David R. Henderson  

Heyne, Boettke, and Prychitko on Congestion Pricing

PRINT
A Swarm of Thoughts on Hive... Meta-Crisis...

I'm using my favorite textbook for my Executive MBA economics course: The Economic Way of Thinking by the late Paul Heyne, Peter Boettke, and David Prychitko. I'm using the 11th edition, which I prefer to the 12th edition. I haven't checked the 13th edition, but I tend to use earlier editions to save my students a lot of money. So the bottom line is that it's possible that the 13th edition doesn't have the problem I'm about to point out with the 11th and 12th.

In an otherwise-excellent section in which the authors lay out the case for congestion pricing of freeways, they write:

There is a way. It's called pricing. Economists call it congestion pricing. Almost everyone calls it tolls for driving and doesn't want to hear anything more about it. "I pay for the roads with my gasoline taxes; I don't want to pay again with a toll." But gasoline taxes pay the cost of constructing the road, not the cost of using them. [italics in original]

There's a better argument. It's that charging a price for using the road is a better way of paying for roads than using taxes. The price doesn't have to be just a high price at congested times. It can also be a low price for uncongested times. Just as hotel room rates, peak and offpeak, pay not only for the cost of using hotels but also for the cost of constructing them, so tolls can pay for the cost of using roads and the cost of constructing them. Indeed, if the tolls cannot pay for both, there's a strong argument that the road shouldn't be constructed.

In other words, the authors shouldn't take the gasoline taxes as an immutable fact. In a book that goes to the root of the problem on so many issues, the authors should also go to the root on this issue. How? By pointing out that, if your goal is efficiency, gasoline taxes should be cut and tolls should be increased. It might even be made revenue-neutral.

One other advantage of tolls over gasoline taxes is that the prospect of tolls, even if it's government that charges them, can cause producers of highways to think about where highways are demanded more. And of course once the highways are already built, tolls can replace gasoline taxes as sources of revenue to maintain the highways.




COMMENTS (19 to date)
Jon Murphy writes:

I've not read their textbook so maybe I am missing something, but perhaps the authors use that argument because they are addressing the objection by the hypothetical person, namely "I already paid gas taxes, why should I pay for a toll, too?"

Prof. Henderson, I think your argument is strong when we are starting from a fresh slate (no road built), but I think they are making the argument that, for a given road already built, congestion pricing still makes sense.

William Bruntrager writes:

It may be true that tolls would achieve the goal of efficiency better than gasoline taxes, but it's not obvious, at least not to me.

In fact, it is my impression that the question of how to pay for things that have a high fixed cost to construct and a low marginal cost to use does not have an answer that is correct in all cases. Demsetz's "Exchange and Enforcement of Property Rights" provides a very interesting discussion of the subject, but I would summarize that discussion as, "It's complicated."

I would be very interested to hear more about the argument for tolls over gasoline taxes.

William Bruntrager writes:

Just one follow-up. The key lines in Demsetz's article are:

"To solve [resource allocation] problems efficiently, we need information which is obtained by excluding nonpurchasers, provided that the additional information is worth more than the exchange and police costs necessitated. In cases where the costs are greater, a zero price can be reconciled with efficiency requirements." [emphasis in original]

So again, it may be true that a toll is more efficient than a gasoline tax, but it depends on the details of the situation.

Tom West writes:

Actually, I have to wonder whether this is another case where the economically efficient solution does not maximize human happiness.

I feel it may be a case like "surge" pricing where people are still upset months that they were able to buy an item that would have been unavailable otherwise because the price had been raised.

Outside of in extremis emergencies, I've never heard anyone complain bitterly that batteries were out in an extended blackout. That just feels right. But be able to buy it at triple the price? Many will be planning to boycott the retailer afterward.

To my mind, the vitriolic hatred of paying for what you use of highways seems to mirror that.

Of course, I suppose it's all about whether you believe government should be trying to maximize human happiness, economic efficiency, or freedom.

Greg G writes:

>---"Of course, I suppose it's all about whether you believe government should be trying to maximize human happiness, economic efficiency, or freedom."

Well said Tom. It's easy to forget that those three values are all important and sometimes conflict with each other.

Zach S writes:

>But gasoline taxes pay the cost of constructing the road, not the cost of using them. [italics in original]

Gas taxes don't even do that in the States. They may pay for certain highways or interstates, but in general about half the cost of roads is from general revenue. This is especially the case at the local level.

I'm in favor of tolling over gas taxes in general. It makes people pay for the cost of congestion and makes alternate modes most cost competitive. Drivers see transit and see subsidies, but ignore their own subsidies.

It wouldn't surprise me if, given a higher cost to drive, there would be a greater shift to transit and a lowering of subsidies there. Hopefully it would help spur a rollback of regulations that make urban construction very difficult, given the increased demand for housing near transit.

David R. Henderson writes:

@Jon Murphy,
perhaps the authors use that argument because they are addressing the objection by the hypothetical person, namely "I already paid gas taxes, why should I pay for a toll, too?"
No need for a “perhaps.” That’s exactly what they are doing.
but I think they are making the argument that, for a given road already built, congestion pricing still makes sense.
Yes. And I’m saying they’re making a bad argument for a good policy and failing to argue for an even better policy: Raise the toll and cut the tax rather than leaving the old level of the tax in place.

Jon Murphy writes:

@David Henderson

Thank you very much. I see what you're saying now!

David R. Henderson writes:

@Jon Murphy,
You’re welcome.

Maniel writes:

And just to return to the obvious point, congestion pricing is an economical and environmentally desirable alternative to 1) pouring ever more concrete to alleviate rush-hour congestion or 2) spending ever increasing amounts of time commuting at low-gas-mileage speeds.

Hazel Meade writes:

Not sure if a better way wouldn't be to examine road construction costs and maybe pursue more free market policies with respect to the construction process itself. Government contracting is always notoriously laden with corruption and cost inflation, and local unions/contractors often monopolize these contracts. I recall reading recently how the cost of building a mile of highway has gone up an extraordinary multiple after adjusting for inflation in the last 50 years.

J Scheppers writes:

Dr. Henderson:

I think you have a great point, but I have many caveats that should be considered.

Congestion harms other drivers, on average each driver pays the average cost of congestion, but they do not pay the marginal price of congestion.

How the property right of who gets charged the marginal price and who receives the payment of the collected charge is so much more important than the weak economic truth that pricing congestion is "the solution".

How much road to build, cross subsidies to transit and pricing policies implemented by government agency clearly are fraught with potential miscalculation or abuse.

The standard argument is that without the toll the project will not get built for 20 years, but the congestion charging does not ensure the user profit maximization is going to result. The frequent perspective from the policy makers perspective is the more concrete the better, which is approximated to mean better roadway customer service.

Re: "the obvious point, congestion pricing is an economical and environmentally desirable alternative..."

Actually I'd say this determination depends on the specific case. Sometimes pouring more concrete makes sense.

On the gas tax vs. congestion charge theme, I just want to interject that it isn't hard to uncover prominent economists proposing raising gasoline taxes as one way to reduce the externality of congestion (Pigou club anyone?).

As someone who lives in a part of the country that only sees real congestion on 7 or 8 weekends each years with home football games, I object to the idea that my federal gasoline taxes ought to be higher because folks in LA, Boston or Washington, DC don't want to pay time-varying tolls.

...things that have a high fixed cost to construct and a low marginal cost to use....
Roads, highways and bridges do not have a low cost to use. Vehicle traffic degrades concrete and asphalt. The heavier the vehicle, the worse the wear and tear. This is in addition to the lost time from congestion.

I used to make my living repairing damaged concrete, so I know how costly it is to keep roads and bridges serviceable.

GabbyD writes:

Hi Dr. Henderson,

Do you know of any paper or work that says a toll a completely fund a road project -- that is without taxation upfront?

ThomasH writes:

Congestion pricing makes even more sense in the case of city streets. I'd leave the gasoline tax alone until there is a comprehensive carbon tax and add road use/congestion pricing on top.

Michael Byrnes writes:

This is a very interesting topic.

I've heard just one convincing argument against tolls/congention pricing, but I think it is a good argument:

Part of why we have things like highways is because it is often productive to have people moving around on highways. Commuting into cities, for example, underlies a lot of beneficial economic activity. Is it a good idea, then, to disincentivize commuting into cites by imposing a toll?

My answer is I think it depends. I think the "right" tolling/congestion pricing scheme couldt be a net benefit to commuters of all stripes. (Right now rush hour commuters pay a sort of toll - they just pay in time rather than money). But I also think a poorly designed tolling system could be a net negative.

ThomasH writes:

Technology exists to actually tax congestion -- proximity to other slow moving vehicles. A time of day toll would help but that's pretty crude and not feasible for urban streets. Congestion pricing should be integrated with dynamic pricing of street parking, too.

David Seltzer writes:

Two examples of congestion pricing. The Chicago Skyway. Traffic is heavy during drive time, but flows smoothly with the advent of toll passes. In Atlanta, where I live, the toll lanes on Interstate 85 are priced as to rush hour and non-rush hour. At peak, the price from my exit to downtown averages $8. Off peak is as low as $.05 cents.

Comments for this entry have been closed
Return to top