Banking crises in the Baltic countries have threatened the nascent recovery of their economies. But their banking sectors have emerged generally stronger as a result of the experience.
The baltic Republics of Estonia, Latvia, and Lithuania are in the vanguard of the transition economies. The first fruits of their reform programs are now being seen in the revival of growth. But there are a number of factors that have threatened to derail the fledgling recovery. All three republics have experienced serious banking crises, which have set in train a process of structural change in their banking systems, and have, in some instances, had adverse political and economic repercussions.
The Baltic countries inherited the Soviet monobank system under which specialized state banks serviced specific branches of the economy. All three countries moved quickly to establish a two-tier banking system with the central bank at the core. None of the Baltic countries had personnel skilled in modern banking practices or an appropriate legal, regulatory, and supervisory framework. Moreover, a strategy had to be devised to handle the remnant of the Soviet banking system. At the same time, the Baltic countries had to face the twin challenges of encouraging the new private banking sector while ensuring that its growth took place in a prudent manner.
Initially, the three countries took different approaches. In Estonia and Lithuania, the specialized Soviet-era banks were reconstituted as state banks and then gradually or partially privatized. In Latvia, in contrast, the Savings Bank was reconstituted as a state bank but the branches of the remaining Soviet-era banks were privatized. Remaining banks were merged, rehabilitated, and then offered for privatization.
All three countries have had extremely liberal policies toward the licensing of new commercial banks. A large number of banks, it was thought, would quickly generate the competition needed to drive down deposit and lending rates and provide the lending needed to support the emerging private sector. Many new private banks were established by enterprises to gain access to a preferential and much cheaper source of funding than was available from existing banking institutions. Little thought was given initially to the implications of this policy for bank soundness and supervision.
How the crises arose
Banking crises, mainly involving private banks, surfaced in Estonia in 1992, in Latvia in early 1995, and in Lithuania in late 1995. There were many causes, some of which—systemic in nature—had been eating away at the fabric of these banking systems for some time. For example, falling inflation was a prominent factor in all three countries, which helped make borrower distress more apparent while simultaneously squeezing banks’ intermediation margins. But in each country, different events led to the crises and different triggers brought them to a head.
In Estonia, the proximate causes of the crisis were the freezing of the assets in Moscow of two important Estonian banks, and the drying up of cheap credit from the central bank, which had previously provided Estonian banks with significant profits and liquidity. In Latvia, the waning of highly profitable trade-financing opportunities, as well as general mismanagement and corruption, set the stage for the crisis. It was set off by the central bank’s requirement that banks be properly audited using International Audit Standards (IAS) principles. Bank profits in Lithuania were also compressed owing to the contraction of lucrative trade financing opportunities. Moreover, the government pressed some banks (both state-owned and private) to lend to the public sector to finance quasifiscal expenditures. Leaks of the results of on-site examinations of two banks, showing deep insolvency, led to runs on those banks and liquidity shortages.
Broadly, there were four systemic factors underlying the crises: poor regulation and supervision, poor accounting and excessive taxation, an inadequate legal infrastructure for lending, and pervasive corruption coupled with weak banking skills and mismanagement. The stresses and strains of the economic transition and stabilization also exposed the banks’ underlying weaknesses. To some extent these factors are interrelated. For instance, the transition environment has unleashed profitseeking in many segments of society, including the banking industry. While much of it reflected entrepreneurial zest, some of it spilled over into illegal and unscrupulous activities. In some instances, weaknesses in bank regulation and supervision created incentives for corruption.
“The Baltic countries had to face the twin challenges of encouraging the new private banking sector while ensuring that its growth took place in a prudent manner.”
The transition from central planning to a market-based system also exposed the structural deficiencies in the banking sector and the regulatory environment. The very tight macroeconomic policy framework pursued in all three countries created an environment that was not propitious for an emerging banking system. Banks, their customers, and bank supervisors were unable to monitor and control the risks inherent in the new policy environment.
Bank regulation and supervision. A contributory factor to all three banking crises was the failure of banking regulation and supervision. In Latvia, deficiencies in the regulatory framework itself contributed to the crises, although there were also some weaknesses in implementation. In Lithuania, the culprit was deficiencies in the implementation of regulations—even when bank supervisors had identified problems, they were not acted upon. In Estonia, the legal, regulatory, and supervisory framework was very underdeveloped at the time of the crisis, but it was less important as a cause of crisis than in the other two countries.
The licensing and regulatory regimes in the three countries did not discourage the entry of foreign banks. The Lithuanian central bank may nonetheless have discouraged foreign banks from entering the local market. Arguably in Estonia—where nine foreign banks have entered the market in recent years—banking discipline may have been more quickly embedded in the system.
Accounting and taxation. Initially, banks in the Baltics continued to use the old Soviet Gosbank chart of accounts. In Estonia, banks were required to use IAS for the first time in 1995, although the better banks had begun doing so in 1993. In Latvia, the introduction of IAS accounting and reporting requirements began in 1994—indeed, this requirement precipitated the country’s banking crisis. In Lithuania, a number of changes in bank accounting and prudential rules have been introduced gradually over the last three years, but full IAS compliance was expected only as of January 1, 1997. The initial absence of and unfamiliarity with IAS-based accounting systems in the Baltics has made it more difficult for bank managers, shareholders, and supervisors alike to accurately gauge the solvency and liquidity problems building up in individual banks. Even though most of the Baltic banks were quick to have international auditors undertake IAS audits, these audits have not served as the early warning signals they were intended to be and often have been ignored altogether by the supervisors.
Perhaps more important, while all three countries moved quickly to introduce loan-loss classification and provisioning rules, in practice these rules were often not applied (loan-loss provisions were not actually booked), as the tax rules did not allow any deduction for loan-loss provision expenses. The distinction between supervisory and tax accounting was an unknown concept in the Baltic countries, making it unattractive for banks to actually book loan-loss provisions. While the better banks nevertheless used profit-and-loss data after hypothetical provisioning to determine dividend payouts (and the more corrupt ones actively use this loophole to drain funds through large dividend payouts from nonexistent profits), all banks—prudent and imprudent alike—were taxed on fictitious profits as a result of this deficiency in the tax regime. The problem was rectified only relatively late in the transition in Estonia and in Latvia, and Lithuania introduced a scheme at the end of 1994, which was to be phased in over three years.
Legal infrastructure. In the Baltic countries, there was initially no legal framework to support bank lending. There was no appropriate legislation relating to bankruptcy and collateral; well-functioning property title, mortgage and pledge registers; or, more generally, a market for land and real estate. Another important omission was corporate governance and accountability provisions for banks that specified the duties and responsibilities of bank shareholders, supervisory board members, and managers. This allowed shareholders to manipulate supervisory board members and, through them, managers to serve their own interests. All of these factors—most of which have been or are being addressed—contributed to the riskiness of bank lending.
Corruption and weak management. In all three Baltic countries, some banks were created as captive funding mechanisms by groups of enterprises and individuals—raising funds directly from the public was cheaper than borrowing from banks. In other cases, owners and managers tried to make quick profits by making high-risk loans or by assuming large open foreign exchange positions. This behavior was encouraged by the knowledge that the supervisory authority was inexperienced and understaffed, and lacked effective enforcement powers. The lack of skills among bank managers and other staff also led to poor decision making.
The policy response
All three countries were, for the most part, ill prepared for the banking crises that erupted. Their immediate responses differed significantly. In Estonia, the government announced very quickly that there would be no bailout. Although Estonia’s currency board arrangement did allow the central bank to provide credit in a banking crisis, the central bank and the IMF took the view that the large scale of an eventual bailout would be inflationary and would undermine the fixed exchange rate. Thus, the central bank liquidated one bank whose problems were primarily caused by mismanagement. Two other banks, which suffered liquidity problems owing to the freezing of their assets by the Moscow Vnesheconombank, were merged, and ownership was taken over by the government. The central bank instituted a licensing review and strengthened supervision. A new Law on Credit Institutions was passed in December 1994, increasing the central bank’s supervision and enforcement capabilities, and requiring all banks to develop internal auditing departments and to be audited annually by external auditors. Starting in 1995, all banks were required to use IAS for their financial statements.
In 1995, Latvia’s central bank initially provided a modest amount of liquidity support for a large private bank that was at the center of the banking crisis, but when it became clear that the bank’s negative net worth had reached 7 percent of GDP, no further support was provided. Drawn-out negotiations between the central bank (which lacked formal enforcement powers) and the bank in question ensued, which allowed the latter’s managers and owners to strip the bank of its assets. Finally, the bank was declared insolvent and the central bank took over its management.
“In the initial stages of transition the risks associated with lending in transition economies combine to overwhelm many banks.”
The Latvian authorities had to deal not only with the immediate management of the crisis but also with the crisis of confidence in the banking sector at large. Urgent changes in the legal, regulatory, supervisory, and institutional framework were made. To restore confidence in the banking sector, the government promised to compensate household depositors who lost funds in failed banks with an initial amount of up to Lat 500 ($1,000) per depositor. During the subsequent three years, depositors were to receive an additional Lat 100 ($200) per year. However, with a new government in place and given the tight state budget, compensation now will most likely depend on recoveries from assets in banks under liquidation. A new Commercial Banking Law—much more detailed and inclusive than the 1992 statute—was enacted in October 1995. The central bank subsequently hired additional supervisory staff, moved to tighten prudential regulations, required banks to establish internal control departments, and arranged for external accounting firms to supplement the work of its own on-site examiners.
In Lithuania, the crisis unfolded over a longer period of time and involved a larger number of banks (both private and state-owned) than in the other two countries, yet the authorities’ response was less decisive than in Latvia and Estonia. The Lithuanian government initially provided unconditional support to troubled banks without removing their managements or suspending shareholders’ rights, thereby signaling that there would be few if any penalties for imprudent behavior. The policy response to the full-blown crisis that occurred later initially appeared more forthright. This time, however, the government’s hands were partially tied by the passage by parliament of a number of emergency pieces of legislation, as well as a new deposit insurance law.
The emergency legislation required the government to lift moratoriums on private banks, which had been imposed earlier, and to ensure that no depositor lost money. In addition, legislation was passed allowing the government to extend up to Litai 300 million ($75 million) in guarantees for interbank borrowing to address liquidity problems in other banks. This scheme was conceived as a substitute for the lender-of-last-resort function of the central bank, which could provide only very limited liquidity support under the currency board arrangement. The scheme did not specify, however, which banks would be eligible for such assistance, again sending a signal to the banking community that government support would not distinguish between prudent and imprudent banks. Parliament also adopted a law requiring the government to provide compensation retroactively to individual depositors in all smaller banks in bankruptcy of up to Litai 2,000 per person ($500).
Notwithstanding these constraints, the government, with the assistance of the World Bank and the IMF, subsequently drew up a detailed bank restructuring plan, which to date has only been partially implemented. This plan envisaged full recapitalization and renationalization of the majority state-owned banks, liquidation or a combination of existing shareholder and government support for private banks, and the transfer of bad loans to a newly created government-owned asset-management institution. Longer-term measures to further strengthen banking legislation, regulation, and supervision, as well as to improve corporate governance in the banks, are also envisaged in the bank restructuring plan. The new government, which was elected in October 1996, is currently in the process of formulating its own policy in the area of banking.
A number of conclusions can be drawn from the Baltic countries’ experiences. They may have implications for banking reform in other countries of the former Soviet Union, especially smaller ones. However, owing to the very specific nature of banking sector distress in transition economies, these conclusions should not be seen as having across-the-board validity for banking crises in more developed economies.
Some banking distress is inevitable. Banking distress is inevitable in countries that have had no recent experience of market-based banking. This comes from the confluence of risk factors that put pressure on the fledgling banking sector. However, it also arises from some of the structural features of the emerging banking systems in the countries of the former Soviet Union, in particular, the existence of a plethora of poorly capitalized banks that are vulnerable because their capital is small and that, because of their size, have not reaped the benefits of portfolio diversification. Also, new banks are often too small to afford the investment in infrastructure that is needed to offer modern services. State-owned banks are invariably overstaffed, and this drives up their operating costs when their salary levels adjust to the higher levels in the private banking sector. These factors and the high-risk nature of bank lending in these countries have fostered high intermediation margins, and the high lending rates this has generated have further added to borrowers’ debt-servicing difficulties.
Banking distress may be desirable. In the initial stages of transition the risks associated with lending in transition economies combine to overwhelm many banks. Furthermore, the intensification of bank regulation—particularly minimum capital regulations and increasing competition—force these banks toward merger or liquidation. Banking difficulties therefore emerge. Such difficulties are, however, a common feature of the structural transition of the banking system and can lead to a much-needed consolidation of overly fragmented systems. Also, when countries’ banking systems are still very small compared with GDP, the resulting costs of banking crises to depositors—compared with the earlier losses imposed by hyperinflation—are fairly limited.
Banking crises in transitional economies can die down relatively quickly. While banking crises erupt quickly, they can subside equally quickly. This reflects, in part, the fact that depositors in the Baltics have become accustomed to banking distress. The more sophisticated ones spread their deposits across many banks to diversify risk. Banking crises are quickly discounted, as evidenced by the sharp rise, and the subsequent sharp fall, in interest rates following the crises in Latvia and Lithuania. Moreover, the three banking systems have shown resilience, reflecting the fact that each country had a core of solvent banks that anchored the system. This resilience suggests that the banking authorities should take a tough stance in dealing with problem banks.
Firm and prompt policy response is needed. Any support provided to banks in difficulty should be conditioned on stern action with respect to the banks concerned (such as the removal of managers, liquidation of the shares of existing owners, etc).
Corruption should never be rewarded. Banks in which severe fraud and corruption are rife should be liquidated early, before they become “too big to fail.” Their shareholders should lose their rights, and their managers should be removed. Banks that have a particular market niche and can be shown to be viable in the longer term can, in principle, be restructured but only under new management and ownership.
Banking crises should be anticipated. While banking distress is inevitable, banking crises should be avoidable if the banking supervision process is geared toward close monitoring of the largest banks, which pose the greatest risk of creating systemic problems. This requires a willingness on the part of the government to refrain from using the banking sector for political and social purposes and to allow bank supervisors to properly discipline the banking sector. Failure to take prompt action when banking distress is uncovered can lead to even greater losses in the longer run, as a number of countries have learned.
Supervisors should send strong signals about appropriate behavior. Heavy emphasis should be placed on tightening on- and off-site supervision to send a strong signal to bankers about the penalties for inappropriate behavior. Banking regulations should not just be “on the books” but should be applied forcefully. The importance of this as a signaling device to bankers prone to fraud and corruption should not be underestimated. Signaling can play a very important role in imposing discipline in banks during the transition years. This applies not only to the intensity of supervision but also to the government’s approach to dealing with banking difficulties when they arise.
Banking distress is likely to be a feature of transition in the countries of the former Soviet Union for some time to come. Governments in these countries should prepare themselves now—by strengthening their supervisory capacity and readying themselves for tough implementation decisions—to deal with the inevitable. Even if banking crises do materialize, they likely will not have such severe effects on the economy as a crisis of similar proportions might have in a more developed, traditionally market-based economy.
This article is based on a paper by the authors “The Baltics: Banking Crises Observed,” Policy Research Working Paper No. 1647 (Washington: World Bank).
THE INTERNATIONAL MONETARY FUND PRESENTS…
Interest Rate Liberalization and Money Market Development
By Hassanali Mehran, Bernard Laurens, and Marc Quintyn
This book brings together the papers presented at a seminar held in Beijing, China, in August 1995 and sponsored jointly by the IMF’s Monetary and Exchange Affairs Department and the People’s Bank of China. The papers were written by central bankers from China, Italy, Korea, Malaysia, Thailand, and Turkey. The Chinese authorities were specifically interested in learning more about the Italian and Turkish models of interbank markets, and about the experiences of neighboring Asian countries with interest rate liberalization. The US experience was also presented, and the introduction to the book draws policy lessons from the experiences presented at the seminar.
US$20.00. Available in English. (paper) ISBN 1-55775-563-9. viii+161 pp. 1996.
TO ORDER, PLEASE WRITE OR CALL:
International Monetary Fund • Publication Services • Box FD-197 • 700 19th Street, N.W. • Washington, DC 20431 U.S.A.
American Express, MasterCard, and VISA credit cards accepted.