Ruminations On Red-Light Cameras

On Saturday, I wrote about Montgomery expanding their red-light camera program to include more intersections. I received an email response to that post, essentially telling me off. Apparently, I presented no detailed arguments against the cameras because my arguments are weak and because I "...prefer running red lights and speeding without consequence and use the "it's just a revenue stream" as your main argument without any recognition of the lives that are saved."

After a little back and forth, I decided to go ahead and post my arguments against the RLC's, as well as some of the data that supports my position. I won't post all, because there are a LOT of studies that I could use.

To start, let's settle on what we're talking about. Proponents of the cameras, mostly city officials or industry groups, hide the revenue stream behind claims that the cameras are a safety issue. The argument goes that the cameras reduce the number of accidents, and therefore save lives. But do they?


The most severe kinds of crashes are "t-bone" crashes, where one car broadsides another. Some studies have shown cameras in some areas reduce the number of those crashes, while other studies have shown the reverse. The evidence is mixed, at best.

2003 study by the Transportation Research Board: "Although nearly every study and crash analysis performed, as discussed in chapters two and three, has had some experimental design or analysis flaw or deficiency, there is considerable “evidence” that RLR cameras do have 39 an overall positive effect."

2007 study for the Virginia Transportation Research Council: "Consistent with the findings of a previous Virginia study (Garber et al., 2005), this study finds that cameras are associated with an increase in rear-end crashes (about 27% or 42% depending on the statistical method used as shown in Tables ES1 and H1) and a decrease in red light running crashes (about 8% or 42% depending on the statistical method used as shown in Tables ES1 and H2). When these results are aggregated across all six jurisdictions, the cameras are associated with a net increase in comprehensive crash costs."

As you can see, even when the t-bone crashes are reduced, the rear-end crashes are increased. Indeed, though studies are mixed as to whether RLC's reduce t-bone crashes, they almost universally agree that the cameras cause an INCREASE in rear-end collisions.

In some areas, the cameras have been effective. In others, not. So, sorry to tell you rabid supporters of the cameras, but their effectiveness is debatable at best.

If I'm right about the revenue being the REAL reason why cities love the cameras, we'd expect to see efforts made by cities to artificially increase the number of tickets issued. And, if we look, we see EXACTLY that.

In Texas, at least one city tried to do this by redefining what "running a red light" meant. Basically, they redefined it so that an RLC ticket could be issued if a vehicle was already in the intersection when the light turned red. MANY fraudulent tickets were issued, and the city experienced significant legal difficulties.

More commonly, we see cities intentionally shortening the duration of yellow lights in order to increase the number of people ticketed for running the red lights. This practice, where not specifically prohibited by state law, is so common as to be termed widespread. The fact that cities would take steps to INCREASE the number of red-light violations merely for the money indicates pretty strongly that safety is just a red herring, and not the real purpose of the cameras at all.

But the question we NOW have before us is this: if RLC's don't work to reduce the number of people running red lights, what CAN we do? The answer to that is as simple as it is time-tested. If yellow lights are lengthened slightly, the number of people who run the red lights drops by up to 80%. Why is this? Because most of the people who run red lights do so by less than a second. By lengthening the yellow light, those violations disappear, and the safety issue is moot.

Georgia tried this by passing a law REQUIRING all intersections with RLC's to have their yellow light set to be one-second longer than the "minimum duration" required by their safety codes. When they did this, cities saw the number of citations drop by 80%. If reducing the number of red-light runners improves safety and reduces the number of accidents, doesn't this mean that lengthening the duration of yellow lights is MORE effective than RLC's?

Yes, it does.

Indeed, some places have noticed that lengthening yellow lights have reduced the number of violations so much that maintaining the RLC's is no longer profitable.

If you want statistical evidence on the longer-yellow-lights thing, here's one from the Texas Transportation Institute.

An interesting tidbit from that study is that most RLC violations occur within half a second of the light turning red, while most t-bone collisions occur from violations A FULL FIVE SECONDS AFTER the light turns red. In other words, most of the tickets issued wouldn't have resulted in t-bone crashes anyway, i.e. no safety issue at all.

I'm just scratching the surface of all the evidence against RLC's, here, and can easily provide more if you really need me to do so. But I think that most people get the point.

So, I oppose red-light cameras because they don't decrease accidents, and because there are simpler, more effective ways of achieving the desired result WITHOUT giving the city a massive revenue stream. Montgomery shouldn't be EXPANDING their RLC program, they should be phasing it out and replacing it with alternatives that work better.