IMF Staff Discussion Notes No. 12/03
April 24, 2012
ISBN/ISSN: 978-1-61635-392-6 / 2221-030X
Stock No: SDNETEA2012003
Large-scale government support of the financial institutions deemed too big or too important to fail during the recent crisis has been costly and has potentially increased moral hazard. To protect taxpayers from exposure to bank losses and to reduce the risks posed by too-big-tofail (TBTF), various reform initiatives have been undertaken at both national and international levels, including expanding resolution powers and tools.
One example is bail-in, which is a statutory power of a resolution authority (as opposed to contractual arrangements, such as contingent capital requirements) to restructure the liabilities of a distressed financial institution by writing down its unsecured debt and/or converting it to equity. The statutory bail-in power is intended to achieve a prompt recapitalization and restructuring of the distressed institution. This paper studies its effectiveness in restoring the viability of distressed institutions, discusses potential risks when a bail-in power is activated, and proposes design features to mitigate these risks. The main conclusions are:
1. As a going-concern form of resolution, bail-in could mitigate the systemic risks associated with disorderly liquidations, reduce deleveraging pressures, and preserve asset values that might otherwise be lost in a liquidation. With a credible threat of stock elimination or dilution by debt conversion and assumption of management by resolution authorities, financial institutions may be incentivized to raise capital or restructure debt voluntarily before the triggering of the bail-in power.
2. However, if the use of a bail-in power is perceived by the market as a sign of the concerned institution’s insolvency, it could trigger a run by short-term creditors and aggravate the institution’s liquidity problem. Ideally, therefore, bail-in should be activated when a capital infusion is expected to restore a distressed financial institution to viability, with official liquidity support as a backstop until the bank is stabilized.
3. Bail-in is not a panacea and should be considered as one element of a comprehensive solution to the TBTF problem. It should supplement, not replace, other resolution tools that would allow for an orderly closure of a failed institution.
4. Most importantly, the bail-in framework needs to be carefully designed to ensure its effective implementation.
- The triggers for bail-in power should be consistent with those used for other resolution tools. They should be set at the point when a firm would have breached the regulatory minima but before it became balance-sheet insolvent. To make bail-in a transparent tool, its scope should be limited to (i) elimination of existing equity shares as a precondition for a bail-in; and (ii) conversion and haircut to subordinated and unsecured senior debt. Debt restructuring under a bail-in should take into account the order of priorities applicable in a liquidation.
- A clear and coherent legal framework for bail-in is essential. The legal framework needs to be designed to establish an appropriate balance between the rights of private stakeholders and the public policy interest in preserving financial stability. Debt restructuring ideally would not be subject to creditor consent, but a “no creditor worse off” test may be introduced to safeguard creditors’ and shareholders’ interests. The framework also needs to provide mechanisms for addressing issues associated with the bail-in of debt issued by an entity of a larger banking group and with the cross-border operations of that entity or banking group.
- The contribution of new capital will come from debt conversion and/or an issuance of new equity, with an elimination or significant dilution of the pre-bail in shareholders. Bail-in will need to be accompanied by mechanisms to ensure the suitability of new shareholders. Some measures (e.g., a floor price for debt/equity conversion) might be necessary to reduce the risk of a “death spiral” in share prices.
- It may be necessary to impose minimum requirements on banks for issuing unsecured debt or to set limits on the encumbrance of assets (which have been introduced by many advanced countries). This would help reassure the market that a bail-in would be sufficient to recapitalize the distressed institution, thus forestalling potential runs by short-term creditors and avert a downward share price spiral. The framework should also include measures to mitigate contagion risks to other systemic financial institutions, for example, by limiting their cross-holding of unsecured senior debt.
Bail-in power needs to be considered as an additional and complementary tool for the resolution of SIFIs. Bail-in is a statutory power of a resolution authority, as opposed to contractual arrangements, such as contingent capital requirements. It involves recapitalization through relatively straightforward mandatory debt restructuring and could therefore avoid some of the operational and legal complexities that arise when using other tools (such as P&A transactions), which require transferring assets and liabilities between different legal entities and across borders. By restoring the viability of a distressed SIFI, the pressure on the institution to post more collateral, for example against their repo contracts, could be significantly reduced, thereby minimizing liquidity risks and preventing runs by short-term creditors.
The design and implementation of a bail-in power, however, need to take into careful consideration its potential market impact and its implications for financial stability. It is especially important that the triggering of a bail-in power is not perceived by the market as a sign of the concerned institution’s non-viability, a perception that could trigger a run by short-term creditors and aggravate the institution’s liquidity problem. An effective bail-in framework generally includes the following key design elements:
- The scope of the statutory power should be limited to (i) eliminating or diluting existing shareholders; and (ii) writing down or converting, in the following order, any contractual contingent capital instruments, subordinated debt, and unsecured senior debt, accompanied by the power of the resolution authority to change bank management.
- The triggers for bail-in power should be consistent with those used for other resolution tools and set at the point when an insititution would have breached the regulatory minima but before it became balance-sheet insolvent, to allow for a prompt response to an SIFI’s financial distress. The intervention criteria (a combination of quantitative and qualitative assessments) need to be as transparent and predictable as possible to avoid market uncertainty.
- It may be necessary to require banks or bank holding companies to maintain a minimum amount of unsecured liabilities (as a percentage of total liabilities) beforehand, which could be subject to bail-in afterwards. This would help reassure the market that bail-in is sufficient to recapitalize the distressed institution and restore its viability, thus reduce the risk of runs by short-term creditors.
- To fund potential liquidity outflows, and given the probable temporary loss of market access, bail-in may need to be coupled with adequate official liquidity assistance.
- Bail-in needs to be considered as one element of a comprehensive framework that includes effective supervision to reduce the likelihood of bank failures and an effective overall resolution framework that allows for an orderly resolution of a failed SIFI, facilitated by up-to-date recovery and resolution plans. In general, statutory bail-in should be used in instances where a capital infusion is likely to restore a distressed financial institution to viability, possibly because, other than a lack of capital, the institution is viable and has a decent business model and good riskmanagement systems. Otherwise, bail-in capital could simply delay the inevitable failure.