As the creation of a PPP involves delegation of some tasks from the government to a private firm, a natural question is whether this can be done at no cost-and if so, under what circumstances. This depends on the extent to which the partners' interests are initially aligned, or can be aligned in the stipulated contract.
PPPs face the immediate difficulty of the existence of information asymmetries between the government and the firm. Hence, these must be considered in contract design. In many situations, during the execution of the contract, the firm is (or becomes) better informed than the government regarding not only some relevant aspects of the activity, but also regarding its own actions that can impact those aspects. For instance, the government cannot observe (or, even if it does, no third party, such as a court of justice, can verify this) whether the firm exerts the specific level of effort that is desirable from the social perspective in building the infrastructure. As providing effort is costly for the firm, but the degree of effort cannot be specified in contracts, a moral hazard problem arises; this is usual when the source of private information is endogenous. That is, the firm has an incentive to shirk from exerting effort during the construction phase to maximize returns.
In addition, the government is unlikely to observe the exact conditions under which the firm manages the activity once the infrastructure is in place. For instance, it may not know whether the service demand or the production cost is high or low. In contrast, the firm will learn this information by the time the project is in operation. This divergence in knowledge levels is the root of an adverse selection problem, as usual when the source of private information is exogenous. That is, the firm has an incentive to cheat, vis-à-vis the government, regarding the conditions under which it actually operates, because this allows it to increase its profits.26
The two information problems are generally linked. This is due to the presence of synergies between project phases, which is one of the main reasons for which various tasks are bundled in a unique activity and entrusted to a single responsible firm. The effort that the firm exerts during the construction phase impacts the conditions it faces during the operation phase. For instance, exerting effort may increase the likelihood of facing a high demand for the service (because the infrastructure is more reliable) or a low cost of production (because the cost is an inner characteristic of the infrastructure). This is why effort provision by the firm is desirable.27
From a standard agency theory, we know that moral hazard is not an issue (and can be handled at no cost) as long as the firm is risk neutral and not protected by limited liability. Nor is adverse selection an issue if contracting occurs ex ante, i.e., when not only the government but also the firm is uncertain regarding the future operating conditions, as is very often the case with PPPs. Under these circumstances, offering a state-dependent compensation scheme will allow the government (1) to prevent the firm from exploiting its informational advantage and (2) to implement the efficient allocation (namely, recommend the efficient output level and give up no rent to the firm). Differentiating the compensation to the firm across states of nature is useful for incentive purposes.
This is backed by the well-known revelation principle (Gibbard 1973; Green and Laffont 1977; Dasgupta, Hammond, and Maskin1979; Myerson 1979). In principal- agent relationships, there is no loss of generality for the principal (the government, in this context) in restricting attention to direct revelation mechanisms. "Directness" of an incentive mechanism resides in whether the agent (the firm, in this context) has no other actions to take other than merely reporting private information to the principal (or, equivalently, picking one particular option within a menu of contractual options, each tailored to a different possible state). To make such a mechanism "truthful," it is necessary to construct it in such a way that the incentive compatibility constraints of the agent are satisfied. This involves motivating the agent to announce the information correctly to the principal, rather than to camouflage it (see, for instance, Laffont and Martimort 2001).
On the one hand, moral hazard requires that the firm bear some risk. The firm is not motivated to engage in costly effort unless it faces a sufficiently significant penalty for non-compliance, while being assigned a sufficiently large reward for good performance. A compensation scheme with this characteristic mirrors the need to transfer, in the words of OECD (2012), a "sufficient and appropriate" amount of risk to the firm. In long-term PPP agreements in which not only the level of output and the compensation to the firm are contractual variables, but also the termination date is stipulated, the firm should be allowed to enjoy the benefits of its effort for a sufficiently long period of time (see, for instance, Iossa and Martimort 2008, and Danau and Vinella 2014).
Adverse selection requires that firm compensation should be sufficiently higher for good performance than bad performance, though not excessively higher. The former requirement discourages the firm from claiming that performance is bad when, in fact, it is good, while the latter prevents the reverse.28
The bottom line is that, as long as no friction arises other than the two information problems described thus far, delegation to a risk-neutral private firm generates no agency costs for the government. The firm can be induced to deliver the efficient level of output without the need to concede any information rent to it. This goal is achieved by designing a sufficiently dispersed compensation scheme, under which the firm breaks even ex ante, obtaining a higher return when the operating conditions are favorable and a lower return when they are not.29 This conclusion might lead one to believe that, after all, it is not very difficult to setup a successful PPP because information issues can be handily circumvented. However, an important clarification is in order.
The conclusion above is drawn under the implicit assumption that both the government and firm fully commit to contractual obligations within the PPP arrangement. However, in practice, the partners are often unable to do so. Difficulties then arise with contract enforcement. Consequently, delegation to the private firm is more problematic and may become costly. One should thus try and understand how the contract, which decentralizes the efficient allocation under full commitment, can be made self-enforcing as it is implemented.