[Michael1]

College Station, TX - General Motors has recently announced the results of a study claiming that 17,000 vehicle crashes have been avoided due to the use of daytime running lights (DRLs). The GM press release is based on a presentation given to the National Highway Traffic Safety Administration (NHTSA) on November 9, 1999. Before we all jump for joy and sing the praises of DRLs, let's take a closer look at GM's news.

First, GM's press release claims that the results were from an "independent study." Well, "independent" is a relative term. The study was independent in that it was not conducted in-house by GM. Rather, GM commissioned Exponent Failure Analysis Associates of San Francisco to conduct the study for them. Immediately, we see that GM has an economic interest in the success of DRLs, and is willing to finance an "independent" study to support their crash avoidance claims. Because of the relationship between the external analysis firm and GM, one must already question the validity of the results.

Next, GM gives a little history of DRLs. In 1990, GM petitioned NHTSA to allow DRLs based on data from Scandanavian countries. The studies from several countries where DRLs are required have been called into question by a report entitled, "Thirty years on: Do motorcar daytime lights reduce accidents?" from Stephen Prower, Research Officer for the British Motorcyclists Federation. Mr. Prower's report questions the methods used by researchers to support the mandated DRLs in the Scandanavian countries and Canada, casting doubt on the DRL proponents' claims of collision avoidance.

The data used in the GM sponsored study compares collision rates of GM, Volvo, Saab, and Volkswagen vehicles before and immediately after the introduction of DRLs on particular models. Immediately, the analysis becomes artificially biased due to the lack of data from non-DRL vehicles. Such data could have been used as a control to filter out other contributing factors to crash avoidance. Such factors might include a change in driving habits, year-to-year weather changes, and other crash avoidances devices, such as anti-lock brakes, becoming more prominent.

A closer look at GM's presentation in DOT's docket reveals that GM claims that DRLs reduce multivehicle collisions --- at night. GM didn't bother to mention this silly statistic in their press release. While there are some drivers who fail to turn on their headlights at night, their numbers are probably very small and quantifying them in a statistical study is nearly impossible. Some DRL equipped vehicles do not have an automatic headlight option. Intuitively, one would assume that such vehicles may be involved in more nighttime collisions since the taillights are not illuminated with the DRLs. GM's claim of a reduction nighttime collisions casts a serious shadow of doubt on the entire study.

GM's press release claims that their DRLs do not cause "disabling glare." This statement is very questionable. Many of GM's vehicles have a DRL intensity at or near the limit of 7000 candela specified by the NHTSA 1993 rule change. However, NHTSA has acknowledged that "discomfort glare" sets in at 2600 candela. Discomfort glare causes drivers to avoid looking into oncoming traffic as well as switching their rearview mirrors to the Night position to avoid the glare from DRLs. Unfortunately, both of these practices run counter to safe driving. Further, in an already bright environment the DRLs add to the visual clutter that drivers must endure and can increase driver anxiety and even provoke road rage.

GM's summary in their presentatation even acknowledges, "the crash reduction benefit of DRLs is difficult to determine with precision." Let's take GM at their word and disregard this inaccurate study. Researchers have spent years trying to massage crash data to support DRL effectiveness, and this appears to be yet another failed attempt.

The GM presentation to NHTSA has been filed at the Department of Transportation's docket management system as NHTSA-98-4124-350.