Abstract
For the latest thinking about the international financial system, monetary policy, economic development, poverty reduction, and other critical issues, subscribe to Finance & Development (F&D). This lively quarterly magazine brings you in-depth analyses of these and other subjects by the IMF’s own staff as well as by prominent international experts. Articles are written for lay readers who want to enrich their understanding of the workings of the global economy and the policies and activities of the IMF.
AS EASTERN EUROPE and the states of the former USSR confront the bad debt problem, they face difficult policy choices. These include when to commit fiscal resources to recapitalize banks, how to change behavior given the incentive system inherited from the past, and under what conditions to pursue bank and enterprise restructuring as an integrated package.
Throughout Eastern Europe, the initial phase of transition has produced a legacy of nonperforming loans in the portfolios of commercial banks. The bad debt problem stems from the deterioration in the financial condition of most state-owned enterprises, reflecting major shifts in relative prices, particularly higher energy prices; substantial declines in domestic demand for capital and military goods; a breakdown of trading arrangements under the CMEA (the former socialist country trading bloc); and a tightening of fiscal subsidies to these enterprises as tax bases shrank. The continuing flow of credit to overindebted enterprises has exacerbated the problem.
Estimates of the volume of nonperforming loans in Eastern Europe (defined here to exclude the states of the former USSR) range between 10 percent and 40 percent of total bank credit, although variation in loan classification criteria make cross-country comparisons difficult. The figures for individual banks can range much higher, and many banks are clearly insolvent under Western loan classification criteria. The stock of bad loans as a share of GDP also varies widely, ranging from 4 percent to over 20 percent.
By contrast, in most states of the former USSR, including Russia, the bad loan problem remains largely hidden because bank lending is conducted at highly negative real interest rates and repayments tend to be rolled over routinely. But as economic stabilization takes hold, the problem could become even more severe than in Eastern Europe, given the larger economic distortions prevailing during the Soviet era and the deeper economic dislocations that have occurred since then. Indeed, the fear that tighter credit and higher real interest rates might constitute an unmanageable burden on enterprises and the banking sector already inhibits governments from embarking on major stabilization.
The growth of nonperforming loans is worrisome as it undermines efforts to restructure existing enterprises and to provide adequate credit to emerging ones. There is still no consensus on how to deal with this problem. Theoretical solutions are largely untested, and difficulties exist even at the conceptual level, reflecting the daunting range of economic and institutional problems facing the economies of the region. Critical questions include whether transition economies should undertake bank and enterprise restructuring as an integral package, and what degree of centralization or decentralization is feasible within that context.
Options for reform
Experience from market economies suggests the problem should be dealt with swiftly; inaction and delay raise the eventual costs as the share of distressed borrowing tends to rise. This situation is particularly harmful for transition economies because the flow of credit to emerging private firms—the main source of output and employment growth in the short term—is reduced.
Admittedly, if strong economic recovery in Eastern Europe could be sustained, for example through the growth of profitable exports to the West, the share of bad loans could decline without government intervention. Preliminary evidence from Poland indicates that the share of bad debt has declined somewhat following positive economic growth in 1992-93. But this is not an argument for inaction. The size of the bad debt problem is in most cases too large to start with, and the likelihood of rapid economic recovery too uncertain. There are, therefore, significant risks associated with inaction.
The bad loan portfolio is not only a result of adverse external shocks; it also reflects the incentive system inherited from the past. Banks had little reason to allocate credit on a commercial basis, and penalties for nonpayment by enterprises were few. Banks had close links with the enterprise sector—often they were merely enterprises’ “treasury” branches. Politicization of credit, poor governance in the enterprise sector due to unclear ownership rights, and lack of prudential regulation and bankruptcy legislation all contributed to a lack of financial discipline.
Dealing with the stock of bad debt in isolation—through recapitalization of banks, for example—without addressing the underlying incentive framework could therefore waste scarce fiscal resources because the problems are likely to recur. Indeed, if no accompanying measures are taken, recapitalization could be viewed as merely rewarding inappropriate lending behavior and exacerbating moral hazard problems.
This has led to the view that an integrated approach to bank and enterprise restructuring is needed because neither can succeed in isolation. But any comprehensive scheme will take time to establish. In some countries, an integrated approach may not be feasible in the foreseeable future, and attempts to construct one could delay any action. In that case, it would be preferable to proceed separately with banking and enterprise reform.
Whichever course is chosen, governments must quickly decide whether and when insolvent banks should be recapitalized or liquidated. Key criteria to guide these decisions should include a realistic assessment of the fiscal costs of adequate recapitalization versus liquidation (for which the major cost is likely to be a partial or full payoff of depositors); and the extent to which recapitalization favorably affects the bank’s lending behavior. This in turn will depend on the credibility of government efforts to establish an environment of greater financial discipline and the capacity of the banks to respond to the new environment.
Centralized versus decentralized
Within the parameters of an integrated approach, solutions range from centralized—in which the government is assigned the key role in dealing with the problem—to decentralized, in which this role is transferred primarily to the creditor banks.
In the simplest form of centralized approach, the government takes over all non-performing loans and replaces them with treasury bonds, guarantees, or cash. The enterprise debt taken over by the government is then sold off, as are the physical assets of the enterprises. The government might also form a liquidation/restructuring agency whose role can vary from the mere auctioning off of the debts and assets of the troubled enterprises to a semipermanent agency entrusted with the restructuring of enterprises that cannot be liquidated rapidly.
A key issue is how to keep the problem from recurring. One option would be to rapidly develop a private banking system, either by privatizing a critical number of banks at the time of transfer of their troubled assets, or by allowing the old state-owned banks to shrink when their claims are taken over and then using any recapitalization resources to jump-start the existing private banks.
The distinguishing feature of a centralized approach is that the government is effectively in charge of dealing with the bad loan problem. But even if politics can be kept out of the process, several problems arise:
Trying to sell off debt into thin or nonexistent capital markets may result in a heavy bias toward liquidation.
The new government agency might then be forced to hold on to the majority of claims, resulting in an ongoing demand for government subsidies.
The incentives for those entrusted with running the new agency would also work against a rapid sell-off of the enterprises under their control.
In practice, the lack of administrative and financial resources to manage such a system has precluded the wholesale adoption of this policy. Proposals to rapidly privatize the banking system have also stalled because the coexistence of a largely private banking sys-tem with a largely public enterprise sector still burdened by an untenable debt level would not necessarily be an improvement over the status quo.
Modified centralization. A number of countries, such as Albania, Kazakhstan, the Kyrgyz Republic, and Romania, are in various stages of considering more manageable versions of the above approach. One option being considered is for the government to form a liquidation or restructuring agency to which the management of a preselected, limited number of the most distressed enterprises, along with their debts, is transferred. These enterprises would be effectively isolated from the banking system. The government would finance downsizing and liquidation, as well as working capital, temporary operating losses, and limited physical restucturing.
The more limited focus of the government agency in this case would allow the agency to be more effective than if burdened by the entire bad debt portfolio of the banking system. Isolating some of the most distressed enterprises would help raise the credibility of government efforts to harden budget constraints through tighter fiscal and credit policies. But because of the limited role of the new agency, the banking system would still be burdened by a significant number of nonperforming loans. Moreover, selecting candidates for the liquidation agency might be a tricky process given the difficulty of assessing the firms’ prospects under unstable macroeconomic conditions.
So far, Romania has proceeded furthest along this path. In July 1993, 30 enterprises in critical financial condition were isolated from the banking system and placed under the surveillance of a restructuring committee. In December 1993, the government endorsed the proposals of this committee to liquidate six of the enterprises and downsize or privatize the rest. It is anticipated that this process will be repeated on a larger scale.
The decentralized approach. An alternative, decentralized approach would minimize the government’s role and instead assign it to the creditor banks in cooperation with the overindebted enterprises. This would require designing a suitable incentive framework for the creditor banks to take the lead in restructuring troubled enterprises. Poland is experimenting in this direction, but it is too early to assess results.
A key objective of the decentralized option is to nudge debt restructuring toward debt-to-equity conversions to facilitate privatization and counter the bias toward liquidation inherent in the thin capital markets characterizing transition economies. In addition, formal bankruptcy procedures are minimized, as the capacity of the courts to handle the large number of cases is limited. Creditor banks are recapitalized at the beginning of the process. The cost of recapitalization is based on prior audit assessments and not on the losses incurred during the transition period. This gives banks an incentive to maximize collection of bad loans. Recapitalization should be announced as a once-and-for-all operation, after which the banks would be accountable for their losses. Finally, the recapitalization should be large enough to ensure the banks’ credibility and accountability.
A functioning regulatory framework with credible enforcement capacity is essential for the success of this option. Strong supervisory boards would be needed to allow monitoring of bank management. Giving managers a stake in the eventual privatization of the banks by linking their compensation to the bank’s value is also essential.
The Polish experience. In Poland, the objective is to restructure and privatize about 2,000 financially troubled companies, which accounted for about 40 percent of the aggregate loan portfolio of nine state-owned banks in 1992. Each bank is expected to take the lead in dealing with 200 to 300 of these enterprises. Through the so-called conciliation process, the banks are given a limited time to deal with troubled customers. They solicit reorganization proposals from other creditors, from out side investors, or from incumbent management. If the debtor is considered nonviable or is unwilling to restructure, the banks are obliged to force the debtor to file for bankruptcy or sell its troubled asset.
For enterprises that fail to reach agreements with their creditors but are too large or politically sensitive to be closed rapidly, a special support mechanism administered by the government—the so-called intervention fund—has been established. The key difference here is that candidates for government intervention are determined after negotiations between creditor and debtor have failed rather than being preselected. Moreover, negative incentives (such as loss of senior creditor status) are built in to discourage banks from excessive recourse to the fund.
“… an integrated approach to bank and enterprise restructuring is needed because neither can succeed in isolation.”
The decentralized option has been criticized for encouraging collusion between distressed enterprises and distressed banks, increasing the risk exposure of banks, and thus possibly bolstering the status quo. Clearly, the success of this approach depends on designing the right incentives for banks, ensuring close monitoring and supervision by the government, and improving banking skills and auditing techniques.
Even if enterprise restructuring proceeds as anticipated, the possibility that the banks may de facto become the major shareholders in most of the distressed enterprises is cause for concern: if banks have to focus on managing troubled enterprises over an extended period, the need to build banking skills may be neglected. The results of Poland’s experience with bank and enterprise restructuring are likely to influence the efforts of other countries in the region. It is difficult to visualize, however, how any solely bank-led scheme could function in the states of the former USSR, where the banks’ major customers are also their main shareholders and the level of commercial banking expertise is low, even in comparison with Eastern Europe.
What role for bankruptcy?
A properly functioning bankruptcy code is the ultimate form of decentralized scheme, involving minimal government intervention. But except in Hungary, bankruptcy laws remain weak or largely unenforced and have had little impact in imposing financial discipline. This reflects, on the one hand, inadequate legal frameworks and shortages of skilled personnel and, on the other, concern that the unrestrained use of bankruptcy could result in an unmanageable number of firms facing liquidation. Moreover, limiting liquidations through the constraints imposed by personnel and legal deficiencies does not appear efficient.
Hungary’s bankruptcy law. Notwithstanding these constraints, Hungary has initiated the use of bankruptcy as the central mechanism of conflict resolution between banks and enterprises. A strict bankruptcy law was adopted in January 1992 that required enterprises to file for bankruptcy or liquidation if they were more than 90 days in arrears on any of their debt. This law has played a key role in tightening financial discipline. As of September 1993, more than 5,000 bankruptcy and 16,000 liquidation proceedings had been submitted to the courts. While only a fraction of these were completed, the stock of inter-enterprise arrears and the number of involuntary creditor enterprises have dropped sharply since the adoption of the law.
In the presence of weak or insolvent banks that have substantial exposure to over-indebted firms, however, the Hungarian law by itself need not end distress borrowing; banks may prefer to continue rolling over debt rather than force the enterprise to declare bankruptcy. Hence, strengthening the banks through recapitalization, coupled with enforceable ceilings on single-borrower loan concentration, can enhance the efficacy of bankruptcy legislation. The Hungarian authorities are also promoting out-of-court settlements, a trend that will economize on the use of scarce legal resources and hence conserve them for the most serious cases.
Priorities in the former USSR
In the states of the former USSR, where inflation is very high, real interest rates substantially negative, and rollover of principal pervasive, it is hard to get a clear picture of the bad loan problem. However, the situation may sharply deteriorate. Once financial discipline is imposed and inflation is reduced, an explosion of bad loans is likely to appear on the banks’ balance sheets. The issue today is more one of containing the damage from the growth of bad loans that will follow as stabilization progresses than of dealing immediately with the existing problem.
In addition to confronting severe macroeconomic instability, most of these countries face a situation in which the number of banks has proliferated, with enterprises being the main shareholders, leading to extensive insider lending. At the same time, commercial banks hold practically no household deposits—they raise funds primarily from the enterprise sector and the central bank, which is not in a position to supervise all banks effectively.
Under these conditions, it is premature to allocate fiscal resources to pursue a comprehensive exercise to clean up loans and recapitalize banks because most banks do not even report a problem, and most household deposits are not at risk from the performance of the commercial bank loan portfolio. The emphasis in the states of the former USSR in the short term thus has to be on strengthening bank supervision, enforcing regulatory restrictions, particularly on capital requirements and connected lending, and enhancing commercial banking skills.
A recent World Bank report on Russia proposes providing incentives for banks to improve their standards and submit to stricter supervision and thus create a separate tier of banks that will be more capitalized, adhere to better banking practices, and be more closely supervised. An important element of this strategy is to impose penalties on banks that do not adhere to the mandated practices and to revoke the licenses of and to liquidate the most problematic banks. The objective is to improve the overall soundness of the banking system gradually through a continuous increase in the number and market share of the better banks. This approach cannot prevent a banking crisis from emerging following stabilization but would lower the costs and facilitate dealing with such a crisis should it occur.
Because other countries in the region share with Russia many of the institutional characteristics of its banking system and face a similar macroeconomic environment, elements of the above approach to banking reform may be relevant to them. Recapitalization of banks that have no potential for developing into modern commercial institutions is not advisable, and bank-led conciliation schemes will not be viable.
Strengthening existing private banks and encouraging the involvement of foreign banks through a direct presence and by twinning arrangements (in which foreign bankers transfer skills to local bankers) may improve credit allocation. Centrally managed restructuring agencies may need to expand and lengthen their scope of operations beyond what would ideally be advisable. And there may be little alternative to relying more heavily on appropriately designed bankruptcy legislation to complement privatization and impose financial discipline after privatization takes place. While each of these options would be difficult to implement, they may all be needed to develop a market-based system of credit allocation after macroeconomic stabilization has occurred.
Mixing and matching
It is important to emphasize that the health of any banking system is largely a reflection of the strength of the economy, which in the case of most transition economies, is still dominated by troubled state enterprises. To the extent that integrated schemes of bank and enterprise reform can speed privatization and restructuring, they should clearly be encouraged. However, if the institutional obstacles to implementing such schemes are too great, it would be preferable to proceed separately with banking and enterprise reform while devising incentives to make the processes mutually reinforcing. Irrespective of the method chosen, reducing the share of nonperforming loans to an acceptable level will be a process, rather than an event, given the difficulty of corporate restructuring, the costs of recapitalization, and the dependence on sustained economic recovery.
For countries attempting the integrated approach, some combination of centralized and decentralized schemes is likely to be required. The weight given to each scheme can be expected to vary with the degree of development of the banking system: the more developed the banking system, the greater the scope for decentralization.
For countries with relatively developed financial systems that have not yet adopted a Polish-style conciliation scheme, pursuing such an option versus strengthening the applicability of bankruptcy legislation is an unresolved issue. Monitoring the experience of Hungary and Poland will provide useful insights for other countries in the region. But ultimately institutional characteristics such as the managerial capacity and independence of the banking system must be weighed against the ability of the courts to cope with the heavier load of bankruptcy cases. Moreover, the two approaches should not be regarded as mutually exclusive. For example, even in Poland, where the conciliation process is being implemented most comprehensively, judicial bankruptcy will continue to be needed, not only for the emerging private sector but also to provide the “stick” for all parties to actually agree on a path of restructuring and debt reduction.
During a transitional period, designing bankruptcy legislation separately for state-owned enterprises and the private sector could be considered (if it does not become a disincentive to privatize), because the legislation that is appropriate to instill the requisite market discipline for the private sector may overburden the judicial system and cause an intolerable level of liquidations when applied to state-owned enterprises. The softer legislation targeted at these enterprises could then be merged with legislation applicable to the private sector over a limited time.
If a centralized approach is undertaken, the number and market share of enterprises selected to undergo restructuring through an official agency should be large enough to improve the quality of the banks’ loan portfolio significantly, but small enough so that takeover by the new government agency leads to perceptibly improved management. Depending on the mix of enterprises selected—those that can be privatized or liquidated rapidly versus those that may require fiscal subsidies for an extended period—creating separate agencies to address each objective may enhance the credibility of each function.