October 29, 2007
Series Editor: Ian Parry
Managing Editor: Felicia Day
Assistant Editors: John Anderson and Adrienne Foerster
This week RFF senior fellow Winston Harrington discusses to what extent the federal government is involved in financing highway construction and maintenance, and whether these decisions are better left to state governments. Given the large amount spent on highways each year, Americans will bear a substantial cost if this money is spent inefficiently rather than on highway projects with a large benefit to road users. Next week, Mark Delucchi of the UC Davis Institute of Transportation Studies will discuss his research on the military and other costs Americans incur to protect oil supplies from the Persian Gulf.
Does the Federal Government Spend too Much for Highways, or too Little?
Certainly, there is ample justification for some federal funding of highways. There is plenty of intercity and interstate travel, both commercial and personal, which means that the benefits of a well-integrated road network are not just local. Just note the number of out-of-state plates you see on interstates and other major thoroughfares. But what's the right amount?
Those who say "too much" note that the Federal-Aid Highway Program makes grants to the states that cover 80 to 90 percent of the costs of qualifying highway projects. Can anyone claim with a straight face that out-of-staters enjoy 80-90 percent of the benefits of the average highway? To this group, which includes a lot of regional planners and anti-sprawl advocates, this is a major subsidy to build roads. With the federal government paying such a large cost share, the argument goes, local and state governments don't have to make the hard choices between whether the project is really justified.
Others say the federal funding share for specific projects seriously overestimates the federal involvement in highways. In terms of total highway spending, the Federal-Aid Highway Program grants for states, which now amount to $30-40 billion, account for only about 25 percent of total spending on highways. Moreover, if federal highway subsidies are excessive, why is it that road use is growing so much faster than capacity? Between 1990 and 2003, for example, road use increased by 2.3 percent annually, compared to a 0.25 percent annual increase in highway lane-miles. One answer is that while your state may get 90 percent of the cost of a new section of interstate, it won't change the total disbursements to your state.
Unfortunately, there is another rapidly growing and often ugly federal influence on highway funding and that is the rampant use of congressional earmarks. In SAFETEA-LU, the most recent highway authorization act, there were 6,000 line-item, named projects, valued at almost 10 percent of the total funding allocated. Half the funding was added at the last minute, during the House-Senate Conference. Those of us who are not lobbyists think the use of earmarks has gotten out of hand, and no more so than in transportation. Earmarks do not increase the funding available to each state; rather, they direct the allocated funds to particular projects. No one has a clear idea how the earmarked projects are selected; they are simply inserted into the legislation at the last minute, without review or comparison with other projects. The potential for poor decisions, not to mention outright corruption, is pretty high.
So what would be a better approach? One increasingly popular option is road pricing. In the Washington area, the Dulles Greenway was built with private funds, and plans are moving forward with private funding of HOT (high occupancy toll) lanes to be added to the Beltway and Interstate 95-395. In addition, there is talk of cities and states selling existing public roads to private operators, as Chicago is considering doing with the Chicago Skyway. On a more experimental level, debate is now underway about implementing road pricing on a large scale in the public sector. For example, New York City is considering implementation of "cordon pricing," charging stiff daily fees for driving into southern and central Manhattan.
Making users pay the full social cost of road use, including the incremental cost of adding capacity, automatically takes care of federal concerns about adequate revenue generation, since out-of-state users will have to pay. However attractive road pricing is in principle, it still faces serious political barriers and practical problems, especially as it becomes more widely used. Unless there is a carefully planned transition, it is likely to generate serious affordability concerns, not only for the poor but for others who just happen to face huge tolls because of previously made choices of where to live and work.
Absent comprehensive road pricing, information about both the local and national benefits of specific transportation projects would be needed to develop appropriate federal highway subsidies. A subsidy equal to the difference between the two would provide the right incentive. Perhaps there would be a way to combine estimates of the nonlocal benefits of such projects with a demand-revealing pricing mechanism, such as a competitive auction among local or state governments, in order to elicit their willingness to accept certain subsidy levels to begin a project.
Views expressed are those of the author. RFF does not take institutional positions on legislative or policy questions.
To receive the Weekly Policy Commentary by email, or to submit comments and feedback,