In the literature on contract design, situations where the contractual parties are unable to commit to their obligations have been labelled as situations of limited commitment. Estache and Wren-Lewis (2008) illustrate that this label can be used to encompass different possible situations. First, with "limited enforcement," the firm may renege on the contract during its execution, even if the government disagrees. Conversely, in a second situation, referred to as "non-commitment," the government may renege on the contract even if this is detrimental for the firm. A third situation is referred to as "commitment and renegotiation," in which the parties commit to their obligations but if they both wish, the contract can be renegotiated at a later stage.30
Examples of PPP projects in which the firm reneges on the contract during its execution and attempts to reach a more favorable deal are pervasive worldwide, as illustrated in Section 1. In institutionally weak contexts, such as in many developing countries, the rule of law often can be circumvented. Thus, contract reneging and, possibly, renegotiation is a likely consequence. For instance, in Ghana, the current monopoly enterprise for fixed telephony entered the mobile business, despite this move being explicitly prohibited. In Tanzania, the regulator failed to enforce regional mobile licenses, and the dominant operator began to expand at the national level (Estache and Wren-Lewis 2008). A large fraction of infrastructure renegotiations in Latin America are found to occur at the firm's initiative.31 Though less common, firms also renege on contracts also in advanced economies. In principle, institutions in these countries are more solid and hence contracts are more easily enforced. For instance, a firm that refuses to produce can be fined heavily. Nonetheless, governments often prove reluctant to engage in litigation, which can be costly and time consuming. As an illustration, in France, the subsidies awarded to urban transport concessionaires have progressively increased (Gagnepain, Ivaldi and Martimort 2013).
There are (at least) two other reasons why governments may accommodate firms' requests. The first set relates to electoral concerns. When high-profile projects generating much media attention and/or projects involving critical infrastructure and services that are essential for the population are at stake, governments may be afraid of a severe backlash if the contractual relationship with the initial partner breaks up and the project's completion is delayed until a new agreement is reached with another partner. In those cases, the threat of imposing sanctions on reticent firms is, in fact, not credible. Governments end up being stuck in the partnership and keep increasing the contractual terms, as appears to have happened in some cases in India.
Corruption and rent seeking might occur as well. Politicians/bureaucrats may be ready to accept bribes from firms, together with other present or future benefits (such as the career promises for friends and relatives), in exchange for a favourable revision of the contractual conditions. In infrastructure projects, corruption may also take the form of softer ex-post price regulation, which allows both firms (through larger profits) and officials (through rent seeking) to benefit at the consumers' expense. Focusing on this form of corruption, Martimort and Straub (2008) show that reliance on a private firm may open the door to more corruption, as compared to public provision. This occurs when the shadow cost of public funds to be borne by taxpayers as long as a public firm receives budgetary subsidies is low relative to the distortion that the price raise induces-to the detriment of consumers-when a private firm is delegated the activity. When officials and bureaucrats are corrupted at various levels in the governmental hierarchy and are biased toward and/or influenced by the private sector, this dynamic would hold even if the taxation systems are largely inefficient. In general, countries with multilevel governments are especially prone to corruption. In many cases, this reflects a weak and opaque institutional framework especially at the subnational level, poor information on comparable local information, and term limits that reduce electoral discipline. Capture and lack of transparency are far from negligible issues, as regional and local governments are responsible for a large part of total national capital investments in an increasing number of countries.32
While it is expected that firms will renege on contracts, in fact, it is equally plausible that governments will lack the desire or ability to commit to contractual obligations. In developing countries, governmental failure to honor contractual terms is even more serious a concern than the firm's failure. This is because, as Estache and Wren-Lewis (2008) stress, the governments' inability to secure investors' remuneration may discourage further large-scale investments, which are desperately needed in those countries, especially in utility sectors (see also Banerjee, Oetzel and Ranganathan 2006). Political risk also heavily challenges public-private contracting in transition economies, such as in Central and Eastern Europe. For instance, in Hungary, transportation projects have been delayed by the repeated changes in political attitude toward PPPs (Brench, Beckers, Heinrich and von Hirschhausen 2005).
In environments characterized by limited commitment, the obvious reason under which either the firm or the government might renege on the stipulated contract is that this may allow for a higher payoff than would be attained if the contract were honored. At the initiative of one or the other party, a new negotiation can occur. If renegotiation succeeds, the partnership continues under a new deal. Otherwise, the partnership breaks up and the project is abandoned. Alternatively, another firm may be required to bring it to completion.
Incentives to renege on the contract arise naturally. To solve information problems, the government needs to design a compensation scheme under which the firm, while breaking even ex ante, is "rewarded" when the state of nature comes out to be favorable and "punished" otherwise. Consequently, one possibility could be that once the true state is observed by the firm and correctly revealed to the government, the firm is unlikely to be happy if the operating conditions are actually bad because, in that case, it receives the lower compensation stipulated in the contract. Another possibility is that the government is unhappy if the operating conditions are good because, in that case, it owes the firm the higher compensation that the contract prescribes. This suggests that, while offering an incentive compatible scheme is helpful to tackle information problems ex ante, this is less safe as a strategy ex post; it may well cause enforcement difficulties.
Two core points are worth making. First, the incentive issues that arise on the firm's side due to the information advantage that it enjoys vis-à-vis the government, do not exhaust the list of incentive issues that potentially challenge the overall performance of PPPs. For a proper arrangement to be set up, it is essential to consider another important temptation-that of refusing to abide by the obligations during the contract's execution. Remarkably, this temptation concerns not only the firm but also the government. Therefore, for successful PPP arrangements, it is generally necessary to find ways to incentivize both partners to behave virtuously.
Second, the contractual payoffs of the two partners underline how difficult it is to ensure the contract is honored. Whether enforcement is problematic depends on what is at stake for each of the partners in the renegotiation process (if any). Incentives to renege may appear even in the absence of information concerns that induce the government to differentiate the firm's compensation across possible states of nature.
A clear strategy is needed, together with a set of instruments, to prevent the two partners from behaving opportunistically. This requires a full understanding of how a hypothetical renegotiation process might unfold and what each party could lose and gain as a consequence.