Skip to main content

Advertisement

SpringerLink
Log in

The Fall and Rise of Debt: Bank Capital Regulation After the Crisis

  • Article
  • Published:
European Business Organization Law Review Aims and scope Submit manuscript

Abstract

Under the Basel capital rules for internationally active banks, subordinated debt has always been permitted to contribute a part of the bank’s regulatory capital requirements. This is a surprising concession to banks, at first sight, since debt, as a liability, cannot contribute to equity (i.e., the surplus of assets over liabilities). The Basel approach is explicable only on the basis that bank capital is designed to protect short-term debt holders, who might ‘run’ on the bank and cause its collapse, rather than bank creditors as a whole. Long-term debt subordinated to the holders of non-subordinated debt could conceivably discharge this function. In the financial crisis, however, subordinated debt singularly failed to discharge this role. Because debt absorbs losses only when the company is put into insolvency (bankruptcy) and states were unwilling to contemplate the bankruptcy of banks, in most cases, debt did not in fact absorb losses as anticipated. Bail-out by the state pre-empted loss falling on creditors. An initial, and wholly to be expected, response on the part of the Basel III authorities was to downgrade the role of even subordinated debt in regulatory bank capital and in the new capital buffers which Basel III introduced. It was to be allowed only where the resolution authorities could write it down or convert it into equity (bail it in) in advance of insolvency. However, in a second phase of reform, bail-in debt seems likely to become a mandatory element in the newly devised procedures for resolving failing banks. Bail-in in resolution is changing from being a concession to banks, which want to issue debt, to a requirement imposed with the aim of giving resolution authorities sufficient room for manoeuvre to reorganise failing banks. Whilst the role of bail-in debt in future regulation is not secure – and has fierce critics – this constitutes a remarkable turnaround for a liability that was once seen as almost beyond redemption.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

Notes

  1. Gorton (2012), Ch XIII; Shin (2010), Ch 8.

  2. Basel Committee on Banking Supervision, Basel III: A global regulatory framework for more resilient banks and banking systems, revised June 2011 (bcbs 189), para 131.

  3. Ibid, § I.B.

  4. Tucker (2014a) estimates (at n. 7) that the definitional changes to CET1 have doubled the effective impact of this requirement, i.e., that in terms of the former definition the increase has been from 2 to 9 %, not to 4.5 %.

  5. The contribution of T2 capital is reduced by half.

  6. Bank for International Settlements, Press Release 03/2011, Annex.

  7. See Financial Stability Board (FSB) 2014 update of list of global systemically important banks, November 2014.

  8. Basel Committee on Banking Supervision, Global systemically important banks: assessment methodology and the additional loss absorbency requirement, November 2011, § IV.C (bcbs 207), discusses the pros and cons of allowing ‘high trigger’ cocos (see Sect. 4 below) to count as well but comes out against such a move.

  9. Basel Committee on Banking Supervision, A framework for dealing with domestic systemically important banks, October 2012 (bcbs 233).

  10. ‘Pillar 2’ of the Basel system (in contrast to the minimum requirements described above which constitute ‘Pillar 1’) establishes national supervisory discretion. This is not a new feature of the Basel system, though it has been used more vigorously by national regulators since the crisis.

  11. Supra n. 2, para 122. Whether it will in fact operate in this way is debatable.

  12. Ibid, § IV.

  13. The contrast is with the standardised approach in which the risk weighting is done within the Basel rules themselves.

  14. Supra n. 2, para 151.

  15. International Monetary Fund (IMF) Global Financial Stability Report, April 2008, Box 1.3.

  16. ‘EU stress test pass rate under fire’, Financial Times, 15 July 2011.

  17. Bank of England, Stress testing of the UK banking system: key elements, April 2014.

  18. This is an exercise of Pillar 2 powers. See supra n. 10.

  19. Bank of England, Stress testing the UK banking system: key elements of the 2014 stress test, at pp. 12–13.

  20. European Central Bank, Press Release, 29 April 2014, requiring CET1 to repair deficiencies shown by the asset quality review and baseline stress test, whilst permitting ‘high trigger’ subordinated debt to contribute up to 1 % RWA to the shortfall resulting from the ‘adverse’ scenario.

  21. Supra n. 2, para 152.

  22. Even at this modest level, the leverage ratio would have done something to constrain leverage in UK banks in the 2 years prior to the crisis, when the 33:1 ratio was exceeded and almost reached 50:1. Bank of England, Financial Stability Report, December 2011, chart 2.4.

  23. ‘Exposures’ takes account of, for example, certain off-balance sheet items.

  24. Supra n. 22, Box 2. All these schemes use book value of equity, not market value. Given that the Basel rules are built on book value, it may have been thought to be introducing an unnecessary level of complexity to move one component of the capital requirements to market value. There may also have been fears that market value can be manipulated.

  25. Financial Stability Board, Adequacy of loss-absorbing capacity of global systemically important banks in resolution, Consultative Document, November 2014. This is twice the number proposed by the Basel Committee, just as the FSB proposes risk-weighted capital requirements for GSIBs of twice the Basel minima.

  26. See supra text attached to n. 3.

  27. Basel Committee on Banking Supervision, Revised Basel III leverage ratio framework and disclosure requirements, Consultative Document, June 2013 (bcbs 251) para 8; implemented by Regulation (EU) No 575/2013 on prudential requirements for credit institutions and investment firms, OJ L 176/1, 27.6.2013, Art. 429.3.

  28. Supra n. 25, at p. 6.

  29. See Federal Reserve (US), Opening statement by Governor Tarullo, in relation to the Federal Reserve proposals to strengthen the capital positions of the largest, most strategically important US bank holding companies, 9 December 2014, available at http://www.federalreserve.gov/newsevents/press/bcreg/tarullo-statement-20141209.htm (last visited 17 May 2015); Prudential Regulatory Authority (UK), Assessing capital adequacy under Pillar 2, Policy Statement 17/15.

  30. Set out in various papers but principally Admati and Hellwig (2013).

  31. Modigliani and Miller (1958).

  32. DeAngelo and Stulz (2013).

  33. Independent Commission on Banking (ICB) (UK), Final Report, September 2011, Figure 4.6, shows that in 2010 Barclays and RBS had senior debt equivalent to about 30 % RWA; the other major UK banks had less, only 12 % in the case of HSBC.

  34. Helpfully summarised by Rochet (2014).

  35. This is probably the issue on which the studies are weakest, but, since they identify a substantial social benefit, they do not affect the argument in this paper.

  36. Miles et al. (2013).

  37. Schoenmaker (2015) suggests, at p. 16, that the Miles et al. recommendation is a leverage ratio. But this does not seem to be so. ‘Our estimate of optimal bank capital is that it should be around 20 % of risk weighted assets. If risk weighted assets are between 1/2 and 1/3 of total assets then even with equity at 20 % of risk weighted assets debt would be between 90 and 93 % of total funding’ (Miles et al. (2013), at p. 30).

  38. ICB, supra n. 33, 4.102ff, esp. 4.114. See also: Final report of the Commission of Experts for Limiting the Economic Risks Posed by Large Companies (Switzerland), September 2010.

  39. As explained below, the 20 % figure implies some element of bail-in of liabilities which do not count as capital.

  40. Even in the absence of subsidies, there is likely to be a ‘debt overhang’ problem in relation to equity issuance by a bank in financial difficulties, leading the bank to adopt a policy of adjusting its equity to debt ratio by scaling back operational activities rather than by raising new capital, i.e., the opposite of what is desirable in public policy terms. A&H, at pp. 172–6, have various suggestions for counter-acting this tendency, for which there is no space to discuss here. Overall, their suggestions are likely to reduce the attractiveness of equity investment in banks and thus raise banks’ cost of equity.

  41. This is something of an overstatement, where, as is the case in banking regulation, equity is defined on an accounting basis, because accounts are produced only periodically (typically half yearly) and are a lagged indicator of the bank’s position. It would be possible (at least for non-mutual banks) to define equity in market terms (i.e., as the current capitalisation of the bank’s shares). Although sometimes advocated (see below), a market measure is open to manipulation more easily than an accounting measure.

  42. Of course, debt may trade in the market at a discount to its face value, reflecting the risk that the company will not meet its obligations – a discount which may be small in the face of an implicit state guarantee of bank debt.

  43. Thus, it is sometimes said the equity is going-concern capital whilst debt is ‘gone concern’ capital, but bail-in debt cuts across this distinction.

  44. Supra text attached to n. 32.

  45. That a bank run is usually a self-fulfilling prophecy, in the absence of governmental or central bank support, is obvious, but this does not make withdrawal an irrational action on the part of the funder which gets its money out ahead of the rest.

  46. Opinions differ about how much modification of standard bankruptcy procedures is needed to make them fit banks and other financial institutions. The answer obviously depends in part on the design of national insolvency systems, i.e., how strongly they have embodied the elements of a ‘rescue culture’. See Jackson and Skeel (2012).

  47. Directive 2013/59/EU, Art. 32, OJ L 173/190, 12.6.2014.

  48. Partly paid equity, where the unpaid amount became due only in distress, was not unknown in nineteenth century corporate financing in the UK, but is virtually unknown today. In any event, the unpaid amount does not count as CET1 capital: supra n. 2, at para 53. This seems correct since the bank carries the risk that the shareholder will not be able to meet the commitment when called upon, whereas with bail-in debt the contribution has already been made.

  49. There is obviously a policy question whether the tax shield should be extended to hybrid debt, but most jurisdictions which allow the relief in principle extend it, at least in relation to regulatory hybrid debt.

  50. Supra n. 40.

  51. FSB, supra n. 25.

  52. Supra n. 33, at para 4.16.

  53. Admati and Hellwig (2013), at p. 187.

  54. Gordon and Ringe (2015).

  55. On cross-border issues, see Davies (2015). Even under the (less preferred) multiple point of entry (MPE) approach a GSIB group is likely to be resolved, not on an entity by entity basis, but on a sub-group by sub-group basis, so the holding of debt by the relevant sub-group holding company becomes an important issue.

  56. For an overview, see Krimminger (2015).

  57. This alternative use of bail-in is recognised in Art. 43(2)(b) of the BRRD.

  58. See Davies (2013).

  59. The first Recital to the BRRD states: ‘During the crisis, those challenges [minimising the negative consequences of insolvency] were a major factor that forced Member States to save institutions using taxpayers’ money. The objective of a credible recovery and resolution framework is to obviate the need for such action to the greatest extent possible.’

  60. Art. 37 of the BRRD sets out bail-in and the tools necessary for purchase and assumption (sale of business and the creation of a bridge bank) in a neutral way. Except for the ‘asset separation’ tool (i.e., the ‘bad bank’ tool) ‘resolution authorities may apply the resolution tools individually or in any combination’ (Art. 37(4)).

  61. Once viable on a balance sheet basis, the bank is expected to have access to central banks’ liquidity facilities if liquidity is not sufficiently available in the private market (as it is likely not to be in the case of the failure of a GSIB).

  62. BRRD Art. 45. The BRRD minimum is to be expressed as a combination of equity (‘own funds’) and bail-in debt (‘eligible liabilities’). However, within that mixture, Art. 45(6) requires the resolution authority to set this minimum on the basis, inter alia, of ‘the need to ensure, in appropriate cases, that the institution has sufficient eligible liabilities to ensure that, if the bail-in tool were to be applied, losses could be absorbed and the Common Equity Tier 1 ratio of the institution could be restored to a level necessary to enable it to continue to comply with the conditions for authorisation’.

  63. Bank of England, Resolution and the future of finance, speech by Paul Tucker, 20 May 2013. Cf BRRD Art. 45(6)(b), supra n. 62, which seems to reflect this formula except for the absence of a reference to how the potential loss should be estimated (the ‘stressed scenario’ in Tucker’s formulation).

  64. BRRD Art. 48(1)(e) clearly contemplates a bail-in of other liabilities once Basel III debt capital and subordinated debt has been bailed-in.

  65. See text attached to nn. 70–74 infra.

  66. Presumably this recommendation will inform Member States’ and the EBRD’s determination of minimum own funds and eligible liabilities under Art. 45 of the BRRD.

  67. When these are included, the range increases to 19.5–23 %, depending on the size of the GSIB buffer: FSB, supra n. 25, Proposed Total Loss Absorbency Capacity (TLAC) term sheet, Item 4.

  68. Ibid, at p. 6. Cf Tucker (2014a), who suggested a minimum of one half.

  69. FSB, supra n. 67, Item 13c. There is some leeway in respect of liabilities which are, or may be, statutorily excluded from bail-in, i.e., these may be held by the entity that has issued the bail-in debt. Structural subordination is arguably unnecessary when the bail-in debt is contractually subordinated to other liabilities and the relevant regulation requires the bank to issue bail-in debt in subordinated form.

  70. Ibid, Items 11 and 12.

  71. Ibid, at p. 7.

  72. Art. 44(2).

  73. See Shin (2010), Ch 8.

  74. Art. 44(3).

  75. But see European Banking Authority, Technical advice on the delegated acts on the circumstances when exclusions from the bail-in tool are necessary, March 2015.

  76. Art. 51. This provision is in addition to the recovery plans required to be drawn up by GSIBs under Art. 5.

  77. BRRD, Arts. 56 and 37(10).

  78. Thus, in § 204, Dodd–Frank defines the purpose of the Orderly Liquidation Authority (the main mechanism for resolving global systemically important financial institutions) as follows: ‘It is the purpose of this title to provide the necessary authority to liquidate failing financial companies that pose a significant risk to the financial stability of the United States in a manner that mitigates such risk and minimizes moral hazard.’ Later, in § 206, it is provided that the FDIC, acting under OLA, ‘shall not take an equity interest in or become a shareholder of any covered financial company or any covered subsidiary’. Most broadly of all, § 214 provides that ‘[n]o taxpayer funds shall be used to prevent the liquidation of any financial company under this title’.

  79. Despite the arguments for using market capitalisation as the trigger point, regulation sticks to accounting measures. The latter may not be as readily available as market measures but they are less capable of manipulation by market players.

  80. Swiss Commission, supra n. 38, at p. 25.

  81. They are triggered when either total capital declines to 8 % RWA or CET1 to 5 %, i.e., at or very near the Basel minima, when it is expected that the resolution authority will intervene. See FINMA (Swiss Financial Market Supervisory Authority), Resolution of global systemically important banks, § 6.2.

  82. On the use of subordinated debt to meet the Basel III requirements, see text attached to n. 3 supra.

  83. FINMA, supra n. 81, § 6.1. Although the Swiss minimum capital requirement for GSIBs is 10 % RWA (plus the applicable GSIB buffer), 5.5 % of that is an enhanced capital conservation buffer. The ‘hard’ minimum is thus 4.5 %, which explains why the Swiss authorities expect to trigger resolution at the 5 % level. Consequently, a coco with a 7 % trigger will be written off or converted ahead of resolution.

  84. Art. 45(14) of the BRRD appears to permit the issuance of cocos as a method meeting the minimum requirements under the Directive, but does not mandate it.

  85. The Basel Committee on Banking Supervision, supra n. 8, sets out the uncertainties which regulators have identified in relation to the operation of cocos. See also Admati and Hellwig (2013), at p. 188 (whose remarks seem better directed at cocos than bail-in debt).

  86. ‘Today, cocos with decently high triggers are likely to be prohibitively expensive, as they would be close to being in-the-money until banks have (truly) built tangible common equity in excess of the full, as opposed to transitional, Basel 3 standard. Eventually, I can conceive that cocos with highish triggers might be issued as a means for the market to maintain control of its own destiny in the shadow of resolution’ (Tucker 2014b).

  87. Goodhart and Avgouleas (2014).

  88. Tucker (2014b), at p. 9.

  89. See supra nn. 62 and 68.

  90. The one third figure is already a compromise. Paul Tucker, who was a leader in the FSB discussion and is perhaps the most outspoken proponent of bail-in debt in resolution, suggested one half as the appropriate proportion of subordinated debt in total capital. See Tucker (2014a), at p. 7.

  91. See supra n. 62.

References

  • Admati A, Hellwig M (2013) The bankers’ new clothes. Princeton University Press, Princeton

    Google Scholar 

  • Davies P (2013) Federal Deposit Insurance Corporation and Bank of England memorandum on resolving globally active systemically important financial institutions. Law Financ Mark Rev 7:304–310

    Article  Google Scholar 

  • Davies P (2015) Resolution of cross-border banking groups. In: Haentjens M, Wessels B (eds) Research handbook on crisis management in the banking sector. Edward Elgar Publishing Ltd, Cheltenham

    Google Scholar 

  • DeAngelo H, Stulz R (2013) Why high leverage is optimal for banks. European Corporate Governance Institute Finance Working Paper 356/2013

  • Goodhart C, Avgouleas E (2014) A critical evaluation of bail-ins as bank recapitalisation mechanisms. Centre for Economic Policy Research. Discussion Paper 10065

  • Gordon J, Ringe W (2015) Bank resolution in Europe: the unfinished agenda of structural reform. In: Busch D, Ferrarini G (eds) European Banking Union. Oxford University Press, Oxford

    Google Scholar 

  • Gorton G (2012) Misunderstanding financial crises. Oxford University Press, New York

    Google Scholar 

  • Jackson T, Skeel D (2012) Dynamic resolution of large financial institutions. Harv Bus Law Rev 2:435–460

    Google Scholar 

  • Krimminger M (2015) Shadows and mirrors: the role of debt in the developing resolution strategies in the U.S., U.K., and European Union. http://ssrn.com/abstract=2560982

  • Miles D, Yang J, Marcheggiano G (2013) Optimal bank capital. Econ J 123:1–37

    Article  Google Scholar 

  • Modigliani F, Miller MH (1958) The cost of capital, corporate finance and the theory of investment. Am Econ Rev 48:261–297

    Google Scholar 

  • Rochet JC (2014) The extra cost of Swiss banking regulation. Swiss Finance Institute, Zurich

    Google Scholar 

  • Schoenmaker D (2015) Regulatory capital: why is it different? Available at SSRN: http://ssrn.com/abstract=2568158

  • Shin H (2010) Risk and liquidity. Oxford University Press, New York

    Google Scholar 

  • Tucker P (2014a) Regulatory reform, stability and central banking. Brookings, Washington, DC

    Google Scholar 

  • Tucker P (2014b) The resolution of financial institutions without taxpayer solvency support: seven retrospective clarifications and elaborations. http://www.cepr.org/sites/default/files/events/papers/6708_TUCKER%20Essay.pdf

Download references

Author information

Authors and Affiliations

  1. Centre for Commercial Law, Harris Manchester College, Mansfield Road, Oxford, OX1 3TD, UK

    Paul Davies

  2. Oxford University, Oxford, UK

    Paul Davies

Authors
  1. Paul Davies
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Paul Davies.

Additional information

Paul Davies: Senior Research Fellow; Allen & Overy Emeritus Professor of Corporate Law.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Davies, P. The Fall and Rise of Debt: Bank Capital Regulation After the Crisis. Eur Bus Org Law Rev 16, 491–512 (2015). https://doi.org/10.1007/s40804-015-0017-8

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s40804-015-0017-8

Keywords

Search

Navigation