Contingent Liabilities: Issues and Practice; Aliona Cebotari; imf working Paper 08/245; October 1, 2008

An important element of the cost-benefit analysis is whether the public sector is better 
placed to bear the risks associated with contingent liabilities than the private sector. 
This requires a comparative institutional analysis aimed at determining whether market 
16
 See at 
http://www.treasury.govt.nz/publications/guidance/costbenefitanalysis
.

---

13
institutions exist and are more effective at risk-bearing than the government.
17
These 
deliberations should take into account the possibility of “inefficient” government 
interventions due to political pressures and constraints, and “government failure” (see 
Stiglitz, 1993). 
One general principle for risk-taking decisions in the context of the institutional 
analysis above is that risks should be borne by those better able to control their source 
and for whom the protection against risk is least costly. The rationale for this lies in the 
need to reduce moral hazard behavior, by giving incentives to those who can mitigate risk to 
do so. This principle has been used, for example, to support the argument for the U.S. 
government’s need to intervene as insurer of last resort against terrorism risk, given that the 
government’s actions and the objectives it pursues could influence the likelihood and 
intensity of terrorist threats (III, 2004). In the case of private-public partnership projects, 
where risk transfer is fundamental, this principle has generally translated into the transfer to 
the private sector of project-specific risks (such as construction, operating, and 
design/technical risks) that it is best placed to manage and the assumption by governments of 
economy-wide risks (such as force majeure, regulatory and political risks), although risk 
distribution has been rightly approached on a case by case basis given the different nature of 
projects (Figure 1 illustrates a rule-of-thumb approach on how risks could be shared in PPP 
arrangements).
18
In cases where neither the public or private partner has an obvious 
advantage in managing particular risks (such as demand, financing, and project default risk), 
countries have followed different approaches in allocating them. In some cases, demand risk 
was fully transferred to the private partner (toll and shadow toll road concessions common in 
Europe), often resulting in subsequent costly renegotiations (Hungarian and Mexican 
motorway PPPs; OECD/ITF 2008), while in others demand risk was retained by the 
government, with sometimes costly outcomes for the government (Colombia).
19
Increasingly, therefore, PPP arrangements involve sharing demand risk between the private 
and the public sector (Chile, Colombia, Korea).
20
17
Good examples of such an analysis include the study by the United States’ Congressional Budget Office 
“Assessing the Public Costs and Benefits of Fannie Mae and Freddie Mac” (May 1996) or U.K.’s NERA report 
“The Economic Rationale for the Public Provision of Export Credit Insurance by ECGD” (April 2000). 
18
For a discussion of the principles underlying risk-sharing arrangements in PPPs see Hemming et al. (2006), 
Irwin (2007), OECD/ITF (2008) and OECD (2008). 
19
During the 1990s, Colombia’s PPP guarantees involved a significant number of demand guarantees, several 
of which were triggered partly as a result of the recession of the late 1990s, resulting in cumulative payments of 
2 percent of GDP by 2004 (Cebotari et al., 2008).
20
Chile, Colombia, and Korea guarantee concessionaire traffic or revenues, but the guarantees are partial and 
include features that ensure sharing of both downside and upside risks (OECD/ITF 2008). 

---

14