Tuesday, February 15, 2005

Opposing red-light cameras

This morning at 8 a.m., I'll be at the Texas House Urban Affairs Committee to testify in support of Rep. Gary Elkins HB 259, which disallows the use of surveillance cameras at Texas intersections for traffic enforcement. See here for earlier Grits coverage of the upcoming hearing. I thought folks might want to see the text of the ACLU of Texas fact sheet I walked around to committee members yesterday, with a few hyperlinks added where I had them to hand.

UPDATE: HB 259 was voted favorably out of committee this afternoon.

Support HB 259 (Elkins) Banning Red Light Cameras
Red light cameras are unfair, they increase injury accidents,
soak the taxpayers, and pave the way for privacy abuses.

Red Light Cameras Increase Injury Accidents
A study released in January of all seven red light camera systems in Virginia found that the number of injury accidents has increased since they were installed.

The New York Times reported last month that "rear-end accidents have shot up at intersections with cameras."

Revenue Generation is Real Motive

Houston officials say they would accept more cash from the Legislature in lieu of red light cameras. The New York Times reported last month: "there has been criticism of the cameras' use to generate revenue from fines … and of revenue sharing arrangements with providers of the technology. Those arrangements, critics contend, have led to the placement of cameras not necessarily where they would promote safety, but where they will rack up the most violations."

Picture Snapped Doesn't Mean Guilt

Red light cameras give the ticket to the wrong person unless the owner of the car is driving. If someone driving a rental car runs a red light, will the rental company pay? If not, why not?

Even if a picture is taken, camera systems may target the wrong people. Houston's EZ-Pass system last year was capturing snapshots of vehicles that failed to pay their tolls, but the Houston Press reported they sometimes sent tickets to the wrong people.

Surveillance Cameras Are Prone To Abuse

Many camera systems zoom in and out and can rotate to view sidewalks or even inside neighboring buildings. Traffic cameras are designed to take high-definition photographs of license plates from a moving vehicle. That means they can photograph inside cars, too, if the operators choose to do so.

In 2003, in response to 9/11, the Texas Legislature made all information about location, specifications and operating procedures from surveillance cameras secret. ACLU believes that bill went too far, since now no one can know what video data the government gathers about them or what is being done with it. It would be unwise to massively expand the scope of government surveillance without fixing that statute to allow greater public accountability. There have been many examples of abuses:

In China, cameras installed ostensibly for traffic enforcement were used to identify and persecute dissidents after the Tiananmen Square uprising.

1 in 10 women were targeted for voyeuristic reasons by male camera operators in a study of London's infamous surveillance camera system.


Ken said...

One time at an intersection I had to run a red light due to a fire truck that was coming from behind. I was in front and it looked like it had to get thru. So I looked both ways and went thru. It turns out that the firetruck turned right instead. The intersection had some type of camera. Not sure of its purpose. But I was nervous. So I can understand the concern over red-light cameras.

Judson Bryant said...

In supporting your position against red light cameras,you cited a study that was done at North Carolina A & T University.
This study was critiqued in the recently released report from the Institute for Transportation Research and Education at North Carolina State Uiversity. The name of the NC Staste report is "Evaluating the Use of Red Light Running Photographic Enforcement Using Collisions and Red Light Running Violations", December 2004. I suggest that this report be carefully read before quoting statistics from the NC A&T report because it states "the model did not appear to fit the data well". In addition it states "Comparisons were made between RLC (red light camera) sites and all other intersection in this data set, with no regard to how the treated RLC sites were chosen......No tests were done to see if the large comparison group acted in a similar manner to the RLC sites."
The overwhelming majortity of rigorous studies done on RLC confirm their effectivness at both reducing the total number of crahes and the severity of those crashes.
Judson Bryant

Gritsforbreakfast said...

Thanks Judson, but I can't find the North Carolina A&T study reference in my posts, and don't recall seeing it. I'd appreciate you pointing me to where I mentioned it.

By contrast, I've not heard of anybody disputing either the Virginia studies, nor the studies cited in the NYT piece referenced above. Those analyses, most clearly in the 7-city VA study, found that while side-impact collisions declined some, rear-end collisions increased even more, resulting in more injury accidents overall.

I'll readily grant, I've not examined the methodologies of any of these studies and can't vouch for them. But neither am I cherrypicking. The findings I've seen reported, at least for independent academic stuff (I've seen some frankly unbelievable claims in consultant's studies funded by the industry), the results seem pretty consistent - the total number of crashes increases, in some cases the number of injury accidents increases, while the argument that more severe crashes are avoided does not play out as dramatically as proponents predict.

As I told the House Urban Affairs committee, bottom line, if the ambulance has to make more trips to red light intersections for injury accidents because of cameras, there's no public policy argument for doing RLCs except revenue.

Judson Bryant said...

I should have used the words "linked to" instead of "cited" when writing about the North Carolina report. You referred to "red light camera systems in Virginia" which took me to a Newspaper.com article entitled "Virginia Study Shows Cameras Increase Injury Accidents". This article about the Va. report stated "This study agrees with long-term findings in Australia and North Carolina". One of the true ironies of this reference to the NC study in the article about the Va. study is that the Va. study includes a very detailed critique of the NC study. I quote a rather technical part of the critique "The pseudo Rsquared values are not comparable to traditional Rsquared values from a linear regression model, and literature on this topic does not give guidance as to what is an appropriate for accepting the model(Green, 2000). Given the fact that these are close to 0.00 (with 1.00 being a perfect fit), Table 13 suggests the models may not be a good fit to the data". The NC report is titled: A Detailed Investigation Of Crash Risk Reduction Resulting From Red Light Cameras In Small Urban Areas. Both the Va. report and the NC State report raise methodological questions about this report. In view of these criticisms, I would urge caution in using the conclusions of this study to guide the development of public policy.
While the Va. study documented an increase rear-end accidents, the NC State University reached a different conclusion:
"In an effort to analyze the effect of RLCs on driver behaviour, two types of analyses were completed. The first type was a before-after collision study. The following three types of improved before-after collision studies were used: accounting for causal factors, a comparison group analysis, and improved comparison group analysis accounting for the halo effect. Each of these studies analyzed four categories of collisions including total, red light running related, angle and rear end. Based on the comparison group study collisions were effectively reduced by 17%,22%, 42% and 25%,respectively.".
With study result differing and with experts differing on the effectiveness of red light cameras, what should the legislature do? Allow cties to operate limited red light cameras programs while closely monitoring both the known positive attributes and the possible adverse unintended consequences. There is no question that RLCs reduce red light running. This lends itself to a more orderly society-less chaos on the public roads. Maintaining this order is one of the prime responsibilities of government.

Anonymous said...

All i have to say is they used to post austin traffic cams for viewing then i had a hard time finding them online. Now they are no longer there. Other texas cities still post them for viewing. And how/why do the local news stations have the feed from them to show us but we can't have it? how are they qualified to see them at their own leisure?