For What the Tolls Pay: Fair and Efficient Highway Charges

Hydrogen cars, expensive oil, fuel efficiency standards, and inflation frighten those interested in maintaining and improving U.S. highways. All of these forces could erode the real value of fuel taxes that now are the largest single source of funding for highway programs and an important source of transit funding as well. Because of this worry, the Transportation Research Board convened a committee to carefully examine the future of the fuel tax.

The committee uncovered both good and bad news. The good news is that there is nothing structurally wrong with the fuel tax that will cause the real value of revenues to decline dramatically over the next couple of decades. The bad news is that it is a very crude way to raise revenues for our highway system. Switching to per-mile fees, the committee concluded, would be a much more efficient and equitable approach.

Looking at the good news first, worries that alternative fuels and improving fuel efficiency will undermine the finance system are definitely exaggerated. Radical improvements in efficiency will take a long time to develop and be implemented, and even less radical improvements, such as hybrid engines, affect fuel consumption very slowly because it takes so long for new models to replace old models in the U.S. car fleet. Moreover, Americans are addicted to oil partly because they are addicted to power. If you make an engine more efficient, they will want it bigger. Consequently, improving technology does not reduce real fuel tax revenues per vehicle mile nearly as much as one might think. Indeed, they have been roughly constant for a long time.

One cannot be quite as certain regarding the future price of oil. There is some possibility that demand may erode because of an upward trend in the price of gasoline. Department of Energy projections (which have been generally consistent with those from other prominent sources) are optimistic that the price of oil will not surge over the next 15 years or so. But it must be admitted that energy experts did not anticipate the recent price increase to over $60 per barrel.

However, the evidence strongly suggests that recent oil price increases are as much the result of geopolitical forces as they are the result of fundamental supply shortages. It is true that China and India are becoming major oil consumers as they grow rapidly, but it is also true that supplies are increasing. There may be limited supplies of the type of oil that we pump from the ground today, but as one expert puts it, the sources of oil will just become heavier and heavier. If light crude runs out, we’ll turn more to heavy crude. If that becomes scarce, tar sands will be exploited more fully, and if they become expensive, we’ll turn to oil shale. In the process, oil will become more expensive, but it will be a slow process. Of course, wars, boycotts, and other disturbances can cause major price spurts that make optimistic forecasts look foolish, but one has no choice but to base long-run forecasts on fundamental trends, and they are not alarming.

The imposition of severe fuel efficiency standards could upset the gasoline-powered apple cart, but new radical regulation seems politically implausible in the near future. Currently, our two political parties are so closely competitive that no one wants to ask the American people to make major sacrifices. We may be addicted to oil, as the president suggests, but as Mae West remarked, “Too much of a good thing can be wonderful.”

Inflation concerns

The possibility of accelerating inflation raises more of a political as opposed to a technical concern. The federal fuel tax is a unit tax. That is to say, it does not vary with the price of gasoline as would a percentage sales tax. Inflation therefore erodes the purchasing power value of the tax. Some, like the Chamber of Commerce (in the National Chamber Foundation’s 2005 report Future Highway and Public Transportation Finance), have suggested indexing the tax for inflation. However, that solution may not be politically sustainable. Politicians at the state and local levels often suspend indexing if it becomes the least bit painful.


Historically, federal and state politicians have compensated for inflation by periodically raising tax rates. There is some question whether this is possible in the severe antitax climate in which we live today, but if this is a problem, it has nothing to do with the basic structure of the fuel tax. It is a political problem afflicting all forms of taxation.

But it should also be noted that politicians have not been strongly pressured by inflation in recent years. First, the inflation rate has been extremely low by historical standards. Second, at the federal level, the government has been able to capture additional revenues for the highway system without raising tax rates. In 1993, the federal gas tax was increased for the express purpose of reducing the deficit. The proceeds were not to be spent on highways or anything else. In 1997, those revenues were redirected into the highway trust fund and are now available to finance highway expenditures. More recently, an ethanol subsidy that was previously financed out of the highway trust fund will, in the future, be financed out of general revenues, thus releasing more resources for highways.

Congress may now have run out of such devices for increasing federal highway funding, which supports about a quarter of all highway spending. It will be interesting to see how Congress reacts in the future, especially if inflation accelerates a bit. In addition, many think that the most recent federal highway bill will more than spend the earmarked revenues that are available, although this is a controversial issue. If true, that, along with more inflation, may pressure Congress to return to its historical practice of occasionally raising the fuel tax when the federal highway program is reauthorized.

Per-mile fees

Although there are few reasons to fear a rapid erosion of fuel tax revenues in the near future, major revenue increases also seem unlikely. Congress and the state legislatures could raise more revenue with the gas tax if they chose to do so, but the political opposition is formidable. That makes it unlikely that enough will be spent in the near future to improve highway quality significantly, and the nation will have to continue to live with the current level of congestion. But relying solely on increased highway expenditures to reduce congestion is probably not cost-effective. Congestion must also be attacked by imposing extra costs on those who cause it.

Whether the nation just wants to maintain the quality of the current system or to improve it, there is good reason to reform our current approach to financing. In searching for alternatives, there is a strong argument for sticking with the established principle that users should pay and that the resulting revenues should be dedicated to highway expenditures. The revenues collected should be related to the costs that the vehicle imposes on the system, including congestion costs. In an extreme version of the principle, all the revenues and no more should be spent on highways, but the present practice of dedicating some revenues to mass transit certainly is defensible, because mass transit expenditures benefit highway users by reducing congestion.

The current fuel tax is only vaguely related to the amount of wear and tear that a vehicle imposes on the road, and it does not vary with the level of congestion. Per-mile fees that vary with the type of vehicle and time of day would be much more efficient and equitable.

Fifteen years ago, it was not possible to think about collecting per-mile fees efficiently. Costs included constructing tollbooths, paying toll takers, and most important, waiting in line at the tollgate. New technology holds the promise of virtually eliminating such costs.

In the immediate future, developments such as the EZpass electronic toll collection system (used on many toll roads and bridges throughout the northeastern states) and license plate imaging greatly increase the opportunities for tolling at low cost. We should exploit these opportunities to the extent possible.

In the longer run, global positioning system (GPS) technology makes it theoretically possible to charge for every road in the country, with fees varying by type of vehicle and the level of congestion. Of course, we may never wish to go that far, and much research is necessary before committing to that path. It is necessary to determine what type of technology is most efficient and to develop safeguards that will assure the public that their privacy will be protected. It is also important to resolve the many problems that will arise as we move from the current system of financing to something completely new. The necessary technology is not costless to develop, but it is very cheap. It is possible that GPS systems will be installed in almost all new cars in the near future, even if they are not required for the purpose of levying a per-mile fee.

The president’s 2007 budget proposal agrees that new forms of highway funding are desirable. It requests $100 million for a pilot program to involve up to five states in evaluating more efficient pricing systems. The necessary research has already started with an experiment in Oregon, and the Germans have initiated a GPS system for levying fees on trucks on the Autobahn, the national motorway system.

An improved pricing system not only has the potential for greatly increasing the efficiency of using existing roads, it can also be helpful in guiding the allocation of new highway investment. If a certain segment of road is yielding revenues far in excess of the cost of building it, it is a pretty good indication that an expansion of capacity in the area is warranted. If, on the other hand, revenues are not sufficient to pay for costs, any request for new construction should be critically examined.

Although such a system holds the promise of implementing the economist’s dream of perfectly pricing the highway system, it would be naïve to believe that a perfect system could ever be implemented. The per-mile fees will be set by politicians operating in a political environment. There will be strong pressures to keep fees low just as there are pressures today to avoid fuel tax increases. In some cases, there will be legitimate arguments for subsidies. For example, the nation may choose to subsidize rural road networks much as it now subsidizes mail service to rural areas.

The equity argument

Many will question charging per-mile fees out of a concern that it will impose a special hardship on the poor. As the notion of charging for road use is discussed more and more, there are many derogatory comments about “Lexus lanes,” as though only the rich would benefit from a reduction in congestion. It can be noted that it is frequently extremely important for poorer people to get to work on time or to pick up their kids from childcare before overtime fees are charged. But such arguments do not resolve the problem. Some people will be worse off as the result of a per-mile fee, and some of the people who are worse off will be poor.

It is not uncommon to face tradeoffs between economic efficiency and a concern for equity. But there are better ways to protect the poor than to prevent a major improvement in the efficiency of our transportation system. If it is determined that fees particularly hurt the poor—and more research on this question is probably warranted, given that the poor also pay the current fuel tax—policies that make the earned income credit or other welfare programs more generous can be considered. If it is deemed desirable to target additional assistance more precisely on poor highway users, a toll stamp equivalent to the food stamp program might be contemplated, although administrative costs would be very high. It may not be worth it to try for very precise targeting. The basic point is that there are other ways to deal with poverty that are more efficient than not charging properly for roads.

Expanding tolling now would acquaint people with the concept. It is easier to start levying tolls on specific lanes when there are alternative lanes that are free. That will make the public aware of the benefits of congestion pricing. If there are no howls of anguish, politicians might be less inclined to oppose road pricing.

Many years ago, the economist William Vickery began extolling the virtues of per-mile fees that would vary with the level of congestion. Having been trained in engineering originally, he went so far as to provide detailed discussions of complex systems that would put wires under the street for the purpose of measuring the distance traveled by particular cars at different times of the day. He died tragically just before traveling to Stockholm to receive the Nobel Prize in economics. At the time, we were on the edge of developing new technology that could turn his dream into a practical reality at low cost. Wherever he is, he must be smiling.

Your participation enriches the conversation

Respond to the ideas raised in this essay by writing to [email protected]. And read what others are saying in our lively Forum section.

Cite this Article

Penner, Rudolph G. “For What the Tolls Pay: Fair and Efficient Highway Charges.” Issues in Science and Technology 22, no. 3 (Spring 2006).

Vol. XXII, No. 3, Spring 2006