Chapter

9. Contingent Liabilities

Author(s):
International Monetary Fund
Published Date:
June 2003
Share
  • ShareShare
Show Summary Details

Introduction

9.1 The financial crises of the 1990s highlighted the shortcomings of conventional accounting systems in capturing the full extent of financial exposures arising from traditional “off-balance-sheet” obligations, such as contingent liabilities, and from financial derivatives contracts. The discovery of the magnitude and role of these obligations in these crises reinforced the need to monitor them. This chapter focuses on contingent liabilities.1 Guidelines for monitoring financial derivatives positions were provided earlier in the Guide.

9.2 Contingent liabilities are complex arrangements, and no single measurement approach can fit all situations; rather, comprehensive standards for measuring these liabilities are still evolving. Indeed, experience has shown that contingent liabilities are not always fully covered in accounting systems. Nonetheless, to encourage the monitoring and measurement of contingent liabilities, with a view to enhancing transparency, this chapter provides some measurement approaches, after first defining contingent liabilities and then providing some reasons for their measurement. More specifically, also provided is a table for the dissemination of external debt data on an “ultimate risk” basis; that is, adjusting residence-based external debt data for certain cross-border risk transfers.

Definition

9.3 Contingent liabilities are obligations that arise from a particular, discrete event(s) that may or may not occur. They can be explicit or implicit. A key aspect of such liabilities, which distinguishes them from current financial liabilities (and external debt), is that one or more conditions or events must be fulfilled before a financial transaction takes place.

Explicit Contingent Liabilities

9.4 Explicit contingent liabilities are those defined by the 1993 SNA as contractual financial arrangements that give rise to conditional requirements—that is, the requirements become effective if one or more stipulated conditions arise—to make payments of economic value.2 In other words, explicit contingent liabilities arise from a legal or contractual arrangement. The contingent liability may arise from an existing debt—such as an institution guaranteeing payment to a third party; or arise from an obligation to provide funds—such as a line of credit, which once advanced creates a claim; or arise from a commitment to compensate another party for losses—such as exchange rate guarantees. Some of the more common explicit contingent liabilities are set out below.

Loan and other payment guarantees

9.5 Loan and other payment guarantees are commitments by one party to bear the risk of nonpayment by another party. Guarantors are only required to make a payment if the debtor defaults. Some of the common types of risks that are assumed by guarantors are commercial risk or financial performance risk of the borrower; market risk, particularly that arising from the possibility of adverse movements in market variables such as exchange rates and interest rates; political risk, including risk of currency inconvertibility and nontransferability of payments (also called transfer risk), expropriation, and political violence; and regulatory or policy risk, where implementation of certain laws and regulations is critical to the financial performance of the debtor.3 Loan and other payment guarantees usually increase the initial debtor’s access to international credit markets and/or improve the maturity structure of borrowing.

Credit guarantees and similar contingent liabilities

9.6 Lines of credit and loan commitments provide a guarantee that undrawn funds will be available in the future, but no financial liability/asset exists until such funds are actually provided. Undrawn lines of credit and undisbursed loan commitments are contingent liabilities of the issuing institutions—namely, banks. Letters of credit are promises to make payment upon the presentation of prespecified documents.

Contingent “credit availability” guarantees or contingent credit facilities

9.7 Underwritten note issuance facilities (NIFs) provide a guarantee that a borrower will be able to issue short-term notes and that the underwriting institution(s) will take up any unsold portion of the notes. Only when funds are advanced by the underwriting institution(s) will an actual liability/asset be created. The unutilized portion is a contingent liability.

9.8 Other note guarantee facilities providing contingent credit or backup purchase facilities are revolving underwriting facilities (RUFs), multiple options facilities (MOFs), and global note facilities (GNFs). Bank and nonbank financial institutions provide backup purchase facilities. Again, the unutilized amounts of these facilities are contingent liabilities.

Implicit Contingent Liabilities

9.9 Implicit contingent liabilities do not arise from a legal or contractual source but are recognized after a condition or event is realized. For example, ensuring systemic solvency of the banking sector might be viewed as an implicit contingent liability of the central bank.4 Likewise, covering the obligations of subnational (state and local) governments or the central bank in the event of default might be viewed as an implicit contingent liability of the central government. Implicit contingencies may be recognized when the cost of not assuming them is believed to be unacceptably high.5Table 9.1 provides a practical way of classifying the types of potential liabilities of the central government.

Table 9.1.Fiscal Risk Matrix with Illustrative Examples
Liabilities1Direct (obligation in any event)Contingent (obligation if a particular event occurs)
Explicit

Government liability as recognized by a law or contract
  • External and domestic sovereign borrowing (loans contracted and securities issued by central government)

  • Budgetary expenditures

  • Budgetary expenditures legally binding in the long term (civil servants’ salaries and pensions)

  • Central government guarantees for nonsovereign borrowing and obligations issued to subnational governments and public and private sector entities (development banks)

  • Umbrella central government guarantees for various types of loans (mortgage loans, student loans, agriculture loans, small business loans)

  • Trade and exchange rate guarantees issued by the central government

  • Guarantees on borrowing by a foreign sovereign government

  • Central government guarantees on private investments

  • Central government insurance schemes (deposit insurance, income from private pension funds, crop insurance, flood insurance, war-risk insurance)

Implicit

Obligations that may be recognized when the cost of not assuming them could be unacceptably high
  • Future public pensions (as opposed to civil service pensions)

  • Social security schemes

  • Future health care financing

  • Future recurrent cost of public investments

  • Default of subnational government, and public entity on nonguaranteed debt and other obligations

  • Liability cleanup in entities under privatization

  • Banking failure (support beyond state insurance)

  • Investment failure of a nonguaranteed pension fund, employment fund, or social security fund (social protection of small investors)

  • Default of central bank on its obligations (foreign exchange contracts, currency defense, balance of payment stability)

  • Bailouts following a reversal in private capital flows

  • Environmental recovery, disaster relief, etc.

Source: Adapted from Polackova Brixi (1999).

9.10 Although implicit contingent liabilities are important in macroeconomic assessment, fiscal burden, and policy analysis, implicit contingent liabilities are even more difficult to measure than explicit contingent liabilities. Also, until measurement techniques are developed, there is a danger of creating moral hazard risks in disseminating information on implicit contingent liabilities of the type set out in Table 9.1. Thus, the rest of this chapter focuses only on the measurement of explicit contingent liabilities.

Why Measure Contingent Liabilities?

9.11 By conferring certain rights or obligations that may be exercised in the future, contingent liabilities can have a financial and economic impact on the economic entities involved. When these liabilities relate to cross-border activity, and they are not captured in conventional accounting systems, it can be difficult to accurately assess the financial position of an economy—and the various institutional sectors within the economy—vis-à-vis nonresidents.

9.12 Analysis of the macroeconomic vulnerability of an economy to external shocks requires information on both external debt obligations and contingent liabilities. Experience has shown that contingent liabilities are not always fully covered in accounting systems. Moreover, there is an increasing realization, when assessing macroeconomic conditions, that contingent liabilities of the government and the central bank can be significant. For example, fiscal contingent claims can clearly have an impact on budget deficits and financing needs, with implications for economic policy. Recognizing the implications of contingent liabilities for policy and analysis, the 1993 SNA (paragraph 11.26) states:

Collectively, such contingencies may be important for financial programming, policy, and analysis. Therefore, where contingent positions are important for policy and analysis, it is recommended that supplementary information be collected and presented….

Measuring Contingent Liabilities

9.13 Contingent liabilities give rise to obligations that may be realized in the future, but because of their complexity and variety, establishing a single method for measuring them may not be appropriate. Several alternative ways of measuring contingencies are outlined below. The relevance of each will depend on the type of contingency being measured, and the availability of data.

9.14 A first step in accounting for contingent liabilities is for economic entities to record all such contingent liabilities as they are created, such as with an accrual-based reporting system. But how should such liabilities be valued? One approach is to record these liabilities at full face value or maximum potential loss. Thus, a guarantee covering the full amount of a loan outstanding would be recorded at the full nominal value of the underlying loan. Some governments have adopted this approach. For example, the New Zealand government routinely publishes the maximum potential loss to the government of quantifiable and nonquantifiable contingent liabilities,6 including guarantees and indemnities, uncalled capital to international institutions, and potential settlements related to legal proceedings and disputes.

9.15 Likewise, the Australian government identifies quantifiable and nonquantifiable contingencies.7 In addition, it identifies “remote” contingent losses (mostly guarantees), including nonquantifiable “remote” contingencies. The Indian government regularly reports the direct guarantees provided by the central government on external borrowings of public sector enterprises, development financial institutions, and nonfinancial private sector corporations.8 The guarantees are presented by sector and at nominal value.

9.16 The maximum potential loss method has an obvious limitation: there is no information on the likelihood of the contingency occurring. Especially for loan and other payment guarantees, the maximum potential loss is likely to exceed the economic value of the contingent liability because there is no certainty that a default will occur (that is, the expected probability of default is less than unity). Theoretically, a better approach is to measure both the maximum possible loss and the expected loss, but calculating the expected loss requires estimating the likelihood of losses, which can be difficult.

9.17 Several alternative methods of valuing the expected loss exist. These range from relatively simple techniques requiring the use of historical data, to complex options-pricing techniques. The actual approach adopted will depend on the availability of information and the type of contingency. If the expected loss can be calculated, an additional approach is to value this loss(es) in present-value terms—expected present value. In other words, since any payment will be in the future and not immediate, the expected future payment streams could be discounted using a market rate of interest faced by the guarantor; that is, the present value. As with all present-value calculations, the appropriate interest rate to use is crucial; a common practice with government contingent liabilities is to use a risk-free rate like the treasury rate. Under this present-value approach, when a guarantee is issued the present value of the expected cost of the guarantee could be recorded as an outlay or expense (in the operating account) in the current year and included in the position data, such as a balance sheet.

9.18 Exact valuation requires detailed market information, but such information is often unavailable. This is particularly true in situations of market failure or incomplete markets—a financial marketplace is said to be complete when a market exists with an equilibrium price for every asset in every possible state of the world. Other means are then required to value a contingency. One possibility is to use historical data on similar types of contingent operations. For example, if the market price of a loan is not observable, but historical data on a large number of loan guarantees and defaults associated with those guarantees are available, then the probability distribution of the default occurrences can be used to estimate the expected cost of a guarantee on the loan. This procedure is similar to that employed by the insurance industry to calculate insurance premiums. Rating information on like entities is often used to impute default value on loan guarantees as well. The U.S. Export-Import Bank employs this method for valuing loan guarantees that it extends.

9.19 Bank regulatory guidelines established by the Basel Committee on Banking Supervision also draw on historical data to measure risks in banks’ off-balance-sheet activities. For traditional off-balance-sheet items like credit contingent liabilities, the guidelines provide “credit conversion factors,” which when multiplied with the notional principal amount provide an estimate of the expected “payout” from the contingent liability. The conversion factors are derived from the estimated size and likely occurrence of the credit exposure, as well as the relative degree of credit risk. Thus, stand-by letters of credit have a 100 percent conversion factor; the unused portion of commitments with an original maturity of over one year is 50 percent; and RUFs, NIFs, and similar arrangements are assigned a 50 percent conversion factor as well.

9.20 Market-value measures use market information to value a contingency. This methodology can be applied across a wide range of contingent liabilities, but it is particularly useful for valuing loan and other payment guarantees, on which the following discussion focuses. This methodology assumes that comparable instruments with and without guarantees are observable in the market and that the market has fully assessed the risk covered by the guarantee. Under this method, the value of a guarantee on a financial instrument is derived as the difference between the price of the instrument without a guarantee and the price inclusive of the guarantee. In the context of a loan guarantee, the nominal value of the guarantee would be the difference between the contractual interest rate (ip) on the unguaranteed loan and the contractual interest rate (ig) on the guaranteed loan times the nominal value of the loan (L): (ip − ig) L. The market value of the guarantee would use market, not contractual, rates.9

9.21 Yet another approach to valuing contingent liabilities applies option-pricing techniques from finance theory. With this method, a guarantee can be viewed as an option: a loan guarantee is essentially a put option written on the underlying assets backing the loan.10 In a loan guarantee, the guarantor sells a put option to a lender. The lender, who is the purchaser of the put option, has the right to “put” (sell) the loan to the guarantor. For example, consider a guarantee on a loan with a nominal value of F and an underlying value of V. If VF <0, then the put option is exercised and the lender receives the exercise price of F. The value of the put option at exercise is FV. When V >F, the option is not exercised. The value of the guarantee is equivalent to the value of the put option. If the value of the credit instrument on which a guarantee is issued is below the value at which it can be sold to the guarantor, then the guarantee will be called.

9.22 Although the option-pricing approach is relatively new and sophisticated, it is being applied in the pricing of guarantees on infrastructure financing and interest and principal payment guarantees.11 But standard option pricing has its limitations as well. This is because the standard option-pricing model assumes an exogenous stochastic process for underlying asset prices. However, it can be argued that the very presence of a guarantee (especially a government guarantee) can affect asset prices.12

Recommended Measures

9.23 The Guide encourages the measurement and monitoring of contingent liabilities, especially of guarantees, and has outlined some measurement techniques. However, it is recognized that comprehensive standards for measuring contingent liabilities are still evolving. Consequently, only the recording of a narrow, albeit important, range of contingent liabilities is specified ahead: guarantees of domestic private sector external debt by the public sector, and the cross-border provision of guarantees. In both instances, it is recommended that the contingency should be valued in terms of the maximum exposure loss.

Public sector guarantees

9.24 In Chapter 5 the dissemination of data on publicly guaranteed private sector debt—that is, the value of private sector debt that is owed to nonresidents, and is guaranteed by the public sector—through a contractual arrangement is discussed.

Ultimate risk

9.25 Set out in Table 9.2 is a format that presents external debt according to an “ultimate” risk concept—augmenting residence-based data to take account of the extent to which external debt is guaranteed by residents for nonresidents. Countries could potentially have debt liabilities to nonresidents in excess of those recorded as external debt on a residence basis if their residents provide guarantees to nonresidents that might be called. Also, branches of domestic institutions located abroad could create a drain on the domestic economy if they ran into difficulties and their own head offices needed to provide funds.

Table 9.2.Gross External Debt Position: Ultimate Risk Basis
End-Period
Gross External Debt (1)Inward risk transfer (+) (2)External Debt (ultimate-risk basis) (3)Memorandum item: Outward risk transfer (4)
General Government
Short-term
Money market instruments
Loans
Trade credits
Other debt liabilities1
Arrears
Other
Long-term
Bonds and notes
Loans
Trade credits
Other debt liabilities1
Monetary Authorities
Short-term
Money market instruments
Loans
Currency and deposits2
Other debt liabilities1
Arrears
Other
Long-term
Bonds and notes
Loans
Currency and deposits2
Other debt liabilities1
Banks
Short-term
Money market instruments
Loans
Currency and deposits2
Other debt liabilities1
Arrears
Other
Long-term
Bonds and notes
Loans
Currency and deposits2
Other debt liabilities1
Other Sectors
Short-term
Money market instruments
Loans
Currency and deposits2
Trade credits
Other debt liabilities1
Arrears
Other
Long-term
Bonds and notes
Loans
Currency and deposits2
Trade credits
Other debt liabilities1
Nonbank financial corporations
Short-term
Money market instruments
Loans
Currency and deposits2
Other debt liabilities1
Arrears
Other
Long-term
Bonds and notes
Loans
Currency and deposits2
Other debt liabilities1
Nonfinancial corporations
Short-term
Money market instruments
Loans
Trade credits
Other debt liabilities1
Arrears
Other
Long-term
Bonds and notes
Loans
Trade credits
Other debt liabilities1
Households and nonprofit institutions serving households (NPISH)
Short-term
Money market instruments
Loans
Trade credits
Other debt liabilities1
Arrears
Other
Long-term
Bonds and notes
Loans
Trade credits
Other debt liabilities1
Direct Investment: Intercompany Lending
Debt liabilities to affiliated enterprises
Arrears
Other
Debt liabilities to direct investors
Arrears
Other
Gross External Debt

9.26 In Table 9.2 residence-based external debt data (column 1) is increased by the amount of debt of nonresidents, not owned by residents, that is guaranteed by a resident entity (inward risk transfer, column 2). Column 3 is the adjusted external debt exposure of the economy. The table is set out in this manner so that external debt on an ultimate-risk basis can be related back to the gross external debt position measured on a residence basis.

9.27 The intention of column 2 is to measure any additional external debt risk exposures of residents arising from contingent liabilities. The definition of contingent liabilities adopted is deliberately narrow. To be included in this definition of contingent liabilities, the debt must exist, so lines of credit and similar potential obligations are not included. The data on the inward transfer of risk covers only the debt of a nonresident to a nonresident on which, and as part of the agreement between debtor and creditor, payments are guaranteed to the creditor(s) by a resident entity under a legally binding contract—the guarantor will most commonly be an entity that is related to the debtor (for example, the parent of the debtor entity), and debt of a legally dependent nonresident branch of a resident entity that is owed to a nonresident. If debt is partially guaranteed, such as if principal payments or interest payments alone are guaranteed, then only the present value of the amount guaranteed should be included in columns 2 or 4. To avoid double counting the same external debt risk exposure, the following should be excluded from column 2: all debt liabilities of nonresident branches to other nonresident branches of the same parent entity; and any amounts arising from external debt borrowings of nonresidents that were guaranteed by a resident entity and on-lent by the nonresident borrower to that same resident entity or any of its branches. This guidance is not intended to exclude debt exposures of residents from the ultimate risk concept, as defined above, but to ensure that they are counted only once.

9.28 External debt is the liability of the debtor economy. However, as a memorandum item, the amount of external debt of the economy that is guaranteed by nonresidents is also presented (outward risk transfer, column 4). The data on the transfer of risk outward covers only external debt on which, and as part of the agreement between debtor and creditor, payments are guaranteed (or partially guaranteed) to the creditor(s) by a nonresident under a legally binding contract—the guarantor will most commonly be an entity that is related to the debtor (for example, the parent of the debtor entity)—and external debt of a resident entity that is a legally dependent branch of a nonresident entity.

9.29 No reallocation of risk is made because of the provision of collateral by the debtor, or because a debt instrument is “backed” by a pool of instruments or streams of revenue originating from outside of the economy. Because the intention of Table 9.2 is to monitor the potential risk transfer from the debtor side, no reallocation of risk is made if the risk transfer is initiated from the creditor side, without any involvement of the debtor—for example, the creditor has paid a premium to a guarantor, such as an export credit agency unrelated to the debtor, to insure against payment default or has purchased a credit derivative that transfers credit risk exposure.

This chapter draws on work at the World Bank.

The European System of Accounts: ESA 1995 (Eurostat, 1996) defines contingent liabilities in a similar way.

Regulatory or policy-based guarantees are especially relevant in infrastructure financing. For more details and country-specific examples, see Irwin and others (1997).

A case in point is Indonesia, where the government’s domestic debt increased from practically nothing, in the period before the crisis (mid-1997), to 500 trillion Indonesian rupiah by the end of 1999, mostly due to the issuance of bonds to recapitalize the banking system. The increase in the government’s stock of domestic debt was accompanied by a rise in its assets, which were received in exchange for issuing bank-restructuring bonds. See also Blejer and Shumacher (2000).

See Guidelines for Public Debt Management (IMF and World Bank, 2001).

New Zealand Treasury, Budget Economic and Fiscal Update (Wellington, annual). As the name suggests, nonquantifiable contingent liabilities cannot be measured and arise from either institutional guarantees that have been provided through legislation or from agreements and arrangements with organizations.

Aggregate Financial Statement (Australia, annual).

See the Ministry of Finance’s annual publication on external debt, India’s External Debt: A Status Report.

For a further discussion of market-value methods see Towe (1990) and Mody and Patro (1996).

Robert C. Merton (1977) was the first to show this.

See Sundaresan (2002) for a detailed exposition on this issue.

    Other Resources Citing This Publication