Airports, highways, and mass transit systems are primarily state and local responsibilities. They are developed and operated by state and local governments (with increasing private-sector involvement) and funded primarily from state and local sources. Yet the federal government, by collecting transportation user taxes and using them to make grants for these systems, both raises the costs and exerts significant control over these state and local activities.

Congress should devolve transportation infrastructure funding and responsibilities to cities and states, ending federal grant programs and their accompanying restrictions. Cities and states have been open to privatization, and most would welcome the flexibility and freedom from costly federal regulations which devolution would give them. Devolving transportation funding would lead to more-productive investment, greater intermodalism, more innovation, and new capital from the private sector.

Conventional wisdom suggests that 21 states are net donors to the federal highway program and the rest are net recipients. But this paper's analysis, taking into account the real costs of federal funding and regulations, concludes that 33 states get back less than they contribute in highway taxes and would be better off if the funds were left in their states to begin with. By adding such major states as Illinois, New Jersey, New York, Pennsylvania, and Virginia to the donor-state category, this assessment could change the political dynamics in favor of devolution.

Abundant evidence now exists that federal transit programs have stimulated investment in unviable rail systems and have needlessly boosted transit system operating costs. The flexibility created by repeal of federal transit regulations would permit changes (such as competitive contracting of transit operations) that could save enough to offset much of the loss of federal operating subsidies. It would be up to cities and states to decide whether to continue to invest in non-cost-effective rail transit.

The only truly federal role in aviation is ensuring safety and facilitating the modernization of the air traffic control system. The latter can best be accomplished by divesting ATC to a user-funded corporation, as 16 other countries have done. Airports should be defederalized; all sizes of commercial airports could make up for the loss of federal grants with modest per-passenger charges. States could decide whether to subsidize unviable general aviation airports.

Overall, the federal government would retain certain coordination and safety-regulation functions in transportation. But it would henceforth leave investment and management decisions to state, city, and private decision-makers.

Table of Contents

I. Introduction

II. Decentralizing Transportation Funding

A. Why Devolve

B. The Case Against Trust Funds

C. State and Local Innovators

III. Devolving Surface Transportation

A. Highways

B. Urban Mass Transit

IV. Defederalizing Airports

V. Political Feasibility: Windows of Opportunity

A. Should Trust Funds Be Taken Off-Budget

B. The ISTEA Opportunity

C. Replacing the Aviation Taxes with User Fees

D. Continued Federal Role

E. Transition Issues

About the Author

ENDNOTES

 

I. Introduction

America's transportation infrastructurefacilities provided by state and local governments for airports, highways, and mass transitis in trouble. As documented in 1988 by the National Council for Public Works Improvement (and NCPWI's findings are equally valid today), these vital systems increasingly suffer from deferred maintenance, obsolescent technology, and inadequate capacity.

Traditionally, diagnosis of such a problem would have brought forth efforts by Congress to solve the problem via new or expanded federal programs. Indeed, the NCPWI report helped generate support for the Intermodal Surface Transportation Efficiency Act of 1991 (ISTEA), which was intended to increase federal investment in transportation infrastructure. Likewise, the Clinton administration, which came into office in 1993, pledged to rebuild America via new spending programs. Continued federal fiscal pressures, however, ensured that neither of these initiatives significantly expanded the federal role.

In November 1994 the voters signaled a major change in course, echoed by mayors and governors across the country. More than ever before, the tasks of Congress are now viewed as downsizing the federal government, balancing the federal budget, getting rid of unfunded federal mandates, and devolving authority and responsibility to states and municipalities. A Roper Center survey for Reader's Digest found 78 percent of voters agreeing with the statement that The federal government should run only those things that cannot be run at the local level. In this new policy environment, how best can the nation's infrastructure challenge be met

II. Decentralizing Transportation Funding

A. Why Devolve

There are three principal reasons for considering the devolution of transportation investment to lower levels of government. First, the responsibility for building, owning, and operating these systems is primarily regional or local, not national. Now that the Interstate highway system has been completed, the federal role in highways can be dramatically reduced, and the federal role in aviation is primarily concerned with the national air traffic control system, not local airports. There is no national interest (as apart from a regional or local interest) in whether San Francisco extends its BART system to the airport or whether Boston puts its Central Artery underground. Second, there are major disadvantages with the centralized federal trust-fund approach to funding transportation infrastructure, as will be discussed below. Third, it is cities and statesnot the federal governmentthat have been most innovative in seeking new and better ways to invest in infrastructure and improve its performance, by making use of public-private partnerships.

The first of these points needs little elaboration, except to cite the growing number of thoughtful policy experts who have come to this conclusion. These include the current director of the Office of Management & Budget, Alice Rivlin. In her 1992 book, Reviving the American Dream, she wrote, The federal government should eliminate most of its programs in education, housing, highways, social services, economic development, and job training. More recently, David Luberoff of the Taubman Center for State and Local Government at Harvard's Kennedy School wrote, Why not eliminate federal gasoline taxes and turn that taxing power over to the states Rivlin and Luberoff have been joined in this conclusion by analysts at the Heritage Foundation and the Cato Institute.

B. The Case Against Trust Funds

The prevailing paradigm for transportation infrastructure in the 20th century has been for a central government to collect tax funds (albeit generally user taxes), deposit those monies in an earmarked trust fund, and then redistribute those funds to lower levels of government to be spent on projects. Much political wrangling takes place over the precise formulas to be used in dividing up these funds, and a large central staff has grown up in each case (Federal Aviation Administration, Federal Highway Administration, Federal Transit Administration) to administer the body of rules and regulations that has evolved.

Conventional wisdom has not questioned this paradigm, despite the existence of serious infrastructure problems. Rather, it has called for increased levels of spending as the answer to the problems. But the conventional paradigm is subject to at least five fundamental problems.

1. The Pork-Barrel Problem

Federal transportation grant programsbe they airport, mass transit, or highwayare plagued by the problem of pork-barrel spending. Members of Congress traditionally derive great benefits from earmarking projects for their districts, regardless of cost-benefit ratios or the relative value of the project compared with alternate uses of the funds. Since trust fund dollars are always limited, this means that every bad project which jumps the queue at the behest of a member of Congress necessarily displaces a better project (better in terms of adding real economic value). Thus, this process systematically wastes scarce transportation infrastructure resources.

2. The Free-Money Problem

Providing federal grants that cover 75 to 90 percent of a project's cost encourages local officials to push for capital-intensive solutions to transportation problemsto build their way out of congestion. In some cases a less-costly solutione.g., an expanded bus system rather than a light-rail systemmay make greater economic sense, but if the federal program makes the costly approach look cheap, it is more likely to be chosen. In other cases, a software approach (peak-hour pricing) might make better sense than a hardware approach (another runway or freeway lane). As Harvard's David Luberoff notes, projects such as Miami's $30,000-a-rider rail system and Boston's Central Artery would never have been built if states and localities had to put up more than a token share of the money needed to fund them. The illusion of free federal money leads to decisions that would not have been taken were the local agency having to make the most cost-effective use of its own resources.

3. The Non-Pricing Problem

Traditional user taxes avoid market pricing. Thus, until the exceptions permitted by ISTEA, federal policy flatly prohibited charging tolls on federally aided highways. Likewise, the way in which the FAA interprets airport pricing policy has discouraged attempts to move toward landing fees that reflect supply and demand for scarce runway space. The results have been serious and costly problems of urban freeway congestion and serious delays at major airports. This creates the impression that the problem is insufficient freeway or runway capacity. In particular cases there may well be insufficient capacity, but the existence of considerable capacity at off-peak times indicates that the congested facility is not being fully or efficiently utilized. Peak-hour pricing would spread out peak loads, thereby reducing (but usually not eliminating) the amount of investment in new capacity required. This would be both economically and environmentally sound.

4. The Ribbon-Cutting Problem

The National Council for Public Works Improvement noted that while nearly 75 percent of current infrastructure capital spending comes from users, only about 50 percent of operations and maintenance funding comes from this source. As a result, maintenance is all too often the stepchild which must fight for annual appropriations. Former New York State Comptroller Edward V. Regan has noted that because politicians get considerable publicity and political credit for cutting the ribbons on new facilities, The incentives, therefore, are for public officials to puposefully starve the maintenance budget. Deferred maintenance will remain a serious problem until the paradigm is changed, and user-funding becomes standard for infrastructure projects.

5. The Innovation Problem

Public agencies tend to be risk-averse and oriented to the status quo. Hence, they are slow to adopt innovations. It is the private sector which is pioneering the introduction of congestion pricing on highways. It is the private sector which is taking full advantage of electronic toll collection to develop the world's first toll road without any toll booths. And it is likely to be the private sector that introduces smart highway technology, targeting upscale customers who desire in-car navigation and two-way communications as a niche market willing to pay for value-added services. Airports, air traffic control, and highways fail to make use of state-of-the-art technology because they are operated by input-oriented public agencies rather than user-friendly service businesses.

These fundamental problems lend support to the idea of changing the infrastructure paradigm to one that, as much as possible, relies on user funding, dedicated revenues, and market pricing. Devolving transportation infrastructure responsibilities to lower levels of government would hasten the adoption of the new infrastructure paradigm.

C. State and Local Innovators

As prospects for increased federal infrastructure investment have given way to likely decreases (as part of budget-balancing efforts), the federal government has attempted to encourage innovative financing and the investment of private capital. The 1991 ISTEA measure included provisions for public-private partnerships and innovative financing. In addition, the Bush administration issued Executive Order 12803 (in 1992) on infrastructure privatization, and the Clinton administration followed up with a complementary measure, Executive Order 12893, in 1994. But little real activity has been generated by these measures.

Instead, it is the states and cities that have been the principal innovators. By the end of 1995, 12 states and Puerto Rico had enacted public-private partnership measures for surface transportation infrastructure, and three private toll projects had been financed and opened to traffic. A growing number of mayors and governors are proposing to sell or lease airports and other infrastructure facilities, seeking to substitute private capital for increasingly limited public capital (so that the latter can be reserved for more inherently governmental needs). This disparity between federal and state/local governments suggests that greater innovation and new forms of private investment would occur if the federal government devolved the responsibility and funding authority for most infrastructure to the state level.

III. Devolving Surface Transportation

A. Highways

In a 1990 analysis prior to the adoption of ISTEA, transportation economist Gabriel Roth summarized the strengths and weaknesses of the Highway Trust Fund. Its main strength is that it has achieved its objective of greatly improving U.S. highways at relatively low cost to its users. But Roth also noted a number of weaknesses.

-Divided Responsibilities. Federal funding and regulations are overlaid on state highway agencies which are the actual owners, operators, and part-funders of the federal highway system. This division of responsibilities hinders sound investment decision-making and businesslike management of our highways.

-Costly Federal Requirements. States must comply with costly and burdensome regulations as a condition of using federal highway funds. Davis-Bacon Act wage provisions, Buy America provisions, and various set-asides requirements increase construction costs by 20 percent. Many other requirements, such as (recently repealed) maximum speed limits and minimum drinking age, do not directly increase highway costs but may conflict with state policy priorities. Still others, such as metric conversion requirements, increase operating costs. In addition, Roth estimates that the administrative costs of federal grant programs are on the order of 1.5 percent at the federal level and 5 to 7 percent at the state level.

-The Free-Money Effect. The availability of 80 percent or more federal funding for a new facility leads to a certain amount of gold-plating of projects funded with federal money compared with comparable projects funded solely with state funds. For example, in Phoenix, AZ those portions of the urban freeway system that are state-funded are relatively austere, whereas the federally funded portions boast landscaping, and one portion was built as a cut-and-cover tunnel with an urban park above it at considerably greater cost. Highway engineers can provide numerous examples of this effect.

-Redistribution Among States. The trust fund distributes federal funds among the states according to complex formulas which significantly redistribute resources. Many states are net donors; others are net recipients. While such redistribution may have been necessary to develop the Interstate system, its continuation is questionable now that this system is complete and operational. Supply and demand more accurately reflects the real need for additional transportation investment than the trust fund's arcane formulas.

-Use for Non-Highway Purposes. Administrations and Congress are routinely tempted to use the trust fund for deficit-management purposes. ISTEA devoted a portion of its fuel-tax increase to federal deficit reduction rather than to the trust fund. In addition, federal budgets frequently appropriate less from the Trust Fund than is collected in a given year, in order to hold down the deficit, thereby accumulating multi-billion dollar balances in the Trust Fund. This practice tends to starve the transportation system of needed investment, even though these funds can only be spent on transportation. Moreover, as of 1996, only 12 cents out of each 18.4 cents paid by highway users in gasoline taxes actually goes for highways; 4.3 cents goes for deficit reduction and another 2 cents goes for mass transit.

-Concealment of Trust Fund Costs. Records of trust fund disbursements to the states reflect the accumulated interest earned on trust fund balances and eventually returned to the states. Over the years since 1956, the trust fund has paid out 16 percent more than states paid in, thanks to this accumulated interest. However, states themselves could have earned at least this much, and had access to the funds at times of their own choosing, had the funds been left with the states in the first place. Hence, the amounts recorded as returned to each state should be adjusted downward by 16 percent to account for the artificial nature of these earnings. In addition, the federal costs of operating the trust fund account for another one percent, requiring another downward adjustment of this amount.

How would states fare if there were no federal fuel taxes and no federal highway grantsi.e., if the money were left to be collected and spent within each state To begin with, there is the estimated 20 percent cost impact of Davis-Bacon and other federal requirements. This number may over-estimate the federal impact, because 19 states have strong state-level equivalents of Davis-Bacon that would still apply if the federal requirement went away and another 12 have weak laws of this type.

For a state with local version of Davis-Bacon, how much would project costs be reduced if the federal requirement went away A 1995 survey of state transportation agencies conducted by the American Association of State Highway and Transportation Officials (AASHTO) turned up very few hard quantitative estimates. The only state that reported such a study was Arizona, which came up with an overall cost increase due to Davis-Bacon of 13 percent. Because Arizona may not be nationally representative (lower labor costs than many other states, right-to-work law), we will assume that the average impact in a state with no mini-Davis Bacon Act of its own is 10 percent. For states with a strong mini-act, we will assume zero percent impact from devolution, and assume five percent for states with a weak mini-act.

What about the balance of the previously estimated 20 percent federal cost impact The other identified regulations making up the original estimate (besides Davis-Bacon) were Buy America provisions and minority/women/small business set-asides (and possiblythough not mentionedmore-costly federal environmental requirements). Nominally, these were apparently being estimated at 10 percent (with Davis-Bacon making up the other 10 percent, for a total of 20 percent). Because some states have regulations in each of these categories, a more conservative estimate of the federal impact would be five percent.

In addition to these direct cost impacts of federal requirements, we have several other impacts:

-Following Roth, we will estimate 1.5 percent federal administrative costs and five percent state administrative costs for dealing with federal aid and its requirementsa total of 6.5 percent. This is a conservative estimate, given that state administrative costs as a fraction of construction outlays have increased from below seven percent in the late 1950s to over 20 percent by the early 1990s.

-We will estimate an average cost increase of five percent due to the free federal money effect, as discussed above.

Summing up these various categories of cost impacts, we find that states with strong mini-Davis-Bacon Acts have 16.5 percent higher costs when using federal highway aid; those with weak mini-acts have 21.5 percent higher costs; and those without mini-acts have 26.5 percent higher costs. In addition, before applying this cost factor, we must reduce the federal revenues returned to each state by the 17 percent noted above under Concealment of Trust Fund Costs essentially, the interest foregone by having the funds on deposit with the feds instead of being invested by each state (at a possibly higher rate of return). Hence, each state only gets back 83 percent of the nominal amount reported by FHWA, and the resulting funds are only worth between 73.5 (100 minus 26.5) percent and 83.5 percent as much as state dollars, because of the higher cost of using federal dollars. This gives us overall adjustment factors of 61 percent, 66 percent, and 69 percent for the three categories of states.

Tables 1 and 2 make use of FHWA data on total amounts paid by each state into the Highway Trust Fund and paid out of the Fund, over the years since 1956. The amounts paid out are adjusted by the appropriate factor depending on the existence and/or type of mini-Davis-Bacon Act in each state to produce an adjusted total paid to each state over this time period. The final number in the tables' right-most column is the adjusted ratio between monies returned to each state and monies paid in by it.

Sources: Federal Highway Administration, Reason Foundation.

 

The result is that only 18 of the 51 states plus the District of Columbia have been net gainers from the Trust Fund's redistribution of funds. All 33 other states would be better off, in dollar terms, if there had been no federal gasoline tax and no federal trust fund. For large states, the Highway Trust Fund is an especially bad deal: California has gotten back only 66 percent of what it puts in, Michigan only 56 percent, Ohio only 64 percent, Florida only 56 percent, and Texas only 58 percent.

Sources: Federal Highway Administration, Reason Foundation.

This analysis suggests that it may be politically feasible to devolve surface transportation funding and responsibility to the state level, reducing the Federal Highway Administration to a small cadre that would maintain uniform standards for Interstate system planning and design. States and metro areas, working with the private sector as they chose, would assume all responsibilities for funding, construction, and operation of highways. The federal gasoline tax would be abolished and states would be free to increase their own gasoline taxes (or other funding sources) to raise the funds necessary to maintain their system at the desired level. The remaining federal ban on tolling Interstate highways would be repealed as part of the change-over.

The benefits of such a change could be very large. Among them would be the following:

-More Productive Investment. As noted above, that portion of our highway system constructed with federal aid costs at least 21.5 percent more than highways constructed without that aid. Thus, the same total dollars invested by states and the private sector could produce more needed infrastructure; alternatively, some states might keep net investment levels the same as at present, freeing the remaining resources for other societal needs.

-Intermodalism. In conjunction with defederalization of airports (discussed below), the devolution of surface transportation responsibilities from DOT to the states and cities would get rid of the rigid modal categories of funding, in which transit funds can only be used for transit, highway funds only for highways, and airport funds only for airports. The present system has made it extremely difficult to fund truly intermodal infrastructure, such as surface transportation access to airports. Decentralizing the funding to where the needs are would facilitate the development of needed intermodal facilities.

-Freedom for Innovation. Overly prescriptive federal regulations and standards have stymied innovation in transportation infrastructure. For example, two decades of federal speed-limit mandates precluded potentially large economic gains from the time savings involved in high-speed heavy-truck-oriented tollways, such as the proposed Chicago-Kansas City Tollway and Colorado's Front Range Toll Road.

-Private Investment. The private sector has stepped up to the plate in the 12 states where the law has been amended in recent years to encourage private investment in surface transportation. Yet because of the higher costs involved, private firms have shied away from public-private partnerships involving federal highway funds. Devolving these responsibilities to the states is more likely to encourage greater private-sector investment than is further attempts to fine-tune the federal public-private partnership provisions.

B. Urban Mass Transit

Since annual federal aid for mass transit was first authorized via the Urban Mass Transportation Act of 1964, the federal government has spent some $130 billion for this purpose, in 1996 dollars. The 1964 act authorized capital grants to assist cities in taking over and rehabilitating mostly bankrupt private transit systems. Further legislation in 1971 added new rail starts to the projects eligible for up to 80 percent federal funding. Another mass transit act in 1974 for the first time authorized operating subsidies, as well, to cover up to 50 percent of a transit agencys operating losses. In 1982 one cent of the federal gasoline tax was dedicated to transit aid (for capital grants). And ISTEA in 1991 further broadened the extent to which highway funds could be shifted to transit projects. The American Public Transit Association estimates that $1.7 billion has been transferred from highways to transit during the years 1992 through 1995 under ISTEAs provisions. Overall, federal, state, and local transit subsidies since 1964 total $310 billion in 1996 dollarsnearly as much as the $329 billion cost of the Interstate highway system.

What have been the results The premise of federal transit aid was that modernized transit systems would reverse the long-term decline in transit ridership, thereby easing traffic congestion, reducing air pollution, and saving energy. Yet as Figure 1 makes clear, the use of transit by commuters has continued to decline, despite the huge sums spent by federal, state, and local governments to subsidize bus and rail services. A 1988 assessment by the Congressional Budget Office concluded the following:

After 25 years of federal aid, transit agencies have modern fleets, and many own considerably more vehicles than they need for rush-hour traffic. Yet most of the equipment in service is underused, and the federal operating subsidies go largely to pay for buses and trains running empty rather than for service improvements or fare discounts.

Indeed, CBO concluded that the ability to have up to 80 percent of capital programs paid for by someone else had biased transit systems decisions in favor of capital expenditures for bus and especially rail systems that are not cost-effective. Figure 2 depicts the results of CBOs calculation of the relative cost per passenger mile of five different transit modes. It is doubtful that many cities would have opted for the high-cost rail alternatives had extensive free federal money not been made available. Now, of course, they are stuck with paying to operate those expensive systems.

But shifting so much resources to rail transit has had the further consequence of reducing overall transit ridership. Figure 3 depicts transit ridership before and after federally funded rail systems were added to the transit systems of a number of major cities. As CBO and others have pointed out, cities that have added rail systems generally reconfigured their bus systems to feed the rail lines. But that has often had the perverse effect of making the bus system (which covers a vastly greater area) less useful for numerous ordinary trips. That, in turn, has served to reduce overall ridership. Again, the CBO report concludes that, New transit systems financed with federal aidparticularly rapid rail projectshave not lived up to their promise. Generally they have lowered the efficiency of transit service by adding expensive unused capacity. The extent of unused capacity is depicted in Figure 4, which shows the measured load factor (fraction of all spaces during hours of operation that are occupied by customers) for the same five transit modes depicted in Figure 2.

Federal transit operating subsidies have also had a negative effect. While productivity in the private bus industry has increased during the period of public transit subsidies (by 8.3 percent in the 1970-95 period), economist Charles Lave found that productivity in public transit systems plummeted during this period. Indeed, Lave estimates that if transit productivity had merely remained constant since 1964, total operating expenses would have been 40 percent lower by 1990sufficient cost-reduction to erase most of the transit operating deficit without raising fares.

DOT economist Don Pickrell links the ready availability of operating subsidies to transits declining productivity, showing that the majority of the funds made available from operating subsidies found their way into increased operating expenditures per vehicle mile, rather than to expansions of service. The federal labor-protection provisions under Sec. 13(c) of the Urban Mass Transportation Act have helped to lock in place inefficient work-rules and to severely restrict the introduction of competitive contracting of bus service, which would have encouraged productivity increases and significant reductions in operating costs.

In response to such findings, a number of transit policy experts have proposed abolishing federal transit subsidies, permitting decisions about such local/regional issues to be made at the local/regional level. And the FY 1996 and 1997 House budget resolutions included the elimination of new-start capital grants and the phase-out of transit operating subsidies by 2002. The previously cited CBO report is relatively sanguine about the consequences of such a move: Replacing [current federal transit subsidies] would require a variety of responses from local governments. Many agencies would be forced to improve efficiency, raise fares, find new local sources of finance, and/or modify services.

CBO looked both at capital grants and at operating subsidies. For the former, they concluded that The expansion needs of all but the major rail systems over the next several years could easily be met by improving the productivity of existing capital. That was before some of the new rail starts of the early 1990s (such as the Los Angeles subway), but since most such systems are being built in stages, the cities in question could decide whether to simply complete those starter rail lines and supplement them with bus service or seek to persuade state and local taxpayers that further rail construction was warranted without federal aid. Without the biasing effect of federal capital grants, more urban areas might opt to build cost-effective busways usable as well by door-to-door shuttle vans (as in Houston, Pittburgh, and Seattle) rather than non-cost-effective rail systems (as in Dallas and Los Angeles).

As for operating subsidies, CBO found that more than 75 percent of U.S. transit ridership is on systems that rely on federal operating subsidies for only 8 percent or less of their revenue. Modest productivity gains could easily compensate for the loss of federal subsidies in those large-city cases. On the other hand, the smallest 200 cities receiving federal transit aid (all but three with populations under one million) carry only 7 percent of transit ridership but account for 27 percent of federal operating subsidies. CBO notes that while cities in this group would have the most to lose from a withdrawal of federal aid, they would potentially have much to gain also: their transit services are now among the least efficient, and pressure to reduce costs could only improve them.

Cities could cut costs dramatically via competitive contracting of transit services. Wendell Cox has reviewed the worldwide experience with competitive contracting of transit service (in which firms bid for the right to operate specific groups of routes on the basis of the least amount of subsidy needed). He finds significant reductions in cost per vehicle mile, ranging from 19 percent in Copenhagen to 25 percent in Australian cities (Adelaide, Brisbane, Perth), 33 percent in San Diego, and 42 percent in London. He notes that competitive contracting has begun expanding from bus to rail, with subway service now outsourced in Stockholm and rail system contracting now planned in Adelaide and Perth. Cox has calculated that the loss of all federal formula funding ($1.9 billion in 1994) could be made up via transit system operating efficiencies averaging 11.3 percent well within the range of what has been achieved via competitive contracting of bus service in the United States and overseas.

Competitive contracting and other productivity improvements would be much easier to accomplish after the end of federal grants, because transit agencies would no longer have to comply with the labor restrictions of section 13(c), which major transit agencies have urged be removed as a costly federal mandate.

Of all transportation modes, urban transit is clearly the most obviously local and the furthest removed from being a federal matter.When this fact is combined with an appreciation of the harm done by federal transit aid, the case for shifting this function to the local level is overwhelming.

IV. Defederalizing Airports

The same general problems that characterize the Highway Trust Fund are also true of the Aviation Trust Fund. There is the same issue of divided responsibilities, with most airports owned by cities or counties and funded largely by users, but with significant federal control over how airports raise and spend their revenues, despite the small fraction of federal funding in most airport budgets. There are costly federal requirements, and long delays in receiving grants for needed improvements. There is also redistribution from large, high-traffic airports to smaller, low-traffic airports; a 1990 analysis showed that many large airports like Boston, La Guardia, Los Angeles, Newark, and San Francisco got back in annual entitlement grants less than 12 percent of what they contributed to the Aviation Trust Fund via airline ticket taxes, while other airports got back far larger amounts (Charlotte got back 36 percent, Dallas Love Field 42 percent, Memphis 31 percent).

There is also great difficulty funding intermodal airport-access projects, since there are very narrow limits on the uses to which federally aided airports can put even locally derived revenues such as passenger facility charges (PFCs). New York City has sought to develop a rail line direct from Manhattan to LaGuardia and Kennedy airports using passenger facility charge (PFC) revenues, but approval by the FAA is far from certain, despite the potential benefits to airport users (and the areas congested highway system). Thus far, the FAA has only approved the use of PFC funds for an 8.4-mi. rail link to the subway station nearest Kennedy airport; this is the first off-airport transit facility to be funded with PFC monies. By contrast, London's privatized Heathrow airport is using airport funds to build a dedicated high-speed rail line to central London, tailored to the needs of airport users.

For these reasons, in parallel with the proposed devolution of surface transportation to the states, airport funding and responsibility should be devolved to the entities (mostly municipalities) which own Americas airports. In 1987 the U.S. Department of Transportation published a study on the possible defederalization of air-carrier airports. It explored the ability of airports to increase revenues enough to make up for the elimination of federal AIP grants, the attitude of airport managers toward defederalization, and the feasibility of a PFC (which at that time was not legal). DOT researchers made use of a survey of the management of the 288 largest airports, as well as interviews with financial experts and a review of airport financial statements.

The study used the commonly accepted typology of airports as large, medium, or small hub airports and non-hub airports. The survey found that 55 percent of large hubs favored defederalization, as did 69 percent of medium hubs and 56 percent of small hubs. These airports account for the large majority of all airline service. For the smallest (non-hub) airports, only 31 percent favored defederalization. That is because federal funds make up a much larger share of the budget of the smallest airports. According to the DOT report, to make up for elimination of federal grants (as of 1985), large hubs would need an average of five percent increase in total revenues, compared with a 20 percent increase for non-hubs.

One of the main conclusions of the DOT report was that a PFC would be a feasible way for airports to make up for the loss of federal grant revenue. Using 1985 figures, Table 3 shows the estimated size of the PFC needed, on average, for each category of airport, to fully replace its federal grant revenues for that year, and an inflation-adjusted estimate for 1996.

*Source: U.S. D.O.T., AThe Effects of Airport Defederalization, DOT-P-36-87-4, February 1987.

Although Congress legalized PFCs in 1990, it did not couple this with defederalizing the airports, even on a voluntary basis. Instead, airports were permitted to levy a PFC of no more than $3, and only with the permission of the FAA following an application by the airport to that agency. Proceeds from the PFC can be used only for a narrow set of FAA-approved purposes (which cannot always be ascertained in advance of the application). Large- and medium-hub airports getting permission to levy a PFC have their entitlement grant amount reduced by half (which only exacerbates the previous degree of redistribution of funds to other airports), but their ticket tax remains the same. Moreover, since the FAA may withdraw its PFC authorization in future years, most airports have been unable to bond the revenue stream from PFCs, making this new funding source of much more limited use than had been expected.

This is a far cry from defederalizationand from what DOT had envisioned in 1987. The DOT report proposed that defederalization be implemented on a voluntary basis, so that only those airports that chose to forego AIP grants would be free to set and use their own PFC. DOT proposed a two-year transition period for such airports, during which time federal grant assurances would remain in place, so as to determine whether any of them would still be essential for safe operation and other national interests. The assumption seemed to be that unless experience proved otherwise, grant agreements would be discontinued following an airport's defederalization.

Nearly a decade after the DOT report, defederalization remains a good idea, especially now that airports and airlines have gotten used to a restricted form of PFC. A modest reform would be to implement the original DOT proposal for voluntary defederalization for large, medium, and small hubs. Those airports opting for this approach would give up AIP grants in exchange for the freedom to set their own PFCs, free of federal restrictions (and hence, presumably, the PFC revenue stream would be bondable). Their existing grant agreements would be phased out after two years.

But the same logic arguing that air-carrier airports are fully capable of supporting themselves via their own resources leads to the conclusion that all of these airports (including non-hubs) could be defederalized, regaining their freedom in exchange for the removal of federal grants and grant agreements.

Freeing airports from federal grant agreements and removing their federal grants should also be accompanied by the repeal of the federal tax on airline passenger tickets. This tax provides about 90 percent of the funds going into the Aviation Trust Fund, which is the source of the AIP grants in question. But the main use of the trust funds revenues is to fund the operations of the air traffic control (ATC) system. As recommended elsewhere, this system should be divested from the Federal Aviation Administration to an independent corporation funded by direct user fees. Some 16 other countries have spun off ATC in this manner. The remaining FAA function of safety regulation, accounting for approximately $900 million in FY 96, would be funded out of general federal revenues, as are the federal government's other safety regulatory activities.

General aviation airports are far more of a state and local than a federal responsibility, but the funding problems of GA airports would most likely be the source of the greatest political resistance to abolishing the AIP program. The GA community is very concerned about the threat of closure of GA airports, as city and county governments eye those land areas for redevelopment into higher-value (from a property-tax revenue standpoint) uses and respond to citizen complaints about noise. But how critical are AIP grants to the economic viability of GA airports Table 4 summarizes the allocation of AIP funds in 1994. As can be seen, the average grant to the 2,450 eligible GA airports is $74,000 hardly enough to make or break that airport. Looked at another way, if a state or community cannot figure out a way to replace $74,000 a year to keep such an airport open (either via some sort of state or local aid or via higher fees and charges to airport users), it truly is questionable whether that airport is serving an economically viable purpose.

Source: AAirport Improvement Program: Updates of Allocation of Funds and Passenger Facility Charges, 1992B94, General Accounting Office, GAO/RCED-95-225FS, July 1995.

 

The average grant amount to reliever airports is significantly higher, at $445,000. Relievers are a special category of (larger) GA airports located in the vicinity of a commercial hub airport. A portion of AIP funds is set aside for relievers, supposedly to reduce congestion at the commercial hub airports. Yet a 1994 General Accounting Office report raised serious questions about this premise. It found that the proportion of GA traffic at congested commercial airports declined by 38 percent between 1983 and 1991. GAO concluded that this decrease can be attributed to an overall decline in general aviation activity, not the presence of reliever airports. Officials at 22 of the 28 reliever airports GAO surveyed said that they considered reliever airports in their metro area to be underused (due to the shrinking GA market). GAO suggested either reducing the number of airports designated as relievers or eliminating the designation of relievers within AIP. Their report concluded by noting that the conditions that the reliever set-aside was created to address do not exist today, largely because of a long and steady decline in general aviation traffic. Thus, there is no good case for continuing special funding for reliever airports.

Those states that wished to do so could easily create a state GA airport grant program, perhaps funded by a state tax on GA fuel. Seven states have had experience administering such grants, under an FAA pilot program under which they administered the AIP grants for GA airports in their states. The General Accounting Office found that states adopted somewhat different criteria and priorities than the FAA, more suited to local needs. It also found that state-level administration eliminated federal-state duplication and permitted airports to obtain project approvals more quickly.

For air-carrier airports, the benefits of the defederalization would be several, and parallel to those of devolving surface transportation to the states.

-More Productive Investment. For air-carrier airports, the present AIP system shifts resources from the busiest, most-congested airports (which collect the largest amounts of ticket taxes) to relatively smaller and less-congested airports. Defederalization would permit the busiest airports to generate and keep funds for expansion. Together with the other policy changes discussed in this paper, it would attract capital (including private capital) to those airports and metro areas which most needed to expand airport capacity.

-Intermodalism. Getting rid of the federal modal categories of grant funding would permit states, cities, and the private sector to devise and fund truly intermodal solutions to transportation problems (such as better road and rail access to airports, where this made economic sense).

-Freedom for Innovation. Ending federal micromanagement of airport finances would encourage new ways of adding value to airports via intensive retail/concessions development and value-maximizing land uses on airport properties. It would also permit airports to use market-based landing fee structures to spread out their peak traffic, thereby making greater use of their present capacity.

-Private Investment. Some cities and counties have wanted to sell or lease their airports to private firms, or to franchise new-terminal development to private firms. They have been deterred by provisions of FAA grant assurances. Defederalization will remove these barriers, permitting private capital to flow into airports and airport-development projects.

The principal objection to defederalization is the argument that a federal grant system and its accompanying regulations are necessary in order to preserve and expand a national airport system. It is not clear what this term is supposed to mean when it comes to air-carrier airports. As noted previously, some cities may be under some degree of pressure to close down money-losing, subsidized general-aviation airfields in order to put the land to use for activities that do not require subsidies and will be part of the tax base. State grants for general aviation airports could help preserve that system of airports, should a state desire to make such expenditures.

But does anyone seriously contend that Albuquerque or Detroit or Norfolk or Reno is likely to shut down or drastically restrict the operations of its air carrier airport, in the absence of FAA grants and grant agreements Every mayor, city council, and chamber of commerce realizes that convenient airline service is a huge competitive advantage for a city. They are not about to kill this goose that lays golden eggs for them. And to the extent that a city or a private firm attempts to raise airport charges (whether landing fees or PFCs) to ridiculous levels, that airport would quickly lose airline service, as passengers drove to the nearest alternative airport and airlines redeployed their resources to more hospitable locations. (Indeed, Santa Barbara, California recently identified exactly this problem: a leakage of 30 to 40 percent of its passengers to airports within 100 miles such as Burbank and Los Angeles due to excessively high airline fares at Santa Barbara. Presenting a careful study of the problem to the airlines, the airport authority persuaded them to institute dramatic cuts in fares.) The problem of local control wreaking havoc on the national airport system is not something we should take seriously.

Cities, counties, and private airport operators know that well-run, accessible, reasonably priced airports are in their interest. They do not need a federal nanny to keep them in line.

V. Political Feasibility: Windows of Opportunity

The general climate of opinion in 1995-96 has favored devolution of responsibilities from Washington, DC to lower levels of government. For example, the Republican Governors Conference in spring 1995 unanimously adopted the Williamsburg Resolve, which strongly supports the devolution of responsibilities from the federal government to the states.

But in addition, two important windows of opportunity have opened for transportation devolution. First, the current federal highway and transit programs (and the underlying federal gasoline tax) expire in 1997, with the expiration of the Intermodal Surface Transportation Efficiency Act (ISTEA). Were Congress to simply do nothing, the gasoline tax would disappear and states would be left to their own devices albeit with some messy transition problems. ISTEAs expiration offers the opportunity for an orderly devolution of surface transportation responsibilities to the states and urban areas.

Second, federal aviation taxes were suspended between January and August 1996. During that period, ongoing FAA programs (air traffic control, safety regulation, and AIP grants) were funded by drawing down the balance in the Aviation Trust Fund. The suspension of these taxes led to serious efforts by the major airlines to replace them with air traffic control user fees. Although Congress has temporarily reinstated the taxes (through the end of 1996), the debate on future aviation funding remains open.

An additional issue arose during 1996: taking the transportation trust funds off budget. This proposal would impede, rather than assist, transportation devolution.

A. Should Trust Funds Be Taken Off-Budget

HR 842 passed the House in spring 1996, with the enthusiastic backing of nearly every transportation interest group. It would take all the transportation trust funds out of the federal budget, on grounds that since these funds are derived solely from transportation user taxes, and can only be spent for transportation purposes, the recipient highway, transit, and aviation spending programs should not be constrained by federal budget-balancing efforts.

As many critics have pointed out, taking these programs off-budget would remove them from needed congressional scrutiny. One of the consequences would be that pork-barrel spending, which plagues all three modes of transportation, would be even less subject to challenge than it is today. In addition, the major inequities between donor and recipient states that are built into current allocation formulas would be preserved. And all of the defects of centralized trust funds, noted in Section II, would remain.

David Luberoff of the Taubman Center at Harvard's Kennedy School says, Instead of taking the trust funds off-budget, Congress should take a cue from the social policy debates and devolve them to the states and localities. Why not eliminate federal gasoline taxes and turn that taxing power over to the states Since having good transportation is a critical factor in ensuring regional economic prosperity, states would have a great incentive to raise their gas taxes to build the infrastructure they really need. Former OMB privatization director Ronald Utt, now at the Heritage Foundation, agrees. The Wall Street Journal editorialized against taking the trust funds off-budget, arguing that instead, when ISTEA comes up for reauthorization in 1997, Congress should consider truly taking the Highway Trust Fund off-budget by devolving the responsibility and revenue base for maintaining roads back to states and communities.

A companion Senate bill to take the trust funds off-budget went nowhere in 1996, but observers expect the idea to resurface in 1997. Within the frame of reference presented in this paper, taking the trust funds-off budget would be a step backwards, preserving the many problems inherent in the status quo.

B. The ISTEA Opportunity

With ISTEA reauthorization in prospect for 1997, the battle lines have been shaping up in 1996. Already a coalition of self-defined donor states has created STEP 21, which would convert surface transportation funds to block grants and guarantee that every state gets back at least 95 percent of what it contributes to the Highway Trust Fund. A more radical devolution measure has been introduced by Sen. Connie Mack (R.-FL) and Rep. John Kasich (R.-OH). Over a two-year period, it would phase out all but two cents of the 14-cent federal gasoline tax, devolving that taxing authority and program responsibility to the states. There would be no federal restrictions on how states could use their gasoline tax monies. The remaining two-cent federal tax would be used to oversee maintenance of the Interstate system, for roadways on federal lands, and for Indian reservation roads.

Officials of several large states have been speaking out strongly in favor of devolution. For example, Ohio Gov. Voinovich's chief of staff Paul Mifsud in 1995 said, We are fighting to the death to get our money back from Washington. Ohio has launched an aggressive campaign to eliminate the federal government as the middle-man and allow the states to retain the gasoline taxes it now collects for the federal government. In 1996 the Florida DOT has worked closely with Sen. Mack on his devolution bill. Strong support also exists in California, where legislation already exists that would automatically increase the state gasoline tax to replace, penny for penny, any decrease in the federal tax. State senator Quentin Kopp, who chairs the Senate Transportation Committee, sponsored joint resolutions in 1995 asking Congress to eliminate all federal fuel taxes and all federal programs for highways and transit. California's 1996 Commission on Transportation Investment formally endorsed full devolution of transportation funding and responsibility to the states. And Gov. Pete Wilson's administration is strongly supporting the Mack/Kasich devolution measure.

How serious are devolution proposals likely to be taken in Congress in 1997 A lot will depend on the positions taken by various interest groups. One of the most important is the American Association of State Highway and Transportation Officials (AASHTO). At its annual meeting in October 1995, AASHTO laid out a strongly decentralist position regarding ISTEA reauthorization. President William Burnett wrote that We are in an era where more and more responsibilities are being shifted from the Federal to the State governments ... Most governors and state legislators support this shift, along with most of Americas voters and business leaders.... They have lost faith in many of the one size fits all solutions.

Burnett has been challenged by the pro-environmentalist, pro-transit Surface Transportation Policy Project, which wants a next-generation ISTEA measure that increases the shift of federal funds from highways to transit. STPPs views are generally supported by transit and urban groups. But at least one major environmental group has endorsed devolution: the Environmental Defense Fund. EDFs Thomas Graff, a member of Californias Commission on Transportation Investment, stated I support devolution of the federal gasoline tax and Highway Trust Fund to the states. This recommendation is fully consistent with my earlier expressed support for moving the principal responsibility for planning and funding of transportation to the regional and local levels wherever possible.

In fact, although groups like STPP would prefer a more pro-transit ISTEA, if their choice boils down to a more pro-highway ISTEA versus devolution, they may well opt for devolution. It seems highly unlikely that urbanized states such as California, Illinois, New York, and Pennsylvania would abandon mass transit in Los Angeles, San Francisco, Chicago, New York, or Philadelphia if the funding and responsibility were devolved to them from Washington, DC. Transit advocates might decide that they had a better shot at maintaining their slice of the transportation pie at the state levelgiven the mechanisms now in place such as metropolitan planning organizations and state transportation investment plans.

The trucking industry is another key player. The American Trucking Associations strongly oppose the diversion of highway user taxes to transit (which increased under ISTEA) and the use of such funds to support the large federal highway and transit bureaucracies. Moreover, despite their historic opposition to tolls, the ATA testified before Congress in late 1995 in favor of a federal privatization measure that would permit highway privatization with tolls, provided tolling were limited to cases of (1) new capacity and (2) major reconstruction. Thus, the ATA might support a devolution measure (with its attendant likelihood of greater use of tolls and privatization by the states) if it meant a higher fraction of fuel taxes ended up being spent for highway purposes. ATA would likely be more comfortable about devolution if some degree of federal pre-emption were included in the measuree.g., to ensure uniform national truck length and weight standards on all Interstate highways.

What about the underlying political arithmetic, based on the relative proportions of winners and losers among the states The conventional accounting has produced the generally accepted notion that only 21 states are donor states. But as Tables 1 and 2 make clear, when the figures are adjusted to take into account the full costs imposed by the federal program, 33 of the 51 participants are actually donor stateshence, net losers from the program. Among those added to the loser category by this analysisand hence potentially recruitable to the devolution sideare Arizona, Colorado, Illinois, Minnesota, New York, New Jersey, Pennsylvania, and Virginia. Clearly, the electoral arithmetic strongly favors devolution if this papers recalculation of winners and losers gains acceptance.

The Mack/Kasich proposal is a major step toward full devolution in surface transportation. But by leaving in place a residual federal gasoline tax, it opens the door to future abuses. Many environmentalists seek massively high gas taxes as a way of penalizing auto use and reducing smog. In future years, it would be much easier to add to a two-cent federal gasoline tax than to impose a brand-new federal tax on gasoline where none existed. As one noted transportation consultant put it, if the federal power to tax gasoline is available, it will be used.... [and] as long as a pile of other peoples money exists in Washington, people will do silly things with it. Hence, a pure devolution would be preferable to the 86 percent devolution proposed by Mack and Kasich. Federal roadway responsibilities on federal lands and Indian reservations would be better met out of general tax revenues or from user fees on those respective lands. Federal safety regulation (e.g., the National Highway Traffic and Safety Administration) should continue to be funded out of general revenues, like other federal safety agencies.

C. Replacing the Aviation Taxes with User Fees

As noted previously, the legal authorization for collecting the 10 percent airline ticket tax, the $6 per passenger international departure tax, the 6.25 percent air-cargo tax, the 15 cent per gallon general-aviation gasoline tax, and the 17.5 cent per gallon GA turbine fuel tax all expired at the end of 1995 and were not collected for the first eight months of 1996. Interim funding of FAA operations by drawing down the balance in the Aviation Trust Fund depleted the majority of that fund's uncommitted balance. Congress's temporary reinstatement of the taxes leaves future aviation funding up in the air.

This uncertain funding outlook has brought forth proposals to replace the existing taxes with some form of user fee. The chairman of the Senate Aviation Subcommittee, Sen. John McCain (R., AZ) has proposed the creation of a special commission representing all aviation interests to develop a user-fee structure. The seven major airlines in June 1996 put forth their own proposal for a system of user fees. And the Reason Foundation proposed a set of air traffic control user fees, to be paid directly to a not-for-profit ATC corporation which would be outside of government and controlled by aviation users.

This situation presents an opportunity not merely to spin off ATC, as some 16 other countries have done successfully, but also to abolish the federal trust fund and grant program. The FAA would become simply a federal safety regulator, funded by general revenues (at about $900 million per year, the current budget level for that portion of FAAs activities). Commercial airports could replace the lost AIP funding via their expanded ability to make use of PFCs. And states could decide whether or not to assist their GA airports via state grant programs.

D. Continued Federal Role

This paper has sought to demonstrate that federal taxes and trust funds are unnecessary and inappropriate for the creation and operation of airports, highways, and urban transit systems, and that the one truly national element of transportation infrastructure the air traffic control system ought to be divested to a user-owned and user-funded corporation. There remains the question of other appropriate federal roles in transportation. Are there certain aspects of transportation which are truly national in scope or otherwise appropriate as federal concerns Four plausible federal functions are examined here.

1. Federal Lands

It goes without saying that for those portions of U.S. territory under the direct federal jurisdiction, the federal government should continue to have the responsibility for providing transportation infrastructure. This would include military installations and defense-related nuclear-weapons facilities operated by the Department of Energy as well as national forests and national parks. The costs of infrastructure for these federal lands should be included in the budgets of the relevant agencies, with the revenues derived either from users (e.g. timber companies needing logging roads) or from general revenues (e.g. for defense installations). None of these purposes requires any tax on general highway or airport users.

2. Safety Regulation

Air travel is clearly a national industry, as is interstate truck, train, and bus service. Uniform federal safety regulation, by such agencies as the Federal Aviation Administration, National Highway and Traffic Safety Administration and Federal Railway Administration, does not conflict with the devolution agenda laid out in this paper. The clear-cut separation of safety regulation from funding and operations is one of the hallmarks of policy reform in such countries as Australia, New Zealand, and the United Kingdom, all of which are cited as models for reinventing government by such bodies as the Clinton administration's National Performance Review. Such safety regulation should continue to be paid for out of general federal revenues.

3. Uniform Standards and Regulations

Probably the most difficult problem for devolution advocates is the issue of (non-safety) standards and regulations. On one hand, one of the legitimate driving forces in favor of devolution is to gain relief from overly costly and intrusive federal regulations. On the other hand, legitimate concerns arise in our major transportation industries over potential state or local restrictions on the functioning of national network operations. For example, local governments might impose a new set of curfews on airport operations, wreaking havoc with the ability of overnight express carriers (e.g., Federal Express) to do business. States which already impose a patchwork quilt of different requirements on truck weights, lengths, and configurations might adopt even more conflicting requirements. It is argued that without the carrot-and-stick approach of being able to grant or withhold federal funding, the federal government will be unable to restrain states or cities from engaging in such costly regulations.

These are reasonable concerns that will need to be addressed in any devolution legislation. But these fears appear to be exaggerated. The underlying federal authority stems not from grant programs but from Congress's power to regulate interstate commerce. This power was included in the Constitution specifically to permit the national interest in the free flow of commerce to override obstacles to such commerce thrown up by individual states.

Thus, Congress would appear to have the authority, under the Commerce Clause of the Constitution, to impose uniform truck weight, length, and configuration requirements on th