Feed on

“Infrastructure” is among the most popular goods for which folks think the federal government ought to have a heavy hand in producing. Let’s think a little bit more about this. A few observations that came to mind while driving to work on a totally un-congested highway that runs right through downtown Rochester and connects the Eastern and Western suburbs to the city.

  1. Infrastructure is not a good, and therefore it makes no sense to argue with blood red face that government needs to do more of it. Infrastructure is a summary term for all of the underlying goods and services that constitute the transportation network within our country, or perhaps even beyond. Therefore, our infrastructure includes not just city streets, sidewalks and bike paths, but urban highways, rural roads, rural highways, bridges, trestles, tunnels, air traffic control and many other related services. It makes as much sense to discuss the government providing “infrastructure” as it does to argue that the government should provide “stuff” but that’s the argument we end up having.
  2. When is there possibly a justification for government involvement in the production and sale of the elements in the infrastructure vector? Only when it is prohibitively difficult to exclude non-payers from enjoying the benefits of those goods. Now, there is an element of public-goodness in everything that is produced and consumed. When I hear a nice song on the radio on the way to work, I am in a better mood and perhaps am a better teacher that day. Since I cannot capture the gains my students enjoy because of the song, then I probably “underinvest” in listening to good music on the way to school. Few folks would argue that the government ought to subsidize me to listen to my music (hilariously, they probably already do). Thus, the case for government involvement in “infrastructure” has to be made on a case by case basis.  The publicness of each of the different elements of the transportation network are vastly different. And note that when I say “publicness” it has nothing at all to do with who or what currently provides it. When economists say the term “public good” it has nothing at all to do with some crude nationalism, or who is producing the good or what you feel about the good, all the term means is the degree of excludability of it (rivalry may or may not matter). That point is so important that I am going to repost it separately in a day or two. Therefore, we must ask, how hard is it to exclude non-payers from enjoying the benefits of a tunnel? A sidewalk? A city-street? A bridge? It is clear that the extent to which preventing non-payers from enjoying benefits is very different across each of these. We’ll perhaps do another post working through each particular item, and what the nature of the benefits are.
  3. The federal gas tax was established in the 1950s as a way to establish a trust fund that is used to build and maintain America’s highways. It stands I think at 18.4 cents per gallon right now. My best guess as to why we taxed gas at the time instead of charging for actual road use (after all, good public finance requires taxes to be derived from the places that are closest to the provision) was that the technology simply did not exist to charge road users for their use, and this was a second best option. I also imagine that there was less variation in fuel efficiency across the 1950s auto/truck fleet than there is today. Leave aside for the time being the ugly public choice issue of just how much the “highway” trust fund has been raided to build bike paths, light rail and the like, and just think about the problem with using a gas tax to fund roads. Presumably, the folks that use roads more should pay more and the folks who contribute more to emissions and urban air pollution should pay more. It’s not at all clear that the federal gas tax as structured does this.The gas tax is an excise tax that is charged on a per gallon basis to all purchasers of gasoline (ignore diesel for now). Ignoring emissions for now, this means that when I purchase a gallon of gas to fuel my Mazda 3 (it averages 30 miles per gallon) that I am paying 18.4 cents to drive the 30 mile round trip to and from the U of R every day. When my neighbor, who has a Ford Explorer, purchases a gallon of fuel and she gets 15 miles to the gallon, she is paying 36.8 cents to make the same trip. Sure her car is a little heavier, but another neighbor I have drives an old sedan that is similar to mine and probably gets the same mileage as the Explorer. In other words, because of differences in fuel economy, you see that people who are identical users of the roads, who impose the same congestion costs on everyone else as each other, pay wildly different taxes to fund the good. That is bad policy, and it also happens to violate horizontal equity considerations, which presumably those who are most interested in being transportation planners are heavily concerned about (I say presumably because I see no evidence that they actually DO care about it through good policy).A second issue raised by the gas tax (among many others that I won’t cover here) is that increases in fuel economy necessarily reduce the revenue that the tax will generate. And of course some of the fuel economy increases have been mandated by the government itself. If President Obama gets his wish and fuel economy doubles in the next 15 years, then that means that the revenues (ignoring behavioral responses) targeted for highways will be half as large in 15 years as they are today.

Both the equity considerations and the revenue limitations of the gas tax point very sharply toward thinking of new ways to fund and maintain highway infrastructure (perhaps even urban roads, but again that’s another discussion). A far superior way to charge for these services is to charge users a per-mile fee that is tied not only to how many miles they drive, but to the real costs they impose on us for driving. Hence heavy vehicles would be charged more than lighter ones, and vehicles that contribute more to congestion are charged more than those that contribute less. This is no longer a technological obstacle but rather a political and psychological one. It is also a consideration that ties into the underlying message of the beginning of this post. It is not at all clear that “infrastructure” is even a public good at all. In the case of intercity transit and perhaps even for much of the transit that happens within cities, it is overwhelmingly the case that excluding the majority of non-payers from enjoying the benefits is possible. And in cases where it is not easily possible, it is increasingly easy to bundle the difficult to exclude goods (such as city sidewalks) with other goods for which it is very hard to exclude non-payers (i.e. homes).


The record of private provision and maintenance of roads and other elements of the infrastructure vector is especially good when compared to how well the governments have provided these services. Advocates of good government and social justice should take notice. Rather than wringing your hands about how many billions less future federal budgets are “shortchanging our crumbling infrastructure” you ought to be celebrating the decreased “publicness” of many elements of the modern infrastructure vector. The point of “public goods” is that these are things that unfortunately we have to rely on coercive collection action in order to get what we otherwise would be able to provide for ourselves. It’s just contracting costs prevent ourselves from being able to do it. Now, if it is not only possible but also likely cheaper and safer and better to have voluntary actions provide services we should all be celebrating it. This means that the pie is bigger.

But we should all be celebrating it for a second and more important reason (at least if you claim to be a “progressive” who is progressive because of a belief in good government, a position I used to take myself). What reason is that? Because now when private voluntary action can take care of interstate highways and bridges and perhaps a few other things, that frees up the government to focus on more valuable things, to focus on things it does well. But the insistence in the modern progressive movement that the government provide “infrastructure” demonstrates nothing more than a religious commitment to the idea that if something used to be done by government it must always be done, or that some things that just “sound” like public goods must be provided by governments. But the many billions of dollars each year that could be saved by allowing voluntary action to take care of some infrastructure can both be refunded back to us and used in perhaps better areas, such as funding basic R&D. Not only does ignorance of this demonstrate an utter lack of imagination, it flies in the face of the many proclamations that I hear about just wanting a better society, a good government, and so on. That’s not at all what radical insistence on government roads demonstrates – it demonstrates to me a dangerously dogmatic approach to wanting the government to do anything at all that suits your interest. If folks truly did respect voluntary forces, private property and free exchange you’d expect a little more reflection on this point.  But they don’t. Just putting a caveat that markets can work great in your otherwise paean to the virtue of government roads and bridges means no more to me than reading those dairy trucks in my area telling me my milk comes from “Local Farmers Who Care” who absolutely know nothing about me. We are what we do, not what we say.

3 Responses to “The Gas Tax, Roads, Public Policy and Public Goods”

  1. Speedmaster says:

    >> “Now, there is an element of public-goodness in everything that is produced and consumed. When I hear a nice song on the radio on the way to work, I am in a better mood and perhaps am a better teacher that day. Since I cannot capture the gains my students enjoy because of the song, then I probably “underinvest” in listening to good music on the way to school.”

    When I argue with people that education should return to the private sector, this argument comes up first, every time.

  2. Michael says:

    “Now, there is an element of public-goodness in everything that is produced and consumed.”

    This is being argued in court, that people are impossing a cost on others for doing (or not doing) X. I come up with this very often, too. No matter how much people benefit from others choices (without which our natural condition is one of despair, intensive labor, and probable starvation), it seems that we have a natural ability to see only the negative that affects us. I guess this is Basiat’s “seen and unseen.”

  3. As for seen and unseen…. if the government had not built all these roads suited to one kind of vehicle, we might have all-terrain vehicles and even aircars and we might have had the Internet (telephone and radio with typewriter: teletype and telephoto) in 1930… who knows?

Leave a Reply