“Easing the Logjam”

That’s the title of another article in yesterday’s Wall Street Journal report, “One Billion Cars.” The WSJ has stumbled onto a truth that, so far, has eluded Virginia pundits and policy makers.

Until recently, expanding highways and roads has been the traditional response to congestion. But in many areas of the world, such expansion isn’t feasible anymore because of lack of funding, opposition from residents or simply lack of room.

The WSJ highlights a variety of transportation alternatives that are being explored around the world. They include (with cutesy WSJ subheads):

  • Rubbing out Rubbernecking. Accidents and stalled cars breed rubbernecking and congestion. Many communities are investing in “incident management” capabilities to get those cars off the roads as quickly as possible.
  • Car-Road Talk. Better traffic light sequencing can move more cars through the same fixed roadway.
  • Paying for a Lane. HOT lanes and congestion pricing will encourage some drivers to find alternatives to driving alone during during periods of peak demand.
  • Alternative-Transit Bonus. Encourage employees to carpool, use mass transit or telecommute through subsidies, vouchers and ride-matching websites.
  • Quick Notice for Drivers. Provide drivers with more real-time information about traffic conditions so they can steer clear of gridlock.

Bacon’s Rebellion has explored each of these alternatives. My point in quoting the Wall Street Journal is to make it clear that these aren’t quirky ideas advanced by some eccentric blogger. Other people around the world are pursuing these ideas. Even the Virginia Department of Transportation is pursuing them. Unfortunately, the Traffic Light Sequencing lobby and the Carpooling lobby in Virginia can’t mobilize millions of dollars in campaign contributions, so, when it comes to funding, their ideas don’t get much of a hearing.

Share this article


(comments below)


(comments below)


12 responses to ““Easing the Logjam””

  1. Ray Hyde Avatar

    Roadway incidents are said to cause 25% of congestion delays. This is an area where we can do a lot, but our abillity to resolve incidents is stymied to some extent by accident investigations that are designed not to prevent future incidents but to resolve liability.

    We already use engine control computer information, but only to verify measurements taken on-scene. Eventually, we should be able to use a simple directional recording accelerometer which would tell us all we need to know about the physical events preceding the crash. Once it is accepted, we could eliminate much of the crash investigation, and thereby speed up the on-scene activities.

    This is an areas where technology could help greatly speed the process. Last night I observed the results of a T-bone intersection crash. There were 4 police cruisers and two emergency vehicles on scene, yet nothing appeared to be happening with any great urgency. Technology is only going to be able to do so much.

    I wouldn’t hold my breath on getting very much out of traffic light sequencing. What looks to be a simple problem for one intersection is a lot harder when you are looking at a network. The grid style street layout now being promoted has a lot more intersections to interact and control. A cynic would say that it is really being promoted because the uncertainty associated with all those intersections will slow traffic down and make it less competitive with walking: it is a perverse form of traffic calming, if you will. And in fact the cul-de-sac feeder system was developed as a superior alternative, for reasons we have apparently forgotten. It may also be that we have carried the new system too far.

    Anyway, light sequencing isn’t going to buy us much, especially if the roads are overjammed. And the same thing goes for driver information: if the information is that there are no usable alternates, why spend the money?

    I suppport hot lanes, but I suspect the results will not be what we expect.

    I don’t see that we gain much by switching from subsidizing auto use to subsidizing transit use, which we know is slower and more expensive. I think you are closer to the right argument whe you say let everyone pay their own location related costs. One of the costs of living in highly dense areas is that you have to increasingly rely on expensive, slow, and inconvenient mass transit.

    We should charge both types o travelers their full costs and be done with it.

  2. Anonymous Avatar

    Who sequences traffic lights, VDOT or a locality?

    It’s a MAJOR problem in the City of Winchester and nothing ever seems to get done – where should I start?

  3. Jim Bacon Avatar
    Jim Bacon

    Anonymous 3:00 p.m., I don’t speak with any authority, but I think you start with the local government. In Arlington, the traffic light-sequencing initiative is locally driven. I suspect that the same is true for Fairfax County, though I am not certain.

  4. Jim Wamsley Avatar
    Jim Wamsley

    As Will Rogers said, “It ain’t what you don’t know that gets you in trouble. It is what you know that ain’t so.”

    “I don’t see that we gain much by switching from subsidizing auto use to subsidizing transit use, which we know is slower and more expensive. I think you are closer to the right argument when you say let everyone pay their own location related costs. One of the costs of living in highly dense areas is that you have to increasingly rely on expensive, slow, and inconvenient mass transit.” Ray Hyde knows transit “is slower and more expensive.” Ray probably compared bus fare to gas cost. The location cost of parking is the biggest expense related to driving. This is a cost whether it comes out of your pocket as a parking fee, or subsidized by VDOT at the park and ride lot, or the merchant or the employer. It is a cost even when zoning requires a number of parking spaces per building.

    When parking is inexpensive, auto use costs less. When parking costs $15 a day transit trips look very attractive. Everyone gains when we “learn new tricks” and subsidize transit use when that trip costs less and subsidize auto use when that trip costs less. This is a switch where everyone will benefit.

  5. Anonymous Avatar

    My understanding is VDOT controls signals in Fairfax County.

  6. E M Risse Avatar
    E M Risse


    Who controls signals and to serve which objectives varies.

    Walk into the closest VDOT office and strat asking questions.

    It is not classified, yet.


  7. Ray Hyde Avatar

    With regard to signals, this is an observation, somebody please tell me I’m wrong. I frequently see new traffic signals located at the entrance to large new developments or groups of developments. The intersection there has frequently been reworked to include turn lanes and gutters, sidewalks etc. probqably at the developers expense. I assume the traffic signal was installed at his expense as well.

    My observation is that the signal favors the development. I notice that when a car approaches from the development side, it immediately trips the light, stops traffic on the main road so the development car can pull out. Please tell me my observtion is wrong.

    Jim Walmsley: here is what I know from personal experience. The last two cars I junked wound up costing me less than $0.30 cents per mile to drive, taking in all my out of pocket expenses.

    According to Ruzeimers the average auto cost is $0.44 cents per mile. A Mercedes is up around $0.92 per mile and a Saturn is down around $0.34. It is capital cost that kills you and that is what kills trains.

    I spent a year working at the Pentagon and used various means to get there: Drive to VRE then Metro; Drive to Vienna, then Metro; drive to Arlington then Metro; drive to my springfield office, then Metro, drive to the Pentagon and pay to park in a comercial lot.

    In each case my out of pocket cost was virtually the same except for driving direct. That cost more but saved me an hour in time. If I add the VRE and Metro operating subsidy and $6.00 an hour for my time plus a 50% surcharge against the car for externalized costs, then it is a wash. I didn’t charge VRE or Metro for their externalities.

    The trip from Virginia Square to the Pentagon costs $0.42 per mile and the average speed is 12.6 mph.

    That is what I know personally.

    Winston and Shirley did a study on transit systems (this is based on 1990 numbers multiply them by 2.5 to get it in todays dollars. )

    According the study of 228 bus companies and 30 rail companies the operating cost per passenger mile for buses was 44 cents per pm for trains it was 37 cents per passenger mile. At that time the direct operating cost for autos was 21 cents per mile. If you exclude NYC from the study, bus and rail figures for the remaining companies are almost twice as high.

    Here is the real killer. For buses the number of passenger miles per mile of revenue operation is only 9. For trains it is only 21. And that doesn’t count all the miles operated when they are out of service. Clearly, there are a lot of buses and trains running around mostly empty and we would be better off to operate them only where they actually pay. What is the difference between 9 passengers on a 40 passenger bus and one passenger in a four passenger car? (Hint, the lumbering bus causes more noise, and more congestion.)

    If you take into account capital costs, external costs, parking space costs, accident costs, etc. etc. etc. cars turn out to be only slightly more expensive than trains or buses, but they offer better, more flexible, generally faster, and more extensive service.

    Im sorry, but subsidizing the trip that cost less makes even less sense than subsidizing the trip that costs more. If transit actually cost less, we wouldn’t need to subsidize it: we could privatize it and run it at a profit. In order for that to happen we have to operate transit only where it actually makes sense.

    The opposite is true for cars. If cars really cost as little as it appears, we wouldn’t be suggesting congestion tolls. But even when we charge cars their full costs they will still fill more needs than transit. The best we can economically hope for is something like an 80/20 split of cars to transit, and even then transit is going to be in addition to autos, not instead of.

    I’d like to see transit work, too (for everybody else), but I’m not willing to lie to myself to make it happen, especially if I know it is going to cost us all money and time. And space, of course, but that is included in the capital costs.

  8. James Atticus Bowden Avatar
    James Atticus Bowden

    Ray Hyde: Do you have your planning numbers posted on the Road to Ruin or anywhere else?

    I’d like to see the basic planning figures – been on my mind for months – for transportation. And then for zoning issues.

    Any suggestions on putting it all together in one place?

  9. E M Risse Avatar
    E M Risse


    Great idea to get all the numbers in one place!

    The problem is there is not yet a widely accepted conceptual framework that matchs economic, social and physical reality.

    See Chapter 16 et. al of The Shape of the Future.

    PROPERTY DYNAMICS coming to an Alpha Neighborhood near you as soon as we can get it ther.


  10. James Atticus Bowden Avatar
    James Atticus Bowden

    I’ll take the numbers as they are EMR. Where is chapter 16 found?

  11. Ray Hyde Avatar

    Ed is right, there is no agreement on what we are talking about, let alone how to measure it. What first brought me to this site was a reference to community metrics which popped up Ed Risse’s name. I took to reading his articles looking for facts, and couldn’t find any, or very few, and none that were verifiable.

    We have a fundamental disagreement as I see it because he thinks, as many do that his way is the only way. I think the only way is every way, and it isn’t going to be optimal for everybody.

    I don’t have all the numbers in one place unless you count a massive hard drive as one place. I’m not even sure I believe all the numbers.

    But you could start like this: walking: speed, 3MPH; max duration, 30 minutes; max load, 60 pounds; max density, 1 per square meter; cost 100 calories per mile plus $1.00 per mile.

    Then do the same thing for bike, auto, bus, rail. Use the observed numbers and not the hypothetical ones.

    Set up a linear algebra equation that gives you the maximum transport capacity per square mile at minimum cost.

    Then figure out what the density of population and jobs would be that would use up that amount of transport capacity.

    Subtract that area from the square mile of transportation capacity, and figure out how much transport you have left.

    Reduce the population and recalculate the density, then and iterate until the population density and transport capacity match.

    Any population density lower, won’t make full use of your transport, any population higher will over use your transport and lead to periodic episodes of congestion.

    Figure out how much congestion you want, figure out how much the transport mix costs, and you have the answer.

  12. James Atticus Bowden Avatar
    James Atticus Bowden

    Ray: You are several steps ahead on the road to solving the problem. I’d be happy to just gather the numbers, as eclectic as they seem, in one place.

    I work on metrics in my day job. I understand how confusing it can be.

    I’d like to chat with you – email or voice – about sharing some of the numbers you have on your hard drive and putting them on an excel spread sheet with many, many tabs.

Leave a Reply