Transport projects involve both construction and operational risks. Construction risks are related, inter alia, to incorrect time estimates, unforeseen ground conditions, failure to obtain necessary services, or protestor actions. Operational risks include demand risk (directly affecting revenues-e.g., in the 1990s Mexican case), interest rate and foreign exchange risks, or risks associated with hydrogeological and weather conditions.
Demand risk in the operation phase is especially problematic. In the majority of cases, reliable forecasts of future traffic flows are difficult to produce.
• One difficulty arises when other transportation modes and facilities are available. Demand can be dramatically influenced by the competition that they induce. For instance, the success of a toll road project depends on whether alternative toll-free roads are available.
• Furthermore, both user needs and, more generally, macroeconomic conditions tend to change over time. Assessing the efficiency of the firm in tackling risks is necessary to establish the exact extent to which these risks should be transferred onto the private firm.
• In transport projects, as the service is sold to end-users rather than to one or few off takers (as is the case of oil and gas, for instance), it is difficult to hedge the demand risk that private firms cannot efficiently tackle, for example, by fixing revenue levels affecting earning and cash flows. It is thus necessary to rely on other instruments to mitigate demand risk in PPPs. For instance, in highway development projects, the public sector can provide the private partners with guarantees on traffic level, in the form of traffic floors or collars (Gatti 2014).
Firms can influence demand for the service in two essential ways: (1) by exerting an effort to build good quality infrastructure, and (2) by exerting an effort to efficiently provide the service. In motorway projects, for instance, the benefit that users obtain depends on the motorway's level of safety. This, in turn, is related both to the quality of the motorway and its maintenance. In railway projects, quality dimensions such as comfort, reliability of services, and on-train services strongly impact demand.
Some quality dimensions are observable and verifiable. For example, consider train punctuality and crash rates in rail concessions; schedule reliability in bus concessions; and congestion levels and mortality rates in highway concessions. These dimensions can be contracted, and, in principle, it is not problematic to design the contract so that the firm holds responsibility for these aspects. Quality targets can be stipulated in the contractual agreement, and the firm can be motivated to meet them through rewards and punishments. This is a common practice in many real-world contexts.
Data collection on verifiable quality dimensions for regulatory and accountability purposes is now widespread for general interest services. Bergantino, Billette de Villemeur, and Vinella (2011) report a few examples in transportation sectors. In the U.S., the Bureau of Transportation Statistics of the Research and Innovative Technology Administration provides detailed information regarding departure and arrival delays for various transportation modes ranging from aviation to maritime, highway, transit, and rail. In France, the Observatoire des retards du transport aérien collects and publishes data on flight punctuality. In Italy, the regulated rail company is currently compelled to disclose information regarding arrival delays.
The effort dimensions are more problematic to measure as they are not verifiable. These are at the root of moral hazard problems.
In addition, adverse selection problems can arise. A firm may effectively hold some private information, for example, regarding the costs of the activity, which it can use to its advantage in its contractual relationship with the government. When this is the case, it becomes necessary to find contractual solutions that address the two information issues at once. In some projects, such as highway projects, the presence of adverse selection is less likely. This is because the marginal cost of providing the service is very small (close to zero) and the private party often faces the same demand uncertainty as the public party. Hence, moral hazard is the main concern in these projects.
Given the difficulty in making precise demand forecasts, the firm's profits are largely uncertain before the operation phase begins. One natural consequence of this uncertainty is difficulties in attracting private investment, especially when projects are big and private sponsors are risk averse. For instance, cross-border infrastructure has received very little attention thus far from private financiers in Europe (EC White Paper 2006). Even if private investors show up, they tend to behave opportunistically. This is possible because, when the right to run the project is awarded, they are generally required to present traffic forecasts, which are then used to define the contractual arrangements. Thus, at that stage, private firms have an incentive to present over-optimistic forecasts to obtain the right to conduct the activity. However, once this is acquired, it may then become clear that traffic flows are lower.
Changing demand parameters leads to costly renegotiation, default, or bailouts. For instance, many highways projects had their contracts renegotiated in Latin America during the 1980s, at the private operators' initiative (Guasch 2004). In a recent motorway project in Hungary, the traffic flows proved to be very low during the operation phase. The private operator responsible for designing, financing, building, operating, and transferring the infrastructure earned very little revenue and therefore stopped repaying its debt. The public partner had to intervene to take over the debt obligations and bailout the concession (European Commission 2004). The list of failed PPPs in transport is extensive and we report some additional cases in sub-section 6below.