do not behave well behind the wheel, and as a consequence, we threaten each other’s lives,”Landrieu told the editorial board of NOLA.com/The Times-Picayune last year. “You don’t need a study to see that in New Orleans.”
Our repeated requests for evidence yielded one study that dealt with crashes. It did not compare crashes before and after cameras were installed, which is how traffic experts say you would learn whether they had any effect on safety.
When we told the city that its records didn’t back up their safety claims, Burns countered that they did show less speeding, which in turn reduced the “likelihood of crashes.”
Police chief says cameras make roads safer
Early last week, during a press conference announcing the beginning of camera enforcement in school zones, New Orleans Police Chief Michael Harrison again touted the safety benefits of the cameras.“An objective analysis of the traffic camera program has found strong evidence that cameras have reduced the incidents of crashes at those locations,” Harrison said.
So we again asked the city for evidence. This time, they had it.
The study compared wrecks at locations with traffic cameras to similar locations without them. It used crash data compiled by the state, which are based on local police accident reports.
The city didn’t provide the underlying data because the state requires they be kept confidential.
The study looked at cameras installed before 2013. Among the 45 sites that fared well, the results varied significantly.
But on average, crashes at camera locations increased by just 1 percent from 2005 to 2015, compared to 24 percent at “control sites” without cameras.
The study concludes there’s “strong evidence that traffic cameras installed in New Orleans from 2008 to 2012 did have a beneficial effect in terms of reducing crashes at camera locations.”
That wasn’t true at all sites. Thirteen of the 59 sites surveyed showed greater increases in wrecks than similar locations without cameras.
The worst locations were the intersection of Harrison Avenue and Marconi Drive, and a school zone on Camp Street at the International School of Louisiana.
The Lens sent the analysis to Helmut Schneider, director of the Louisiana Transportation Research Center at LSU. He said the city appears to have done solid work, but some key information is missing.
Since the conclusions are based on a comparison to similar sites without cameras, a lot depends on how those sites were picked.
So it “all depends on whether the matched locations were done correctly,” he wrote in an email.
The document the city provided explains that comparison sites were chosen based on the number of crashes and injuries as well as crash severity. Comparison sites had to be between 2,000 feet and five miles from a camera.
Schneider said the Office of Performance and Accountability should also have taken other factors into account, such as traffic volume and the type of street.
“Not every intersection is the same. Is it a four-lane, a two-lane, divided, not divided?” he said.
In a written statement, Office of Performance and Accountability Director Oliver Wise said the city did not have access to complete traffic data for the sites.
He said the office considered using types of roads to determine the control sites, but decided crash, injury and severity data would provide better comparisons. Among other reasons, Wise said, the crash data provided “a proxy for traffic volume that would otherwise have been unavailable.”
No clear trend in other studies
Nationally, studies on the effectiveness of traffic cameras have varied. A study of 14 cities by the Insurance Institute for Highway Safety found red-light cameras reduced fatal red-light running crashes by 24 percent.A number of studies, including one in 2005 by the Federal Highway Administration, have concluded that traffic cameras posted at stoplights result in decreased “T-bone” wrecks that often result from people running red lights. But rear-end wrecks, which are usually less severe, increased.
In a report released last month, the National Transportation Safety Board concluded, based on a number of national studies, that automated traffic enforcement programs are an “effective countermeasure to reduce speeding-related crashes, fatalities, and injuries.”
But a report by the Chicago Tribune found that red-light cameras appeared to have caused more injury-causing accidents at intersections where there had been few crashes.
As for speed cameras not posted at stoplights, the Insurance Institute for Highway Safety found they led to lower rates of speeding, and as a result a reduced likelihood of crashes. But a Phoenix study found they had no statistical effect on crashes.
This story was originally published by
The Lens , a nonprofit publication focused on New Orleans and the Gulf Coast.
