[cwyers]

So, that's an open question -- do the sorts of infractions that red light cameras catch make roads less safe? Does the installation of red light cameras make roads more safe? As it turns out, no. The installation of red light cameras in Chicago has actually increased rear-end collisions that result in injury by 22%.[1]

And that's what I mean by moral hazard. Revenue generation by these means changes the city's incentive. Rather than having traffic laws and enforcement designed to increase safety, you have laws and enforcement designed to turn the greatest part of your city's population into offenders as possible, in order to maximize your revenue generation, at the expense of public safety.

And it's an injustice because the well-off simply pay their fines and move on, while the poor end up trapped in a neverending cycle of fines and punishments, because the inability to pay the first fine leads to a cascade of involvement with a court system pressured to produce revenue, not justice or fairness.

1) https://arstechnica.com/tech-policy/2014/12/major-chicago-st...

[wcummings]

>As it turns out, no. The installation of red light cameras in Chicago has actually increased rear-end collisions that result in injury by 22%.

You're misrepresenting the results of the study. The study recommended the continuation of the program. The rear-end collisions were less severe than types of collisions it reduced, it was a safety improvement overall.

From the findings section of the actual study (emphasis mine):

>Quantitative studies conducted in this project demonstrate significant safety benefits of the current RLC program. As a result, it is appropriate to recommend continuation of the program. Most of the intersections have experienced an improvement in safety, particularly in terms of severe angle and turn crashes, albeit with an accompanying increase in less severe rear-end crashes. The safety benefits extend beyond the immediate vicinity of the RLC intersections, evidenced by a significant spillover effect. However, some intersections appear to experience no significant safety impact. Recognizing that crashes are the result of complex interactions amongst many factors, and subject to considerable randomness, these deviations should be used as opportunities for detailed investigation and learning to design and deploy more effective automated enforcement programs

http://www.transportation.northwestern.edu/docs/research/RLC...

Similar benefits have been demonstrated in other cities. Doesn't mean it will work in every city, but there is evidence it works in Chicago.

[cwyers]

If you want to accuse me of "misrepresenting" things, you should have your facts right. The study I was quoting that finding from was conducted by Texas A&M at the behest of the Chicago Tribune. The study you're quoting from was conducted by Northwestern University for the Chicago Department of Transportation.

[wcummings]

Fair enough, I incorrectly assumed you were citing the one released last week, not 3 years ago. Either way, increased collisions (or one type of collision) doesn't necessarily mean decreased safety.

[cwyers]

I don't know if I consider "fair enough" and a backhanded insult to really be an apology for unfairly assuming maliciousness on my part, but okay.

I would perhaps phrase that as "citing the study released by an impartial third-party, not by the people who are engaging in the moral hazard and trying to justify their behavior" instead.