Maintaining existing roads and bridges in the United States requires an annual outlay of $20 billion more than current local and federal spending. Additionally, expanding the road network to accommodate the growing economy may require $40 billion per annum more than currently budgeted for road construction. To meet that demand, some states are turning to the private sector for construction and management of toll roads.

Changing attitudes

Planning and construction of private toll roads grew 7 percent worldwide between October 1997 and October 1998. By that time, 475 roads valued at $228 billion were being planned, and construction had started on 188 of them, mostly in Asia and the Far East. China's investment in toll roads exceeds that of either North America or Latin America. In order to stimulate growth and commerce among the states, federal policy generally discouraged the use of tolls until the 1900s. Most toll roads were built in the years following World War II by the states or by special state authorities.

In 1956, the Federal Aid Highway Act provided for federal funding of the interstate highway system --- a non-toll system. The lack of encouragement for toll roads has meant that, by 1989, such roads comprised just 4,657 miles of the 3.8 million miles of streets and roads in the United States and only 2,695 miles out of the 44,759 miles of the interstate system.

In 1991, the Intermodal Surface Transportation Efficiency Act signaled a shift in federal policy. Tolls are now permitted on all federally funded roads except those that are part of the interstate system. Additionally, private toll roads can receive federal money in the form of states lending their federal aid to private developers. The new transportation law, TEA-21, continues that policy.

The change in federal attitude has prompted some states to look at the efficacy of permitting the building of private toll roads. For example, Georgia used tolls to finance the construction of Georgia 400, an extension connecting I-285 to I-85. The road, which collected $55 million in tolls over its first four years, has been proclaimed a success by the state DOT. In the late 1990s, Texas began developing the President George Bush Turnpike around Dallas to connect with four freeways and the Dallas North Tollway.

Modern toll roads, however, have little in common with their predecessors. For instance, today's toll projects involve public/private partnerships; they always include a non-tolled alternative road; and ownership always shifts to the public, either immediately or after some period. The private sector has been tapped to help in the construction and operation of toll roads for a number of reasons: * Many of the resources made available through the federal Highway Trust Fund are channeled to other uses; * Tight government budgets prohibit capital funding of new roads; * The trend of the last 20 years has been toward smaller government, exposure of government to market forces and reduced regulation; * Technology in toll collection and monitoring of roads for accidents or congestion has boomed; and * Better construction materials allow for the substitution of capital (technology) for labor, yielding lower costs that lead to higher returns on investment.

California takes the lead

According to the Denver-based Reason Foundation's "1999 Privatization Report," 15 states have legislation permitting private toll roads. This year, South Carolina began construction of its first project, and Virginia has approved several new projects. In Washington, voters approved a plan to privately develop a second span of the Tacoma Bridge.

Still, many states are not completely comfortable with the idea of private participation in the toll road business. For example, Washington is considering six projects for public-private development, but progress has been slow. Minnesota and Arizona also have considered privatization, but, again, progress is not clear. In general, there is substantial opposition to use of tolls especially when most other roads are not toll-financed.

Despite that discomfort, however, private toll roads are making a run into the public consciousness. In California, for instance, the 10-mile 91 Express Lanes were constructed within the median area of SR91 to connect with the 55 freeway near Anaheim and to run east-west to the border of Riverside County. (During the 1980s, traffic on SR91 increased at an average annual rate of 8 percent, growing from 91,000 vehicles per day in 1980 to 188,000 in 1989.)

Before construction of what Californians call "the 91," both Orange and Riverside counties had supported plans to build additional capacity for SR91. But it was not until 1989, when the state passed a law permitting private companies to build and operate four private toll facilities, that the California Private Transportation Company (CPTC), a subsidiary of Houston-based architectural firm CRSS, proposed to build and operate the 91. The CPTC was seeking a project with few right-of-way requirements, relatively easy environmental approvals and sufficient demand so that profits could be realized fairly early. The 91 was expected to be relatively low cost, at $88.4 million, a figure well within the capability of private finance.

Additionally, the local population was affluent, congestion was a problem, and time-savings were valued enough to make investment in a private toll road reasonably attractive. Electronic toll collection, the company reasoned, would keep labor operating costs low.

The CPTC built the 91, which has no intermediate entrances or exits, and then ceded ownership back to the state in exchange for a 35-year lease to operate the road. The company has complete pricing flexibility outside of its commitment to permit toll-free use for vehicles with three or more occupants. Additionally, it is subject to the maximum profit constraint of 17 percent return on equity.

According to the California Transportation Department, the 91 has cut commuting time substantially. In fact, the department estimates that peak traffic periods in the morning and afternoon have been reduced by one hour, and entrance ramp backups that, on occasion, necessitated a half-hour wait, have been greatly reduced or eliminated.

The 91 represents the first full implementation of congestion pricing in the United States. Additionally, it is the world's first fully automated toll road. On the eastbound highway, peak drive time occurs Monday through Friday from 3 p.m. to 7 p.m. During its first year of operation, the toll during those hours was $2.50. The lowest toll of 25 cents was in force from 9 p.m. to 7 a.m. daily. The CPTC guarantees freely flowing traffic, or it refunds charges.

In 1996, the 91 accommodated more than 5.7 million vehicles with an average occupancy of 1.65 passengers, a relatively high figure for California. More than 20 percent of peak time drivers used the toll route, andmore than 20 percent of those were in carpools with three or more occupants. On weekdays, 25,000 vehicles traveled on the 91. By the end of the first year, the 91 project had operating revenues of $7.07 million, $730,000 more than its operating expenses. By 1997, weekday volume grew to 30,000, and gross revenues rose to $14 million.

In January 1998, the CPTC began to charge tolls discounted by 50 percent based on its franchise agreement on carpools with three or more occupants. By August of that year, the company reported that the 91 had broken even for the year in the first six months. Today, the toll road carries more than 250,000 vehicles per day --- so many that, at peak hours, it is operating beyond its design capacity.

The Greenway

On the other side of the country, the Dulles Greenway, a 15-mile extension of the Dulles Toll Road, connects the Beltway (I-495) around Washington, D.C., with Dulles International Airport. The Greenway does not employ congestion pricing or offer off-peak travel discounts, nor does it have HOV lanes. Opened in 1995, it was specifically designed to get travelers to the airport and to provide a link to Fairfax and Loudon counties in Virginia.

The original toll road opened in 1984 with tolls set at 7 cents per mile. Sufficient demand existed that the road was profitable. That profitability, plus growth in the nearby suburbs, convinced Virginia to build the extension. Its DOT, however, decided not to build a public road and awarded the franchise to the Toll Road Corporation of Virginia (TRCV).

The TRCV will operate the Greenway for 40 years, after which the road becomes state property. The Greenway, meanwhile, is subject to utility-style regulation by the state's corporation commission with a target return on equity of 21 percent.

The Greenway opened with an initial toll of $1.75. Traffic during the first year was only 10,000 vehicles per day as opposed to the predicted 30,000. The cost posed problems for motorists who were reluctant to pony up $1.75 when the original Dulles Toll Road still was charging only 85 cents for about the same distance. Additionally, tolls on the Greenway are not prorated, so a motorist exiting at one of the seven interchanges before the end of the highway still has to pay the full amount.

In spring 1996, the toll was reduced to $1, and ridership grew to 17,000. A year later, the toll was increased to $1.15 where it has remained. Daily ridership has grown to between 30,000 and 35,000, but toll collection remains below anticipated levels.

The road also has been subject to extensive regulation. For example, Greenway officials wanted to raise the speed limit on the road from 55 to 65 miles per hour, an approval process that took substantial time and required an act of the Virginia legislature. Furthermore, state regulators and lenders have to approve toll restructuring.

Private problems

In general, consumers have shown some reluctance to pay tolls that do not vary with distance and/or time of day. Private toll roads also encounter substantial difficulties in negotiating for land acquisition and rights-of-way, and in dealing with environmental matters.

Private developers lack the ability of government authorities to use eminent domain provisions to acquire necessary land. Negotiations for the Dulles Greenway, for example, took much more time than anticipated, adding to the project's cost. The 91 Express Lanes did not have that problem because the required land was in the median of the existing highway.

Private toll roads also cannot issue tax-exempt securities --- an obstacle that raises developers' interest expenses and, ultimately, the required tolls. Additionally, expensive projects that are contingent upon stimulation of land use or induced traffic have a lower chance of success. Equity investors presumably want returns on their investment, and bondholders require their interest payments, but traffic forecasts and traffic development are both uncertain. The substantial risk to lenders probably means that interest expenses will be high, requiring hightolls, and further reducing the likelihood of success.

Government regulation of private toll roads also presents an obstacle since it can stifle price flexibility, reducing the benefits of for-profit operation. Regulation also involves negotiation and paperwork, adding to the cost of operation. Simple incentive regulation, which sets the maximum prices that can be charged, could be an appropriate type of regulation. Liability for accidents or other problems is another roadblock for private investors. Public highways have the advantage of sovereign immunity, which serves as a shield against liability.

A few facts

Understanding a few facts can help local government officials make better decisions about the value of tapping the private sector for a toll road project: * A private road requires reliance on existing demand and not on future demand induced from land use development. Initial costs are high, while future returns in their discounted form are usually insufficient. * Metropolitan roads that serve high peak-time traffic are more likely to be financially viable than intercity roads. * Prices need to be demand-sensitive to both distance and time of day. Prices should vary to avoid congestion. * Private toll roads that require a large initial investment and a long time for completion are risky. * Most private investments have alternative uses if the intended use fails. There is no alternative use for a private road that fails financially, so the investment becomes a sunk cost. * One private entity should build the road and operate it for a long period of time. * Success in completing the process of planning and final construction of the road depends upon committed political champions --- governors, county supervisors and influential legislators. * Private developers are more likely to succeed when no rights-of-way multiple ownerships exist, and environmental problems are small. Delays involve substantial interest costs to private developers.

Edwin Blackstone and Simon Hakim are senior fellows at the Privatization Research Center and professors of economics at Temple University, Philadelphia. They can be reached by e-mail at or by telephone at (215) 204-5037.