A Simple and Effective Answer for the TRB Workshop
By James Walker, President, JCW Consulting
This was a paper that was submitted to the Transportation Research Board Annual Meeting in 2000.
This problem has a simple and effective answer, one which has been well known for nearly sixty years.
The default posted speed limit, in the absence of very unusual circumstances that the driver cannot perceive, must be set at the 85th to 90th percentile level of free-flowing traffic under good conditions. The list of engineering references supporting this proven methodology is very long. Here are a few highlights:
What to do has been well known since the 1940's. You tend to get the best overall safety, the smoothest traffic flow, the most appropriate choice of road types, etc. when limits are set at the 85th-90th percentile speed of free-flowing traffic. The problem is to get the decision past two major opposition consituencies.
First, education is needed to persuade those that do not understand the safest methodologies, which includes most state legislators. Those in charge of making the speed laws at the state and local levels are frequently very ill-informed about how to set statuatory limits to achieve the greatest safety and efficiency. The best law on the subject would be one that requires periodic legitimate engineering surveys, with no upper limit.
Second, strong statements from the engineering community are needed to deflect the opposition from the constituencies that will lose money when the safer and more effective 85th-90th percentile limits are adopted. This includes the insurance industry that will no longer be able to surcharge their safest 60th to 90th percentile drivers, the drivers that are least likely to cause accidents or make claims. It is enormously profitable for the insurance companies to surcharge their safest drivers, and proper 85th-90th percentile limits will make it impossible for them to enjoy this inappropriate and counterproductive source of invalid premium-surcharge revenue.
In some areas, police departments get a portion of ticket revenue ($10/ticket for State Police in Michigan) and this can make police agencies oppose any change in posted limits that will reduce speeding ticket volume. Court systems at state, county and local levels will lose a lot of speeding ticket revenue with 85th-90th percentile limits, so their opposition must be blunted with strong statements to the public from the engineering community that safety and traffic efficiency should override ticket revenue.
Most highway speed limits in the USA are posted between the 30th and 50th percentile of free-flowing traffic speeds under good conditions. This means that posted limits arbitrarily define 50% to 70% of all drivers as violators. The public will never accept that the safe driving habits of over half of all drivers should be illegal. Until this counterproductive practice is stopped, drivers will not respect posted limits.
All posted speed limits need to be set with science, generally at the 85th-90th percentile speeds, to achieve the highest levels of safety, the smoothest traffic flow, and the most appropriate choice of road types. They must not be set with arbitrary decrees from unqualified groups, many of whom have revenue motives.
Respectfully submitted to the workshop by James C. Walker, President, JCW Consulting.
The author has been a student of proper traffic engineering and speed management principles for over thirty years. He has made presentations before the Michigan State Safety Steering Committee and the MDOT Committee in charge of revising construction zone speed limits. He has testified several times on speed management issues before the Michigan legislature. He has published several editorials on the subject.
Speed limits should be based on sound traffic-engineering principles that consider responsible motorists' actual travel speeds.
Typically, this should result in speed limits set at the 85th percentile speed of free-flowing traffic (the speed under which 85 percent of traffic is traveling).