Congestion Pricing

by Tom Temple

Dec 13, 02:14 PM

Looking out over a snowy Cambridge right now, I see total gridlock. A few cities, most notably London (and New York to some extent) are trying to fix the problem by implementing a “congestion tax”. I think it’s a great idea. I think it should be dynamically priced and implemented anywhere where it could be made to work logistically (e.g. any highway that already has a toll system).

The basic idea is this: When you use busy roads, you do not have to pay for the marginal delay that you put on all the other users. So to internalize this externality, the people in charge of the roads can charge a congestion tax to account for this marginal delay. When the system is running close to capacity, this marginal delay can be several times greater than the delay experienced by an individual. (If this doesn’t make sense, think about how far a delay might propagate before finding a gap. You might only need 4 seconds of road time, but that might mean delaying thousands of people by 4 seconds.)

I’m not talking about some sort of capitalist system where the guys who own the road can charge some sort of fair market price. I don’t think we could ensure adequate competition. The congestion priced system instead finds the balance between time and money, rather than between demand and price. Although It is important to note that the amount of the tax should vary according to the demand. At 3am Tuesday, the tax should be zero, while at 5pm on Friday, it could be $50 or more.

But there are a few concerns that a civic minded person might have. For one, the roads are a public good and shouldn’t be reserved for the wealthy. Right now, the wealthy are being taxed with time and the the poor1 are being subsidized. I don’t consider that especially problematic, but it’s important to realize that there is already a congestion tax—you just pay with minutes instead of dollars. What is nice about a dollars tax is that you can use dollars to buy stuff while a minutes tax just pisses it away.

1 Poor: People who drive but wouldn’t want to trade dollars for minutes at whatever turns out to be the market rate. This would certainly include me.

In particular, you can take the congestion tax and dump the (considerable) proceeds into public transportation. That is pretty obvious. A more interesting idea (and the one that prompted this post) would be to divide up the proceeds and give everyone credits which could be used for the tolls, or redeemed for cash. This would promise a certain level of road access regardless of the ability to pay while at the same time not being wasteful like the current setup.

Comments:

  • Michael
    Dec 13, 11:14 PM

    Good idea. Perhaps some of the proceeds could also be put toward some mandatory driver re-training, too. Lord knows enough of the drivers in the greater Boston area could use it. A technique I think would be particularly effective is to require each driver to spend at least two days per week for one full year, commuting by bicycle.

    Implausible, no doubt, but I think it would help.

  • Jon Shea
    Dec 14, 02:34 PM

    A tax to correct the negative externalities of an activity is called a Pigovian tax. You pretty much nail the situation, Tom.

    In economics, the term public good usually refers to a good that is non-rivalrous and non-excludable, such as national defense. Roads are both rivalrous and excludable. You mean to say that roadways are state-owned, or publicly-owned.

    Also, I strongly favor abstracting public revenue away from public expenditures. While there’s a certain emotional appeal to throwing all of the profit from a congestion tax at public transportation, what are the odds that the transportation tax earns exactly the best amount of money to spend on PT? We’re much better off if we tax everything in the most efficient way possible to earn the level of revenue we want our government have, and then spend that money in the most efficient way possible. There’s no reason to believe that various moves of revenue and expenditure or magically linked at efficient levels.

  • Tom
    Dec 14, 04:22 PM

    I agree with you about the revenue—expenditure link, Jon. It is conceptually appealing to imagine a mapping from taxes to expenditures that makes it explicit who is paying for what. If we were going to start completely over with this whole taxes-thing, we’d be able to do what you say and it’d be efficient, if only for a short time.

    That said, I think as a practical matter if the goal is to keep the current balance of taxes to payouts on an individual level, it is impossible not to link them. First consider that this new revenue needs to be channeled into new expenditure (if we define expenditure so as to include savings). Let’s assume that the old system was fair. Unless we got lucky with this new tax, we now have created some new unfairness. If we want to preserve fairness, we need to shape the expenditure to mitigate this. Hence, PT and congestion credits. There are clearly other ways to do this shaping, but there is no way that doesn’t involve transportation and is as precise.

    Even if the old system was unfair, losers are always louder than winners and so changing the tax distribution is very hard politically.

  • Tom
    Dec 14, 04:26 PM

    To clarify, I don’t mean to say that all of it needs to get spent on transportation. Just enough so that we maintain the balance of winners and losers. If it’s not zero sum (which I should hope it’s not), the DC part can go into the general budget.

Comment: