Internalizing the Externalized: The Case of Roads

Samuel Staley's picture

In a previous blog post, my discussion of externalities, public goods and roads spurred an unexpectedly lengthy set of posts and repostes. In this article, I want to address a trickier topic: Whether road users have effectively shifted the burden for paying for roads to non-users and whether the reason we pay for roads out of general taxes is a result of that lobbying effort.

Some planners have referred to this process (if I interpret their comments correctly) as "externalizing" the costs of providing services and goods. Essentially, the idea boils down to whether a private good is paid for by the general public when the general public doesn't benefit from its provision; it's a version of privatizing the benefit while socializing the cost. At its core, this is a question of subsidization.

I'll continue to use roads because that's the case in which the term emerged, and this topic seems to create the most controversy in the planning community. And, let me acknowledge from the outset the obvious: The vast majority of roads are subsidized in the sense they are funded by government. The question is the degree of subsidization by non-users and whether those subsidies should continue.

On the surface, the case for automobile users externalizing the cost for roads makes sense. Tolls, gas taxes, and other user fees made up just 52.5 percent of the revenues spent on highways nationwide in 2008 according to the Federal Highway Administration (Table HF-1).  The share ranged from just 28 percent in Arizona to 82.3 percent in West Virginia. Moreover, the U.S. General Accountability Office recently found that federal general fund transfers to the Highway Trust Fund have been significant enough to make every state a "donor" state based on pure dollars flowing in and out of the account.

But, this breakdown is a little too facile. Not all roads serve the same purpose or lend themselves to user-fee applications. Moreover, the public benefit properties of roads have been recognized for centuries (and even discussed by Adam Smith in the Wealth of Nations). The U.S. Constitution explicitly enables the U.S. Congress to authorize funding for Post Roads. The economic returns to the investment in the Interstate Highway System ran into the double digits during its early decades (financed by the gas tax) as the efficiencies of improved mobility translated into stronger bottom lines for U.S. business. Road building dramatically curtailed urban traffic congestion to the point only Los Angeles faced severe regional traffic congestion in 1982. The returns to highway investments have declined in recent decades. While the causes are not well known, many analysts believe the increased politicization of road funding and diminishing marginal returns to investments as a national network play important roles.

As a practical matter, if a strict user fee approach to road finance had been in place prior to the mid-20th century, quite simply very few roads would have been built, let alone improved or adapted to changing economic circumstances. A fascinating short history of U.S. tollroads and their economic demise in the 19ths century can be found in Daniel B. Klein's extensive work and his summary at with John Majewski.) The economic and social implications of severely curtailed road building would have been decidedly negative.

But, as the saying goes: Then was then, and now is now.

Personally, I support privatizing the entire road system. I've been publicly supportive of a distance-based fee (probably based on vehicle miles traveled) that is set to ensure free-flow travel. I believe that a price at this level would be "market clearing," eliminate the need for public subsidies, rationalize modal choice, improve land-use efficiency base on more transparent transportation costs, and generally improve urban economic competitiveness. (By implication, our failure to price roads has led to a general oversupply of roads although the local circumstances vary.)

Nevertheless, the reality is that systemwide road pricing is not quite technically feasible, currently impractical, and economically infeasible for most of the road system in 2010. The technology is close (within 20 years), as the recent road-user fee pilot project in Oregon demonstrated, but we aren't quite there yet. Limited access highways? Yes. Collector roads? Maybe. Local roads? No.

Which gets us back to the original issue at hand: If mobility has general social and economic benefits, and roads are key elements of providing that mobility, than the idea that we fund roads largely because a narrowly defined group of users have manipulated the process to off-load those expenses to nonusers is pretty weak. We tend to forget that the basic skeleton (and design) of most urban road systems predated the widespread use of the automobile. Moreover, in the case of federal interstate highways at least, the gas tax covered virtually all the construction costs until recently. Indeed, a recent Reason Foundation analysis of a re-purposed Highway Trust Fund suggests that gas taxes could cover the needs of the federal Interstate Highway System if current gas tax revenues were rededicated to that function. So, in the broad sense, users paid (and could continue to pay) for federal interstates. 

But, roads are more than federal interstates.Local and neighborhood roads provide different levels of service, and we don't yet have the technology to apply real user fees to these roads. Local roads are still largely public goods for practical purposes.

So, are road users externalizing their costs? Road investments have historically predated the widespread use of the automobile and modern-day urbanization in concept if not scale. The grid street pattern pre-dated the automobile as well as the system of roads knitting rural villages and towns to urban areas.

I don't believe road users externalized the costs of roads as much as governments recognized their social benefits (for commercial purposes as well as passengers) and the practical inability of the private sector to provide those facilities and services. As a consequence, governments dedicated funds to building what was considered social infrastructure that served rural and urban purposes.

Now, the rationale for non-user based public expenditures on roads and highways is weakening, and weaker by the decade with each new spurt in technology. While we can expect intense lobbying by current users to retain the subsidies they have, that's not the same as arguing that the reason we subsidize public infrastructure is because of special-interest lobbying for a redistribution of funds away from non-users to support their own narrow projects. The more productive discussion, I believe, is over whether the rationale for public funding of roads exists any longer and how to phase it out.

Sam Staley is Associate Director of the DeVoe L. Moore Center at Florida State University in Tallahassee.



Reply to HCGoddard

Thanks for the link to Weitzman, which I will read when I have time.

I did not mean to say that economists always have tunnel vision, only that, when they calculate the ROI of investments in controlling global warming, they are forced to have tunnel vision by their methodology of converting environmental costs to money costs and discounting future costs to calculate PV - which forces them to discount costs to virtually zero if they are a few years in the future.

"if one assumes a scenario of catastrophe, that is, infinite damages, discounting infinite damages at a positive discount rate still yields infinite damages,"

No one is assuming infinite damages. From Stern to Nordhaus, they assume finite damages and quibble over the discount rate.

"Stern is using a low discount rate for the reason that it is not ethically proper for a present generation to project its time preference (discount rate) on to unborn future generations. As a result, he gets a result that suggests we should take drastic action now."

You are saying that he comes up with the conclusion that we should control global warming on ethical grounds, and then he chooses the discount rate that gives him the conclusion he wants. Reminds me of one of my old mathematics teachers, who got equations to come out right by using what he called "the finagler's constant."

I agree with conclusion that Stern reaches on ethical grounds. I don't think it makes sense to finagle with the numbers to pretend you have reached that conclusion on the basis of an economic calculation.

In fact, even with the finagling, his numbers imply an unethical conclusion: that we should value the cost of global warming in 3000 years at less than one cent per year.

"Using a tax on carbon is the same as cap and trade. That is, starting with the proposition that we cannot properly evaluate the future damages of climate change, a widely accepted view, then what we have to do is choose some physical target, such as no more that 2 C degrees warming..."

I think you misunderstand me here.

I am criticizing the common idea that we should optimize output by calculating the environmental costs of different forms of production and charging a tax that internalizes them. You agree that we cannot do this when you say that we cannot properly evaluate the future damages of climate change.

I am saying that, instead of calculating the cost of climate change, we should set a cap where the damage is tolerable. You agree about that also.

We can keep emissions below that cap with a tax, a trading scheme, or some other form of regulation. I actually support cap-and-trade rather than a tax because it is more feasible politically, but that is irrelevant to the point I am making here.

Here, I am just saying that we need to control global warming by capping emissions at a level that avoids damage, not by calculating the total cost of damage and charging a tax based on that cost.

But what does your statement that "we cannot properly evaluate the future damages of climate change" say about the calculations of ROI on investments in controlling climate change? All those calculations depend on giving a money value to the future damages of climate change and using a discount rate to determine their PV.

Based on what you say now, I think you must conclude that we should stop trying to calculate the ROI on these investments. Instead, we should decide on ethical grounds that it is imperative to control global warming, and we should cap emissions at some safe level.

Of course, once we make that decision to cap emissions, it does make sense for economists to calculate which are the most cost-effective ways for us to reach that level.

But because ""we cannot properly evaluate the future damages of climate change," it does not make sense for economists to make calculations like those that Wodehouse cites, saying that there is an ROI of only .9 on investments in controlling climate change.

And because ""we cannot properly evaluate the future damages of climate change," it also does not make sense to accept Stern's calculations, even though they use a discount rate that gives us the conclusion we want.

Charles Siegel

Weitzman On Economics of Global Warming

I read the Weitzman paper at and I think he makes a good point.

It is similar to a point I made on one of my earlier posts:

"Even if we do discount these future costs, many economists have argued that it is economically justified to control global warming as a sort of insurance policy, Global warming can have a range of possible costs over the next century or two, and the worst plausible costs - hundreds of millions of people dead and billions of people displaced from their homes - are so serious that it makes sense to spend a few percent of the world's GDP as insurance against them."

Weitzman adds an important technical point: that there is a fat-tailed PDF, which is a greater danger than a long-tailed PDF. He also points out correctly that it is dangerous to take a wait-and-see approach because there would be long delays before remedial actions had any effect.

Thanks for the reference to this paper.

Charles Siegel


Prepare for the AICP* Exam

Join the thousands of students who have utilized the Planetizen AICP* Exam Preparation Class to prepare for the American Planning Association's AICP* exam.
Starting at $245

Essential Readings in Urban Planning

Planning on taking the AICP* Exam? Register for Planetizen's AICP * Exam Preparation Course to save $25.
Habitat blockitecture tower

Build the world you want to see

Irresistible block set for adults when placed on a coffee table or desk, and great fun for kids.

Stay thirsty, urbanists

These sturdy water bottles are eco-friendly and perfect for urbanists on the go.