Tuesday, October 6, 2009

Montgomery County OLO Report Rebuttal

The Montgomery County Government has recently been touting a glowing report by the Office of Legislative Oversight. To summarize the report: everything about the program is perfect, everyone is happy, everything is wonderful, pay no attention to the man behind the curtain. Since this is a highly biased report, we will now give you our highly biased rebuttal.

The County has presented this as an objective report, but the OLO is in fact not an independent body, but is directly beholding to the county Council. According to the OLO website, "the OLO Director is a non-merit appointee of the County Council" -- basically a political appointee whose serves at the pleasure of a council whose members have all invested a great deal of political capital in the speed camera program. As such the OLO report is the County Council reporting on itself, so it is not surprisingly that the County Council gave themselves a nearly perfect review.

The report claims a 23% reduction is accidents within 1/2 mile of speed camera sites within a year of the cameras being activated. It mentions that part of the selection criteria for sites was 'Annual number of property damage, personal injury, and fatal automobile collisions (within 1/2 mile)', which indicates that there was a built in 'regression to mean' factor built into the study: if a site was selected because there had been an unusual increase in accidents the year before, that number would tend to return to the average. It is also the case that it was dealing with accident reports, which only take place when the incident is reported to police. If accident reports are not taken in the same way, or with the same frequency, the validity of results becomes questionable. If police were no longer deployed in a location because human traffic stops were replaced by a speed camera, minor accidents might never be reported because the drivers would have moved on before any police arrive. The report could not avoid two important statistics regarding the most highly published types of accidents:
"A small number of the collisions in the injury/fatality category involved fatalities. In the four years prior to camera activation, the County experienced an average of two fatal collisions per year in the vicinity of future MCPD speed camera sites. Three collisions resulting in a fatality occurred in the year after camera activation."
"In the four years before camera activation, the County experienced an average of 15 collisions per year involving pedestrians or bicyclists within one half mile of future MCPD speed camera sites. During the year following activation of speed cameras, 22 collisions with pedestrians and bicyclists occurred in the same areas."

It is known that several traffic fatalities have occurred in close proximity to MoCO speed cameras in 2009 (although the exact criteria used for counting such incidents is unclear) so it is entirely possible that these figures were not flukes and do represent a trend. However both of the previous unfortunate facts were followed by disclaimers : 'OLO notes that the number of events in this category is too small to draw any definitive conclusions about the relationship between speed cameras and the prevalence of fatal collisions.' and 'Given the small number of incidents and the annual variations, data from more than one year is necessary to establish a meaningful correlation between the presence of speed cameras and the frequency of pedestrian and bicycle collisions.' Similar 'excuses' were not present in sections where other samples for the same period of time were used to claim the success of the program.

The report makes no mention of the fact that traffic fatalities increased in MoCo from 2007 to 2008 from 48 to 51, a 6% increase, at a time when the county added 24 new fixed speed camera sites, and at a time when nationwide traffic fatalities dropped 10%. Nationwide traffic fatalities dropped an additional 9% in the 1rst quarter of 2009, the 12th consecutive quarterly decline nationally. This is despite the fact that only a few states use speed cameras now, and most of those only in limited areas. It also fails to mention that nationwide traffic fatalities per vehicle mile reached their lowest level ever in 2008, according to the National Highway Safety Administration, the continuation of a 30 year nationwide trend. In a sense the OLO report lacks a 'control group' and fails to compare the results in new speed camera zones with those in other areas, where other types of safety improvements were made instead.

The report made an effort to show the percent of vehicles traveling 1-10 and 11+ mph above the speed limit. However it makes no effort to measure traffic congestion, or what % of traffic was traveling 10mph below the limit, an indication of the level of traffic congestion. The report did indicate that there was apparently a 4% reduction in traffic volume, and stated that some people may have sought alternate routes to avoid the cameras (or traffic congestion caused by cameras?).

The report made no effort to compare their results with those achieved through other types of traffic calming. We previously did, and found that the city of Gaithersburg had succeeded in reducing average speeds by about 20% using other measures. The OLO report indicates that the average reduction in speeds at all camera sites was a mere 6%. (The OLO report makes it clear that most citations were given to people going 1-2mph over the ticket threshold, which is the reason why a 6% average decrease, apparently made a large difference in the % of drivers exceeding that threshold), or 11% as the maximum speed increase achieved by speed cameras. So if speed reduction was a goal, the report fails to demonstrate that the cameras are the only or best way to achieve them. The OLO report also fails to take into account the fact that at some speed camera sites there were new safety devices or additional signage added other than the speed cameras themselves (added either by the county's own choice, or in response to critics of the program) and that these may have affected the results at those locations.

The document acknowledges only 10 cases where a defendant was found not guilty, claiming a 99.7% conviction rate. In fact, the report does not acknowledge ANY cases where tickets were dismissed because citations were issued in error, even though several such cases have been documented in the mainstream media -- including at least 40 cases of duplicate citations being mailed months apart with incorrect timestamps, drivers receiving citations for another person's vehicle, and one case of an innocent driver being falsely clocked at 100mph. In fact the OLO Report confirms that 9 % of violations which were rejected either by ACS or by Police were determined to be "No Violation Occurred" or "Operator Error": a total of 24,868 speed camera photos were in fact generated in error(olo report pg 35). Yet mysteriously none of those cases when to court. Of course there is no way to know exactly how many erroneous citations slipped through the system and the drivers either paid them without question or could not prove their innocence beyond reasonable doubt. The report also does not acknowledge that the part of the reason is that drivers were instructed that they could not plead not guilty unless someone else was driving their car, even though defendants have reported to the media that they were told this by a judge.

The report did state that "MCPD attributes the high conviction rate to the coordination with the District Court in the Safe Speed programs early implementation stage. When the program began, MCPD demonstrated the speed camera technology for the District Court judges" -- in other words, the county began presenting its case to district court judges before the alleged violations took place without the later defendants having the opportunity to hear and know what information was presented to those judges. The report also confirms how court hearings are taking place: with 100-200 hearings being scheduled in a single courtroom on a single day once per month, giving each defendant only a few minutes to present their case (known as District Court 'Speed Camera Day').

The report documents some of Chevy Chase Village's results. However in May of this year we used CCV's own monthly police reports to show that there had been no reduction in accidents at all. The Village's response to that report was to stop publishing monthly police reports to their website: No new reports were posted to the Chevy Chase Village since their May report (see screen snapshot). Now that their raw data is no longer being subjected to public scrutiny, PROBLEM SOLVED!

The government cannot provide objective oversight for itself. That requires a vigilant public. It is the duty of the citizens and the media to question the information which the government produces and to ensure that it tells the whole story, not just the parts which help provide cover to county and state lawmakers for their votes.