Police traffic enforcement is not a recoupable cost


Even if you changed the speed limit to a very reasonable 20 MPH, it would still need enforcement. A campaign in New York City and the UK is called “twenty’s plenty”. Photo by ddartley. 

Just throwing this out there… There’s a lot of talk about speed cameras (and red light cameras) being a “money grab”, that it’s about making money more than improving street safety. (See my comprehensive review of what it means for Chicago that Governor Quinn signed legislation authorizing the City to install speed cameras.)

Let’s say you’re the administrator of a city that needs to do two things: 1) make streets safe to walk and cycle on, 2) receive revenue for needed services (like crime investigation, libraries, or road resurfacing).

To make streets safer, you can reduce the incidence of speeding, and reduce high speed traffic. You could make this happen in several ways with one or more of these strategies:

  1. Redesign the streets. Add curb extensions, landscaping, raised crosswalks, road diets, and more traffic calming techniques.
  2. Use human police officers to enforce existing laws. Possibly reduce the speed limit and enforce that as well.
  3. Use cameras to issue citations to people who speed in their automobiles.

Each one of these has a cost of doing business. For (1) redesign the streets, you have the cost of installation and maintenance.  For (2) human police, you have the cost of paying police officers their wages and pensions (and their cars and other materials). For (3) using cameras, you have the cost of installing and operating cameras, and the systems to verify violations and issue citations. For both (2) and (3), you have the cost of courts that deal with the traffic violations (mainly for those who appeal their citation). There are other costs for each of these I’m not listing.

As the city administrator, you’ve got to cover these costs. Which one will pay for itself? Using speed cameras. And it will probably have a lot of money left over after covering the installation and operation.

I posit that using speed cameras has the highest calculable return on investment. The benefits realized by (1) and (2) are about having safer streets, meaning fewer crashes and fewer injuries. And (2), enforcement by police officers, returns revenue via tickets to drivers – but does this revenue cover the enormous cost of wages, pensions, and operations? The return is probably much lower than the rate of a machine that works tirelessly from 5 AM to 11 PM near parks, and from 6 AM to 10 PM near schools.

If the city administrator chooses to reduce speeding with strategy (3), have they thrown in the towel on using street redesign (which also has hard-to-measure quality of life benefits) or police officers?

It may be that potential revenues are a motivation to install speed enforcement systems, but might it also be that speed cameras have shown to reduce the incidence of speeding, and that speeding is related to injury severity in crashes, and that speeding is related to causing crashes?

Transportation planning is a major arena for doing cost-benefit analysis, but determining the costs and benefits, and adjusting them so they are comparable and on the same scale, is not a very transparent or discussed process.

10 thoughts on “Police traffic enforcement is not a recoupable cost”

  1. Enforcement by police officers has other benefits that aren’t considered here, many of which have nothing to do with traffic safety.  If an officer makes a traffic stop, it is routine to do a computer lookup on the license plate and driver’s license.  This screening sometimes catches people driving stolen cars or driving with suspended or revoked licenses.  It also catches people with outstanding warrants for many types of crimes.  The officer may observe something unusual about the car and its occupants, then get a radio call about suspects from a recent crime.  The driver may have been driving erratically and reek of alcohol while talking to the officer, something a speed camera would never detect.

    Speed cameras may offer more bang for the back if you’re just looking at speed enforcement.  If you’re looking at overall public safety, those cameras miss a lot.

    1. Thank you for catching what I missed. In no way should my omissions be considered a comment about the value of police work. I think there’s an overwhelming opinion that speed cameras are essentially a ploy for money and I am sort of playing devil’s advocate here, exploring that “ploy”. 
      That being said, based on the Tribune’s analysis, and on the discussions on the Second City Cop blog (which is hard to take as a primary information source because everyone is anonymous, but that information is very unique), traffic stops have decreased. And it’s known that the police force has gotten smaller. 

      I wonder what the effectiveness of blue light cameras is in changing public safety (for better or for worse). 

  2. I think ANY form of speed enforcement has indirect cost benefits. If there is less speeding, there are fewer crashes, and the crashes that do occur are less severe. Over time, this drives down insurance rates, providing a savings to anybody who drives in Chicago.

    1. I’d love to see a strategy like this:

      1. Let’s make our streets safer! Kids are getting hurt, well, everyone’s getting hurt! Let’s stop that!
      2. Hey, we’re going to have a multi-prong approach to reduce speeding, one of the major causes of crashes and injuries. Here’s what we’ll do: We’re going to hire new police officers, retrain existing ones, and focus on these three traffic violations. We’re also going to redesign 1,000 miles of neighborhood streets. We’re also going to test a “twenty’s plenty” project in 20 neighborhoods – your neighborhood can apply to be one of the first 20! Lastly, we’re going to set up some speed cameras where the number of crashes is highest, and the rate of severe crashes is highest. 
      3. 6 months and 12 months later, we’ll do a really good study evaluating the effectiveness of each strategy!
      4. How about it!

      1. Have you ever heard of an OODA loop? It stands for Observe, Orient, Decide, Act. It was developed by a man named John Boyd for dogfighting aircraft but it has all kinds of uses for any kind of decision making. The big idea is that you are always in the loop someplace and it repeats whether you want it to or not. It’s a bit like continuous improvement but it has a lot more rigor behind it from a logical perspective. Basically the more iterations you have of trying to solve a problem, the better your solution will be at any given moment. Software companies use it for fixing bugs and you can see OODA in action with the daily updates you get to your antivirus program or OS. Government planning could operate in the same away (outside of combat) but too many people in government tie their fate to subjective perception rather than objective outcomes.

          1. It sounds valuing data is a subset of the “observe” aspect in an OODA loop. I think what is lacking in a lot of decision making is the “orient” part of the loop. You can generate a ton of data and then figure out how to make something 5% more efficient but unless you keep asking why you are making something 5% more efficient in the first place you can get lost. OODA is also meant to describe competitive environments that are in flux, which is most things in government.

            If you can figure out where somebody is making their observations and decisions in the process you can be a more effective advocate by trying to influence those steps instead of just assuming that data or objective data will carry the day.


  3. Hi Steve, your experiment is an excellent idea, but modify it slightly and it can become a scientific trial. Then it would be the 1st ever scientific trial of cameras, of 20 limits and of road redesign (at least as far as I know).

    eg find 20 neighborhoods suitable for 20mph and select them into pairs. Then randomly select 1 from each pair to be 20mph, the others to be left as is.

    Same with cameras. Find as many locations as possible, pair them, and then randomly select 1 from each pair to to have a camera, the other left alone.

    You’d have to be careful that traffic doesn’t avoid the cameras, thereby confounding the results, but expect you, and everyone else, might be very surprised by the results!

    See: http://speedcamerareport.co.uk/02_scientific_trials.htm

Leave a Reply

Your email address will not be published. Required fields are marked *