Setting the "right" toll for a targeted toll facility or network is straightforward in theory but often difficult in practice. First, setting tolls based on congestion, road damage, and other costs can be difficult, as the right prices are hard to determine and implement. Second, there is the potential conflict between the price that optimizes the use of the road or network and the price that generates sufficient revenue to pay for it, as the optimal price can vary over the life of the road.56
For targeted tolling and pricing, the most common tolling methods use a limited set of rates that do not vary by time of day and thus may not cover the full costs of system use. The cost of building, maintaining, and operating a road or network can be combined with information on traffic levels and theories on price elasticity, as well as other revenue goals and policies, to determine appropriate pricing. But setting pricing rates is politically challenging and incorporates a much greater range of issues than simple cost calculations. Some kinds of private participation may facilitate easier toll rate adjustments over time, but even concession arrangements typically include guidelines or restrictions on price setting.
Setting rates for a comprehensive road pricing system could present more challenges. Determining the optimal pricing system would require information on levels of congestion, its variation by location and time, and the amount of road damage a vehicle does. This in turn could lead to a complex set of prices that might be confusing to users. Prices set on this basis are not likely to generate exactly the revenue needed for maintenance and expansion of the existing system. Over time, the price signals could guide investment to bring these closer in line, but initial applications would likely be more problematic.
Identifying the optimal set of prices on a network is also complex and challenging. In particular, recent research indicates that setting the same price on all lanes of the same freeway may not be appropriate. A Brookings Institution report examined ways to set prices, recognizing that drivers are not homogenous. It identifies the so-called limited two-route HOT network as providing the highest consumer gain. This option calls for tolling all lanes but at two different toll levels (one free-flowing and one more congested), with carpools free in either lane.57 Extending these complications out to full road pricing would compound the complexity.
In addition, some jurisdictions may be tempted to set prices on particular roads to maximize revenue from out-of-jurisdiction travelers. For example, if a road traverses only part of a state, has a high share of out-of-state travelers, and there are few if any alternative routes, state officials may be tempted to set prices higher there in order to maximize revenue from non-residents.
Another hurdle is the interplay of comprehensive charges that could be implemented at various government levels. Technology can allow simultaneous federal, state, and local mileage charges via the same platform, but administering the distribution of funds and operating costs for the system would be more complex than current targeted tolling systems.
In spite of these challenges, pricing does not have to be perfect to be successful. While charging optimal price levels will be difficult, pricing is highly likely to lead to more effective management of the road system and to generate revenues that are closer to costs than the current surface transportation funding system produces. As one analysis of real-world experience with road pricing stated: "There is enough knowledge on the monetary values of externals costs to start with pricing reform in the transport sector. The remaining gaps are being filled rapidly by ongoing research."58 While this statement may be overly dismissive of the challenges, costs, and complexities associated with implementing pricing, enough is known to move to the next level of planning for the introduction of comprehensive pricing in the United States at some point in the future.