3 Domestic Lending in Foreign Currency

Charles Enoch, Dewitt Marston, and Michael Taylor
Published Date:
September 2002
  • ShareShare
Show Summary Details
Fernando L. Delgado,, Daniel S. Kanda, and Greta Mitchell Casselle, 

The availability of foreign currency loans to domestic borrowers has expanded as a natural result of increased global liquidity and the liberalization of domestic financial systems. These loans have also expanded because of negative incentives resulting from implicit government guarantees on the exchange rate or the expectation of governmental support of certain financial institutions and because of a supervisory environment that did not fully incorporate all relevant risks into lending decisions. Moreover, some stabilization policies have often resulted in relatively higher interest rates on domestic currency loans, which along with somewhat stable exchange rates made it attractive for domestic agents to borrow in foreign currency even when their businesses essentially dealt in domestic currency (see Bank for International Settlements, 1998). Banks generally transfer currency risk to customers who commit to debt service payments in foreign currency, regardless of the currency denomination of their revenues, in the expectation that the currency risk element implicit in domestic interest rates will not materialize during the term of the loan. Concerns about the potential sudden deterioration in bank portfolios during currency crises have led policymakers to consider prudential instruments that would discourage banks from lending to agents whose revenue is not denominated in the currency in which they borrowed, while encouraging them generally to manage foreign exchange and credit risks more actively.

In economies whose currencies are not internationally accepted, economic factors that undermine credibility in the domestic currency will condition the assessment of risks and will further affect the choice of the currency of financing. The issue of bank domestic lending in foreign currency can then be re-addressed as one of borrowing in a currency accepted by others for international transactions by agents whose revenue is denominated in a currency only accepted domestically.1 The continuous uncertainty about potentially high expected depreciation may lead to equally high real interest rates in domestic currency, a factor that in turn makes borrowing in a foreign currency dangerously attractive and raises the likelihood of systemic banking problems. Box 3.1 illustrates the limits to which bank lending in foreign currency can reach following a sudden depreciation of the domestic currency.

This chapter focuses on when supervisory limits to banks’ domestic lending in foreign currency are appropriate, considering both economic and supervisory grounds, and when internal policies and prudential measures are acceptable to address these risks. Particular attention is given to characterizing the nature of risk exposure and differentiating between prudential limits and discriminatory standards or indirect capital controls. Following the chapter’s introduction, the second section describes the nature of the risks involved in foreign currency lending, whereas the next section discusses the incentives to assume currency and credit risk and the need for prudential guidelines. The following section describes best practices in terms of risk management guidelines, internal policies, and prudential and supervisory measures. The final section provides conclusions based on the preceding analysis.

Risks of Banks’ Domestic Lending in Foreign Currency

Compound Credit and Currency Risk

The compounding of credit and currency risks inherent in banks’ domestic lending in foreign currency results from counterparty exposure.

Box 3.1.The Effect of U.S. Dollar Lending in Ecuadorian Domestic Markets on the Banking Crisis

Ecuador maintained the exchange of the sucre within a bank that was kept relatively stable for several years, while fiscal imbalances resulted in rates on sucre-denominated deposits and loans that were substantially higher than those denominated in U.S. dollars (see the first figure). As a result, a large number of households and corporations, even those without dollar-denominated income, decided to borrow in U.S. dollars, on the expectation that reduced interest rates would compensate the possible exchange rate depreciation during the life of the loan. Banks also favored domestic lending in U.S. dollars in order to reduce their consolidated depreciation during the life of the loan. Banks also favored domestic lending in U.S. dollars in order to reduce their consolidated foreign exchange exposure, since a sizable share of their funding was in dollar-denominated deposits through their off-shore subsidiaries.1 By January 1999, over 70 percent of total banking sector credit was denominated in U.S. dollars.2

The compound credit risk created by the high volume of the U.S. dollar-denominated lending was not appropriately assessed by bank risk management systems, nor did the bank supervisory authority introduce appropriate regulations. As a consequence, reserves and provisions of the banks were grossly inadequate to confront a large depreciation of the sucre. From January to December 1999 the sucre/U.S. dollar exchange rate depreciated by 191 percent (see the second figure).

Interest Rates

(Annual percentage)

Nominal Exchange Rate

(Sucres per U.S. dollar)

Nonperforming Loan Ratios (on-shore)

(In percent)

A number of other factors, ranging from external shocks affecting the productive sector3 to the freeze of most of the banking system deposits, also contributed to reducing liquidity of Ecuadorian corporations and resulted in a substantial increase in the nonperforming loan ratio, which jumped from 10 percent in December 1998 to over 40 percent in December 1999. Nonperforming loan ratios of U.S. dollar-denominated portfolios increased well above those denominated in sucres (see the third figure), illustrating how the compound risk involved in foreign exchange lending to domestic borrowers could be a major force behind banking crisis.

1Early in 1996, Ecuador introduced limits on banks’ foreign exchange net open positions.2Although there is no statistical data, it is estimated that approximately three-fourths of borrowers in U.S. dollars did not have U.S. dollar-denominated income.3These included El Nino floods, white spot epidemic affecting shrimp production, and other similar shocks.

This exposure may result from a counterparty cash flow that is predominantly in domestic currency—for example, in the case of import financing of capital goods in foreign currency for the production of nontradable goods, which only generate domestic currency revenue. Import financing expressed in foreign currency is customary, especially when financial institutions are concerned about potential convertibility problems at the time of settlement. However, by not refinancing or hedging the obligations, the borrower remains exposed to an exchange rate risk, which translates into a credit risk for the financing institution. Counterparty exposure also results from risk that the domestic currency market value of the collateral backing the obligations to the bank declines below what is required to cover the obligations—for example, the use of real estate assets as collateral for the financing of an export activity. In this case, the borrower does not face direct exchange rate risk; however, the bank is exposed to a potential credit risk that would be triggered by industry- or firm-related adversities, when, in addition, the value of the collateral declines below the value of the related obligation. Because the same demand factors support domestic activities and asset prices, it is not unusual that countries experience both types of effects simultaneously.

Other Risks

Other risks particular to bank domestic lending in foreign currency include currency and transfer risks. Currency risk (a type of market risk) results from the probability of immediate large losses associated with the impact of exchange rate instability inherent in the foreign exchange business.2 This type of risk may also affect those institutions that lend in foreign currency with a balanced foreign exchange position. For example, this situation would occur if a bank sets a loan-loss provision because of increased currency risk, which would subsequently decrease the value of its assets, thus creating an open position. Therefore, the incorporation of an exchange rate risk may turn an ex ante balanced foreign exchange position into an ex-post-uncovered open position.3 This type of risk has been analyzed particularly in connection with open foreign exchange positions (Basel Committee on Banking Supervision, 1980).

Transfer risk derives from lending in a currency other than that of the country in which the borrower resides. This risk arises from the ability of borrowers to obtain the foreign exchange necessary to repay the debt. It includes the risk that, due to lack of foreign currency reserves (usually caused by a currency or debt crisis), a borrower’s local currency holdings and cash flow could not be converted into a sufficient quantity of foreign currency to repay their loans.4

Foreign currency loans also entail additional risks, which are the same potential concerns seen in any typical lending product. Those common risks may be captured in three broad categories: credit, market (interest rate and liquidity), and compliance (internal policies and procedures, operational and legal) risks. Because of their commonality with other lending products, these risks will not be addressed in this chapter.

Borrower and Lender Incentives to Assume Credit and Currency Risk

Credit Risk, Exchange Rate Risk, and Arbitrage

Ideally, complete and continuous arbitrage across currencies should lead to perfectly equivalent ex ante and ex post financial costs of borrowing in domestic or foreign currency in conditions of perfect competition, complete information, full flexibility of prices, and full capital mobility. However, the timing, duration, and magnitude of exchange rate instability have proven extremely difficult to anticipate across exchange rate regimes—a situation that complicates the measurement of foreign exchange risk. Thus, even in the best of conditions, perfect arbitrage would not be possible at all times or within short-term periods.5

In this context, risks may not be fully internalized by economic agents. Banks may not fully assess credit risk, and borrowers may misperceive exchange rate risk. However, even if supervisors have reason to believe that the lack of internalization of risks is a severe problem, inappropriate attempts of correction may lead to disproportional limits to arbitrage opportunities. In the extreme, innovation in the financial market may be arrested, the allocation of financial resources may be suboptimal, and competitiveness may be diminished, especially if limits are introduced in the context of market-unfriendly government measures.6

Arbitrage could also be affected by the design of monetary instruments. Countries restrict coverage of deposit insurance for some types of foreign currency deposits, especially if they are not significant.7 Countries issuing currency accepted for international transactions and many transition economies prefer reserve requirements on foreign currency deposits held in domestic currency, unlike countries experiencing protracted stabilization efforts and currency weakness.8 Countries perceiving a recovery in demand for home currencies may prefer nonsterilized intervention to sterilization (see Marston, 1995). This may result in unbalanced incentives to borrow or lend in either domestic or foreign currency.

Borrower Incentives to Demand Foreign Exchange Credit

A borrower’s preference for financing in either currency may be related to strategic decisions or to various cost incentives. The latter may result from imperfect arbitrage, policy-induced costs, or the inadequate incorporation of risks into the borrower’s cost-benefit considerations.

Strategic incentives

In more developed financial markets, swap opportunities result from differences in comparative advantage for economic agents when borrowing in foreign currency and in domestic currency. This makes it sensible for both types of borrowers to exploit this advantage by borrowing and swapping the obligations.9 This development could only result from allowing borrowers access to financing in different currencies.10 Other strategic incentives to borrow in foreign currency for borrowers with domestic currency revenues may be related to the time structure of repayments and the possibility of indirect natural hedging.

  • Time structure incentives. Because exchange rate depreciation directly affects the domestic value of the principal, the domestic value of repayments increases throughout the repayment period. For loans in foreign and domestic currency in equivalent terms, an increase in the rate of exchange rate depreciation matched by a higher interest rate would result in a more acute skewness of the repayment schedule for foreign currency loans expressed in domestic currency.11

  • Indirect natural hedging. Economic sectors benefiting from an exchange rate depreciation compensate higher financial costs with additional revenue from a favorable switch in demand. Domestic providers of tradable goods and services to the domestic market would hedge against an increase in financial costs by borrowing in foreign currency, since a depreciation allows them to capture additional domestic and foreign demand.

Strategic incentives do not require supervisory response. Borrowers may have legitimate reasons to take advantage of a loan with back-loaded repayments, since further back loading constitutes a reasonable escape valve in times of distress. Economic sectors benefiting from the depreciation of the exchange rate could make legitimate and reasonable use of the hedging protection that results from borrowing in foreign exchange. However, inflationary processes may justify additional supervisory concerns. Since a back-loaded debt repayment schedule becomes even more attractive, uncertainty surrounding an inflationary period may lead borrowers to use back loading as a means to facilitate the delay of default.

Cost-benefit incentives to foreign exchange borrowing

Even in reasonably efficient financial markets, arbitrage is not perfect as a result of the economic distortions affecting financial cost equivalence across currencies.12 Imperfect arbitrage will be reflected on a lag between the incorporation of expectations into interest rates and actual exchange rate changes. Even agents envisaging an exchange rate depreciation may prefer to borrow in foreign currency, especially on a short-term basis, if they perceive that the exchange rate would not change according to expectations at least until after the loans are repaid. Thus, borrowing in foreign currency might still be preferred even in times of increasing risk. In periods of liquidity problems, better informed agents modify their risk preferences faster, and they may reduce the range of alternatives for less-informed agents to adjust their own risks subsequently.

Continuous and uninterrupted convertibility and the associated market arrangements ensures the presence of a degree of credibility in internationally accepted currencies. As a result, overall risk neutrality between financing alternatives in different currencies is achieved. Unbalanced incentives are more likely in situations of protracted and severe instability combined with external vulnerability. To compensate for the long-lasting expectations of a large depreciation or even inconvertibility (transfer risk), governments frequently embark on a policy mix that combines a stable exchange rate with higher interest rates. For countries with a long history of instability, slow convergence of the real exchange rate to its equilibrium level in the context of high domestic interest rates may result in more costly domestic currency lending, regardless of the exchange rate regime.13

Lender Incentives to Supply Foreign Exchange Credit

The motivation to lend in foreign currency is related to the availability of foreign exchange resources and risk management considerations. The sources of financing may be long-term lines of credit, short-term capital inflows, or foreign currency deposits.14 In all these cases, availability of resources may prevail over risk management considerations in developing economies, as a result of one or more of the following factors:

  • Lack of medium-term sources of financing in domestic currency. Foreign currency borrowing by banks or intermediated or guaranteed by the government normally complements domestic savings and may be the only available option for long-term financing in less developed financial markets. In such markets, and in the absence of government intermediation or guarantees, the banking system may well face the dilemma of lending in foreign currency without coverage or not lending, which may leave unsatisfied a particular demand for financing (especially long-term financing).15

  • Dollarization. High inflation, negative real interest rates on domestic-currency-denominated assets, and frequent exchange rate devaluations can contribute to the rise of the ratio of foreign currency deposits over total deposits (see Baliño, Bennett, and Borensztein,1999). The larger availability of deposits in foreign currency leaves few options to domestic banks than to on-lend these resources domestically.

  • Speculative capital inflows.16 Banks and foreign investors may have a common interest in channeling foreign exchange flows in some periods, attracted by high interest rates on a nonrisk-adjusted basis in terms of foreign currency. However, these high interest rate differentials may be the result of a low risk premium on foreign exchange resources, resulting in turn from the discount of implicit government guarantees, as a government embarked on enhancing its credibility would resist any possibility of default on these transactions. On the assumption that exchange rate stability would be preserved within the investment maturity time frame, such a decision artificially increases the availability of foreign exchange resources, especially at times of low interest rates in developed markets.17 Even in cases where exchange rate flexibility is nominally allowed, the government may limit it to a minimum precisely in periods when the currency is under pressure because of its impact on the real sector and on expectations.18

The Nature of the Problem

Banks’ domestic lending in foreign currency is not a problem as such. As was stated previously, the availability of multiple sources of financing in terms of currency facilitated the development of swaps, related to the development of other derivative products in international financial markets. Because of strategic motives, this type of lending may exceed the borrowing requirements of economic agents whose revenue is in foreign currency. The preceding section illustrates that the problem of banks’ domestic lending in foreign currency arises when the allocation of credit is not based on an adequate assessment of risks, which in turn would make this lending unduly inexpensive in times of exchange rate stability. In the extreme, this distortion may lead to a significant allocation of lending to borrowers whose activities appear profitable only because of the apparent low cost of credit.

Table 3.1 provides some indicators for a sample of countries with available data related to the analysis in this chapter. First, the share of foreign currency loans appears closely related to the share of foreign currency deposits, but there are cases where wide differences between both variables are observed (such as Belarus, Estonia, Hungary, and Mexico). Second, the share of exports to GDP does not show a close relationship with lending in foreign currency, which could be a rough signal that this type of lending is not predominantly allocated to activities generating foreign currency revenue.19 Third, for some countries showing a large availability of foreign loans to banks to GDP (such as El Salvador, Malaysia, and Thailand), the share of foreign currency loans made by banks domestically appears surprisingly modest. Fourth, the last column shows casual evidence of the importance of cost-benefit factors: The apparent significance of the number of years out of the last five when borrowing in foreign currency has been less costly than borrowing in domestic currency, considering the interest rate differential at the beginning of the period and the exchange rate depreciation in the subsequent year.

Table 3.1.Indicators of Bank Foreign Currency Operations, 1999
Share of Foreign


Deposits to

Total Deposits
Share of Foreign


Loans to

Total Loans
Exports of

Goods and


Loans to

Number of Years

when Foreign

Currency Loans

were less costly1
El Salvador8.613.818.211.75
South Africa3.
Source: International Financial Statistics, EDSS, and BIS.

For the years 1995-1999, considering end-of-period interest rate differentials and subsequent annual exchange rate depreciation.

Source: International Financial Statistics, EDSS, and BIS.

For the years 1995-1999, considering end-of-period interest rate differentials and subsequent annual exchange rate depreciation.

Internal Policies, Supervisory, and Prudential Best Practices

The Need for Guidelines

Because lending in foreign currency has not been a matter of concern for industrial countries, industrial-country bank regulations have not served as a reference for developing economies. In the absence of a framework to address this problem, developing countries have frequently resorted to exchange restrictions to forcefully limit these operations during crises. Practices followed by the countries reporting to the IMF’s Annual Report of Exchange Arrangements and Exchange Restrictions (see Appendix) show that, while degree and form of restriction on capital transactions varies, such restrictions are more common among developing and transition economies than among industrial countries.

The way in which developing countries and transition economies have addressed this issue depends also on several factors, such as the level of dollarization, the availability of foreign exchange, and concerns related to trade flows. Heavily dollarized economies can only impose some quantitative limits. Other developing countries allow only on-lending of credit from abroad to discourage domestic banks from competing to attract deposits in foreign currency from residents. Concerns about the continuous availability of loanable resources in foreign exchange seem to be behind “official approval requirements” as the favored restrictive tool. When limitations are based on the use of resources, the motivation appears more related to concern about affecting trade flows than to an intention to restrict the use of these resources to economic agents generating foreign currency revenue.

The trend toward the removal of capital controls and the globalization of financial flows will tend to limit the use of these restrictions. Recently, prudential mechanisms have been used to induce an effective internalization of foreign exchange risks in banks’ decisions. This mechanism has been used with a high degree of improvisation and without direct practical references from industrial economies. This section presents guidelines that banks and supervisors could consider to manage this risk. Although most of the measures included in the guidelines are common to industrial and developing countries, they would be particularly relevant in countries with a large availability of foreign exchange resources from dollarization and short-term capital inflows relative to savings in domestic currency (particularly long-term).

The following set of guidelines are derived mainly from the limited experience of measures taken by countries after a banking crisis where domestic lending in foreign currency has played a major role and from extrapolating banks’ exchange risk management and prudential measures to bank customers. The guidelines have been classified in two broad groups: internal policies and procedures for risk management; and supervisory and prudential guiding principles for banks’ domestic lending in foreign currency.

Internal Policies and Procedures

The evaluation of bank policies, practices, and procedures and a bank’s adherence to them, together with adequate management information systems, should be assessed in ensuring adequate internal oversight of domestic lending in foreign currency.20 Fundamental to the success of banks’ domestic lending in foreign currency is the strength of banks’ respective management information systems and risk management systems.

Regarding management information systems, banks should determine that such systems monitor: (1) the relative and absolute exposure related to foreign exchange to unhedged domestic borrowers, including the share in total lending and on total foreign currency deposits and other funding; (2) open positions and maturity structure of foreign currency assets and liabilities, including net lending in foreign exchange to domestic hedged and unhedged borrowers, and corresponding bank hedging positions; (3) hedging positions (including “natural hedges”)21 of borrowers in foreign exchange at an individual level, and the aggregated foreign exchange exposure of the corporate and household sectors and their corresponding hedging positions; (4) compliance with limits and procedures set in the risk management system; and (5) the direct and indirect impact of exchange rate fluctuations on future earnings.

Banks’ risk management systems in this area share the same building blocks as the management of overall foreign exchange risk—namely, written policies and procedures; sound internal controls, accounting procedures and rules resulting in accurate valuation of assets and liabilities; formulas to allow the accurate measurement of the various forms of risk; steps to ensure that unexpected shocks do not unduly undermine the individual institution; and the availability of timely and accurate information on each specific risk (see Abrams and Beato, 1998).

Written policies and procedures should include specific provisions to manage the compound risk of domestic lending in foreign currency. Among the specific main guiding principles for risk management in banks’ domestic lending in foreign currency are: (1) limits to unhedged foreign exchange open positions should be set “net of provisions” for counterparty risk derived from bank domestic lending in foreign currency to reduce the chances of a sudden exposure of a bank’s position in case of a counterparty’s default (the corresponding capital requirements would apply); (2) credit in foreign currency to activities generating foreign currency revenue should be categorized according to the standard classification criteria of a bank’s asset portfolio, which should include an assessment of the eventual impact of the exchange rate on profitability driven by consequent changes in relative prices; (3) proper counterparty risk evaluation systems should be applied in the assessment of credits in foreign exchange to nonforeign exchange earners, interest rate spreads applied should reflect the compound risk,22 additional guarantees should be required if necessary, and conservative loan classification criteria should be applied particularly during periods of high exchange rate volatility; (4) appropriate limits on foreign currency loan concentration should be made; (5) collateral denominated in currencies different from the one in which the loans are expressed should be valued conservatively, allowing for a margin for exchange rate movements and providing for reappraisals in the face of eventual but substantial exchange rate changes; (6) maximum use of available hedging mechanisms should be pursued, as should their frequent reassessment, both for the bank and borrowers in foreign exchange; and (7) the viability of projects should be assessed independently of the currency of financing.

As part of internal risk management, banks should have hedging techniques appropriate to address exposures from foreign currency lending to domestic borrowers. In addition, banks should introduce clear written procedures that critically evaluate a borrower’s ability to withstand adverse exchange rate changes, and protect banks from excessive exposure to direct exchange rate risk.23 Limits on foreign currency lending to various types of borrowers should be clearly specified in written policies. Moreover, the Basel Committee on Banking Supervision recommends establishing a “maturity ladder” approach to future cash inflows and outflows over a series of specified time periods, taking into account the complicating factor of differences in currencies.24 In addition, banks should analyze the results of “alternative scenarios,” including the effects of higher interest rates or depreciation of the exchange rate.25

Sophisticated risk management systems include provisions for stress testing loans, foreign exchange exposures, and other activities.26 Stress testing provides banks with insights into levels of risk consequent on various changes in the economy, like currency depreciation, which can then be used in evaluating total risk exposure. The results of stress testing allow bank management to decide whether potential losses identified through the process are acceptable. Stress testing models for monitoring foreign currency loans to domestic borrowers could include, among other things, vulnerability to scenarios such as a sharp depreciation in currency, foreign exchange movements of international reserves (and foreign exchange in regional currencies), and related changes in other relevant prices. Banks should consider placing either absolute limits or management action triggers on the size of losses they will tolerate due to foreign currency lending to domestic borrowers. Depending on the potential losses revealed by the stress tests and the anticipated likelihood of occurrence, bank management should consider appropriate risk-reducing actions.

Supervisory and Prudential Guidelines

This section contains a discussion on the negative effects of an outright prohibition (or limitation) to a bank’s domestic lending in foreign currency.27 Currently a large number of countries (see Table 3.3 in the Appendix) resort to prudential measures—including prudential currency matching mechanisms—aimed at eliminating the risk-enhancing incentives to borrow in foreign exchange. To the extent that these measures tend to correct incentive biases arising from other sources, they should not be considered as capital control devices.

Specific measures along these lines should include the following: (1) incentives for unhedged domestic borrowers to borrow in foreign exchange that are introduced by an inconsistent monetary and exchange rate policy mix or the design of monetary instruments (for example, differential reserve requirements) should be discussed between supervisors and monetary authorities, to avoid unintended negative effects on loan allocation; (2) prudential currency-matching mechanisms, such as additional capital requirements or additional provisions, may be used to offset currency biases introduced by the overall macroeconomic policy in coordination with the monetary authority; or required in periods of prolonged discrepancy between expected exchange rate changes embedded in interest rates and actual exchange rate changes;28 (3) value-at-risk models with a proven track record should be used if banks have sound risk management systems, skilled staff, and are made subject to stress tests;29 if this is not the case, a generic provision on all foreign exchange loans to unhedged domestic borrowers should be applied preferably; (4) while domestic lending in foreign currency should not be prohibited, lending in foreign currency to sectors naturally hedged against exchange rate fluctuations should probably benefit from a better loan classification; and (5) the banking system should be allowed without restriction (other than prudential regulations) to invest abroad foreign exchange resources from their depositors.

To be able to promptly and accurately assess the compounded risk generated by domestic lending in foreign currency, the supervisory authorities should monitor, on an aggregated basis, the trend of the share of foreign currency loans of the banking system to total loans and to total foreign currency deposits (and other funding) over certain time periods—at least on quarterly basis. Accounting system requirements should be set up so as to allow the supervisory authorities to track the counterparty risk assumed by banks as a result of foreign exchange lending to domestic unhedged borrowers. In addition, selected information should be disclosed to the public to promote market discipline. Supervisors in some Latin American countries provide quarterly bulletins that detail individual bank and aggregate bank data, including the amount of foreign currency liquid assets, loans and liabilities (see Edmonds, 2000). Supervisors should also monitor corporate foreign currency debt (on an aggregated and disaggregated basis) since significant currency depreciation could put severe pressure on those banks whose clients have large external burdens.

Concluding Remarks

After analyzing the risks involved in banks’ domestic lending in foreign currency, the following concluding remarks could be made:

  • Banks and supervisory authorities should be aware of the compounding of risks involved in foreign exchange-denominated lending to domestic borrowers who are not hedged against currency risk. For a particular country, macroeconomic volatility, the consistency of the monetary and exchange rate policy mix, the scope to diversify risks, the quality of bank management, and the quality of supervision determine the complexity and the potential severity of problems arising from banks’ domestic lending in foreign currency. Supervisors should be alert to policies that undermine the safety and soundness of the banking system (Basel Committee on Banking Supervision, 1997).

  • Special care should be used in redressing perverse incentives and establishing the adequate internal policies and procedures, oversight, and regulatory systems in countries whose currencies are not internationally accepted, as the compound risk introduced by foreign exchange-denominated lending to domestic borrowers who are not hedged against currency risk is more likely to materialize.

  • The domestic borrower’s demand for foreign exchange-denominated loans is determined by preferences that ultimately respond to a set of economic incentives. Some of these incentives contribute to the stability of the banking system (e.g., borrowing in foreign exchange to hedge against negatively correlated risks), while some other incentives result from market imperfections (e.g., those resulting in imperfect arbitrage) or inadequate policies (e.g., inconsistent monetary and exchange rate policy mix) and thus have a negative effect on the stability of the banking system. It is primarily by redressing these latter perverse incentives that the problem of compound risk introduced by foreign exchange-denominated lending to domestic borrowers should be addressed.

  • Prohibiting foreign exchange-denominated lending to domestic borrowers would be difficult to enforce, would result in nonoptimal allocation of financial resources, could hamper financial innovation, and would introduce a competitive disadvantage for domestic banks.

  • To the extent that negative incentives could not always be eliminated and new ones may emerge, the banks and the supervisory authorities should develop adequate systems to deal with the risks

  • introduced by foreign exchange-denominated lending to domestic borrowers. These systems should be based on sound internal policies and procedures for banks, adequate oversight by the supervisory authorities, and a comprehensive prudential regulation framework (see Table 3.2 for a summary of these mechanisms). To the extent that prudential measures are aimed at eliminating asymmetrical incentives to borrow in foreign exchange, prudential currency-matching mechanisms should not be considered as capital control devices, since asymmetric prudential mechanisms are appropriate for asymmetric risks. To the extent that different mechanisms are applied for different types or risk, these instruments would not imply discriminatory standards for different stages of development. Because of the dynamic nature of foreign exchange risk, these mechanisms should be adaptable to changing circumstances, but they should not be expected to change frequently to avoid obscure signals.

Table 3.2.Risk Management, Internal Policies and Procedures, and Supervisory and Prudential Guidelines for Banks’ Domestic Lending in Foreign Currency
Foreign Exchange Position.Loan Policies and ProceduresOther Specific Measures
Risk management guiding principles
  • - Position limits net of provisions for counterparty risk should be incorporated.

  • - Maximum use of available hedging and frequent reassessment should be exercised.

  • - Loan classification should account for economic impact of possible adverse exchange rate move.

  • - Projects should be viable regardless of currency of financing.

  • - Collateral should be valued conservatively, allowing for margins and reappraisals after exchange rate movements.

  • - When demand for domestic currency is low or declines, loan classification may still reflect a preference for allocating foreign currency loans to domestically oriented sectors potentially benefiting from an exchange rate devaluation.

  • - Banks should be allowed to invest part of foreign exchange resources abroad in first quality, liquid instruments.

  • - Inadequate incentives introduced by the design of monetary instruments should be continuously reassessed.

Governance guidelines
  • - Written policies and procedures, along with sound internal controls should be employed.

  • - Accounting procedures should allow accurate valuation of assets and liabilities.

  • - Formulas should allow an accurate measurement of various forms of risks.

  • - Timely and accurate information on each specific risk should be included.

  • All banks should:

  • - Adopt written procedures on evaluation of borrowers’ ability to withstand adverse exchange rate changes, along with written procedures on limits on foreign currency lending should be used.

  • - Avoid facilitating currency switching to refinance loans as a device to delay potential defaults.

  • - Monitor trends and maturity structure of hedged and unhedged share of foreign currency loans to total loans.

  • - Ensure that stress testing indicates absolute limits and management action triggers

  • - Differentiate by currency in “maturity ladders’ of future cash flows.

  • - Adopt a policy favoring foreign currency lending to borrowers that follow hedging practices.

  • - VAR models of risk management should be used if systems and skilled staff allow it

Supervisory guidelines
  • Banks should:

  • - make provisions to incorporate corrections in the cost of borrowing for exchange rate risks that may have been neglected.

  • - evaluate interest and exchange rate incentives introduced by macroeconomic policy.

  • Banks should:

  • - monitor share of foreign exchange loans to total loans.

Prudential Guidelines
  • - Supervisors should require additional capital requirements for foreign exchange positions.

  • - Supervisors may also require provisions if unexpected discrepancies between interest rate and exchange rate movements are prolonged.

  • - Supervisors may also require additional capital requirements when inadequate incentives from macroeconomic policy are evident (fixed exchange rates, currency boards).

Appendix: Current Practices for Domestic Lending in Foreign Currency

Patterns Observed in the Restrictions Used to Limit Banks’ Domestic Lending in Foreign Currency

Main restrictions on banks’ domestic lending in foreign currency, as reported for the IMF’s 1999 Annual Report of Exchange Arrangements and Exchange Restrictions related to the Article VIII,30 show the following patterns (see Table 3.3):

Currency-related patterns

  • Industrial countries do not report restrictions. This is consistent with the fact that banks’ domestic lending in foreign currency are perceived as a problem mainly by developing economies. Continuous and sizable cost-benefit incentives for borrowers to increase foreign exchange borrowing are unlikely in these economies.

  • Dollarized economies tend to have no restrictions. The group of countries without restrictions includes Argentina, Bolivia, Ecuador, Indonesia, Peru, Poland, Russia, and Uruguay, all of them showing some degree of dollarization. The large availability of foreign exchange resources severely limits the possibility of effectively restricting the domestic allocation of these resources significantly. Some countries set quantitative limits to the amount that can be lent domestically, without restricting the use of resources (Bahrain, Honduras, Jordan, Lebanon, and Mexico).

  • Some restrictions aim at favoring the use of domestic currency over foreign currency. In some cases, only the on-lending of credit from abroad is allowed (Algeria and Brazil). Residents may relate this to an intention to discourage domestic banks from competing to attract deposits in foreign currency.

Capital-Control Related Patterns

  • Outright prohibitions are rare. Even in developing economies, considerations of the negative effects on arbitrage and efficiency seem to have prevailed in limiting the use of outright prohibitions to this type of operation. Countries reporting such a limitation are Dominica, Macedonia, and Pakistan. Some countries only allow foreign currency loans to nonresidents (Belarus, Bangladesh, and Sri Lanka) and some others only for trade-related purposes (Morocco).

  • An approval by an official entity is the main control mechanism. A large number of countries require an approval by an official entity, which allows the government to impose ad hoc restrictions when deemed necessary. This apparently aims at having the possibility to adapt restrictions to specific needs related to the evolution of capital flows. Countries that show these restrictions include The Bahamas, Benin, China, Cyprus, Fiji, Niger, Papua New Guinea, San Marino, Togo, and Zimbabwe.

Patterns Related to the Use of Resources

  • Some countries restrict the use of resources to foreign exchange payments, while others allow foreign exchange borrowing to finance foreign exchange payments regardless of the source of revenue of the borrower. Countries reporting this practice include Korea, Moldova, Morocco, Ukraine, and Vietnam. Rather than hedging considerations, these restrictions may be explained by an intention to limit foreign exchange financing to facilitate trade.

  • Some countries restrict the use of resources based on the source of revenue of the borrower. Some countries report that natural hedging considerations are the main reason to restrict banks’ domestic lending in foreign currency, such as Croatia, Dominican Republic, Egypt, Philippines, St. Kitts, and Turkey.

Table 3.3.Restrictions on Foreign Exchange Lending
No RestrictionsOutright

Other Specific

AlbaniaBank of Albania may impose credit ceilings on outstanding stock of credits for each commercial bank.
AlgeriaBanks and financial institutions may on-lend foreign funds borrowed abroad.
BahamasExchange control approval is required to make loans to residents.
BahrainLending is limited to 1 5 percent of each bank’s capital base.
BangladeshLending is subject to prior approval by the Bank of Bangladesh.
BelarusPermitted for settlements with nonresidents.
BelizeUnspecified restriction reported.
BeninDomestic lending denominated in foreign exchange or purchases of foreign currency-denominated securities issued in Benin require prior authorization by the Ministry of Finance.
Bosnia and HerzegovinaUnspecified restriction reported.
BrazilAll contracts, securities, or other documents, as well as any obligations executable in Brazil that require payment in foreign currency, are null and void. Consequently, banks are prohibited from granting foreign currency loans within Brazil. However, this regulation does not apply to the on-lending of external foreign currency loans.
Burkina FasoThe same regulations apply as for lending to nonresidents.
BurundiUnspecified restriction reported.
ChadUnspecified restriction reported.
ChinaLending is mainly subject to review of qualifications by the PBC and to asset-liability ratio requirements. Borrowers need to register ex post the transaction with the SAFE and should get a permit from the SAFE to repay the principal.
ColombiaUnspecified restriction reported.
CroatiaDomestic commercial banks may not give foreign exchange credits to resident natural persons. Commercial banks may, under certain circumstances, give foreign exchange credits to juridical persons for activities they are registered for. Foreign commercial banks may extend credit in foreign exchange to both natural and juridical persons.
CyprusAuthorized dealers may grant certain short-term credit facilities in foreign currency (e.g., discounting bills of exchange) to residents without reference to the Central Bank of Cyprus. For other lending to residents in foreign currency, prior approval of the Central Bank of Cyprus is required. Such approval is usually granted for the following purposes: to finance transit trade, to provide working capital for resident oil companies or for industries operating in the industrial free zone, to meet the financial needs of Cyprus Airways, and to finance other desirable productive projects.
Czech Republic
DjiboutiUnspecified restriction reported.
DominicaThese transactions are generally not permitted.
Dominican RepublicBanks authorized to offer services may grant loans in dollars to the sectors exporting goods and services to finance activities specific to their functions and to the importing sectors to cover payments abroad for the acquisition of goods and services.
EgyptBank lending in foreign exchange restricted to borrowers with foreign exchange revenue, and other related exchange controls.
Equatorial GuineaUnspecified restriction reported.
EritreaUnspecified restriction reported.
EthiopiaDomestic banks may grant personal loans to staff members of international institutions.
FijiThe banks and nonbank financial institutions may not lend foreign currency to any resident of Fiji without the specific permission of the Reserve Bank of Fiji.
GeorgiaUnspecified restriction reported.
GhanaUnspecified restriction reported.
GuineaUnspecified restriction reported.
GuyanaUnspecified restriction reported.
HondurasFinancial institutions may lend SO percent of their foreign exchange deposits locally in foreign exchange.
JordanLicensed banks are permitted to lend up to 50 percent of their foreign exchange deposits.
KazakhstanAuthorized banks are entitled to provide credits in foreign currency to resident and nonresident juridical persons only, in accordance with noncash procedures. There are no controls on credits extended to natural persons.
Republic of KoreaThere are no controls on loan ceilings, but there are some restrictions on the use of loans.
LebanonNo restrictions. However, banks may only on-lend up to 65 percent of FCD.
Former Yugoslav Republic of Macedonia
MalawiUnspecified restriction reported.
MalaysiaAuthorized dealers and merchant banks are allowed to lend in foreign currency to residents.
MexicoThere are limits on the amount that banks may tend to individual borrowers and on the open foreign exchange position of banks.
MoldovaAuthorized banks are permitted to lend in freely convertible currencies only to importers that are resident juridical persons.
MoroccoMoroccan banks may grant foreign exchange loans to residents for the financing of foreign trade operations and investment.
NigerThere are no controls on these operations when they involve commercial credits; the prior authorization of the MOFERP is required for loans, financial credits, and purchases of securities issued abroad.
Papua New GuineaSubject to presentation of commercial invoices for bona fide transactions.
ParaguayUnspecified restriction reported.
PhilippinesThe following foreign currency loans may be granted by foreign currency deposit units (FCDUs) of commercial banks without prior BSP approval: (1) private sector loans, if they are to be serviced using foreign exchange to be obtained outside the banking system; (2) short-term loans to financial institutions for normal interbank transactions; (3) short-term loans to commodity and service exporters, and producers/manufacturers, provided that the loan proceeds are to be used to finance the import costs of goods and services necessary in the production of goods. Regular units of commercial banks may also grant foreign currency loans to residents involving trade transactions. Commercial banks may grant: (1) private sector loans if serviced using foreign exchange to be sourced outside the banking system; (2) short-term loans to financial institutions for normal interbank transactions; and (3) short-term loans to commodity and service exporters and to importers. In practice, this regulation does not restrict FCL.
RussiaOnly authorized banks are allowed to lend to residents in foreign exchange.
SamoaUnspecified restriction reported.
Sierra LeoneBanks are not engaged in this lending.
San MarinoThe granting of financial credits of a considerable amount is subject to authorization.
Slovak Republic
Solomon IslandsUnspecified restriction reported.
South AfricaUnspecified restriction reported.
Sri LankaCommercial banks may grant foreign currency loans to experts.
St. Kitts and NevisMinistry of Finance approval is required, which is granted only in the case of projects generating foreign exchange to service the loan. The purchase of locally issued securities denominated in foreign currencies requires Ministry of Finance approval.
SudanUnspecified restriction reported.
SwazilandUnspecified restriction reported.
TajikistanUnspecified restriction reported.
ThailandCommercial lending to particular industries denominated in foreign currencies can be partially (50 percent) included as foreign assets in order to recognize the potential risk that banks may not be fully repaid as exchange rate risk is heightened.
TogoLocal lending in foreign exchange or purchases of securities issued locally and denominated in foreign exchange requires prior authorization by the MEF.
TunisiaResident banks may freely extend credit to finance import and export operations. They may also lend their foreign currency surpluses to other resident banks and to their correspondent banks, in exchange for loans in another currency with the same maturity.
TurkeyResident banks may not extend credits to residents in foreign exchange except to exporters, investors, financial leasing firms, Turkish entrepreneurs working abroad, residents who are conducting business related to international tenders held in Turkey, and residents who are conducting business related to defense industry projects that have been approved by the Undersecretariat of the Defense Industry.
TurkmenistanUnspecified restriction reported.
UkraineThis may be done only for financing a limited range of “critical imports.” The provision of foreign exchange to service private sector loans contracted by residents may not be made at an interest rate higher than 20 percent.
VietnamPermission is given only for import purposes. Domestic lending in foreign currency may be done only for trade-related financing.
Democratic Republic of the Congo (Zaire)Unspecified restriction reported.
ZimbabweSubject to exchange control rules and regulations.
Sources: IMF’s 1999 Annual Report of Exchange Arrangements and Exchange Restrictions (AREAER); IMF, Exchange Arrangements and Exchange Restrictions Database; and Baliño, Bennett, and Borensztein, 1999.

Other restrictions different from those defined by the banking license, prudential regulations applied to domestic loans, and capital/open position limits.

Sources: IMF’s 1999 Annual Report of Exchange Arrangements and Exchange Restrictions (AREAER); IMF, Exchange Arrangements and Exchange Restrictions Database; and Baliño, Bennett, and Borensztein, 1999.

Other restrictions different from those defined by the banking license, prudential regulations applied to domestic loans, and capital/open position limits.


    AbramsRichard andPaulinaBeato and1998“The Prudential Regulation and Management of Foreign Exchange Risk,”IMF Working Paper No. 98/37 (Washington: International Monetary Fund).

    BaliñoTomásJ.T.AdamBennett andEduardoBorensztein1999“Monetary Policy in Dollarized Economies,”IMF Occasional Paper No. 171 (Washington: International Monetary Fund).

    Bank for International Settlements199868th Annual Report (Basel: BIS).

    Basel Committee on Banking Supervision1980Supervision of Banks’ Foreign Exchange Positions (Basel: BIS).

    Basel Committee on Banking Supervision1997Core Principles for Effective Banking Supervision (Basel: BIS).

    CassardMarcel andDavidFolkerts-Landau1997“Risk Management of Sovereign Assets and Liabilities,”IMF Working Paper No. 97/166 (Washington: International Monetary Fund).

    EdisonHali andB. DiannePauls1993“A Re-Assessment of the Relationship Between Real Exchange Rates and Real Interest Rates: 1974-1990,”Journal of Monetary EconomicsVol. 31 (April) pp. 16587.

    EdmondsHoward2000Bank Supervision as a Tool to Reduce Risks in Foreign Currency Funding to Emerging Markets Office of the Comptroller of the Currency (Washington: U.S. Office of the Comptroller of the Currency).

    Folkerts-LandauDavid andCarl-JohanLindgren1998Toward a Framework for Financial Stability World Economic and Financial Surveys (Washington: International Monetary Fund).

    FrankelJeffreyA. andKennethFroot1990“Exchange Rate Forecasting Techniques, Survey Data, and Implications for the Foreign Exchange Market,”NBER Working Paper No. 3470 (Cambridge, Mass.: National Bureau of Economic Research).

    GarciaGillianG.H.1999“Deposit Insurance: A Survey of Actual and Best Practices,”IMF Working Paper No. 99/54 (Washington: International Monetary Fund).

    IzeAlain1995“Reserve Requirements of Foreign Currency Deposits,”Monetary and Exchange Affairs Department Operational Paper No. 95/1 (Washington: International Monetary Fund).

    JohnstonBarry andInciOtker-Robe1999“A Modernized Approach to Managing Risk,”IMF Policy Discussion Paper No. 99/6 (Washington: International Monetary Fund).

    MacDonaldRonald1995“Long-Run Exchange Rate Modeling,”IMF Staff PapersInternational Monetary FundVol. 42 (September) pp. 43789.

    MacDonaldRonald andJunNagayasu1999“The Long-Run Relationship Between Real Exchange Rates and Real Interest Rate Differentials: A Panel Study,”IMF Working Paper No. 99/37 (Washington: International Monetary Fund).

    MarstonDavid1995“Short-Term Absorption of Capital Inflows,”Monetary and Exchange Affairs Department Operational Paper No. 95/3 (Washington: International Monetary Fund).

    McCallumBennett1994“A Reconsideration of the Uncovered Interest Parity Relationship,”Journal of Monetary EconomicsVol. 33 (April) pp. 10532.

    MeeseRichard andKennethRogoff1988“Was It Real? The Exchange Rate-Interest Differential Relation Over the Modern Floating-Rate Period,”Journal of FinanceVol. 43 (September) pp. 93348.

    MeredithGuy1998“Long-Horizon Uncovered Interest Parity,”NBER Working Paper No. 6797 (Cambridge, Mass.: National Bureau of Economic Research).

The authors would like to thank the comments and assistance from the staff of the IMF’s Monetary and Exchange Affairs Department (MAE), in particular participants in the MAE seminar on the subject, and from Richard K. Abrams, Daniel Dueñas, Claudia Dziobek, Charles Enoch, Huw Evans, Peter Hayward, Alain lie, David Marston, Elizabeth Milne, Thordur Olafsson, Mark Stone, and Delisle Worrell, and from Natalie Baumer for her editorial assistance.

Hard currencies of worldwide acceptance include the U.S. dollar, the euro, the yen, and the British pound. Currencies with regional acceptance could be hard currencies relative to those only accepted in the home country. Pegged currencies, to the extent that they are subordinated to a third party’s currency, are in principle weaker, as their sustain-ability depends on the availability of the third party’s currency in central bank reserves.

Currency risk is also denoted as exchange rate risk for the purpose of this chapter.

Discrepancies of ex ante and ex post open foreign positions could be dramatic as illustrated by the case of Chile in the early 1980s.

A measure of transfer risk is a comparison of a country’s ability to meet its external debt obligations from foreign currency reserves, cash flow, credit lines, salable assets, and its access to new foreign currency borrowing. However, it also relates to market arrangements to access foreign exchange and to capital controls.

A recent study by MacDonald and Nagayasu (1999) shows that a relationship between real exchange rates and real interest rate differentials is more of a long-term nature.

In March 1999, the Central Bank of Egypt restricted bank lending in dollars to borrowers with dollar revenues in parallel with the imposition by the government of trade restrictions and exchange controls.

This is true for 21 countries out of a sample of 61—see Garcia (1999).

In Eastern European economies, quick removal of inadequate incentives in favor of foreign currency deposits permitted some reversal of dollarization in the context of adequate macroeconomic policies. This may prove difficult for countries experiencing a long-standing preference for foreign currency. See Ize (1995).

Thus, a company whose revenues are in domestic currency may still borrow in foreign currency and, by swapping the obligations, contract a liability in domestic currency.

As the scope of the chapter is related to unhedged positions, swaps are not further analyzed.

The ratio of initial-final repayment quotas for domestic currency loans against foreign currency loans increases from 1.069 to 1.166 for a rise in the lending interest rate in domestic currency from 15 to 30 percent paired by an equivalent acceleration of exchange rate depreciation at a 5 percent interest rate in foreign currency.

In other words, uncovered interest rate parity may not hold, among other reasons, on account of (1) imperfect markets, including incomplete liberalization of the capital account, insufficiently open financial systems, price rigidities, significant transaction costs, lack of competition, and limited information; (2) time-varying risk premiums, lack of risk neutrality, or irrational expectations of some sort (Frankel and Froot, 1990); (3) continuous policy shocks as a reaction of the monetary authorities to undesired exchange rate fluctuations (McCallum, 1994); (4) predominant impact of real disturbances on exchange rate volatility (Meese and Rogoff, 1988); or (5) slow convergence to equilibrium in the long term (Metedith, 1998; and Edison and Pauls, 1993). The list is by no means exhaustive. It is intended to include the main factors chat may be relevant for the operational aspects of domestic lending in foreign currency.

Deviations from purchase power parity (PPP) may last on average between six to eight years. See MacDonald (1995).

Banks’ domestic lending in foreign currency is frequently related to dollarization, as the most obvious example of its unavoidable widespread practice, connected with the ample availability of deposits in foreign currency by residents. However, it is by no means the only case, neither the most dramatic. The collapse of the banking system in Chile in the early 1980s was the result of widespread lending in foreign currency financed with external loans. Chile converted 47 percent of bank loans originally denominated in foreign currency into domestic currency, and at the same time turned bank private external debt into sovereign debt. Dollarization remained low throughout that period. Cost-benefit incentives related to a fixed exchange rate regime had a more important role in encouraging this practice.

If the government provides direct intermediation or guarantees for foreign credit lines, it may still undermine credibility in the domestic currency by increasing the public debt, even if the additional debt is hedged in international markets through swap or swap options. Industrial economies such as Germany, Japan, and the United States do not issue sovereign foreign currency debt, and the United Kingdom has only issued such debt in European currency units. Canada, Belgium, Denmark, and New Zealand have stopped issuing it. See Cassard and Folkerts-Landau (1997).

See Johnston and Orker-Robe(1999) for a discussion of policy mix consistency and how it could affect capital flows.

In addition, it is usually difficult to ascertain which factors are at work: See Marston (1995).

Or alternatively, it may limit interest rate fluctuations. In Costa Rica, in the early 1990s, direct central bank control of interest rates on domestic currency deposits led to higher returns on foreign currency deposits, contributing to a rapid increase in foreign currency denominated deposits and encouraging foreign currency lending.

A cross-section regression using this data shows that lending to agents that do not generate foreign currency revenue (proxied by foreign currency loans to total loans standardized by exports over GDP) is explained by both the availability of foreign currency deposits and the length of the period when borrowing in foreign exchange was less costly. These results are only illustrative.

See Principles 11, 12, and 13 in Basel Committee on Banking Supervision (1997).

Borrowings for which the adverse exchange rate impact on domestic currency obligations is compensated by a positive impact on revenue and profitability.

Appropriate pricing systems should be applied in which lending rates to unhedged borrowers are inclusive of the implicit or explicit incremental counterparty risk hedging costs for the banks.

For example, such procedures might include monitoring the size and trend of the borrowers’ foreign exchange profits and cash flow, and requesting additional collateral or constituting appropriate provisions on unhedged foreign exchange loans to borrowers with domestic currency-denominated income, according to the magnitude of exchange rate depreciation.

Basle Core Principles 6, 12, and 13 apply, in this case, related respectively to capturing market risk with respect to capital adequacy, maintaining appropriate systems to measure, monitor, and adequately control market risks, and establishing an overall risk management process.

Recent events in Korea provide an illustration. Prior to the Asian financial crisis, Korea had a very crude liquidity requirement, like many other developing economies. In the context of technical assistance, the IMF recommended and Korea adopted a cash flow approach along the lines described above. Banks are now required to analyze the maturity of their assets and liabilities in foreign exchange and ensure that negative mismatches do not exceed certain limits.

Stress testing may involve the results of performance under extreme, perhaps improbable scenarios. Line personnel and a bank’s auditing unit should both conduct stress testing to provide independent oversight and integrity to the process.

In addition, these measures ate usually difficult to implement effectively and uniformly, thus aggravating their negative impact.

Raising capital requirements domestically for foreign exchange operations, which in turn raises bank costs that will be passed on to customers, may provide an incentive to shift foreign exchange borrowing from on-shore to off-shore institutions. Hence, there is a limit to how much arbitrage opportunities should be introduced through this mechanism.

Supervisors are responsible for evaluating each bank’s overall foreign exchange risk profile; the potential impact on future earnings of foreign exchange fluctuations, including the effect induced through mismatching maturities in foreign-denominated assets and liabilities (taking into account the compounded risk of net unhedged domestic lending in foreign currency); and the bank management’s ability to measure and manage these risks (including the appropriateness of vector autoregression (VAR) models and stress test to incorporate the links between the macroeconomic policies and the compounded risk of domestic lending in foreign exchange). General supervisory principles call for more frequent information on foreign exchange operations given that changes in risk exposure can take place rapidly (see Folkerts-Landau and Lindgren, 1998).

Some additional information was incorporated to the list, mainly from the Occasional Paper, “Monetary Policy in Dollarized Economies” (Baliño, Bennett, and Borensztein, 1999).

    Other Resources Citing This Publication