Romania: Selected Issues and Statistical Appendix
Author: International Monetary Fund
Publication Date:
09 May 2006
eISBN:
9781451832860
Language:
English
Keywords:
ISCR; CR; exchange rate; Romania; EU accession; product; banking system; excess demand; consumer lending; financial system; credit measure; Loans; Bank credit; Currencies; Consumer credit
Download PDF (482.6 KB)

This Selected Issues and Statistical Appendix paper assesses Romania’s external competitiveness by reviewing recent developments in a range of standard indicators and estimating equilibrium real exchange rates. The results suggest that, although Romania’s historical cost advantage vis-à-vis trading partners has eroded since end-2004, on account of a strong real exchange rate appreciation, some undervaluation still remains. Furthermore, evidence suggests that the recent weak output and export performance in some of the traditional exporting sectors mainly reflects the transition toward higher value-added products.

Abstract

This Selected Issues and Statistical Appendix paper assesses Romania’s external competitiveness by reviewing recent developments in a range of standard indicators and estimating equilibrium real exchange rates. The results suggest that, although Romania’s historical cost advantage vis-à-vis trading partners has eroded since end-2004, on account of a strong real exchange rate appreciation, some undervaluation still remains. Furthermore, evidence suggests that the recent weak output and export performance in some of the traditional exporting sectors mainly reflects the transition toward higher value-added products.

II. Credit Growth: Developments and Prospects5

A. Introduction

32. One of the most prominent features of Romania’s recent economic environment has been the rapid pace of private-sector credit growth. Over the past five years, annual credit growth has ranged between 30 to 50 percent in real terms, and credit is still growing strongly—as of February 2006, real y/y credit growth stood at around 49 percent, of which lei-denominated (real) credit growth was 69 percent, and foreign exchange-denominated credit growth was 30 percent. This phenomenon is not unique to Romania. Most economies throughout the Central and Eastern European (CEE) region have experienced similar lending flows, with the result that the ratio of private-sector credit to GDP has increased markedly for almost all of these countries, albeit from a relatively low base.

33. This chapter outlines the key features of credit growth in Romania, placing recent developments within a broader regional context. Romania, like many neighboring countries, is in the midst of a profound transition. After decades of often misdirected development, the key goal of these countries is to bring living standards in line with those of Western Europe. Success is not guaranteed,6 and will require a determined macroeconomic and structural policy effort on the part of the authorities.

34. Catch-up growth in Romania, as in other countries, will entail dramatic change, and part of this process will involve the expansion of Romania’s underdeveloped financial sector. Indeed, recent credit developments are linked in part to a broader region-wide issue—the large-scale inflow of foreign savings. These funds will play a key role in accelerating Romania’s convergence with the EU. However, with a bank-dominated, underdeveloped financial system, large-scale inflows will also have clear implications for the pace of credit growth. Section B of this chapter examines Romania’s recent credit-growth experience within a broader regional framework. From the experience other CEE countries, rapid lending growth appears to be part of an ongoing process of financial development, in which a growing banking sector not only reflects the increased availability of resources, but also helps ensure that these resources are channeled efficiently. Moreover, the empirical evidence suggests that this process is far from complete.

35. However, rapid change is not without risk. Moving quickly to a new equilibrium, although welcome, may often entail increased macroeconomic and financial-sector vulnerabilities. Section C describes in detail some of the key features of Romania’s recent credit growth, highlighting areas in which rapid growth may be making Romania more susceptible to macroeconomic and financial shocks. Section D summarizes the authorities’ recent policy response to these potential risks, and analyzes their likely impact. Section E provides some conclusions.

B. Credit Growth and Convergence

36. Romania’s recent credit growth is broadly comparable to the experience of other CEE countries. As outlined in Cottarelli and others (2003), most countries in Central and Eastern Europe have experienced extended periods of rapid lending growth (30-50 percent) as they moved closer to the European Union. More recently, growth rates have been particularly high in the Baltic countries, Hungary, and Bulgaria; whereas in Poland, private-sector credit growth has eased from the relative high levels recorded previously. Romania, in contrast, is a relative latecomer—Cottarelli (2003) classifies Romania as a “sleeping beauty,” given that growth did not start to take off in Romania until after 2002.7

37. As in most CEE countries, however, the stock of credit in Romania is still very small by worldwide standards—at end-2005, total credit to the private sector stood at 20.9 percent of GDP, compared to a Euro-area average of around 95 percent. In this context, the recent expansion of private-sector credit may simply reflect a “normal” process of financial deepening, in which the stock of credit (as a proportion of GDP) naturally expands to meet the developing needs of a market economy. And as in other countries, this process has been accelerated by a familiar pattern of financial-sector liberalization, fiscal consolidation and, especially, capital inflows.

38. Therefore, the question of whether or not Romania’s credit growth is a matter of concern, is related to the issue of what an appropriate credit stock should be for a country like Romania. This issue has been taken up by a number of recent studies8 that have attempted to estimate equilibrium credit-to-GDP ratios for transition countries. Drawing on the methodology of Schadler and others (2005) (Box 1), and using recently-revised Eurostat data, Table 1 below presents some updated results from this analysis. As illustrated, Romania’s credit ratio is still somewhat below that of other CEE countries, and significantly short of equilibrium—compared to an estimated equilibrium ratio of about 54-60 percent, the credit/GDP ratio for Romania is still only around 21 percent.

Table 1.

Credit to GDP Ratios in transition countries, 2002-2005.

(In percent)

article image
Sources: Eurostat; IMF, International Financial Statistics; Cottarelli, Dell’Ariccia, and Vladkova-Hollar (2003); Schadler, Drummond, Kuijs, Murgasova, and van Elkan (2005); and IMF staff calculations.

Data for 2Q05.

Equilibrium value predicted based on estimates of the long-term cointegrating relationship.

Deviation of the actual from the predicted level.

For 2002, Eurostat data for local currency long-term government debt is unavailable for Slovenia and Romania. The predicted credit ratio in the VECM is based on the average real 10-year government bond yield in the other CECs.

Data for end-2005.

Estimation of Equilibrium Credit-to-GDP Ratios

Following the model of Schadler and others (2005), the equilibrium relationship is estimated within a vector error-correction model (VECM) that includes three variables. The key variable under study is (i) the ratio of nominal bank loans to the non-government sector relative to GDP (credit ratio). Second, (ii) the long-run real interest rate on government bonds (rlti) serves as a proxy for the cost of credit, where the 10 year government bond is deflated by annual inflation 3 years ahead. Deflating by contemporaneous inflation during a period of sustained disinflation would likely have biased downward measured real interest rates. Finally, (iii) the log of purchasing power-adjusted per capita income (ln(ppsinc)) represents the overall financial health of households and corporations—this can be viewed as a proxy for a borrower’s ability to service debt and take on new loans, and accords with actual bank lending practices where, given imperfect information, banks rely on observable measures of repayment ability. Without such market imperfections, the importance of this variable in predicting credit might be reduced.

Given that many transition countries are likely to have been persistently out of equilibrium over the entire sample period, including such countries in the sample would likely bias the estimated results. So, the model’s data sample is taken from the original 11 members of the euro-area as well as Greece. Euro-zone data are taken from the following sources: bank loans to the nongovernment sector are from the ECB; 10-year government bond yields, HICP inflation, nominal GDP, and per capita GDP measured in PPS are from Eurostat. While some empirical studies of credit volume use indicators of financial liberalization and banking sector competition, these are not available as time series for the countries included in the Schadler (2005) framework. Moreover, these and other supply-side factors are likely to influence the dynamic adjustment in credit, rather than the equilibrium credit to GDP ratio.

Based on the maximum eigenvalue and trace tests, there is a single cointegrating relationship between the three variables that is significant at the 1 percent level:

creditratio = 32.52 ln(ppsinc) − 1.85 rlti.

This estimated long-run relationship indicates that the credit ratio is positively related to per capita income and negatively related to the real rate of interest. The coefficient on the income term can be interpreted as a semi elasticity: where a 10 percent increase in per capita income raises the credit to GDP ratio by about 3 percentage points in the long run. A rise in the real interest rate by 1 percentage point lowers the equilibrium credit ratio by nearly 2 percentage points. Applying this model out of sample to the set of transition countries provides an estimate of their long-term equilibrium credit ratio—the credit stock to which these countries will eventually converge as they move closer to the EU. Detailed results and discussion of the methodology are outlined in Schadler and others (2005), Annex 5.2.

39. Similar to other countries in the region, the stock of credit in Romania has closed steadily on its predicted long-term value. The rate at which most transition countries have approached equilibrium has been relatively modest, although the Baltics have progressed somewhat more rapidly over the past few years. And while Romania is indeed a “late riser”—as the current stock of credit is significantly below that displayed by other countries—this difference in part reflects Romania’s relatively low income level (and lower equilibrium level of credit).

40. This longer-term structural view of credit growth has significant policy implications, as lending will most likely continue to grow as Romania converges with the EU. When interpreting the dramatic growth rates of recent years, and when considering the optimal pace of credit growth in the future, the authorities should keep in mind that credit developments will partly reflect an underlying process of financial deepening. This process, in turn, has two key dimensions. Over the medium term, Romania is moving from one state in which the economy (or some sectors) are financially underserved, to a new state in which the depth of financial services will better correspond to the economy’s fundamentals (representing movement toward an equilibrium level of credit). At the same time, the equilibrium level of financial depth will itself evolve as Romania gets richer.

41. Indeed, there is a significant literature suggesting that financial development and longer-term economic growth are closely intertwined. In principle, the direction of causality may run in both directions. As the economy grows, the demand for financial services also expands, prompting an eventual supply response. On the other hand, a more developed financial sector will also help ensure an efficient allocation of resources, boosting investment, productivity, growth, and welfare. Overall, there is a significant body of theoretical and empirical research that supports the latter view.9 In this light, efforts to unnecessarily constrain the expansion of credit, if successful, may risk damaging Romania’s longer-term growth prospects.

42. Unfortunately, however, rapid credit growth may entail substantial shorter-term risks. In terms of macroeconomic stability, rapid credit growth can facilitate the expansion of excess aggregate demand—potentially adding to inflationary pressures, and perhaps prompting a widening of external imbalances. In terms of financial stability, rapid credit growth may place undue strain on banks’ ability to assess risk, leading to poor lending decisions, falling asset quality, and potential overexposure to financial risks. Moreover, macroeconomic and financial risks are often interrelated. On the one hand, macroeconomic instability (in the form of inflation or external imbalances) can contribute to financial instability, especially when banks and borrowers are exposed to interest and exchange-rate risk. On the other hand, a vulnerable financial system may add to macroeconomic instability, as markets often react suddenly when adjusting investment portfolios or currency holdings, with significant effects on the real economy.

43. In this context, policy makers will often face a difficult dilemma. There is no simple way to determine whether an observed rate of credit growth is a cause for concern, especially against a background of ongoing structural change. However, a necessary starting point for any policy response should be a comprehensive assessment of credit growth, with a particular focus on the evolution of potential macroeconomic imbalances as well as any adverse trends in the overall resilience of the financial system. This is the topic of the next section.

C. Recent Developments

44. Credit growth over the past few years has reflected a dramatic increase in the demand for new loans, particularly in the household sector. Indeed, growth rates for household lending in Romania are relatively high by regional standards (Figure 1). With an increased level of overall confidence and an improved ability to service debt, this sector has started to address its longstanding demand for durables and real estate. And more recently, households’ willingness to take on debt has been boosted further by a significant increase in incomes, combined with a drop in personal taxes and falling domestic interest rates. However, it should be noted that these high growth rates reflect, in part, a comparatively small initial level of household lending. As illustrated in Table 2, the stock of total credit extended to households has risen steadily since 2000, from a low level of ½ percent of GDP to the current level about 7½ percent (36 percent of total credit).

Figure 1:
Figure 1:

Growth of Credit to the Non-government Sector, 2002-05 1/

(Average year-on-year percent change)

Citation: IMF Staff Country Reports 2006, 169; 10.5089/9781451832860.002.A002

Sources: National central banks; and International Financial Statistics, IMF.1/ Data starting in 2004 for Slovak Republic.2/ Includes credits to non-profit institutions serving households (NPISH), except for Latvia, Lithuania, and Slovenia, where credit to NPISH is included under credit to non-financial corporations.
Table 2.

Romania: Basic Economic Indicators, 2000-05

(Annual percent change, unless otherwise indicated)

article image
Sources: National authorities; and IMF staff estimates.

45. Lending in Romania has also been facilitated by a ready supply of credit, reflecting buoyant capital inflows, which have been channeled in large part through the banking system. These inflows have in turn reflected high global liquidity, low worldwide interest rates, and increased investor confidence associated with Romania’s impending accession to the European Union.

46. The currency composition of credit flows has varied over the past few years, but a significant and rising portion of credit has been denominated in foreign currency (Figure 2). Again, this is especially evident in the evolution of household credit. For consumer loans to households, about 32 percent of the end-2005 credit stock was denominated in foreign currency, whereas the corresponding fraction for household mortgage lending was about 88 percent. In comparison, the foreign-exchange share is about 61 percent for lending to private companies. Demand for foreign exchange-denominated loans has been supported in large part by a steady, or appreciating, nominal exchange rate; combined with the fact that, until recently, real interest rates on local currency loans have been relatively high.

Figure 2.
Figure 2.

Romania: Credit Flows, 2002-05

(percent of GDP, rolling 12-month basis)

Citation: IMF Staff Country Reports 2006, 169; 10.5089/9781451832860.002.A002

Source: National Bank of Romania; IMF staff calculations

47. The banks that have been extending loans in Romania are generally well funded, and have financed the recent credit expansion through a mix of deposit mobilization and a fall in their net foreign assets. Deposit growth has been rapid, owing to rising household incomes and buoyant expectations, but has been supplemented significantly by direct borrowing from foreign parent banks. Moreover, figure 3 below illustrates that the banking system has a considerable excess supply of funds, particularly in local currency, which they have generally placed at the central bank—the return from holding funds at the central bank dropped significantly toward the end of 2005, as the NBR scaled back its effective sterilization rate, but the demand for local-currency loans has yet to fill the gap.

Figure 3
Figure 3

Romania: Credit and Bank Liabilities

(In percent of GDP)

Citation: IMF Staff Country Reports 2006, 169; 10.5089/9781451832860.002.A002

48. Looking at potential macroeconomic risks, the expansion of credit in 2005 mirrored a marked increase in aggregate demand and inflationary pressures. For the most part, this reflected a sharp boost in private consumption—supported by lower taxes, higher wages, and a downward trend in interest rates. As outlined in the main Staff Report, y/y inflation in 2005 stood at 8.6 percent; higher than the authorities’ target of 7.5 percent (with a +/− 1 percent band). This result would have been even more disappointing without the mitigating impact of a nominal appreciation over the course of the year. Excess demand pressures have also impacted the current account deficit, which widened to 8.7 percent of GDP in 2005 compared to 8.5 percent in 2004. This imbalance is slightly above the estimated sustainable deficit for Romania, which is about 8 percent of GDP, and so raises the concern that Romania may be increasingly vulnerable to sudden shifts in foreign investor sentiment.

49. As for financial-sector vulnerabilities, Romania’s banking system seems relatively sound. As illustrated in Table 2, financial-soundness indicators (FSIs) suggest a healthy and robust financial system—rates of return on equity and capital adequacy ratios are high and non-performing loan (NPL) ratios remain moderate.10 Banks are also very liquid and appear well positioned to absorb the direct impact of interest-rate and exchange-rate movements. Moreover, Romania’s banking system is dominated by foreign-owned institutions, which are able to provide substantial financial support in the event of problems, and which are also able to provide significant transfers of credit-assessment and portfolio-management skills.

50. However, the substantial share of foreign-denominated lending raises the possibility of an indirect risk to the financial sector. While NBR stress tests suggest that the banking system’s direct exposure to adverse exchange-rate movements is limited, they still face an indirect exposure through their loan portfolios. Many companies, and most households, that have taken on foreign currency-denominated loans do not receive foreign-currency income—receiving instead income that is denominated in local currency. This mismatch has not been readily apparent in an environment of a steady or appreciating currency. But potentially adverse movements in the future could impact borrowers’ ability to service their foreign-currency debt. Such indirect exposures are difficult to assess, and in a climate of rapid credit growth, there is the possibility that banks may not be appropriately pricing or provisioning for this risk. It is this feature of the financial sector that has raised the most concern among the authorities.

D. Policy Responses: Measures and Impact

51. Over the course of 2005, the authorities put into effect a series of prudential-style measures aimed at addressing credit-related risks. These measures were primarily introduced to help reduce the currency-mismatch risk associated with excessive foreign-currency lending. As outlined in Table 2, at end-2004 the total ratio of foreign-currency deposits to lei deposits stood at around 40:60. The corresponding ratio for forex loans, on the other hand was the reverse; i.e. a ratio of 60:40. The measures, therefore, focused mainly on limiting banks’ foreign-currency exposure to unhedged borrowers, as well as increasing the coverage and level of required reserves on foreign-currency liabilities. The key measure in this regard was a requirement limiting credit institutions’ overall foreign-exchange lending to unhedged borrowers to less that 300 percent of the banks’ own funds (Box 2).

52. In addition, the authorities tightened loan classification norms for credit institutions, explicitly requiring banks to consider foreign-currency risk when classifying their loans to individuals. In effect, the new norms introduced in September 2005 required banks to downgrade the classification of unhedged borrowers, regardless of their financial position or collateral. This latter measure had an immediate impact on reported NPL figures and provisioning requirements, as many foreign-currency loans that had previously been classified at satisfactory were automatically reclassified as substandard; from a level of 8.1 percent at end 2004, the NPL ratio increased to 9.4 percent in September 2005. However, in the final quarter of 2005 the banks rapidly managed to bring the ratio back down to 8.2 percent, by shifting their household-lending portfolios toward local-currency loans.

53. Over the short run the measures have had a significant effect, especially on the currency composition of credit growth. The 300-percent limit on exposure to unhedged borrowers had an immediate impact. From a total of 39 banks, 13 exceeded the exposure limit when the new measures were introduced. Over the immediate run these banks, representing 43 percent of total foreign-exchange loans at that time, were forced to sharply curtail their lending in foreign exchange. The impact of higher foreign-currency reserve requirements, however, has perhaps been less immediate. Typically, the full effect of reserve-requirement shifts is felt only after a 3-4 month lag. So, the impact of the end-2005 reserve-ratio hikes may spill over into the first half of 2006. Following the introduction of the measures, the y/y growth rate in foreign currency credit fell from 56 percent in September 2005 down to 30 percent in February 2006.

Figure 4.
Figure 4.

Romania: Credit Flows, 2001-06

(In percent of GDP, seasonally adjusted, annualized 3-month moving average)

Citation: IMF Staff Country Reports 2006, 169; 10.5089/9781451832860.002.A002

Source: Staff estimates.

54. In terms of credit flows, the measures have resulted in a dramatic switch away from foreign-currency loans in favor of local-currency lending. This impact can be seen in the 12-month (moving average) flows shown in Figure 2, where forex denominated flows start to fall after September 2005. It is even more apparent from a chart of 3-month flows. From a peak of 5.1 percent of GDP in August 2005, the three-month annualized flow of foreign-currency lending dropped to -1.7 percent by end-December. Local-currency credit flows, on the other hand, increased from 3.7 percent of GDP to 7.0 percent over the same period. The net effect was a fall in the overall flow of credit, from 8.9 percent of GDP to 5.3 percent in December. It should be noted also that the shift in borrowing toward local-currency credit was also assisted by a monetary loosening over the course of 2005, which has resulted in a drop in real lei interest rates, and a narrowing of the gap between local- and foreign-currency rates.

55. The impact has been most marked on the composition of consumer lending. As illustrated in figure 6 below, the surge in household credit flows throughout the early part of 2005 was driven in large part by a rapid expansion of local-currency consumer loans. In contrast, the expansion of foreign-currency consumer lending was relatively steady. Following the introduction of the credit measures, however, this pattern has reversed. Lei-denominated flows have remained broadly steady or have increased slightly, whereas the fall in household lending has chiefly reflected a contraction in foreign-currency consumer credit. Compared to the wide swings in consumer credit flows, mortgage lending to households has remained relatively stable—although even in this sector the new credit measures have prompted a switch into local-currency lending.

Figure 5.
Figure 5.

Lei credit flows to households, 2002-06

Citation: IMF Staff Country Reports 2006, 169; 10.5089/9781451832860.002.A002

Figure 6.
Figure 6.

Credit Flows to Households

(percent of GDP, 3-month m.a., LHS)

Citation: IMF Staff Country Reports 2006, 169; 10.5089/9781451832860.002.A002

Source: NBR; IMF staff calculations1/ Spread between RON-denominated loan rates and EUR-denominated loan rates to households.

Credit-Related Measures.

  • (September 2004). The reserve requirement for foreign-currency liabilities with maturity less than 2 years was increased from 25 percent to 30 percent. The requirement on liabilities with maturity greater than 2 years remained at zero.

  • (February 2005). The 30-percent reserve requirement on foreign-currency liabilities was extended to liabilities with maturity greater than 2 years, if contracted after 24 February 2005.

  • (July 2005). The reserve requirement was extended to all foreign currency-denominated liabilities, regardless of their maturity or contract date. The measure was implemented in two stages:

    • i) For the 24 July – 23 August maintenance period, a 15 percent ratio applied to foreign-currency liabilities with maturities greater than two years, which were raised before 23 February 2005.

    • ii) For the 24 August – 23 September maintenance period, the full 30 percent ratio applied to these liabilities.

  • (August 2005). To encourage a switch away from foreign-currency credit, the reserve requirement on RON-denominated liabilities, with maturities less than 2 years, was lowered from 18 percent to 16 percent.

  • (August 2005). Regulations on limits to household-lending risk were tightened. The new regulations set a monthly debt-service ceiling equal to 40 percent of the net monthly income of the borrower, and covered the sum of all commitments (mortgage, real-estate, consumer loans, and other similar contracts). Moreover, the monthly debt service ceilings for consumer and real-estate credits were limited to 30 and 35 percent of monthly net income, respectively.

  • (September 2005). For credit institutions granting foreign exchange-denominated loans, exposure to unhedged borrowers was limited to 300 percent of the creditor’s own funds. In this context, only borrowers with foreign currency income (natural hedge) were considered to be hedged.

  • (December 2005). The reserve requirement for foreign-currency liabilities was increased to 35 percent from 30 percent.

  • (January 2006). The reserve requirement for foreign-currency liabilities was increased from 35 percent to 40 percent.

56. Looking forward, however, the effectiveness of these measures may ease over time. Demand for credit remains strong, so lenders have a continued incentive to find alternate channels for funding. In discussions with staff, local banks generally pointed out that the measures would be mostly ineffective in halting foreign-currency credit to large corporate borrowers, as these could easily borrow directly from foreign banks, often with the assistance of the banks’ local subsidiaries or branches. As for lending to other clients, they noted that the impact of the 300-percent ceiling would likely diminish as the banks adjusted their strategy to the new environment. For example, of the 13 banks that had been constrained initially by the new measures, five had already turned to their owners by end-2005 for an increase in capital so that they could resume foreign-currency lending. This process was cumbersome, and could take a number of months. But eventually, local banks predicted that lending would most likely resume—although it was agreed that higher foreign-exchange reserve requirements would make this particular line of business more expensive, and that lower local-currency interest rates would continue to make lei-denominated loans increasingly attractive for borrowers. It should be noted in this context that, although monthly foreign currency-denominated flows had turned sharply negative in the final quarter of 2005, they had become positive once again by February 2006.

57. As for Romania’s macroeconomic risks, a key question is whether these measures will also help restrain the growth of excess demand—i.e. whether the measures will help curtail the overall pace of lending, rather than just its currency composition, and whether a lower level of credit growth can be relied upon to ease pressure on prices and the current account. This has been a recent research topic within the Fund (Hilbers and others, 2005).

58. Evidence suggests that prudential-style credit measures, by themselves, are generally not well suited to deal with macroeconomic stability issues. This is borne out by the recent experience of Bulgaria and Croatia, where similar credit measures have been mostly ineffective in stemming those countries’ widening external imbalances (Box 3). The conclusion also has more general empirical support, as shown in Table 3 below which examines the relationship between demand and bank credit growth across a broader set of countries. The table shows a series of fixed-effects regressions covering five CEE countries11 over 2000-04. On its own, household lending growth has a significant impact on private consumption (Model 1). However, when the model is augmented to control for disposable income, the coefficient for bank lending becomes statistically insignificant (Model 2). This suggests that it is current income flows that are the more important determinant of consumption growth, rather than loan supply. Indeed, when we start with a specification that includes income only (Model 5), and then add a measure of lending growth, the explanatory power of the model actually decreases. Looking at a different set of models that examine the impact of total private-sector lending on total domestic demand, loan growth does appears statistically significant (Models 6-10), even controlling for income. However, the coefficient is small and dominated by that for disposable income—the differences between the two sets of models may suggest that bank lending may be more important in shaping the pace of investment growth, rather than consumption. In sum, therefore, there is reason to doubt the efficacy of prudential-style measures as stabilization instruments. Excess aggregate demand, as mirrored by rising incomes, is often associated with an increase in credit demand—and in a world of open capital flows and porous financial markets, this demand has typically been accommodated by an elastic supply of funds, from various sources. It does not follow, therefore, that isolated prudential-style efforts to restrict the supply of bank loans will necessarily help limit aggregate demand. In practical terms, experience has shown that, in the face of continued and growing demand, funding will tend to find its way to those who want it. And from a policy perspective, stemming this process would require an unwieldy and comprehensive regulatory regime that would likely become more and more distortionary over time.

Table 3:

Determinants of Demand Growth, 2000-04

article image
Note: t-statistics in parentheses indicate *** significant at 1 percent level; ** significant at 5 percent level; and * significant at 10 percent level.Source: IMF. Republic of Slovenia, Selected Issues 2005. (IMF Country Report 05/254)

59. Therefore, in addressing the macroeconomic risks posed by excess demand, the first-best approach is generally to tackle the key causes of demand directly, using more traditional macro instruments such as fiscal and incomes policy, as well as monetary policy. From the experience of other countries, there appear limits as to what prudential-style policies can do in the absence of an appropriate monetary- and fiscal-policy framework. Moreover, to the extent that foreign-currency lending has been encouraged by historically stable exchange rates, increased exchange-rate flexibility can reduce perceptions of low currency risk and help produce a more appropriate credit mix. For example, increasing the flexibility of the exchange rate and allowing domestic interest-rate differentials to narrow helped reduce foreign currency-denominated bank lending in Poland in the early 2000s.

60. Furthermore, undue reliance on credit measures may risk slowing the development of a sound and efficient financial system. As noted earlier in this chapter, financial development and long-term economic growth are closely related, so the authorities should be wary of depending too heavily on measures that attempt to limit the supply of particular types of credit—these are often distortionary, with unintended and undesirable side effects, such as impeding competition and encouraging circumvention through non-bank and foreign institutions. This latter phenomenon, in which lending shifts to other non-bank channels (that are less well-supervised) can occur very quickly, and was a particular feature in Croatia. The authorities have taken steps to address this issue in Romania (Box 4), but any measures that focus primarily on the banking system risk generating incentives for regulatory arbitrage.

The Impact of Credit Measures in Croatia and Bulgaria

Croatia:

In January 2003, faced with booming credit and a mounting external imbalance, the central bank (CNB) introduced a number of direct measures to limit the supply of credit—banks with lending growth greater than 4 percent per quarter were obliged to purchase an amount of CNB bills, at penalty rates, equal to twice the amount of excess loans. The CNB also increased the minimum liquid foreign-exchange asset requirement on bank’s foreign-exchange liabilities to discourage foreign borrowing. While domestic bank credit did decelerate in 2003, it became clear that enterprises were easily able to switch their borrowing from domestic to foreign banks (local banks typically directed their corporate customers to their parent banks abroad). Enterprises also increased the use of leasing and other forms of financing. Consequently, external borrowing in 2003 was about 2½ times higher than in 2002, and the share of foreign debt in financing corporate investment rose sharply. Although the current account deficit fell in 2003 owing to a bumper tourism season, import growth remained strong and the trade balance deteriorated further. The measures were dropped in the beginning of 2004. Credit growth did not bounce back immediately, suggesting that some of the fall in credit growth in 2003 may have been driven by a fall in credit demand, rather than restrained supply, but external debt continued to increase. In 2004, the CNB introduced a 24-percent, unremunerated marginal reserve requirement (MRR) on new bank borrowing from abroad. Faced with a further increase in bank foreign liabilities, the CNB raised the MRR in two steps to 40 percent in the first half of 2005. The CNB then added a second 55-percent tier and closed some loopholes in January 2006. Evidence on the effectiveness of the latest measures is not yet available.

The credit limits also had a negative impact on financial-sector development, as they encouraged the rapid growth of unsupervised and unregulated leasing companies, and reduced the transparency of banking statistics—banks engaged in a number of activities designed to circumvent the limits, such as asset swaps, collateralization, and accelerated write-offs of nonperforming loans.

Bulgaria:

Facing surging credit flows, in 2004 the central bank (BNB) introduced a number of liquidity-draining measures, such as tightened reserve-requirement provisions. When credit growth continued, however, the BNB announced in March 2005 that banks with lending growth greater than 6 percent per quarter, from a fixed base, would be subject to an unremunerated deposit requirement equal to twice the excess credit expansion. Initially scheduled to last only 12 months, this restriction has been extended until end-2006.

The effect of these measures was limited—Bulgaria’s open capital account has permitted large businesses to access credit abroad, while allowing domestic banks to redirect their credit to households. The measures have also contributed to the rapid growth of partially-unsupervised non-bank intermediation. On the whole, financial flows to the private sector were little changed year on year. Household credit rose as a share of total bank lending, and firms increasingly financed themselves through bonds, leasing and capital inflows.

In terms of macroeconomic stability, at the time of the first review it had been expected that the measures would help bring down the current account deficit. Instead, continued excess demand has caused the deficit to more than double, from 5.8 percent in 2004 to 11.8 percent in 2005, and the data for January 2006 show a further deterioration to 12.8 percent.

It should be noted that, for both of these countries, it is difficult to assess the precise impact of the credit measures, as it is impossible to know what would have happened had they not been put in place. However, on the basis of available experience, it appears that their effectiveness so far in restraining domestic demand has been somewhat disappointing.

The Non-Bank Financial Sector in Romania

The Romanian authorities’ efforts in 2005 to restrict the activities of banking institutions raised the concern that credit flows would simply be redirected through the less-regulated non-bank sector, which has been reported to be growing rapidly. Although comprehensive and consistent data on the size and activities of the non-bank sector are not yet available, there does not seem to have been a dramatic shift to this sector as yet.

Growth in the leasing sector in 2005, although healthy, was less than in 2004 and broadly in line with general credit demand. According to the two main leasing organizations—the Romanian Leasing Association (ASLR) and the Association of Banks’ Leasing Divisions (ALB)—the value of total leased assets increased by 45 percent in 2005 from about 1.3 to 2.0 billion euros (2.½ percent of GDP). Growth in 2004 was somewhat higher, at 75 percent, but from a lower base of around 830 million euros. These figures are only approximate, however, as the two associations do not include all leasing companies.

Similarly, the 2005 growth rate for lending by the main consumer finance companies was substantial, at about 75 percent, but this reflects a relative small base. As shown in the chart, credit trends in 2005 were broadly in line with 2004, although there was a marked shift away from loans in which the finance companies served as intermediaries for banks, in favor of loans where the finance companies offered loans on their own account.

Looking forward, the authorities have taken steps to prevent a large-scale shift in resources to the non-bank sector; by moving this sector under the supervisory and regulatory authority of the NBR. Pending passage of the appropriate legislation, the NBR in January 2006 passed a long-anticipated emergency ordinance requiring all non-bank credit institutions to register with the central bank. In order to be included in the registry, these institutions must be incorporated as joint stock companies with share capital of at least EUR 200,000. They must also submit detailed financial information to the NBR and set up specific provisions for credit risk. It is anticipated that, on receiving this information, the NBR will establish a two-tier supervisory regime, with larger, more systemically-important institutions receiving greater scrutiny. The details of these regulations are still being discussed. However, it is envisaged that the NBFI sector will be subject to most of the same core requirements as the banking sector; including the 300-percent limit on foreign-exchange lending to unhedged borrowers, as well as the 40-percent ceiling on household debt service as a percent of monthly income. The new system should be in place by mid-2006.

A03ufig01

Loans by major consumer-credit companies, 2003-05

(by source of funding, Mill. RON)

Citation: IMF Staff Country Reports 2006, 169; 10.5089/9781451832860.002.A002

Source: National Bank of Romania

61. Ideally, therefore, prudential-style measures should be aimed primarily at addressing specific financial-sector risks—ensuring that the risks associated with credit growth are managed appropriately by lenders. Indeed, this should be the case regardless of the rate of credit growth. However, to the extent that rapid lending growth is also associated with concerns about macroeconomic risk, the experience of other countries suggests that prudential-style measures can play a potentially useful role, but only in support of more traditional macroeconomic instruments, as part of a comprehensive policy response. Looking forward, therefore, the effectiveness of recent efforts to limit the financial and macroeconomic risks associated with recent credit growth will depend on the authorities overall policy package—the credit measures adopted in 2005 appear to have helped ease a worrying trend of increased foreign-currency lending, but their impact on excess demand may be more limited. This latter issue will require the use of more traditional instruments, including a coordinated tightening of interest rates, and fiscal and incomes policies.

E. Conclusion

62. Over the past few years, credit growth rates in Romania have been among the highest in the region. This chapter has noted that rapid lending growth has been fairly typical within most CEE countries as they moved closer to Western Europe, and that the high growth rates of recent years reflects the fact that Romania is a relative latecomer to this process. In a broad sense, credit growth should be welcome, as it reflects an underlying process of financial deepening that promises to help raise longer-term growth and living standards. Moreover, based on the experience of other countries in the region, this growth is likely to continue well into the foreseeable future.

63. However, while rapid credit growth may be a predictable part of the convergence process, recent developments have raised concerns about increased financial-sector and macroeconomic vulnerabilities. Looking first at the financial-sector, Romania’s banking system seems sound, liquid, and well positioned to absorb the direct impact of interest- or exchange-rate movements. But, the growing proportion of foreign currency-denominated lending suggests a substantial and increasing exposure to indirect risk, as adverse currency movements in the future may impact the ability of unhedged borrowers to meet their obligations. This type of risk is often hard to assess, and the authorities’ recent credit measures have addressed the problem directly—the measures appear to have prompted a significant switch away from foreign-currency loans in favor of local-currency credit.

64. In tackling macroeconomic risks, prudential-style credit measures cannot be a substitute for more traditional stabilization instruments. The experience of other countries suggests that such measures, by themselves, are not very effective in stemming the consequences of excess demand. Instead, stabilization will require a more comprehensive response, including a coordinated tightening of interest rates, and fiscal and incomes policies.

References

  • Backé, Peter, and Tina Zumer, 2005, “Developments in Credit to the Private Sector in Central and Eastern European EU Member States: Emerging from Financial Repression—A Comparative Overview,” Focus on European Economic Integration 02/05, (Vienna, Oesterreichische Nationalbank).

    • Search Google Scholar
    • Export Citation

  • Backé, Peter, Balázs Égert, and Tina Zumer, 2005, “Credit Growth in Central and Eastern Europe: Emerging from Financial Repression to New (Over) Shooting Stars?” Mimeo.

    • Search Google Scholar
    • Export Citation

  • Cottarelli, Carlo, Giovanni Dell’Aricca, 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 the Balkans,” IMF Working Paper 03/213 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Duenwald, Christoph, Nikolay Gueorguiev, and Schaechter, Andrea 2005, “Too Much of a Good Thing? Credit Booms in Tranition Countries: The Cases of Bulgaria, Romania, and Ukraine,” IMF Working Paper 05/128 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • Hilbers, Paul, Inci Otker-Robe, Ceyla Pazarbasoglu, and Gudrun Johnsen, 2005, “Assessing and Managing Rapid Credit Growth and the Role of Supervisory and Prudential Policies,” IMF Working Paper 05/151 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

  • International Monetary Fund, 2005, World Economic Outlook, 2005 “Are Credit Booms in Emerging Markets a Concern” (Washington).

  • Kraft, Evan, and Ljubinko Jankov,, 2005. “Does Speed Kill? Lending Booms and their Consequences in Croatia,” Journal of Banking & Finance, Vol. 29 pp. 105-21.

    • Search Google Scholar
    • Export Citation

  • Neagu, Florian, Angela Mărgărit, Mihai Copaciu, Irina Racaru, Romulus Mircea, and Arpad Andrassy, 2006, “Creditul neguvernamental în România: Perspective şi implicaţii,” Case Study No.15 (Buchurest, National Bank of Romania).

    • Search Google Scholar
    • Export Citation

  • Schadler, Susan, Paulo Drummond, Louis Kuijs, Zuzana Murgasova, and Rachel van Elkan, 2005, Adopting the Euro in Central Europe: Challenges of the Next Step in European Integration, IMF Occasional Paper No. 234 (Washington, International Monetary Fund).

    • Search Google Scholar
    • Export Citation
5

Prepared by Andrew Tiffin.

6

See Section III, Real Convergence Prospects.

7

Bulgaria, Croatia, Estonia, Hungary, Latvia, Poland, and Slovenia are classified as “early birds” with rapid credit growth starting in the mid- to late-1990s; whereas Bosnia & Herzegovina, Serbia, and Lithuania are classified as “late risers” with growth starting before 2002. Other “sleeping beauties” include Albania, the Czech Republic, Macedonia FYR, and the Slovak Republic.

8

See Cottarelli and others (2003); Schadler and others (2005).

9

See IMF (2005).

10

The reported NPL ratio includes loans that have been classified as substandard, doubtful, and loss.

11

These include the Czech Republic, Slovakia, Slovenia, Lithuania and Estonia.

Same Series

Other IMF Content

Other Publishers

Cover IMF Staff Country Reports
Romania: Selected Issues and Statistical Appendix
Author: International Monetary Fund
Volume/Issue:
Volume 2006: Issue 169
Publisher:
International Monetary Fund
ISBN:
9781451832860
ISSN:
1934-7685
Pages:
115
DOI:
https://doi.org/10.5089/9781451832860.002
  • View in gallery

    Growth of Credit to the Non-government Sector, 2002-05 1/

    (Average year-on-year percent change)

  • View in gallery

    Romania: Credit Flows, 2002-05

    (percent of GDP, rolling 12-month basis)

  • View in gallery

    Romania: Credit and Bank Liabilities

    (In percent of GDP)

  • View in gallery

    Romania: Credit Flows, 2001-06

    (In percent of GDP, seasonally adjusted, annualized 3-month moving average)

  • View in gallery

    Lei credit flows to households, 2002-06

  • View in gallery

    Credit Flows to Households

    (percent of GDP, 3-month m.a., LHS)

  • View in gallery

    Loans by major consumer-credit companies, 2003-05

    (by source of funding, Mill. RON)

Growth of Credit to the Non-government Sector, 2002-05 1/

(Average year-on-year percent change)

Romania: Credit Flows, 2002-05

(percent of GDP, rolling 12-month basis)

Romania: Credit and Bank Liabilities

(In percent of GDP)

Romania: Credit Flows, 2001-06

(In percent of GDP, seasonally adjusted, annualized 3-month moving average)

Lei credit flows to households, 2002-06

Credit Flows to Households

(percent of GDP, 3-month m.a., LHS)

Loans by major consumer-credit companies, 2003-05

(by source of funding, Mill. RON)