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

Box 3. Estimating the Expected Cost and Market Value of Guarantees 
The methods for estimating expected costs or market value range from educated guesses, to market or 
historical data, to quantitative models, such as option pricing and simulations (these are discussed in 
more detail in Annex II). 
The choice of method depends generally on the availability of data and the cost of the method. 
Valuations based on market data are used when the borrower issues debt quoted in the market or when 
it is relatively easy to find comparable companies that do so. If these are difficult to find, simulation 
methods are frequently used instead, but these may be time consuming and expensive to develop, and 
hence may be cost effective only if the guaranteed amounts are large. Otherwise, if the guarantees are 
relatively small, an option pricing model may be a good substitute (Hagelin, 2003). In cases where data 
are so limited that neither method is feasible, a simple classification of guarantees into high, medium, 
low, or very low default risks (with associated probabilities of default)—based on available information 
or educated guesses—could be employed to assess expected losses. Any reasonable approach will 
produce better estimates of the cost of loan guarantees than the cash-based approach that will always 
assume zero cost in the budget year. 
Countries that price guarantees, therefore, generally use all of the above methods. Sweden, for example, 
uses market data, option pricing and simulations for pricing guarantees. Chile, Colombia, and Peru use 
simulations to estimate contingent liabilities associated with minimum revenue guarantees under PPP 
arrangements. Chile also used option pricing methods to value exchange rate guarantees under PPP 
arrangements, when these were provided. The Federal Deposit Insurance Corporation in the U.S. uses 
expected loss estimates derived from historical and institution specific loss data, and Korea uses 
investment rating agencies to assess the likelihood of payments on explicit contingent debts.