In almost all countries, financial safety nets are considered an integral part of the financial infrastructure and are seen as necessary for promoting the stability of financial systems by enhancing confidence in the banking system. Most financial safety nets have two key elements, namely, a lender of last resort, usually the central bank, and a deposit insurance scheme.66
A major problem with any financial safety net is that it undermines market discipline. To counteract this effect insolvent banks must be allowed to fail. Furthermore, the cost of failure should be borne first by the bank’s owners/shareholders, at least to the extent of their investment, and then by the bank’s larger creditors. Small depositors may be protected under a deposit insurance scheme, where one exists, although in the case of a systemic crisis broader deposit guarantees may be offered. Furthermore, in an effective environment a failed bank’s senior officers would be ousted.67 A gradual phasing out of public sector support would help assist market forces to operate, but would need to be based on the development of the market and be instituted carefully and at an opportune time, as abrupt changes in public policies may have adverse effects. Such changes would need to be accompanied by transparent public explanations of the new policy.
Components of Financial Safety Nets
Lender of Last Resort
Lender-of-last-resort policies typically have three primary objectives: (1) to protect the integrity of the payments system; (2) to avoid runs that spill over from bank to bank and develop into a systemic crisis; and (3) to prevent illiquidity at an individual bank from unnecessarily leading to its insolvency. A central bank may also have a role in ensuring adequate liquidity in financial markets generally, but will also have a responsibility to ensure that its monetary policy objectives are not vitiated by its last-resort lending.68
While central banks have a variety of tools that can be employed to achieve these objectives, three are most frequently employed. The first instrument is lending through the discount window to target aid to specific banks. The second and third tools, open market operations and public announcements, respectively, can be used to support the financial system as a whole. The relative importance of these tools varies across countries and according to circumstances, and their use can be influenced by such factors as the monetary policy stance of the central bank,69 the financial system’s institutional structures, and the country’s exchange rate arrangements, as well as the degree of market segmentation and widespread weakness in the system.70
The very existence of a lender-of-last-resort facility may weaken risk management incentives for banks, causing them to lend more than they otherwise might, and to maintain less liquidity than they would otherwise find appropriate. This tendency can be further exacerbated if lender-of-last-resort support is available at a subsidized rate of interest. The best practices for central banks in normal times were laid out over a century ago, and have changed little since then. (See Box 2.)71 The key practice is for the central bank to lend only to solvent institutions and to be prepared to let insolvent institutions fail.
Typical Practices for the Lender of Last Resort in Normal Times
-
Be available to the whole financial system, but only to solvent, although illiquid, institutions.
-
Lend speedily.
-
Lend only for the short term.
-
Lend only at a penalty rate.
-
Lend only if the loan is collateralized.
-
Accept collateral at a conservative value in normal times.
-
Allow individual institutions to fail and be closed.
It is frequently difficult to distinguish between illiquidity and insolvency, even in normal periods. This problem can become all the more difficult when there are concerns that denying liquidity support may result in widespread confidence problems that may in turn have the potential to create a systemic crisis.72 The globalization of banking and finance, the impact that the increased use of derivatives can have on bank liquidity, and growth in international capital flows make the demands on lender-of-last-resort systems more difficult and complex to assess. To alleviate some of these problems, the lender of last resort needs to have access to relevant supervisory information, which necessitates close and continuous contacts with the supervisory authority.
To preserve the incentive structure and to prevent lender-of-last-resort support from becoming long-term funding for the banks or turning into a source of central bank losses, central banks would normally lend short term, with collateral valued at its precrisis price levels, and at a penalty rate. However, even short-term collateralized lending needs to be conservative, since the condition of a failing bank frequently deteriorates rapidly. In such circumstances, continued lender-of-last-resort support may allow an insolvent bank to accumulate further losses. Thus, while the lender of last resort is protected by the collateral it takes, the bank’s other uninsured creditors may be made worse off as a result of the central bank’s actions.
Central banks are also at times faced with the issue of whether to support just the banking system or the whole financial system. Some countries have adopted a policy of making the discount window available only to depository institutions (which, by their nature, are particularly vulnerable to runs). However, in some cases, the banks receiving central bank assistance are, in turn, expected and sometimes encouraged to pass on this benefit by acting as lenders of next-to-last resort to their solvent customers. Those that advocate that the lender of last resort rely solely on open market operations, rather than lending to specific institutions through the discount window, argue that liquidity assistance to banks through open market operations will filter through to the whole financial system.73 However, such filtering may not occur in times of stress when confidence is low and the market becomes segmented. Many central banks therefore combine use of the discount window with open market operations.
In times of systemic crises, the central bank as lender of last resort attempts to assure the public that it will act firmly and limit the scope of any financial disturbance. While there are strong arguments for limiting such support to solvent institutions that need short-term liquidity in a crisis, the need to reassure the public may mean it will be necessary to provide support to banks that turn out to be insolvent. But in many cases, such lending goes beyond the functions of a central bank and therefore needs to be guaranteed explicitly by the government. It is important for a central bank in this position to minimize, for example through offsetting open market operations, the impact on its long-term goals for monetary policy.
In providing emergency liquidity assistance during crisis periods, central banks generally have sought to lend only on a short-term basis, but often without the penalty that they charge for liquidity assistance in normal times, since banks may be weakened due to problems that were not of their own making. When a major part of the banking system is insolvent, it is preferable that a comprehensive bank restructuring plan be designed and implemented, with new capital coming from the government and private sources, not the central bank, and any public costs recognized explicitly (see Alexander et al., 1997). The government may decide that the central bank should provide support until a systemic restructuring strategy is in place, and perhaps thereafter; but such credit would normally be explicitly guaranteed by the government.
Deposit Insurance
It is widely agreed that the primary objective of a system of deposit insurance is to provide a safe asset to small savers74 while avoiding the moral hazard that market discipline will be weakened. The protection of small depositors, while leaving large creditors at risk, will increase household confidence, help protect the payments system, and thereby provide a measure of stability for the banks. These objectives as well as the basic structure of deposit insurance schemes are conventionally defined in law and regulation.
Protecting deposits may also enable smaller and newly established banks to compete with larger, well-established banks that may be the beneficiaries of an implicit, “too big to fail” guarantee. Thus, it may help to counteract tendencies toward concentration in the banking industry, which in turn may make the banking system more competitive through the possibility of entry by new banks. While some countries have not yet enacted laws or regulations for resolving insolvent banks, the creation of a deposit insurance scheme makes it essential to have such instruments in place. Further, a formal scheme that offers limited coverage can reduce government outlays when political considerations would otherwise compel the authorities to protect all the depositors of failed banks. In most cases, banks meet the cost of the deposit insurance in normal times.
It is desirable that the deposit insurance scheme be sufficiently funded to deal with incidental bank failures and that any disbursal of funds to depositors be executed without delay. Resources may be obtained ex ante by having banks contribute to a fund that accumulates to a target level or by imposing an ex post levy on surviving banks as the need arises. It is also often argued that a deposit insurance scheme should charge a “risk-based “premium that corresponds to the degree of risk taken on by the bank, to ameliorate the adverse selection that accompanies a flat rate premium.75 When a flat rate is charged, lower-risk, well-managed banks are likely to subsidize the excesses of the higher-risk, poorly managed banks that are more likely to benefit from deposit insurance. However, measuring the risk profile and pricing risk is often difficult. All banks should be members of a deposit insurance system, since otherwise only the weak, high-risk banks will have the incentive to join, negating some of the efficiencies that arise from a broader risk-sharing across banks.
It is rarely possible to ensure that depositors, especially large depositors, retain some incentive to monitor banks and that the banks in turn have incentives to maintain sound practices, unless reliable information about the extent of coverage, procedures governing the use of deposit insurance funds, and the financial viability of the scheme is regularly and publicly disclosed. Some form of co-insurance, whereby the deposit insurance scheme pays only a percentage of the deposit insured or perhaps covers 100 percent of deposits to a certain limited threshold, is advisable. Above this limit, leaving some risk with the depositor is also helpful. Since the existence of deposit insurance is often accompanied by some increased incentive to take on riskier activities, it requires strong and professional bank supervision to monitor banks’ risk-taking activities.
A system of graduated, calibrated early intervention by the supervisory authority will help to restore problem banks to soundness or allow for their closure at minimal cost to the scheme, and with minimal damage to public confidence. Such a system would enable the deposit insurance scheme to cope with the number of failures that occur in normal times and even with most periods of multiple failures (see Box 3), through a combination of an existing fund and ex post assessments on all remaining banks.76 However, a workable scheme cannot be expected to handle the costs of a systemic crisis involving pervasive failures. Once a widespread crisis is in progress, the government may deem it necessary to institute a full guarantee, either anew or to override an existing scheme that has limited coverage, despite the moral hazard referred to above. However, it would normally only do so after other options have been rejected and the cost has been fully taken into account, and this only for a limited period.
Exit Policy
Ensuring that banks approaching insolvency leave the market quickly and cleanly is one of the most important aspects of banking supervision policy.77 An effectively implemented exit policy for weak and insolvent banks that demonstrates that banks will be allowed to fail is essential to counteract the adverse impact on risk-taking incentives created by even well-designed lenders of last resort and deposit insurance schemes. It underscores market discipline by penalizing management, owners, and large creditors. Furthermore, such a policy limits the potential losses that might otherwise accrue to the public sector. While a central bank always needs to be aware of the danger that bank closure might trigger systemic problems, the earlier action is taken, the less the impact on the rest of the system is likely to be.
Typical Practices for a Successful Deposit Insurance Scheme Under Normal Conditions and in Systemic Crises
In normal times, the scheme should
-
be explicitly defined in law and regulation;
-
resolve failed depository institutions promptly;
-
impose limitations on coverage;
-
have wide membership;
-
pay deposits quickly;
-
have adequate sources of funding to avoid insolvency;
-
offer risk-adjusted premiums (when risks can be accurately measured);
-
have accurate information and disclosure of an insured bank’s financial condition;
-
grant no decision-making authority for bankers within the deposit insurance scheme;
-
take prompt remedial actions; and
-
have close relations with the lender of last resort and the supervisor.
In a systemic crisis, a deposit insurance scheme should
-
extend coverage temporarily; and
-
obtain government backing.
An effective exit policy is not possible without an adequate and effective legal framework and a supervisory authority that has the will, autonomy, and powers to implement a firm policy. In countries where the supervisory authority is still in the process of developing sufficient independence, autonomy and skills, it may be difficult to withstand political pressures to exercise forbearance. In those circumstances, it may be desirable to have rules providing for the use of obligatory graduated corrective measures, ultimately leading to mandatory closure.78 When capital falls below a certain critical level, there is an incentive for management to try to assume more risk in the hope of benefiting from the higher return; hence, there is frequently a strong likelihood that the bank will, upon closer scrutiny, prove to be insolvent. Early intervention therefore limits the damage to stakeholders in the bank. A rule-based system for intervention in banks can, however, increase the incentives for “window dressing” by bankers, and even bank supervisors, who could be reluctant to intervene if such intervention were to trigger a mandatory response.
Rule-based exit policies can be applied to individual banks. However, in distressed banking systems such a policy would not be viable, as it could lead to a large number of more or less simultaneous bank closures. In these circumstances, care must be taken that the necessary mechanism is in place to manage closures and recovery plans of undercapitalized but viable banks. Such a mechanism could be either inside the supervisory authority or in a separate body.79 Once the independence, authority, and quality of banking supervision have been enhanced, and the pressure to exercise forbearance has eased, a more discretionary system can be considered.
Conservatorship
When the problems in a bank threaten its viability, the imposition of conservatorship by the supervisory authority needs to be considered. Conservatorship can be described as the assumption of control, by or on behalf of the supervisory authorities, of a bank that is facing serious problems.80 Under conservatorship it may be necessary to suspend the bank’s obligations temporarily, so that time can be gained to assess whether the bank can be put back on track while protecting the interests of the bank’s creditors. The conservator establishes the bank’s current net worth.81 If viability seems possible, the conservator may manage the bank until new, qualified shareholders and management can be found. If not, the conservator will initiate liquidation proceedings. Experience in many countries shows that banks placed under conservatorship are rarely restored to profitability without major restructuring and injection of new capital. Ideally, a bank should therefore spend as little time as possible in conservatorship.
It is essential that the supervisory authority take action well before the bank fails. Once a bank starts to experience difficulties in its asset portfolio, its first tendency is to capitalize accrued interest, to underprovision, or to sell remaining good assets, in order to present an artificially high level of capital. During the decline of the bank’s condition, the risk of dissipation of assets or looting rises rapidly. Therefore, the decision to impose conservatorship or close the bank needs to have immediate legal effect, notwithstanding possible appeals against such decisions. In the majority of cases, banks that have been placed under conservatorship or similar form of supervisory control do in fact prove to be insolvent.
Closing a Bank
Once it is clear that a bank cannot be restored to profitability, or that the use of conservatorship is not feasible owing to either a shortage of qualified persons or a lack of supervisory resources, its license has to be withdrawn, and the liquidation process initiated. The decision that a bank is to be liquidated can result from (1) the supervisor’s judgment that the bank is insolvent; (2) the violation of other licensing criteria; (3) missed interest payments or other financial obligations that spur creditors to file a bankruptcy suit; or (4) bank owners’ voluntarily desire to close the bank. The first reason for liquidation is the most problematic. The closer a bank approaches insolvency, the stronger will be the incentives to hide information from the supervisory authorities and the markets. This makes it difficult for the supervisor to establish insolvency irrefutably, and makes the timing of the bank closure difficult. The supervisor risks taking corrective action too late, in which case it will be reproached for not taking timely measures, or it risks taking action at a time when the bank will be able to mount a plausible case that closure was unnecessarily imposed and caused material damage to shareholders. While in many countries the decision to liquidate is for the courts, as it affects the interests of creditors and shareholders, it is important that the supervisor have powers to initiate the process and to restrict the bank’s business pending the court’s decision.
When the bank is insolvent, the loss will need to be allocated among creditors. Most legislative systems have set priorities for the satisfaction of different categories of creditors. Where there is no deposit insurance system, priority settlement to household depositors is sometimes recognized. To ensure problems of moral hazard are kept in check, the primary brunt of the bank’s failure needs to be borne by the bank’s shareholders. The technical procedure for liquidation is normally laid out in the law, and requires it to be performed by a professional liquidator. This function generally does not belong within the supervisory authority.
Because confidence in a bank can be terminally damaged by a bankruptcy suit brought by creditors, especially if the suit is unjustified, and because there is a need for speedy action, special insolvency procedures for banks are often established. The supervisory authorities may then be able to delay a decision on the bank’s bankruptcy, while they investigate whether the bank is solvent or not.82