Crashes Caused by Red Light Cameras Hidden Since May 2012 in St. Petersburg.

CORRECTION: The percentage increase in crashes was 20%, and the absolute number was 35 (A prior version reported otherwise).    redlightcameradatamayLast Spring, the City of St. Petersburg began crowing about how its revenue-producing “red light camera” program has reduced “red light related” and “red light rear end” crashes.  But they forgot to leave out one thing: the total number of crashes at these intersections went up!  The Tampa Bay Times today reported on data received from the city staff this week showing that red light camera crashes were up from the period up until October 2012 and that the city concealed that fact to the City Council! The Sun Beam Times now reports that this is the same behavior displayed in June, when key city staffers were reporting glowing results to the Times about “red light related” crashes going down in the first six months of operation.  The fact is that the primary purpose in the red light camera program has been to raise money, and a secondary goal has been to improve safety, which it has not done. In fact, the city is obviously more than willing to make the roads more dangerous to raise money;  they knew in May 2012 that the total number of crashes at red light camera intersections had gone up 20% in the first 6 months of the program!

Last May the City had its first six months of data on crashes at red light cameras (November 2011-May 2012).  Instead of offering an honest assessment on the actual increase in total crashes at these intersections, the City staff used deceptive tactics to mislead the City Council and the taxpayers.  They did this by reporting data on “red light related” crashes and also in comparing crashes to a prior 3 year average instead of compared to the prior year. Joe Kubicki, City Director of Traffic and Parking,  reported that the number of “red light related crashes” had fallen 45% and “red light rear end” crashes had fallen 60%.  However, they analyzed only a subset of crashes at these same intersection: red light “related” crashes.  There is a problem in reporting “red light related crashes”: it is not a precise measure and is subject to “fudging” based on the subjective opinions of the government agents that report it.  The person making the report gets to arbitrarily decide when a crash is “red light related” and categorize it to make their red light camera program look good. A 2003 study by the Texas Traffic Institute at Texas A&M University concluded that understanding the true number of red light “related” cases is “difficult at best” and “impossible at worst”:

“In short, an accurate identification of the true number of red-light-running-related crashes is difficult at best (as it may require a manual review of the crash report) and impossible at worst (should the original report be destroyed or not contain sufficient details of the crash). Unfortunately, many agencies currently report the frequency of “red-light-running-related crashes” but do not explain the method by which they identify these crashes. This practice makes it very difficult to confidently compare the crash data or countermeasure effectiveness that is reported by different agencies. Extrapolation of findings to other locations and combining measures of effectiveness is practically impossible without a consistently applied definition of a “red-light-running-related” crash.”

In addition the city chose to massage the data in their favor.  The data for the first six months of the Red light camera (RLC) program compared to the same period in the prior year showed a total increase in all crashes at the intersection of 20% (up from 178 to 213) (see chart above).  However, to minimize this effect, they chose to compare 2011-12 data to a three year average of data to make it appear the total crash rate was only up 3.8% from 205 (three year average) to 213. All time periods reported were from November to May beginning in 2008 to 2012.  By using the total crash average from 2008-2011, they could more easily hide the slight increase in total crashes and crow about an apparent large fall in a number that has been criticized as “impossible” to calculate: red light “related” crashes.  It should be noted that the number of crashes at other “high crash intersection” without red light cameras stayed exactly the same: 170 in the six month period prior to and after installation of the cameras. Thus it would be impossible to blame the 20% increase at RLC intersections on an overall increase in crashes in the city.  The most probable conclusion to draw from the data is that the placement of red light cameras is the direct cause of these new accidents.  This subjects the city and the taxpayers to legal liability since it will be quite easy to argue in a court of law that the red light camera was the cause of the accident* and **.

Several clear facts emerge from the red light camera debacle here in St. Petersburg:

  1. The City of St. Petersburg is far more interested in raising money than cutting spending.
  2. The City of St. Petersburg will do anything to raise money, even at the expense of harming public safety.
  3. The City will hide the truth from its citizens to achieve its goal of raising money.

The politicians in charge of the City of St. Petersburg are demonstrating that they cannot be trusted with the most basic and important functions of government: stewardship of city budget and public safety.  They have set raising money above all other priorities including the public safety.  How many people were harmed in those 35 new crashes that arguably never would have happened?   How many lawsuits are now occurring because of them and how are they affecting all our auto insurance rates? The City needs to immediately abandon the red light camera program and remove these cameras.  They need to apologize to the public and the Mayor should consider disciplinary action against city staffers that chose to hide this information from the public and the City Council. But make no mistake, the City Council is not blameless.  They approved the program.  They also did not look at or demand the data that was there. If a single and busy citizen (this reporter from the Sun Beam Times) can uncover this data in June, why couldn’t the individual City Council members do the same?


 *In the interest of full disclosure, this author has had a ticket at a red light camera for a safe right turn on red, after not coming to a complete stop at an empty intersection. This is one of the more common tickets issued.

** This author does testimony for such cases on plaintiff and defense sides of personal injury cases and knows the lawyers would welcome such facts.

facebooktwittergoogle_plusredditpinterestlinkedinmailfacebooktwittergoogle_plusredditpinterestlinkedinmail

2 Replies:

  1. James C Walker

    The city is not concerned with higher crash rates caused by the red light cameras. Red light cameras often increase the crash rates and the city was aware of that fact before they installed any cameras and before they considered adding more cameras for even more increased crashes. The ONLY concern of the city is the MONEY to be collected.

    The city is not concerned with the ethics of hiding the higher total crash rates from the public. If ethics played any part in the city’s decisions, they would probably never have installed the cameras to risk the higher crash rates. Now that the higher crashes are documented, proper ethics would demand that the city voluntarily end the program that has caused more crashes. But ethics play no part in these decisions, the ONLY item of concern to the city is more MONEY.

    St. Petersburg is also aware that simply adding one second to the yellow intervals would almost certainly drop the violation rates below the level needed to just pay the high camera costs. This added safety factor with many fewer violations would gut the revenue stream, so the chances the city would voluntarily improve safety with longer yellow intervals is zero.

    The ONLY thing of concern to the city is MONEY, MONEY, and MORE MONEY.

    If St. Petersburg voters want to end the red light camera program they will need to vote out every camera supporter in city government and replace them with active opponents to the cameras who will not rest until the cameras are entirely gone. Start by repeatedly contacting the city officials to demand the cameras be removed and let them know this issue will affect your future votes.

    James C. Walker, National Motorists Association

Comments are closed.