1 ANNEX, Recommendation of the European Systemic Risk Board of 21 September 2011 on lending in fo…

In forcesince 21 Sep 2011

The financial stability concerns arising from excessive foreign currency lending in some Member States have been debated in several fora in the last few years.

At the level of the Union, foreign currency lending to the non-financial private sector has been most prevalent in the central and eastern European (CEE) countries. In these cases, it has led to a build-up of substantial currency mismatches on non-financial private sector balance sheets. The reasons for the prevalence of foreign currency lending stem from both demand and supply-side factors including, among others, positive interest rate differentials and access to funding from parent banks.

High levels of foreign currency lending may have systemic consequences for these countries and create conditions for negative cross-border spillover effects. In some cases, foreign currency lending reached excessive levels and contributed to the reinforcement of credit cycles, having potentially affected asset prices. For foreign currency loans, credit risk includes market risk for all unhedged borrowers, as instalments are affected by exchange rates. These borrowers will tend to behave similarly, and at the same time, due to negative developments in the exchange rate. Moreover, the dependence on parent banks for funding and, in some cases, reliance on the foreign currency swap markets, form an additional layer of liquidity and refinancing risk at times of crisis. Finally, the high level of integration of financial groups creates yet another channel for cross-border contagion in the case of the crystallisation of risk due to foreign currency lending.

Due to the potential for cross-border contagion and the possibility of circumvention of national measures, when taken unilaterally and not subscribed to by other Member States, the ESRB drew up recommendations.

The goals of the ESRB recommendations are aligned with the risks identified: (i) to limit exposure to credit and market risks, thus increasing the resilience of the financial system; (ii) to control excessive (foreign currency) credit growth and avoid asset price bubbles; (iii) to limit funding and liquidity risks; and (iv) to improve risk pricing. The recommendations apply to foreign currency lending, defined as all lending in currencies other than the legal tender of the relevant country. Whenever relevant, the recommendations only cover unhedged borrowers, i.e. borrowers without a natural or financial hedge, meaning agents that are exposed to a currency mismatch.

To address credit risk, recommendations include: (i) increasing borrowers’ awareness of risks embedded in foreign currency lending, by guaranteeing that they are given adequate information; and (ii) ensuring that new foreign currency loans are extended only to borrowers that are creditworthy and capable of withstanding severe shocks to the exchange rate. The use of debt-to-income and loan-to-value ratios is encouraged. Whenever foreign currency lending is inducing excessive overall credit growth, more stringent or new measures on foreign currency lending should be considered.

To tackle mispricing of risks associated with foreign currency lending, authorities should require institutions to (i) better incorporate these risks in their internal risk pricing and internal capital allocation; and (ii) hold adequate capital, under the Second Pillar, for foreign currency lending due to the non-linear relationship between credit and market risks.

Authorities should closely monitor and, if necessary, consider imposing limits on funding and liquidity risks associated with foreign currency lending, paying particular attention to concentration of funding sources, currency and maturity mismatches between assets and liabilities and the resulting reliance on foreign currency swap markets.

The recommendations should be applied at an individual, sub-consolidated and consolidated level, as appropriate. Member States should contribute to impeding regulatory arbitrage by applying reciprocity towards other Member States that have implemented measures to limit risks associated with foreign currency lending. Supervisory actions may also be discussed within the colleges of supervisors.

The presence of foreign currency lending is significantly diversified across the Union. While for most western European countries foreign currency lending accounts for a relatively negligible share of total loans, it is relatively high in central and eastern European (CEE) countries [CEE countries are Bulgaria, Czech Republic, Hungary, Poland, Romania, Slovenia, Slovakia, Estonia, Latvia and Lithuania and third countries such as Croatia and Serbia.] and Austria (see Chart 1).

In countries with a high share of foreign currency lending, the phenomenon is often visible in both lending to households and to non-financial corporations. In contrast, in countries where foreign currency lending accounts for a relatively low share of total loans, non-financial corporations tend to borrow more in foreign currencies compared with households. This may be linked to the presence of export-oriented companies as well as an overall degree of trade openness.

The risks to financial stability are predominantly high in countries with a large stock of foreign currency loans granted to unhedged borrowers. Households and some non-financial corporations (i.e. small and medium-sized enterprises (SMEs) active in the country’s domestic market) in particular tend to be unhedged (i.e. exposed to a currency mismatch) as their income is generally in local currency.

Exporting non-financial corporations may, by contrast, be less sensitive to swings in the exchange rate as they have more opportunities to hedge against the currency risk [Hedging against currency risk can take different forms, including natural hedging, when a household/non-financial corporation receives income in foreign currency (for example remittances/export receipts), and financial hedging, which presumes a contract with a financial institution. The latter is often considered as unavailable to households and some SMEs, mainly due to relatively high costs. Including the unhedged non-financial corporations – for which there are no data available – would, most likely, not change the sample of countries considered in this annex.]. Therefore, the remaining analysis concentrates on countries with considerable share of foreign currency lending to households [Bulgaria, Latvia, Lithuania, Hungary, Austria, Poland and Romania.].

The currency structure of foreign currency lending also differs across Member States (see Chart 2). In the majority of countries analysed (Bulgaria, Latvia, Lithuania and Romania) foreign currency loans have been extended predominantly in euro which seems to be a natural choice given the Union membership and, in particular, regimes of exchange rates fixed to euro. On the other hand, in some countries the dominant role was played by other currencies, especially Swiss francs (for example Hungary, Austria and Poland).

Focusing on countries with a higher share of foreign currency lending to unhedged borrowers (proxied by lending to households), several common features may be identified. First, the share of foreign currency lending has increased since December 2004 across essentially all countries (see Chart 3), except for Austria. At the same time, the share of foreign currency deposits held by the non-financial private sector in these countries increased slightly or, in some cases, declined (except in Latvia, where foreign currency deposits increased notably). These asymmetric shifts in favour of foreign currency lending could be a basic sign of rising currency mismatches on the non-financial private sectors’ balance sheets. Moreover, they indirectly point to the existence of incentives driving foreign currency lending in Member States. In some countries, the share of foreign currency lending to the non-financial private sector increased further since the global financial and economic crisis hit the Member States, while in others it remained broadly unchanged. In several countries, this increase has occurred in the environment of falling credit demand.

When searching for sources of financing of credit growth in these countries, the loan-to-deposit (LTD) ratio can be used as a rough indicator of domestically available sources of financing. A strong increase in the LTD ratio in turn indicates a heavy dependence on foreign capital to finance loans in these economies (see Chart 4). In some CEE countries, foreign capital was funnelled primarily via borrowing from parent companies of financial institutions granting credit [Henceforth, the terms ‘institutions’, ‘financial institutions granting credit’ and ‘financial institutions’ will be used interchangeably and mean all financial institutions that are able to grant credit. These are mainly banks, but all other non-banking institutions that are able to grant credit are included.] operating in these countries and also by tapping the wholesale money markets abroad.

There are several factors driving foreign currency lending, both on the supply and the demand side. On the supply side, rapid foreign currency credit growth in the CEE regions was, to a large extent, a consequence of easy access to wholesale funding (facilitated by positive global liquidity conditions and financing from foreign parent entities). On the demand side, interest rate differentials seem to have played the major role. Even though common reasons can be identified, their importance is likely to differ across countries.

Leaving aside numerous individual supply and demand-side factors, the expansion of foreign currency lending in some of the CEE economies constituted part of a wider phenomenon of foreign-financed demand and/or asset price booms. Further, most of the Member States with a high share of foreign currency lending are converging economies, with often significant catching-up potential. The real convergence process in these countries relied to a large extent on inflows of foreign capital, as domestic savings were insufficient.

In the CEE countries referred to, foreign currency loans have been financed to a large extent by cross-border borrowing in the form of credit lines from parent institutions residing in the rest of the Union. Other credit institutions with ample domestic currency deposit bases tapped the foreign currency swap markets.

Wherever there was a shortage of domestic funding sources, institutions relied on foreign funding [In Hungary and Romania, parent funding accounted for around 50-70 % of the total banking sector’s foreign liabilities. For further details, see Walko, Z., ‘The refinancing structure of banks in selected CESEE countries’, Financial Stability Report, No 16, Oesterreichische Nationalbank, November 2008.] (see Chart 5). The lower stage of development of national capital markets in CEE countries – compared with the early euro area countries – could also have played a role. Specifically, the relative scarcity of longer maturity local currency debt instruments – which could serve as pricing benchmarks or be used to raise long-term funding – could have discouraged institutions from engaging in long-term local currency lending. High costs of securitisation for domestic currency instruments were another factor which contributed to the fact that banks obtained their funding for mortgage loans in foreign currency.

Moreover, financing within an international financial group constituted a relatively cheaper source of funding as compared to those available to local banks outside such groups. This further strengthened other factors promoting foreign currency lending, such as interest rate differentials and margin benefits.

Due to foreign funding availability and by transferring the exchange rate risk to borrowers, institutions have been able to offer loan products with interest rates significantly below the interest rates on domestic currency loans. In some countries (for example, Bulgaria, Latvia), characterised by a high share of foreign currency deposits, institutions could have been motivated to extend foreign currency loans by having access to a large and stable domestic funding base in foreign currency (predominantly euro). In addition, fixed or pegged exchange rate regimes eliminated costs related to exchange rate hedging [This was the case in Bulgaria, Latvia and Lithuania, which have currency board arrangements or have their local currencies pegged to the euro.].

Credit expansion was facilitated by the integration of European financial markets, reflected, inter alia, in the growing presence or heightened activity of already present foreign financial institutions in these economies’ financial systems.

Except for Austria, the share of the foreign banks’ assets in the overall banking sector assets of the seven countries dealt with in analytical part of this annex is close to or higher than 60 % (see Chart 6). Parent institutions’ engagement in foreign currency funding of their subsidiaries has been to a large extent motivated by higher profitability of credit activities in catching-up economies and pursuit of higher market shares in those countries. A large share of foreign-owned banks in domestic financial sectors of the CEE countries has thus created an additional channel of capital inflows, directed mainly to credit markets.

The abovementioned high share of foreign-owned banks in the CEE countries’ financial sectors, together with their substantial growth potential, contributed to a build-up of competition pressures in the credit markets, mainly concentrated on the housing loan market [This preference arose since the origination costs of mortgages are relatively low, a long-term relationship with customers is established (cross-selling opportunities), and mortgages are typically large and have long maturities, therefore facilitating a rapid growth of banks’ assets. Moreover, institutions favoured mortgage loans as these were perceived as less risky than other types of loans due to their collateralisation.]. As a consequence of higher competition, institutions expanded their product range by offering foreign currency mortgage loans which allowed them to provide households with cheaper credit. An attempt to offer products with lower interest rates was also one of the factors behind Swiss franc loans expansion in some of the CEE countries and Austria. Banks offering Swiss francs and JPY loans could compete for market share by offering lower debt service costs than banks offering euro loans.

The impact of competition pressure was two-dimensional. On the one hand, in a competitive environment, more conservative institutions were ‘forced’ to enter the foreign currency lending market in order not to lose their market share, which might have coincided with an easing of credit standards. On the other hand, because of significant interest rate differentials, institutions could also set higher profit margins and fees compared with domestic currency loans and thereby improve their financial results (which also put additional competitive pressure on banks not offering foreign currency loans). In the case of foreign currency indexed loans, institutions made additional profit from exchange rate spreads, when converting loans instalments from/to domestic currency.

Interest rate differentials between the analysed countries and the main advanced economies in Europe were the primary driver behind strong demand for foreign currency lending in the CEE region and Austria (see Chart 7, Chart 8 and Chart 9). Foreign currency loans became particularly attractive in the segment of long-term loans (for example mortgage loans), in the case of which the effect of interest differential on initial monthly repayment is larger than in the case of loans with short-term maturities. In the case of fixed exchange rate regimes, foreign currency lending tended to be cheaper due to a variety of factors, including lower risk premia (for example credit, liquidity).

The fact that euro lending reached the highest levels in economies operating under fixed exchange regimes could be due to several reasons, including lower liquidity premiums on euro debt instruments and the low perceived exchange rate risk which could have motivated higher demand for euro lending in these countries. Some borrowers might have been unaware of the risks they were engaging in when taking out a foreign currency loan. Even those who were informed might have taken unhedged foreign currency positions as they assumed that these were implicitly guaranteed by the existing currency regime. To some extent, these assumptions seemed to have been positively validated during the recent crisis, especially in CEE economies with currency board arrangements or pegged exchange rate regimes as they did not devalue, although in Latvia maintaining the peg required a Union/International Monetary Fund (IMF) supported programme, mainly due to pro-cyclical fiscal policies and liquidity squeeze in global financial markets.

Exchange rate developments could also have supported demand for foreign currency lending in some countries with floating exchange rate regimes [A review of various studies of foreign currency loans showed that foreign currency volatility is the most robust determinant of foreign currency lending, apart from foreign currency deposits and real exchange rate and inflation volatility: see Hake, M., ‘Determinants of foreign currency loans in CESEE countries: a meta-analysis’, presentation at the 69th East Jour Fixe of the Oesterreichische Nationalbank, June 2011.]. In Austria, low historic euro/Swiss franc volatility contributed to a perception of low exchange rate risk. In the CEE economies with floating exchange rates, borrowers were attracted to foreign currency loans by prolonged nominal exchange rate appreciation and expectations of further appreciation. Borrowers’ expectations of nominal exchange rate appreciation were to some degree self-fulfilling [Loans were mostly denominated in or indexed to foreign currency and the funding was in (or was transformed into) foreign currency, but borrowers received loans in local currency. This means that institutions were selling foreign currency funds, sourced from parent companies or wholesale markets, or received under swap contracts, on the spot market, exerting an upward pressure on domestic currencies.]. The appreciation exacerbated external imbalances which built up as a result of strong domestic demand growth.

The perceived risks of euro lending and borrowing in some of these countries were likely to have been affected by the expectations of near-term euro adoption. Such expectations supported both the assumption of ‘zero’ exchange rate risk in the case of countries with fixed/pegged exchange rate regimes and the presumption of a sustained nominal exchange rate appreciation trend in the case of countries with floating exchange rate regimes.

While this section focuses on the main risks arising from foreign currency lending, it is acknowledged that there are also benefits stemming from both financial integration and sustainable levels of foreign currency lending.

Banks engaged in foreign currency lending are exposed to indirect exchange rate risk (as a component of credit risk) through currency mismatches on their clients’ balance sheets. A significant depreciation of the local currency translates into an increase in the local currency value of outstanding debt (also in relation to the value of collateral) as well as in the flow of payments to service the debt. As a consequence, the debt-servicing capacity of unhedged domestic borrowers deteriorates, leading to a significant weakening in the financial condition of the private sector. The reduction of borrowers’ ability to service loans [Domestic currency depreciation may even reduce the borrower’s willingness to pay, because, for example, the value of the loan exceeds the value of collateral. This mechanism is however more prevalent in markets (for example a large part of the United States residential mortgage market) where banks restrict their recovery efforts to collecting on the collateral and do not pursue repayments from other assets and income of the borrower.] and a lower recovery rate affects the loan portfolio quality, increases banks’ loan losses and puts pressure on earnings and capital buffers. While not included in the scenario of the EBA’s Union-wide stress test, the EBA emphasised in its report that in some Member States the main risk may be an adverse currency movement associated with an impact on foreign currency loans [See ‘2011 EU-wide stress test aggregate report’, European Banking Authority, 15 July 2011, p. 28.].

Calculation of the exact extent of exchange rate (and interest rate) risks of foreign currency loans is difficult. The traditional risk calculation methods do not take into consideration that foreign currency bank loans to unhedged borrowers combine market and credit risk in a highly non-linear way [This issue was investigated in a study led by the Oesterreichische Nationalbank and conducted by a working group of the Basel Committee Research Task Force. See Breuer, T., Jandacka, M., Rheinberger, K. and Summer, M., ‘Does adding up of economic capital for market- and credit risk amount to conservative risk assessment?’, Journal of Banking and Finance, Volume 34(4), 2010, pp. 703-712.]. Academic literature illustrates how standard risk management approaches treating different types of risks separately may lead to a substantial underestimation of the overall risk involved. Simply adding up the separately measured exchange rate and default risk components underestimates the actual level of risk by a factor of several times.

Finally, the interest rate risk profile of foreign currency loans differs from the risk profile of domestic currency loans. This can be detrimental to the quality of foreign currency loans if the interest rate cycles of the foreign currency diverge from that of the domestic economy. However, the extent of the exchange rate and foreign interest rate risks differs significantly for different currency pairs, as well as due to pricing regimes prevalent in each country.

In countries with fixed/pegged regimes, the exchange risk of foreign currency loans did not materialise during the crisis, as local currencies did not devalue and remained pegged to the euro. As a consequence, borrowers in foreign currency did not suffer from currency devaluation but rather benefited from euro interest rate cuts.

This article is long: the rest is at the official publication.