- Rethinking Federal Highway Spending
- Making Cost-Effective Highway Improvements
- Congestion Tolls vs. Urban Growth Boundaries
- What Is "Sustainable Mobilityï¿½"
- Speaking Engagements
Much of the heated debate about reauthorizing the federal surface transportation program this year has concerned the level of federal spending. The House Transportation & Infrastructure Committee, strongly backed by highway groups, initially sought a $375 billion program over six years, a huge increase in federal funding from the six-year $218 billion of TEA-21. But after White House veto threats, the Senate passed a $318 billion bill and the House just $275 billion, and the summerï¿½s compromise discussions focused on something near $299 billion, with signals suggesting that the White House could live with that figure.
But this entire debate has presumed that if Congress increases federal highway spending by, say, $100 billion, the net result will be $100 billion more highway spending by the time the money actually reaches the states. What if that assumption is wrong?
Good evidence that it is wrong is presented in a new study from the Government Accountability Office, released late in August (www.gao.gov/new.items/d04802.pdf). Sen. Harry Reid (D, NV) asked GAO to examine how federal-aid highway program grants influence the level of state and local highway spending. What GAO found is strong evidence for a "substitution effect," especially during the 1990s in the wake of large increases in federal aid thanks to ISTEA and TEA-21. For example, from 1998 to 2002, while federal highway investment increased by 40 percent in real terms (from $21.9 billion to $30.7 billion per year), state and local highway investment actually decreased by 4 percent (from $37 billion to $35.7 billion per year). GAO's model, designed to isolate the effects of changes in the economy and other factors, found that "state and local governments have used roughly half of the increases in federal highway grants since 1982 to substitute for funding they would otherwise have spent from their own resources."
Because Congress has given states considerable flexibility in spending the federal funds, as well as ensuring that each state gets back over 90 percent of what it puts in, "the federal-aid highway program is to some extent functioning as a cash transfer, general purpose grant program," concludes GAO. And though not mentioned by GAO, the huge increase in earmarks over the past three reauthorizations also suggests that the program has become less and less about transportation investment and more and more about handing out funds.
GAO reviews several alternatives Congress could pursue in light of these findings. To reduce the substitution effect, it could increase state matching requirements and add other requirements for continued state investment. Or it could go the other way, concluding that states know best and devolving greater authority to them. It even suggests the possibility of "devolving not only the federal program, but the revenue sources that support it," as in the "turnback" proposals of 1998.
This report raises serious questions about the current federal role in the highway system. While it's probably too late for it to affect this year's reauthorization, it should be required reading for those thinking ahead to the next round, six years hence.
Speaking of investments in highway infrastructure, to what extent are state DOTs spending these scarce funds wisely? That's the question University of North Carolina at Charlotte researcher David Hartgen sought to answer for North Carolina, in a recent policy study for the John Locke Foundation. Hartgen analyzed 349 major road projects constructed between 1990 and 2003 in that state. (You can find the report at www.johnlocke.org/policy_reports/2004100649.html).
The 349 projects cost $7.34 billion, about one-third of the total North Carolina state highway program during this time period. In addition to the cost of each project, the study estimated the trips (in vehicle-miles) each would serve over 20 years, its basic measure of effectiveness. Cost-effective projects were defined as those with a high level of vehicle-miles per dollar spent. The best projects turned out to be things like urban freeway widenings (from four to six lanes) and adding climbing lanes. Non cost-effective projects included new exits on highways in rural areas, new two and four-lane arterials, and new freeways serving less than 35,000 average daily traffic.
If NCDOT had delayed or deleted projects costing more than 5.3 cents/vehicle-mile (twice the statewide average), it would have saved $2.5 billion over the study period. That would have been enough to cover the highway maintenance shortfall that occurred during those years.
There's no denying that what some call urban sprawl (and others call low-density suburbs) affects people's transportation choices. Though suburbanization of jobs tends to follow suburbanization of housing, thereby shortening some commutes, it's plausible that those living in suburbia travel longer and further when they make non-work trips. To the extent that "sprawl" is seen as creating a transportation problem, urban planners advocate measures such as urban growth boundaries, while economists favor road pricing. So it was only a matter of time until somebody decided to compare the two, asking which one has a greater impact on travel times.
The stunning answer, from Alex Anas (SUNY Buffalo) and Hyok-Joo Rhee (Seoul National University of Technology) is that to get the same travel time improvement from an urban growth boundary as from congestion tolls would entail a deadweight loss 70 times the efficiency gains from tolls. This is the outcome of an urban land-use and transportation model the two developed and summarized in a recent paper.
Their model is interesting in several respects. For one thing, it models both commuting trips and discretionary trips, rather than focusing only on the former. Since the latter make up the majority of all trips, and are likely to be increased by suburbanization, this is very important. Second, they explicitly take into account the suburbanization of employment, rather than sticking with old-fashioned centralized employment models of urban areas.
Their basic modeling compares an urban area that funds roads via a head tax and deals with sprawl by creating a greenbelt with one that allows suburban growth but pays for roads via a congestion toll. Because the urban area is modeled as a circle, the greenbelt is (realistically) huge, encompassing 47 percent of the potential urban land and entailing large increases in densities and land rents within the boundary. When they run the model, they find that congestion tolling reduces the extent of road-building and makes the distribution of jobs and residences more compact. This creates efficiency gains of 0.22 percent of average income. By contrast, the urban boundary (designed so as to produce the same reduction in travel) produces an efficiency loss 69.6 times larger, and equal to 15 percent of average incomes.
The above modeling assumes that residents value the greenbelt at zero. The authors argue that this is not unrealistic, given that there are "vast open spaces" outside the greenbelt. But they go on to model a second case in which people do value the greenbelt itself. They find that the greenbelt would be efficient at the margin if each worker were willing to pay about $8 per year per acre of greenbelt. But since the greenbelt is huge (about 17,000 acres), this is hardly likely to be the case.
I will confess that the math in this paper is beyond me, but if youï¿½d like to review it for yourself, you can request a copy from Anas (firstname.lastname@example.org).
Several months ago, the World Business Council for Sustainable Development released Mobility 2030: Meeting the Challenges to Sustainable Mobility. This 178-page report was developed over a several-year period by consultants working with 12 sponsor companies, including most of the world's largest auto companies and several leading energy companies. You'd be forgiven for suspecting that this is just another exercise in political correctness by big business. Actually, having been involved in some of the early meetings that helped shape its content (and having now read the final product), I conclude that this is a fairly sensible piece of work.
While it never precisely defines "sustainable mobility," the report defines a dozen indicators (reduced emissions, greater safety, economizing on resource use, etc.) and concludes that especially with the projected growth of auto and truck use in developing countries, some big changes are needed over the next 30 years. It sets forth seven goals and after considering possible means of achieving them, actually devotes a (short) chapter to the problem of real-world institutions, which are subject to all kinds of political and other limitations that will make it hard to implement many of the reportï¿½s recommendations, especially in the Third World.
The report is also notable for not simply embracing every politically trendy remedy. For example, while noting the long-term potential of hydrogen fuel cells, it presents a well-reasoned discussion of the need to assess the overall emissions of every combination of fuel and propulsion system, from well-to-tank and from tank-to-wheels. And when looking at alternatives to individually owned autos, it looks seriously at car-sharing and paratransit, rather than proposing large-scale expansion of rail systems, which are neither affordable nor effective in many urban settings. And I was pleased to see a strong endorsement of both HOT lanes and congestion pricing as serious proposals for dealing with the enormous problem of urban traffic congestion.
All in all, a fact-based and thoughtful contribution to our thinking about the future of mobility. (www.wbcsd.org/web/publications/mobility/mobility-full.pdf)
I'm two-third through a very hectic fall/winter speaking season. Upcoming engagements still on the calendar are as follows:
- Nov. 12: Atty Gen. Bill Lockyer workshop, San Bernardino (CA highway finance)
- Nov.18-19: TRB Fuel Tax Committee, Washington, DC (meeting)
- Dec. 1: ACPA, Marco Island, FL (Toll Truckways)
- Dec. 8-9: ARTBA PPV Conference, Washington, DC (Toll Truckways)
- Dec. 13-16: Sacramento, SF, LA (California highway finance)
- Jan. 5: Team Florida, Orlando (Managed Lanes)
- Jan. 10-12: TRB annual meeting, Washington, DC (mega-projects panel)
Hope to see you at one or more of these!
In the last issue, I wrote about an important new paper on the benefits of Longer Combination Vehicles in the western states. This report, the Western Uniformity Scenario, is now posted on the Federal Highway Administration web site. Go to: www.fhwa.gov/policy/otps/truck/wusr/wusrindex.htm.
Finally, in case you missed it, Tony Downs this summer published a revised edition of his classic book on traffic congestion, now titled Still Stuck in Traffic. I'm pleased to report that it strongly endorses HOT lanes. It also makes the point that his well-known "triple convergence" phenomenon does not mean we should not add capacity to overloaded freeways. "[E]xpanding existing road capacity does produce significant benefits, even if it does not end or greatly reduce peak-period congestion."