We invite our members to submit their questions on anything that they would like our thoughts on. We’ll give you a Strong Towns answer or find an expert who can. This week, Adam from Miami, Florida, asks:
I recently had the opportunity to argue with our county's lead traffic engineer on why municipalities are not able to determine their own maximum speed limits on their city streets, less than the default statewide 30mph for residential or commercial districts. Our streets, however, are narrow and we have dense foliage that comes right to the streets in many places with no sidewalks.
Her answer was that our streets have an 85th percentile of people that drive under the speed limit already, but not so far under that lowering it makes sense. In her words "Before and after studies have shown that there is no significant change in prevailing speeds when the speed limit is changed. Drivers will continue to travel at speeds that they feel are safe and prudent, despite the posted limit." I think that this is simply not the case.
My question is, how do I argue against a traffic engineer that isn't interested in the effects on the community of her LOS, and who believes that posted limits have no effect?
This question stumbles into what I consider to be one of the primary points of incoherence in the traffic engineering profession.
The traffic engineer believes:
- That people will drive at whatever speed they feel is safe and prudent, AND
- That traffic engineers are incapable of influencing speeds through design.
This makes the traffic engineer into one of the world’s most impotent professions. They design driving lanes with consideration of design speed, but never based on desired speed. They then plead helplessness in influencing how fast people drive. It is as if they compartmentalize different parts of their brain and allow no cross communication, despite the obvious tools for good at their disposal.
Let’s deal first with the 85th percentile speed. I totally agree with the quote that you have attributed to the engineer. The typical person will drive the speed they feel comfortable driving, regardless of the posted speed limit. There will always be the individual that would drive 5mph on the highway if it were posted that way as well as the individual who would drive 60mph through a school zone if they were late for work. These are the deviants and you really can’t design for them. They are the ones where enforcement is needed. The vast majority of people, however, will drive the speed which they feel safe driving.
Just so everyone knows how the 85th percentile is calculated: they send a young engineer (I used to do this) out to a place where they want to do a speed study. The young engineer listens to talk radio and writes down the speeds that drivers are traveling over a set interval...say, two hours. After that period of time, the data is reviewed and the speed limit is the speed which only 15% of drivers exceed. So 85% of drivers are driving at that speed or slower.
This is really important because a lot of people – including a subset of engineers, sadly – believe you can just set a speed limit artificially low, get the local cops to enforce the heck out of it and that will make things safer. Let’s say the 85th percentile speed was 50mph but the city wanted people to drive 30mph and so the lower speed limit. Lots of people would adjust and drive 30mph, but a lot of people would – for a variety of reasons – continue to drive at the speed they feel comfortable (50mph). Maybe they didn’t notice the sign. You can say this is inattentiveness, but it is reality, and the design of the roadway is communicating that the higher speed is safe. The difference in speed here is the greatest danger, especially in a place where you have turning traffic (as is the case in most 30mph zones).
So your engineer is right: people will drive what they feel is safe and prudent. Where your engineer is wrong is in believing they are held captive by that reality. The engineer is not powerless. Safety in this situation is not someone else's silo.
A while back I did a little contest where I asked people to guess the speed of the street based on the design. This was along one highway that went from 55mph down to 40mph and then to 30mph before returning to 40mph then 55mph. I clipped each photo so readers couldn’t see the adjoining land use, just the roadway the traffic engineer had built. The result: totally random guesses. Nobody was able to tell the speed. You can read that post here.
In other words, the design of the roadway communicated nothing to the driver. Everything was communicated through signage. Drive slower. Drive faster. Look out for children. It is as if the roadway design is so taken for granted that it is never even considered.
Yet, it is obvious that people respond to the design of a roadway. If you ever get a chance – if you are in some remote area or are driving really, really late at night – try driving 20 miles per hour on a highway. It will feel really strange. Everything about a highway from the wide lanes to the wide shoulders to the clear zones and the sweeping curves signal to the driver that is it okay to drive fast. Driving slow actually feels unsafe in this type of environment.
Now take the opposite: a place as you describe where we would like people to drive more slowly. The same signals to drive fast go out to the driver when we have wide lanes, wide shoulders, clear zones and sweeping curves. We can post a slower speed limit and counteract that impression to a degree, but driving slow in these environments will not feel natural. It is a conscious act instead of one that is automatic or instinctual.
The answer, therefore, is not a slower speed limit but a slower design. You actually need to build the street to communicate the need for a slower speed. That means doing things that are counterintuitive to the traffic engineer. We need to narrow lanes. We need to use shoulders for parking, not as “recovery” area. (Note that so-called "recovery areas" aren’t needed when speeds are low because drifting is not going to be frequent or, except in the very rare instance, fatal.) We not only don’t want a clear zone, we actually want street trees and other things that communicate to the driver that they are in a complex, urban environment. And most importantly, we want people. That means benches and planters and crosswalks and bike racks and all the things that encourage people to be present.
So your engineer is right: people will drive the speed they feel comfortable. You need to demand that your engineer build you a street where drivers feel uncomfortable going more than 20 mph. If you do that, you’ll have one of the key components of a productive street and you’ll be on your way to building a strong town.
After I finished this post, I received an email from Adam from Miami with a link to this article on the 85th percentile. Here's a quote:
Basically, a study from 1964 remains the main argument to build more highways and freeways with faster speeds where the ends justify the means. Even if the means ignore vulnerable groups such as pedestrians and cyclists. Even if the study is now also used to serve the automobile in densely-populated urban areas, far from any freeway.
Here, I think, might be the flash point. The 85th percentile calculation is a sound approach if your only objective is to move cars quickly and efficiently. That is the approach we use on interstate highways, for example, where (sorry cycling enthusiasts) there should be no expectation of anything other than automobiles.
This logic is really destructive when it is applied to local streets where there is -- or should be -- a high degree of complexity. The 85th percentile calculation does ignore the needs of pedestrians and cyclists, not to mention local businesses, where slower speeds are needed.
Does this mean the 85th percentile approach is incorrect or outdated? No. Stop attacking the logic behind the 85th percentile approach because it is sound.
Here is what is messed up: when engineers apply highway design logic to local streets. For that situation, a different body of knowledge is needed, one that lowers speeds dramatically and -- if you really want a safe and financially productive place -- puts the needs of pedestrians and cyclists (also known in economic terms as consumers and property owners) ahead of that of drivers.
The 85th percentile calculation, level of service and forgiving design are all engineering concepts that have been developed for highway design. They work brilliantly in that environment. They have no application, however, to local streets. Zero. None. Applying them to local streets is dangerous, expensive and just plain dumb.
Another quote from that article:
Why did the engineer cross the road? Because that's what they did last year.
It's time for the engineering profession to stop being so dogmatic and just plain stupid and start engineering solutions for the complexity of problems our cities face, not just traffic.