Chile stands out for routinely updating and publishing estimates of the risks and costs created by PPP-related contingent liabilities. But it is not unique in developing and maintaining sophisticated measurement techniques. Colombia has measured the risks of PPP-related contingent liabilities since the late 1990s (Lewis and Mody 1997; Government of Colombia no date). And governments in Australia and South Africa do some sophisticated quantitative analysis behind the scenes.
Should countries with limited administrative capacity attempt similar quantification? Quantifying contingent liabilities requires skills that not all ministries of finance have, at least in abundance. So ambitious attempts at quantification can run into problems. With the help of consultants, Turkey developed a sophisticated model for measuring the risks created by government guarantees for energy projects, but Jenkins (2008) reports that the government has let the model fall into disuse. It now uses only a simpler (less powerful) credit-scoring model for the limited purpose of estimating the likelihood of default in the next year.
The appropriate amount of quantification depends in part on the nature of the contingent liabilities. Chile has two dozen concessions with revenue guarantees that in sum create sizeable fiscal risks. Whether these guarantees should be offered is also contentious, so the results of the quantification could change the decision whether to offer them. In addition, the revenue guarantees are similar enough to each other to create economies of scope in measurement and valuation: once you have quantified one revenue guarantee, you don't have to do much more work to quantify the next. Measuring and valuing guarantees is less likely to be warranted when there are only a few small, diverse PPPs or when the government bears only those risks that it is unambiguously best placed to manage.
But limited administrative capacity by itself is a poor reason for not estimating the cost of a proposed guarantee of a significant risk not under the government's control. True, the government may have limited analytical resources and other pressing priorities. But an estimate doesn't always have to be complicated to be useful. For example, for PPPs that sell services to the public sector (such as independent power projects), a simple way of analyzing fiscal obligations is to treat the PPPs as public projects for accounting purposes-as in financial reporting Australia and other countries that have adopted International Financial Reporting Standards. And, if the government has no idea of the risks that the guarantee would create, it would do well either to get advice from external advisers or to avoid offering the guarantee.
Cost-benefit analysis of projects and comparison of the costs of PPPs and publicly financed projects provide useful information for decision makers and can help ensure that PPP-related contingent liabilities are incurred only for good projects. They, too, are useful only to the extent that they may influence decisions: if the decision to use a PPP has effectively been made before the comparative analysis of cost is undertaken, there are limits to the value of the analysis. However, even then, the analysis may influence the design of the project and inform the management of contingent liabilities associated with it.
At least as important as quantitative analysis is good qualitative analysis, based on common sense, economic and other theory, and a knowledge of standard practice (as well as a willingness to challenge standard practice). Here, the involvement of expert PPP units and the development of guidelines and standardized contractual terms seems useful. Moreover, involving the ministry of finance in this analysis should help ensure that possible future fiscal costs are properly considered.