Roads, traffic, and the importance in decision analysis of carefully examining your goals

Sandeep Baliga writes:

[In a recent study, Gilles Duranton and Matthew Turner write:]

For interstate highways in metropolitan areas we [Duranton and Turner] find that VKT (vehicle kilometers traveled) increases one for one with interstate highways, confirming the fundamental law of highway congestion.’

Provision of public transit also simply leads to the people taking public transport being replaced by drivers on the road. Therefore:

These findings suggest that both road capacity expansions and extensions to public transit are not appropriate policies with which to combat traffic congestion. This leaves congestion pricing as the main candidate tool to curb traffic congestion.

To which I reply: Sure, if your goal is to curb traffic congestion. But what sort of goal is that? Thinking like a microeconomist, my policy goal is to increase people’s utility. Sure, traffic congestion is annoying, but there must be some advantages to driving on that crowded road or people wouldn’t be doing it, right? (Just to be clear: I’m serious here. This is not intended to be some sort of parody of economic reasoning. I do believe that people venture out in traffic for a reason.)

Let me put it another way. Suppose that (a) Los Angeles could immediately double its existing road system, at zero cost in dollars, effort, open space, pollution, etc., and (b) if that were done, there would be exactly twice as many cars on the road at the same exact level of congestion. If this (inherently impossible) scenario came to pass, this would be a good thing, no? Twice as many people would be able to get what they want, which is to be out on the roads.

Or, here’s another one. Suppose NYC could immediately and costlessly double the frequency of its subway trains, and suppose this would (i) draw lots of people out of their cars into the subway and (ii) draw out new drivers, so the existing roads would be as crowded as possible. If the goal is to curb traffic congestion, we’ve gotten nowhere, but if the goal is to improve quality of life, we’re in great shape. I ride the subway (sometimes). If there were twice as many trains, I wouldn’t have to wait as long for a train. I also ride on the roads. The traffic congestion doesn’t bother me so much, I can still get around.

If you don’t like the NYC subway example, pick your favorite alternative, for example consider a suburban bus line that changes its frequency from once per hour to once per 10 minutes. Once per 10 minutes is great for bus riders—even if it does nothing to alleviate congestion, you don’t have to wait that extra 50 minutes at the bus stop.

My point here is not that we should be necessarily be expanding highways and transit systems: there’s a cost to subsidizing transportation, a cost in pollution and an opportunity cost because the money and resources are not spent elsewhere (maybe not such a cost right now with 9% unemployment but it’s an issue at other times). Nor am I knocking congestion pricing, an idea that seems to make a lot of sense, whether or not roads and transit are increased. My key point is that, yes, traffic congestion is annoying, but I don’t think it makes sense for “curbing traffic congestion” to be a primary policy goal. A policy that leaves traffic congestion unchanged can still help out lots of people, inside and outside of cars.

Why am I blogging this?

I’m not a transportation expert. But I think the above discussion illustrates a general principle in decision analysis, which is that it is important to consider carefully what your goals are. In the above example, traffic congestion is a problem, but I think it’s a mistake to set the reduction of congestion as a goal in itself, or to dismiss various actions because they don’t provide progress toward that particular goal.

10 thoughts on “Roads, traffic, and the importance in decision analysis of carefully examining your goals

  1. Maybe the issue is that decision makers do have “curbing traffic congestion” as a primary policy goal, and use it to justify investment in new highways and public transit. The underlying problem may be the policy goal, but the fact that the means of achieving it won’t work is also worth addressing.

  2. Reading Sandeep and your post I considered a different policy intent. If you want to reduce traffic, you can close streets and lanes. The congestion will stay the same, with drivers opting out of driving cars.

    In the same vein, increasing public transport does not reduce congestion, but since congestion is constant, the relative part of public transport in the total person-kilometer amount rises.

    I believe both these policies increase public AND private utility, if you account correctly for the costs of private vehicles, urban sprawl and the effect of traffic on urban life and design.

  3. I highly recommend Anthony Downs’ book Stuck in Traffic. This study is just confirming something that Downs has written about for years, why building more roads can’t solve a congestion problem. Downs main point, which is echoed by Andrew here, is that one could think of congestion as an economic solution to the problem of scarce resource so it is not a problem at all.

  4. On a very primitive level the problem with this nonsense is that the number of people is conserved, so at best it can be true only for small changes. If more people are using public transport and getting where they want to go faster and with less stress, then even if congestion stays constant, then there are more people satisfied.

    • The *point* is that the number of people isn’t conserved, at least locally. That’s the active mechanism behind the law here. Build a new highway, and communities pop up to fill it. Metros increase population density, etc.

  5. When you’re considering traffic congestion, its “dual” factor is the geographical patterns in cost of real estate.

    Consider two extremes.

    1. There are no paved roads or public transportation. People end up living in dense neighborhoods and price of an apartment is directly correlated with average income offered by jobs in its proximity.

    2. Roads are uncongested, public transportation is cheap and widespread, and gasoline is free. Many people buy houses in the exurbs and end up commuting 1.5 hours one way, because a house in the exurb + wear & tear of the vehicle + opportunity cost of 3 hours/day spent driving comes out cheaper than an apartment downtown.

    Starting from the point somewhere in between, most naive attempts to reduce traffic congestion result in moving the city closer to point 2. Which is probably not the direction in which the society as a whole should want to go.

  6. And here’s another point that is related more directly to the question “why curb traffic congestion”.

    The throughput (in vehicle-kilometers per hour) of any road is related to the density in an inverse U pattern.

    Traffic jams occur well above the critical density. It is in the interest of city planners to keep that from happening (and ideally to keep all roads below critical at all times), because that maximizes the efficiency of the transportation system.

Comments are closed.