Skip Navigation Planning & Markets
Subscribe Submission Requirements Editorial Board Archive Links Search Home


Support for semi-privatized highway policy via franchising and concessions has been rising. Direct government provision has produced neglected and bad roads, partly due to public choice problems,(13) further encouraging planners to favor semi-privatization schemes (Chetwynd, 1993).(14) Concession toll highways are considered to be more efficient, faster to build, a way to alleviate negative externalities (and perhaps some public choice problems), and empirically successful in the United States (Lutz and Bartlett, 1995). They are often profitable too. Government agencies have themselves tried to capitalize by selling tollway rights. For instance, a portion (4.4 miles) of a New Jersey turnpike was sold "to a state authority to raise money" in order to ostensibly help balance that state's budget (Boroughs and Collins, 1992).

Gabriel Roth argues that under public provision users "face increasing congestion", and the system is suffering from deterioration of pavements and public choice impediments (Roth, 1991). Indeed, the climate of public opinion may be changing, with many people viewing toll roads as choice-enhancing solutions (Haldane, 1997), and many businessmen interested in participating. In Canada, the construction industry has been lobbying for admittance into the highway market, although in Alberta semi-privatization is evidently viewed as "radical" (Jenkinson, 1993).

The lack of confidence in direct government provision by Chileans follows an international trend favoring less reliance on public transportation infrastructure, which has been criticized by scholars across the political spectrum. For instance, Henry Holmes argues that it has promoted "economic injustice and environmental degradation" and have "fractured" communities (Holmes, 1996). While some praise public transportation systems over extending automobile use, such as in the case of the light rail system in Los Angeles (Gofman, 1994), many others chide them as wasteful failures of government planning (de Moraes, 1995).(15)

The Key Concern: Traffic Congestion

One of the primary negative externality targets of Chilean highway concession policy is congestion reduction. The social costs of congestion are great and growing worldwide, and public mass transit systems are not averting them. Thus, semi-privatization and reliance on markets has become attractive (Samuel, 1997). For instance, California's congestion problems have reached dramatic proportions (Manasian 1990). Perhaps hundreds of thousands of hours are wasted daily in traffic jams in Los Angeles alone (e.g., Moe, 1994), where emission levels are two and one half times their level under free-flowing traffic conditions and losses to time and productivity amount to $14 billion annually (Schiller, 1998).(16) Robert Poole cites a Texas Transportation Institute study suggesting that "Americans wasted $48 billion in 1992 stuck in urban traffic congestion" (Poole, 1996). Dawn Sauter notes that "an estimated 70 percent of the nation's rush-hour commuters must navigate stop-and-go traffic conditions" and, citing Bruce Ingersoll and Alice Reid, the total cost of congestion comes to $820 a year per capita in Washington D.C. (Sauter 1997, pp.1-2). In Chile, congestion is hardly trivial, and improvements to infrastructure are needed. Massive and prolonged congestion still occurs in Santiago, even on freeways like Avenida Kennedy, where vehicles may be stalled well over an hour due to incapacity to manage heavy rainfall and roadway emergencies.(17)

Cars stuck in traffic jams cause up to three times the pollution they would otherwise cause, but planners are impeded from resolving them on account of knowledge and public choice problems, particularly special interest pressures.(18) For example, a knowledge problem arises in predicting traffic flows and knowing the profitable (optimal) amount of road to build. Indeed, in Mexico, inadequate knowledge led planners to set inappropriate toll rates, resulting in fiscal disaster (Emmons and Brand 1993, pp.7, 11). A growing number of transport analysts are writing about public choice problems associated with government provision and its monopoly power over roads (see Peter Samuel, 1997). For instance, if "telematics" were adopted (i.e., policies of "smart cars" and "smart highways"), rent seeking auto makers would be able to boost profits from each sale since cars would have more electronic gadgets. Clearly, special interest groups or firms like General Motors, Lockheed Martin, and Bechtel have been handsome beneficiaries of large government grants to establish platoon driving, which would enable a driver to read a magazine while the car's electronics take him to his destination automatically (Robb-Nicholson, 1996). Nevertheless, despite these planning pitfalls, Santiago planners hope to direct infrastructure concessions to improve the quality of life, saving commuting time by reducing congestion with electronic toll collection and congestion pricing.(19)

Toll Roads and Toll Collection

Toll roads are certainly not a novel concept. In the United States, government-run road building has relatively recent roots, beginning around World War I (Weingroff, 1996). Toll roads were utilized successfully in Britain two centuries ago (Malone, 1990), and spontaneously emerged in California in 1850, with about 150 being operated through 1902. Accordingly, they may be a means of "enlisting private-sector entrepreneurship into the field" to control congestion problems (Klein and Yin, 1994). However, semi-private toll roads are not a panacea. Some are not successful, as evidenced by the Dulles tollway near Washington, D.C. which flounders under low quantity demanded.(20) Sauter considers the Dulles failure to have been caused by planning and forecasting errors (which were not fully disassociated from the political process). She also thinks, that the Dulles tollway serves as an example of how the price system does work to allocate resources efficiently (Sauter 1997, pp. 89-119, especially 118).

A pervasive technological problem with toll roads has been the cumbersome nature of toll collection via toll booths or coin drop machines. However, recent scanning technology has relieved this burden. In Orange County, California, antennas are used to scan microchips in cars traveling as fast as 100 mph, and highway users are automatically charged the appropriate toll (D'Souza, 1995). Similar "smart card" technology has been considered in Great Britain (Malone, 1990), and some have discussed smart cars and smart highways as possible technological breakthroughs that would improve toll collection (Rhode, 1992). However, reducing lag times at toll booths with scanner technology will not be a sufficient means to alleviating congestion problems in and of itself (Riley and Fresco, 1996).

Congestion Pricing

What seems to be needed is congestion pricing, i.e., charging higher tolls for roadway use during peak traffic periods. Congestion and other roadway problems emerge because road use is not priced (Giese and Aron, 1993). Scholars like Poole and Roth contend that roads would be better if privatized and congestion pricing were utilized (Poole, 1996), which was proven successful when tried in Southern California (Schiller, 1998). Despite some political opposition, congestion pricing could be a sensible policy. Kim Clark offers that, "peak traffic pricing could serve as one of those terrific examples in which the invisible hand stirs up benefits for everyone-- faster commutes, cleaner air, financial rewards for investors, and savings for taxpayers" (Clark, 1997).

Mohring (1998, pp.189-191) reminds us that, transport policies of subsidies, reserved bus lanes, at least a limited use of meters, and especially congestion pricing (which would have the same effect as staggering work hours, p. 215) could increase social efficiency. Indeed, HOV lanes and mass transit, which are essentially "bribes" and "grants" to consumers (p. 192), could be efficiently displaced by congestion pricing. Thus, he argues that "introducing congestion pricing to an urban road network would substantially increase the efficiency with which it operates" but he laments that "many problems must be solved to ensure that congestion pricing could make (almost) everyone better off" (Mohring 1998, p. 212).

Randall Holcombe argues that infrastructure policy which ignores the incentives and institutions it creates will have adverse effects. For example, congestion may not be alleviated by well-intentioned highway planning. This would certainly be true when planning reduces opportunities for application of the Coase Theorem (21) (although Holcombe does not mention it ), and thus planning might produce inefficiencies and negative externalities. Plus, planners may be misdirected (being constrained by the knowledge problem) in deriding "urban sprawl" since forcibly congesting downtown areas may be worse than sprawling. Thus, he suggests that tolls would be an effective market mechanism for reducing congestion (Holcombe, 1990). In Holcombe's view, private toll roads would be more efficient to run, would likely be built faster, and would ease congestion problems. Using tolls on government roads would also serve to alleviate congestion when applied appropriately. Funds to build roads must come from somewhere, and tolls seem to be a reasonable alternative to other taxes. Moreover, tolling roads also provides a post-construction opportunity to see if the project really is self-sustaining.

Even some who otherwise advocate top-down transportation planning have recognized the sensibility in coupling planning with market-based congestion pricing (Spiegler, 1995). Others suggest arranging an admittedly difficult "intelligent alliance of state and market" where governments lead while the price system is relied on to reduce congestion and social costs (Flagg, 1992).

However, a few critics are leery that pricing might well rationalize increased automobile use and bring greater environmental harm. They say this despite acknowledging that (a) electronic tags for scanning are 99.7% accurate, (b) congestion pricing would be beneficial, and (c) overall automobile use might be reduced (Raubner, 1994). There might be policy concerns about practicality too. Norwegian experts have estimated that peak hour prices would have to be four to five times as great as off-peak rates in Oslo for congestion pricing to have a significant affect (Robb-Nicholson, 1996). Conformably, Héctor Gutiérrez argues that there are limits to the effectiveness that tolls can have in eliminating congestion (Gutiérrez 1996, 146-147). Similarly, since accidents and other traffic incidents account for about half of all highway congestion (Schrank, Turner and Lomax, 1993), the most congestion relief that can be reasonably sought by means of congestion pricing is a 50% reduction.

Nevertheless, market solutions such as tolls and concessions might be optimally utilized through prudent planning. Some enhancements suggested by Gutiérrez include bidding to obtain the lowest toll charges, eliminating government guarantees, reinforcing property rights, and contingent clauses in contracts which permit governments to purchase concessions and allow expansion of capacity (Gutiérrez1996, pp.146-147). Something else to be avoided, according to Pablo Serra, is fixing toll rates, since that would cause "excessive rigidity" and thus lower social welfare (Serra 1996, p.225). Furthermore, potential public choice problems must be kept in check. Some toll road operations, such as in Orange County, California, have been criticized because the compensation packages for bureaucrats overseeing the tollway projects seem very high (Platte, 1995), perhaps indicating political favoritism and rent seeking (22) along with other inefficiencies.

page 13



USC Seal

Main Page | Subscribe | Submission Requirements | Editorial Board | Archive | Links

ISSN 1548-6036

Copyright 1999-2000
University of Southern California
Los Angeles, California 90089-0626