II. Credit Boom in Ukraine: Risks for Banking Sector Stability14
A. Introduction and Summary
22. Rapid credit growth in Ukraine has raised concerns about risks of distress in the banking sector. The credit boom contributed to financing the strong recovery of the Ukrainian economy but average annual real loan growth of 46 percent since end-2000 or 5 percentage points of GDP was among the highest in transition economies. While catching up from a very low degree of financial intermediation and monetization, the credit to GDP ratio has now reached a level comparable to countries with much stronger structural and institutional environments in terms of banks’ risk management capacities, banking supervision, creditor rights, accounting and auditing standards, transparency, and legal framework. The rapid credit expansion therefore entails risks in terms of loan quality, in particular due to the large share of foreign currency-denominated loans and the relatively weak financial conditions of some Ukrainian banks.
23. Recognizing those risks, the National Bank of Ukraine (NBU) has recently taken numerous steps to tighten prudential regulations and banking supervision. The NBU raised the minimum capital adequacy ratio to 10 percent, effective March 2004, tightened related-party lending rules, strengthened the definition of capital while facilitating the issuance of new capital, established interest rate ceilings for banks borrowing abroad, introduced a real time monitoring system for interbank activities, and put in place a new risk assessment methodology for its supervisors.
24. However, further institutional improvements are needed. A number of additional measures are already in progress and should swiftly be pursued. Amendments to the Banking Act which would require identification of ultimate banks owners and further tighten related-party lending—a performance criterion under the Fund-supported program—have been submitted to parliament. Revisions in provisioning requirements for foreign currency loans have been discussed with banks. Introduction of minimum risk management standards for banks is envisaged and setting up a credit bureau is under way. Until these measures are put in place and fully bear fruit, the NBU should consider raising the capital adequacy ratio to 12 percent and raising the general provisioning ratio back to 2 percent. Closely monitoring banks risk assessment and management practices and acting promptly in response to excessive risk taking is key. Strengthening accounting and reporting standards for corporates, which still inhibit the ability to properly identify risks, and eliminating institutional weaknesses that impact seizure of collateral and enforcement of contracts are other crucial areas for reform.
25. Macroeconomic policy adjustments can support the containment of credit risk. In light of emerging inflationary pressures and an unwinding fiscal deficit, a tighter monetary policy stance would contribute not only to containing inflationary risks but also help stem credit risk. Allowing for more exchange rate flexibility would contribute to raising the awareness of foreign exchange risks from unhedged foreign currency borrowing.
B. Causes and Characteristics of the Credit Boom
26. Credit growth in Ukraine has been rapid since the decline in real GDP was reversed in 2000 and has only decelerated somewhat during 2004. Real loan growth averaged 45.9 percent per annum since 2000 and peaked at 60 percent in November 2003 (Figure 1, Table 1). Since then nominal and real credit growth has slowed to rates below 60 percent. The ratio of credit to GDP more than doubled between 2000 and end-June 2004 from 12 percent to to 29 percent. The acceleration, which culminated in an 8 percentage points increase during 2003, seems to have come to an end in 2004 but growth rates were still high at 7 percentage points in the first half of the year.
|Credit growth 1/||43.4||61.8||40.9||47.5||63.7||55.2|
|o/w credit growth to households||35.1||40.1||44.6||133.4||179.4||122.0|
|Real growth of credit||24.2||36.0||34.5||48.1||55.5||55.2|
|Credit to GDP ratio||9.9||12.3||14.4||19.3||27.0||29.4|
|Share of credit to total bank assets||47.1||52.7||58.4||64.9||67.9||68.7|
|(In percentage points)|
|Change of credit to GDP ratio||1.1||2.4||2.1||4.8||7.7||6.7|Figure 1.Ukraine: Real Credit Growth, 1999-2004
(In percent; y-o-y) 1/
Citation: 2005, 20; 10.5089/9781451839043.002.A002
Sources: National Bank of Ukraine; and staff estimates.
1/ Credit to the economy deflated by changes in the CPI.
Credit by economic sectors, currency, and maturity
27. While loans to enterprises still make up the bulk of bank lending, consumer loans have risen significantly over the past two years more than doubling their share to 13½ percent. Trade and agriculture have also experienced above average credit growth (Table 2).
|(In percent of total loans)|
|Loans to individuals||5.8||5.0||5.0||7.9||13.2||13.3|
28. Loans denominated in foreign currency continue to play an important role accounting for 37 percent of all loans(Table 3). In particular, households have turned to foreign currency credits which now account for nearly 60 percent of their bank debt. Most enterprises which borrow in foreign currency, according to anecdotal evidence, are not hedged and often do not have foreign currency income, except for exporters. Data on the share of unhedged borrowers is not available but one can assume a share of at least one quarter of total loans if one considers that all household loans are unhedged and exporters have a share of 43 percent in enterprise loans (equal to the share of exports in private demand).
|Fx credits to total credits 1/||44.8||41.0||41.1||41.1||38.4||37.3|
|Fx credits to households to total credits to housholds||7.6||14.7||29.9||40.0||55.3||59.1|
|Fx deposits to total deposits||44.4||39.0||32.9||32.2||32.2||30.5|
|Fx credits to fx deposits||104.6||115.6||141.8||138.5||137.5||137.8|
|Long-term deposits to long-term loans||25.4||32.6||57.7||77.2||64.6||69.1|
|Short-term deposits to short-term loans||125.6||109.6||99.6||94.6||111.7||118.1|
|Long-term fx deposits to long-term fx loans||20.3||30.3||43.7||61.8||50.6||51.2|
|Long-term credits to total credits||22.4||18.0||21.7||28.2||45.0||49.3|
|o/w Long-term fx credits to total credits||13.9||9.4||11.8||14.5||23.2||24.3|
|Long-term deposits to total deposits||5.5||6.1||13.8||24.3||32.2||36.2|
|o/w long-term fx deposits tot total deposits||2.8||3.1||8.1||14.3||19.2||23.0|
|(In millions of hryvnias)|
|Diff. between long-term fx deposits and loans||-1,309||-1,288||-1,883||-2,321||-7,773||-9,611|
|Banks’ other long-term fx funding||424||422||572||860||2,302||3,007|
|Banks’ net foreign assets||1,434||2,663||2,494||656||-1,924||-2,985|
29. A remarkable shift to loans with longer maturities has occurred but at the same time the maturity mismatch between foreign currency denominated loans and funds has widened (Table 3). Half of loans were long term (maturity greater than one year) at end-June 2004 compared to only 18 percent in 2000. With longer term deposits rising even more strongly, the maturity mismatch between total deposits and loans narrowed continuously until end-2002. However, over the past 1½ years the maturity gap has widened, particularly between foreign currency deposits and loans. At end-June 2004, long-term foreign currency loans exceeded long-term deposits and other long-term foreign currency funds by Hrv 9.6 billion ($1.8 billion) or 75 percent of banks’ capital.
30. During the lending boom, banks have extended loans at very high, though rapidly falling, real lending rates and spreads. Lower operating and provisioning costs have enabled banks to cut lending rates so that spreads have fallen by 18 percentage points since end-2000 (Figure 2).
Figure 2.Ukraine: Real Bank Lending Rate and Spread, 1999-2004
Source: National Bank of Ukraine; and staff estimates.
1/ Deflated by CPI inflation.
Origins of the credit boom
31. The strong build-up of NBU’s net international reserves has been the prime source for the expansion of base money and, eventually, credit and broad money (Table 4). Under the de facto fixed exchange rate policy of the NBU, balance of payments (and current account) surpluses, have turned the negative NIR position around in 2001 and led to an accumulation of about US$ 8.1 billion (Hrv 44.5 billion) until end-June 2004.15
|(In millions of hryvnias; y-o-y)|
|I. Broad money growth (=II.+III.+IV) 1/||10,081||13,502||19,115||30,173||34,929|
|Share of broad money growth explained by:|
|II. Base money growth (=m∆B)||8,884||12,059||15,387||19,543||27,059|
|II.a Change in NIR||11,613||17,266||14,666||28,759||44,567|
|II.b Change in NDA||-2,729||-5,207||720||-9,216||-17,507|
|III. Change in multiplier (=∆mB)||855||1,050||2,790||8,169||5,863|
|III.a Change in R/D ratio||172||2,052||1,274||101||-1,705|
|III.b Change in C/D ratio||683||-1,002||1,516||8,068||7,568|
|IV. Combined effect (=∆m∆B)||342||393||938||2,461||2,007|
|(In percent; y-o-y)|
|I. Broad money growth (=II.+III.+IV)||45.5||41.9||41.8||46.5||44.2|
|Share of broad money growth explained by:|
|II. Base money growth||40.1||37.4||33.6||30.1||34.2|
|II.a Change in NIR||52.4||53.5||32.1||44.3||56.4|
|II.b Change in NDA||-12.3||-16.1||1.6||-14.2||-22.2|
|III. Change in multiplier||3.9||3.3||6.1||12.6||7.4|
|III.a Change in R/D ratio||0.8||6.4||2.8||0.2||-2.2|
|III.b Change in C/D ratio||3.1||-3.1||3.3||12.4||9.6|
|IV. Combined effect||1.5||1.2||2.1||3.8||2.5|
|Balance of payments (In percent of GDP)||4.9||7.5||2.7||4.6||na|
|Current account (In percent of GDP)||4.7||3.7||5.8||5.8||na|
32. In the past, the NBU did not actively sterilize these NIR accumulations but reductions in NDA were brought about mainly through higher government deposits. Despite a reduction of NDA by nearly three quarters between 2001 and end-June 2004, base money has grown on average 34 percent in nominal and 28 percent in real terms. The issuance of central bank CDs and deposits, reverse repos, and sales of government papers from the NBU portfolio to counterbalance such developments were only very sparingly used since 2001. Only in light of the recently emerging inflationary pressures, has the NBU shifted to a more active sterilization policy. However, the volumes issued so far since June 2004 have fallen significantly short of the NIR increases. The NBU’s contribution to the reduction in the money multiplier since 2000, by lowering the ratio for reserve requirements, is now also being reversed through various measures.
33. The additional supply of liquidity through the NBU has been matched by a rising demand for broad money so that inflation has generally been kept in check and velocity fallen. Remonetization is attributed to the reestablished confidence in the hryvnia after the 1998/99 crisis, a decline in noncash payments and taxoffsets, increased confidence in the banking sector, as well repatriated capital encouraged by the improved macroeconomic outlook. In addition to a higher overall real demand for money, these factors have caused the cash to deposit ratio to fall and raised the money multiplier so that broad money growth has surpassed base money growth. The major input to money growth has however been the accumulation of NIR as the breakdown in Table 4 shows.
34. Banks have funded the expansion of credit mostly through additional deposits from households and enterprises but increasingly also through borrowing from abroad (Table 5). With the stock of credit surpassing the stock of deposits—at a ratio of 113 percent—, the faster credit than deposit growth has required additional resources. Banks have resorted to foreign short-term funds, which has turned their net foreign asset position negative in September 2003. In response to the risks of these developments, the NBU has introduced interest rate ceilings on foreign borrowings for corporates, banks, and nonbank financial institutions in August 2004. Another source has been additional capital, which has however increased much more slowly than banks’ total assets, so that capital asset ratios declined until April 2004. Net claims on the government and on the NBU are relatively small and have varied year by year.
|Banks’ balance sheet:||(In percent of GDP)|
|II. Net foreign assets||2.0||1.5||0.3||0.1||-0.7||-1.0|
|III. Net credit to government||0.5||-0.3||-0.1||0.4||0.6||0.4|
|IV. Net claims on NBU||0.8||1.8||1.3||1.4||1.8||3.0|
|V. Other items, net||-3.6||-3.9||-3.2||-4.3||-5.2||-5.5|
|Credit to the economy (I.-II.-III.-IV.-V.)||9.9||12.3||14.4||19.3||27.0||29.4|
|(Ratio to GDP, change over previous year, in percentage points)|
|Credits to deposits||103.7||110.0||113.5||113.6||115.2||112.7|
C. Risks for Banking Sector Stability
35. Rapid credit growth is one of the most robust leading indicators for banking distress even though the majority of lending booms has not resulted in banking crises. Numerous studies have found that periods of significant and accelerating credit growth often preceded banking crises.16 The likelihood for a banking crisis to follow a lending boom is estimated to be as high as 20 percent, depending on the data set and methodology used. Prominent examples include the Scandinavian banking crises in the early 1990s, the Asian financial crisis in 1997/98, and Mexico’s banking crisis in 1994. As depicted in Figure 3 and Table 6 the ratio of credit to GDP increased rapidly in those countries, averaging 5.2 percentage points per year in the five years leading up to the crisis.
Figure 3:Credit-to-GDP Ratio in Banking Crisis Countries 1/
Sources: IMF International Financial Statistics; and staff estimates.
1/ t = year of crisis.
36. There are several reasons why excessive lending booms entail risks for future banking distress. First, risk assessments suffer due to the vast amount of new loans extended. Loan officers may be overburdened and agree to riskier loans which are not appropriately priced. Second, the perceived risks of loans are underestimated during lending booms because the risk assessments are based on the current strong economy and rising values of underlying collateral. As a result, in most countries lending is strongly procyclical: in upswings, lending is extended much faster than real GDP and in recessions it contracts stronger than output. Third, lending booms can facilitate “ever-greening” when new loans are used to service existing debt. And fourth, such type of credit supply behavior can also prevent banks from further diversifying their loan portfolio.
37. The risks for financial sector stability are predominantly linked to the speed of credit growth rather than the stock of credit or money. While the rapid expansion of money and credit may very well be a sign of “catching-up”, remonetization, and deeper financial intermediation, and does not necessarily have negative macroeconomic (inflationary) implications, it entails important risks for the stability of the financial sector. The vulnerabilities of the banking sector depend on the risks that banks have taken on board in relation to their financial cushion. Numerous factors contribute to this, such as the quality of banks’ credit policies, the quality of collateral, the level of provisions, diversifications of loans, foreign exchange risk, maturity mismatches, and access to other income sources.
38. Nevertheless, the level of financial intermediation impacts the costs for the economy from financial sector distress. An economy where the banking sector and the level of credit are sizable, would be more strongly affected from distress in the financial system than an economy where only a small fraction of individuals and enterprises uses banking services. Empirical studies have however not yet attempted to distinguish the costs based on the degree of financial intermediation which has varied strongly in banking crises countries, from as low as 10 percent in some sub-Saharan African and Latin American countries to as high as 120 percent in Thailand. The costs of full-blown banking crises have been estimated to range up to 50 percent of GDP, depending on many factors, such as the severity and length of the crisis and the responses by the authorities.17
Risks for the Ukrainian banking sector
39. The rapid increase in loans has exposed the Ukrainian banking sector to significant credit and exchange rate risk that could result in financial distress if the economy were to suffer negative shocks. Even though the macroeconomic performance has been strong and many vulnerability indicators for currency crises have been improving, such as the level of public debt, the current account balance, and the level of reserves, the lending boom is a serious concern and needs to be monitored carefully. The main factors that constitute this concern are discussed below.
40. The lending boom presents a severe challenge for banks’ risk management and loan practices which are still evolving. The sheer volume of additional loans—having increased sevenfold over the past five years—must have placed great strain on banks’ ability to evaluate loan applications with due care. Risk management practices are still evolving in many Ukrainian banks, in particular since there are only few foreign banks in Ukraine that bring international practices to the sector (see Table 7 for the structure of the Ukrainian banking system).
|Number of banks||153||152||157||158||157||158|
|o/w: 100% foreign-owned||7||6||7||7||6||6|
|Share of assets of largest 10 banks||55.3||52.5||54.1||53.7||54.2||54.9|
|Share of assets of largest 25 banks||71.4||66.8||71.0||71.7||71.5||72.3|
|Number of bank with assets less than $150 million||145||141||140||132||127||127|
|Regulatory capital to risk-weighted assets||15.5||20.7||18.0||15.2||14.8||15.2|
|Capital to total assets||16.2||15.6||14.9||12.3||12.1||12.1|
|Credit growth (year-over-year)||61.3||40.5||47.3||63.4||61.4||55.0|
|Credit to GDP ratio||12.4||14.5||19.4||27.0||28.2||29.7|
|Change of loan to GDP ratio (in percentage points)||2.4||2.1||4.8||7.7||1.2||1.5|
|Loans in foreign currency to total loans||41.4||41.3||39.5||38.5||38.4||37.4|
|NPLs to total loans 1/3/||29.6||25.1||21.9||28.3||28.0||27.5|
|NPLs (excl. timely serviced substandard loans) 2/||8.5||8.2||7.9|
|Loans classified as doubtful and loss to total loans||16.8||9.9||7.2||7.2||6.9||6.6|
|NPLs net of provisions to capital 3/||68.0||62.9||63.8||143.8||150.5||147.8|
|Specific provisions to NPLs 3/4/||38.4||39.2||39.6||22.7||21.8||21.5|
|Specific provisions to total loans||11.3||9.6||8.7||6.4||6.1||5.9|
|Earnings and Profitability|
|Return on assets (after tax; end-of-period)||0.1||1.2||1.2||1.0||1.0||1.0|
|Return on equity (after tax; end-of-period)||0.5||7.5||8.0||7.6||7.5||7.6|
|Net interest margin to total assets||6.3||6.5||5.6||5.8||5.0||4.9|
|Interest rate spreads (in percentage points; end-of-period)|
|Between loans and deposits in domestic currency||28.5||18.9||14.9||9.9||8.7||6.8|
|Between loans and deposits in foreign currency||10.2||5.4||5.3||4.0||5.6||5.5|
|Between loans in domestic and foreign currency||21.3||18.0||10.5||6.9||5.0||4.5|
|Between deposits in domestic and foreign currency||3.4||4.6||0.9||1.1||2.0||3.2|
|Liquid assets to total assets||20.8||15.3||13.5||15.3||17.3||15.5|
|Customer deposits to total (non-interbank) loans||59.1||87.6||87.6||86.6||87.2||89.5|
|of which: foreign currency deposits to total deposits||44.4||32.9||32.2||32.2||30.3||30.9|
|Sensitivity to market risk|
|Net open positions in foreign currency to capital||32.9||22.3||21.5||17.7||17.3||17.9|
|Foreign currency loans minus foreign currency deposits to capital||49.6||47.2||49.2||59.1||68.3||55.1|
41. Fundamental structural weaknesses, including lack of transparency, also remain in the enterprise sector which impedes risk assessment. The framework for creditor’s rights and insolvency continues to be weak even though the recent adoption of the Mortgage Law and the Registration Law are important steps. Significant differences also remain between national accounting and auditing standards and international standards which make proper credit risks assessments for banks more difficult. Moreover, the current legal and regulatory framework for corporate governance is weak and seen as an impediment to investment. Transparency of state enterprises is another important aspect in particular since their bank borrowing has also increased, although more slowly than for the rest of the economy.
42. In addition to these difficulties, there are signs that signal imprudent behavior of some banks. Related-party lending has been a widespread practice but the NBU has stepped up its efforts to contain it (see Section III for more details). Moreover, some banks have reportedly attracted deposits by offering very high deposit rates and funded themselves short-term at high rates on the interbank market to on-lend to risky borrowers.
43. The share of foreign currency lending remains high (Table 7). Most borrowers, including the rising share of households, are unhedged and do not receive cash-flows in foreign currency. Given the NBU’s de facto fixed exchange rate policy over the past few years, a strong perception has developed among enterprises and individuals that it is safe and cheaper to borrow in foreign currency at lending rates, approximately 700 basis points below hryvnia loan rates. Another risk is the rising maturity mismatch between foreign currency loans and foreign currency deposits and other foreign currency funding which is mostly short-term.
44. Compared to other transition economies that experienced rapid credit growth, structural and institutional conditions are much weaker in Ukraine. In a recent IMF Working Paper, Cottarelli, Dell’Ariccia, and Vladkova-Hollar (2003) find that factors, such as the degree of structural reforms, the degree of foreign (or private) ownership, and strength of creditor’s rights have contributed to lending booms. Since Ukraine lags behind on this front to many transition countries while its credit-to-GDP ratio is well within the range of countries with stronger structural environments (Figure 4)18 and the speed of credit growth among the highest of transition economies (Figure 5), this gives rise to concern that the credit boom could be unsustainable. An indicative example is the banking crisis in Latvia in 1995 which came amidst strong credit growth against the background of fraudulent activities and weaknesses in the regulatory framework, in particular regarding connected lending and interlocking ownerships patterns between banks and enterprises.
Figure 4:Transition Economies: Credit-to-GDP Ratio and Institutional Reform, 2003 1/
Sources: IMF International Financial Statistics; EBRD Transition Report 2003; and staff estimates.
1/ Private sector credit.
Figure 5:Credit Growth in Transition Economies, 2001-03 1/
Sources: IMF International Financial Statistics; and staff estimates.
1/ Bank credit to the private sector.
45. Compared to prominent crisis countries, the stock of credit in Ukraine is still much lower but differences in the speed of credit expansion have narrowed. Ukraine’s credit-to-GDP ratio of 29 percent at end-June 2004 is still less than half of that in the crisis countries shown in Figure 6. Moreover, the episodes of credit expansion in the crisis countries were more prolonged. The annual increase in credit to GDP averaged about 5 percentage points over the five years before the crisis. This compares to 4 percentage points in Ukraine, but credit growth has accelerated to 5 percent of GDP over the past 3½ years. Even though the credit boom in Ukraine can be attributed to catching-up from a very low level of financial intermediation and has contributed to funding the strong economic recovery, the speed itself inhibits the above discussed risks since institutional and structural reforms have lacked behind.
46. Relatively high levels of adversely classified loans continue to plague banks. The high share of loans classified as “substandard”, doubtful,” and “loss of 27.5 percent at end-June 2004 reflects to some extent strict loan classification rules but nevertheless credit quality remains a concern (Table 7). Loans are classified according to three criteria: (i) status of debt servicing, (ii) financial conditions of the borrower, and (iii) collateral. In addition to past due criteria which are stricter than in many countries, loans automatically fall in the category “substandard” if the loan is to a “class C” borrower even when the loan is serviced timely. This includes all loans to borrowers without credit history. Other “class C” borrowers exhibit profitability below industry average, but this is not viewed as a sign of higher risk but rather tax evasion. A jump in that category occurred in August 2003, when the NBU tightened its classification rules. NPLs peaked at end-2003 at 28.3 percent but have since fallen gradually as have the lowest two classification categories. Adjusting NPLs by the timely serviced loans reduces the ratio from 27.5 percent to 7.9 percent but does not lift concerns about credit quality. The revision is based on an NBU survey among banks which finds 94 percent of “substandard” loans being repaid on time. A better understanding of the repayment risks for loans in that category, however, would also require a better understanding of the “true” financial conditions of the borrowers. Even at 8 percent the level of NPLs is higher than in many other transition economies (Figure 7), except those which have similarly tight classification rules such as Kazakhstan and Poland.
Figure 7:Asset Quality in Emerging Europe, 2003
Source: IMF Global Financial Stability Report.
1/ Provisions to NPLs not available for Estonia.
47. The buildup of cushions, in form of capital and provisions, has not kept pace with the rapid credit expansion. The fall in the ratio of regulatory capital to risk-weighted assets has only been reversed in April 2004 (Table 7). The ratio stood at 15.2 percent at end-June 2004, but for the largest 10 banks, it was only 1½ percentage points above the minimum requirement. Moreover, the NBU indicated that the quality of capital in some banks has been impaired since it included accrued income and revaluation gains on fixed assets. However, a new regulation dealing with these issues came into effect in July 2004. The level of provisions to total loans has fallen steadily despite a tightening in loan classification in mid-2003 which has moved a substantial share of loans from the “watch” to the “substandard” category (Table 7).
48. Profitability of Ukrainian banks continues to be low. Since tax avoidance is reported as a major factor behind the low official profit numbers, it is difficult to assess the true ability to replenish bank capital from retained earnings. Investments in new technology and infrastructure, the level of NPLs, high operating costs, the relatively large shares of fixed assets, and the lack of other income sources than interest income are other factors behind the low profits. Profitability indicators lag behind other transition economies (Figure 8).
Figure 8:Profitability of Banking Systems in Emerging Europe, 2003
(In percent) 1/
Citation: 2005, 20; 10.5089/9781451839043.002.A002
Source: IMF Global Financial Stability Report.
1/ Return on assets 2002 for Latvia.
49. The underdevelopment of capital markets and restrictions on international capital flows have also contributed to a concentration of risk in banks’ assets. Due to the lack of other investment opportunities, additional customer deposits have been used entirely to expand the loan portfolio raising its share in total assets to 68 percent in 2003 compared to 53 percent at end-2000, which is high in comparison to other transition economies (Table 8). Investment in securities has remained low at less than 6 percent and foreign assets have dropped to 7 percent (compared to 13 percent in 2000). The government’s decision to borrow on the Eurobond market rather than domestically has also contributed to this trend.
50. The high level of real interest rates during the lending boom could potentially lead to credit losses. The economy has more than quadrupled its real bank debt since end-1999 by borrowing at average annual real lending rates of 17 percent. Even with a real GDP growth of 7 percent per year, some share of loans has financed high-risk investments for which it is unclear whether risk premiums are appropriate. Since the lending boom facilitates access to new loans, the true level of non-performing loans may easily be underestimated.
51. It is unclear whether loans are adequately priced. Real lending rates and spreads have come down to 9.7 percent and 10.9 percent respectively at end-2003, which is still high but comparable to episodes in more advanced transition economies or other emerging market countries that have also been plagued by structural banking sector problems (Table 9). Given the lack of data it is impossible to disentangle how much of the interest rate level can be attributed to remaining operating inefficiencies and how much is a risk premium. Thus, whether loans are adequately priced is difficult to determine. However, aggressive lending practices (rising loan portfolios and falling lending rates) could be indications for excessive risk taking if pursued by banks with weak capital positions and profitability.
|Transition economies||Real bank lending rates 1/||Spreads 2/|
|Other emerging markets|
52. Rating agencies also have concerns about rapid credit growth. “Recend growth in banking sector assets (primarily loans) has also raised a number of concerns. These include worries about asset quality…. Loan growth has placed further pressure on banks’ capital positions, raising concerns about many banks’ ability to absorb a sharp deterioration in asset quality, whilst also restricting opportunities for further growth” (FitchRatings, The Ukrainian Banking System, March 16, 2004). At present, ten banks are rated, accounting for nearly one third of banking sector assets, and all receive low speculative grades (Table 10) reflecting the low confidence in their financial strength.
|Long term||Short term||Individual bank|
|First Ukrainian International Bank||B-||B||D|
|Long term||Short term||Financial strength|
D. Policy Options
53. Even though the credit boom shows first signs of deceleration, a combination of macroeconomic and regulatory and supervisory policies should be used to contain risks for the financial sector. 19 Allowing for more exchange rate flexibility and curbing strong money growth can go a long way in addressing exchange rate risk and stemming the credit boom. Tighter monetary policy is also called for in light of emerging inflationary pressures. Regulatory and supervisory responses are crucial complementary policies. The NBU has already significantly stepped up its efforts in this area but continuous progress will be needed. Developing domestic securities markets and thereby providing banks with alternative investment and business opportunities is another strategy that contributes to banking sector stability over the long run.
Monetary and exchange rate policies
54. A greater degree of exchange rate flexibility could contribute to moderating the general lack of exchange rate risk perception of borrowers in foreign currency. The NBU’s de facto fixed exchange rate policy has successfully brought inflation down to single digits and stabilized expectations, but has also created an environment in which borrowers continue to take on exchange rate risk with more and more consumers resorting to the “cheaper” form of foreign currency borrowing. More flexibility in the exchange rate, without allowing excessive short-run fluctuations, should contribute to changing that risk perception.
55. The role that monetary policy can and should play for financial stability has received increased attention over the past years(see Box 1). Traditionally, most central bankers viewed monetary policy to contribute to financial stability through its macroeconomic objectives. However, even at times of low inflation, financial vulnerabilities can build up and call for pre-emptive actions to head off potential financial instability. Views still differ on the weight and timing of financial stability considerations. It remains a judgment call in weighing the potential costs that the distress in the financial system could cause versus output costs from curbing an economic boom (without inflationary pressures) through tighter monetary policy and other costs such as moral hazard (see Box 1).
Box 1.Should Financial Stability be an Explicit Central Bank Objective?
The appropriate degree of activism on the part of central banks when pursuing financial stability objectives through monetary policy levers is being discussed critically by policy-makers and academics. The schools of thought range from the view that monetary policy serves financial stability best by focusing on providing low inflation and stable macroeconomic conditions to the view that monetary policy should be more proactive to address those financial vulnerabilities that arise from macroeconomic imbalances even at times of low inflation.
The view that monetary policy should not be guided by financial stability objectives is based on the understanding that financial distress derives primarily from failures of individual institutions at the micro-level that could then spread to other institutions through money markets or payments and settlement systems. A clear division of labor between supervision and regulation with a financial stability objective and monetary policy with a price stability objective follows from this view. Too much monetary policy activism could cause higher variability in output.
The supporters of a more proactive stance of monetary policy to avoid the build up of financial vulnerabilities argue that many systemic crises, including the crises in the Nordic countries, Mexico, and Asia, arose from unbalanced macroeconomic conditions such as rapid credit growth, poor risk assessments, weak financing constraints, and buoyant asset prices while inflation was mostly contained. Banking crises were the result of common exposures to macroeconomic risk factors rather than bank or group specific exposures. Central banks should thus respond to indicators of future financial distress, such as rapid credit expansion and asset price growth, by tightening monetary policy conditions.
On this spectrum of views, a lower degree of activism in monetary policy is supported for example by Ferguson (2004) of the Federal Reserve System while Icard (2004) and Borio and Lowe (2002) from the Bank of International Settlements back a more proactive stance. Both sides agree that “the real question may not be so much whether financial stability should be a central bank objective, but rather how policy makers should weight the objective in reaching policy decisions.”
56. For Ukraine, taking steps to stem base money and broad money growth is desirable from both the financial stability as well as the price stability front. In light of the strong balance of payments and rapid accumulation of international reserves, a more pro-active monetary policy is needed in particular should looser fiscal policy create additional liquidity and put more pressures on prices. The NBU has already reintroduced the use of liquidity absorbing instruments such as reverse repos, NBU certificates of deposits, and deposit taking and has tightened reserve requirements. However, all operations aimed at absorbing liquidity have not yet fully matched increases in international reserves.
Regulatory and supervisory policies
57. Establishing a regulatory framework and supervisory practices that avoid that a deterioration in the credit quality puts significant strains on banking sector is a key objective independent of the speed of credit growth. The NBU has already undertaken many efforts in these areas, in particular over the past 6 months. The new higher minimum capital adequacy ratio of 10 percent took effect on March 1, 2004. Related-party lending was tightened. Amendments to the Banking Act that would require identification of ultimate bank owners and further tighten related-party lending, including by eliminating related-party lending at favorable terms, were submitted to parliament. The definition of capital was strengthened, in particular by eliminating accrued income from capital and tying the use of revalued fixed assets to audit results. New resolutions were issued that facilitate raising new capital. A real-time online monitoring system for interbank activities was established and a new risk assessment methodology for supervisors was put in place and is envisaged to be introduced as guidelines for banks. Interest rate ceilings on borrowing abroad were established. The setting up of a credit bureau is underway with the support of the World Bank.
58. Until these policies fully bear fruit, additional discretionary measures are useful to address specific risks arising from the lending boom which can be reversed at a later point in time. Regulators have been discussing such “specific options” in conjunction with the strong procylclicality of loans and bank profitability which is being observed in many countries.20 Particular options that the NBU should consider include higher minimum capital adequacy ratios, higher or differentiated provisioning rates, and further tightened related-party lending rules.
59. While the increase in the minimum capital adequacy ratio to 10 percent already goes a long way in providing an additional cushion for asset quality deterioration, a further increase to 12 percent should be considered. Strictly enforcing the new minimum CAR, effective March 2004, as well as the tighter definition of capital will contribute to banks with excessive credit portfolio growth and much slower capital growth to rebalance their expansion strategy somewhat. However, given the potential risks already accumulated in some banks and the time it takes to strengthen risk management practices, even higher capital cushions may be needed. Many transition economies have minimum CARs above 10 percent or have lowered them only later in their transition process (Table 11). For example, Lithuania reduced its initial minimum CAR from 13 percent to currently 10 percent. For the Ukrainian banking system, a further increase in the minimum CAR should also have no major negative impact on the competitiveness of the banking sector since capital controls inhibit major bank business going abroad.
60. Raising requirements for general reserves and specific reserves for unhedged foreign currency lending would explicitly ask banks to provision for significant risk taken on board during the lending boom. An increase in the general provisioning requirement back to 2 percent—the level from which it was lowered to 1 percent in January 2003—would be a conservative approach to credit risk and can be reversed once banks’ risk management practices and the NBU’s supervisory practices improved further or there are clear indications that risk taking has not been excessive. Moreover, levying specific provisioning requirements on foreign currency denominated loans to unhedged borrowers would help address banks’ credit risk to this particular group of borrowers and reduce incentives for foreign currency borrowings.21 The NBU is currently discussing this option. Higher provisioning rates for specific types of credits, which carry higher than average risks, is a standard practice found for example in the United States for credit card loans.
61. Tightening further related party regulations is also crucial to stem credit risk. While related-party lending always inhibits particular risk factors that need to be carefully managed, there is a heightened risk during a lending boom when such loans can mask true risk if used for ever-greening and can lead to higher concentrations of banks’ loan portfolios. The NBU has already introduced measures to contain related-party lending at favorable rates by requesting it to be fully matched by set aside capital. Moreover, credit exceeding the related-party as well as single exposure limits will be deducted from capital. Adoption of the draft amendments to the Banking Act, which will help identify ultimate banks owners and fully eliminate related-party lending at favorable terms, is crucial for further risk containment in this area.
62. Institutional weaknesses also need to be tackled. Stronger accounting and auditing practices for corporates would facilitate proper assessment of risks. A simplification of foreclosure rules, e.g. based on contract without court intervention, could raise the intrinsic value of the underlying collateral. Also procedural impediments, such as the role of the State Execution Service should be revisited.
Development of securities markets
63. Developing domestic securities markets would contribute to providing alternative investment opportunities to banks and improve pricing by establishing a yield curve. Making the existing restructured securities more marketable and creating a supportive infrastructure, possibly including a primary dealers system, are two of the policy options. However, for such measures to be successful a conducive public debt management strategy and practices with a view to securities market development are needed. Recommendations to that effect were made during the FSAP, including creating benchmark issues by narrowing the range of government securities, reducing the auction frequency, overissuing at times, and making auction procedures more transparent.22
Bank for International Settlements (2001), Marrying the Macro-and Microprudential Dimensions of Financial Stability (Basel).
Borio, Claudio, Furfine, Craig, and Philip Lowe (2001), “Procyclicality of the Financial System and Financial Stability: Issues and Policy Options”, in: Bank for International Settlements, Marrying the Macro-and Mircoprudential Dimensions of Financial Stability (Basel).
Borio, Claudio, and PhilipLowe (2002), “Asset Prices, Financial and Monetary Stability”, Bank for International Settlements Working Paper No. 114 (Basel).
Cottarelli, Claudio, Dell’Ariccia, Giovanni, and Ivanna-Vladkova-Hollar (2003), “Early Birds, Late Risers, and Sleeping Beauties: Bank Credit Growth to the Private Sector in Central and Eastern Europe and in the Balkans”, IMF Working Paper 03/213 (Washington: International Monetary Fund).
Demirguc-Kunt, Asli and EnricaDetragiache (1997), “The Determinants of Banking Crises: Evidence from Developing and Developed Countries,”IMF Working Paper 97/106 (Washington: International Monetary Fund).
Del Mar Cacha, Maria and R. ArmandoMorales (2003), “The Role of Supervisory Tools in Addressing Bank Borrowers’ Currency Mismatches”, IMF Working Paper 03/219 (Washington: International Monetary Fund).
Drees, Burkhard and CeylaPazarbasioglu (1998), The Nordic Banking Crises: Pitfalls in Financial Liberalization?, IMF Occasional Paper 161 (Washington: International Monetary Fund).
Eichengreen, Barry and CarlosArteta (2000), “Banking Crises in Emerging Markets: Presumptions and Evidence”, Institute for Business and Economic Research, Center for International and Development Economics Research, Paper C00–115.
Ferguson, Jr Roger W. (2004), “Should Financial Stability Be An Explicit Central Bank Objective?” in: Ugolini, Piero, Schaechter, Andrea, and Mark R. Stone (eds.) Challenges to Central Banking from Globalized Financial Systems, IMF (Washington: International Monetary Fund).
Gavin, Michael and RicardoHausmann (1996), “The Roots of Banking Crises: The Macroeconomic Context,” in: Hausmann, Ricardo and Liliana Rojas-Suarez (eds.), Banking Crises in Latin America, (Baltimore: Johns Hopkins University Press), pp. 27–63.
Gourinchas, Pierre-Olivier,Valdés, Rodrigo, and OscarLanderretche (2001), “Lending Booms: Latin America and the World”, NBER Working Paper 8249 (Cambridge, Massachusetts).
Hardy, Daniel C. and CeylaPazarbasioglu (1998), “Leading Indicators of Banking Crises-Was Asia Different?”, IMF Working Paper 98/91, (Washington: International Monetary Fund).
Hoelscher, David S. and MarcQuintyn (2003), Managing Systemic Banking Crises, IMF Occasional Paper 224 (Washington: International Monetary Fund).
Icard, André, “Comments: Should Financial Stability Be An Explicit Central Bank Objective?” (2004), in: Ugolini, Piero, Schaechter, Andrea, and Mark R. Stone (eds.) Challenges to Central Banking from Globalized Financial Systems, IMF (Washington: International Monetary Fund).
Prepared by Andrea Schaechter.
The strong foreign reserve accumulation has continued since and NIR stood at US$10.7 billion on September 10, 2004.
Eichengreen and Arteta (2001) find robustness of these results by testing the findings of earlier studies by Gavin and Hausmann (1996), Kaminsky and Reinhart (1999) and Gourinchas, Valdes, and Landerretche (1999). Other papers that support the importance of lending booms for banking crises are, for example, Drees and Pazarbaioglu (1998), Hardy and Pazarbasioglu (1998), Demirguc-Kunt and Detragiache (1997).
For an attempt to estimate the fiscal costs of selected banking crises see Hoelscher and Quintyn (2003).
This is the case even if one adjusts Ukraine’s GDP for the underestimation of the shadow economy, which is however also present in many other transition economies.
For a discussion on the regulatory policy options to deal with procylicality of bank lending and financial stability see Borio, Furfine, and Lowe (2001).
See for example Bank for International Settlements (2001). Establishing rules according to which prudential regulations change over the cycle is also being discussed. More-forward looking dynamic provisioning requirements to smooth their procyclicality is one example. Additional provisions would be set aside in economic (lending) booms, when loans are extended and risk is acquired, rather than in downturns when the loan quality deteriorates. However, while some countries, such as Spain, have moved in this direction, there are still concerns being raised in accounting circles as the concept would reflect future values rather than realized values and reduce transparency.
Given the low share of foreign currency lending in most industrial countries, there are yet no “best practices” or BIS recommendations on provisioning regulations for partially dollarized economies. Nevertheless, different policy options are discussed in a recent IMF Working Paper by del Mar Cacha and Morales (2003). Specific-to-group provisions are the favored prudential approach in line with current proposal to improve International Accounting Standards and proposals by the Accounting Task Force of the Basel Committee.
See Chapter V for a more detailed discussion.