LSE - Small Logo
LSE - Small Logo

Blog Admin

May 7th, 2013

Higher capital requirements do not necessarily make banks safer


Estimated reading time: 5 minutes

Blog Admin

May 7th, 2013

Higher capital requirements do not necessarily make banks safer


Estimated reading time: 5 minutes

David Murphy#Following the financial crisis, regulators have come up with tougher requirements for banks, forcing them to hold higher levels of common equity capital. David Murphy explains that banks that are distressed are usually suffering from a liquidity shortage rather than actually being insolvent, and that higher capital requirements on their own will not end bank runs. Regulators should instead focus on ensuring confidence in banks in order to avoid liquidity crises.

It can sometimes seem as if there is one tool to improve financial stability – call for higher capital requirements. From Basel III through to the most recent UK Financial Policy Committee report, the official sector has been unwavering. Many commentators have applauded this, and urged even more action (the recent book by Admati and Hellwig, the Banker’s New Clothes, is a prominent example here). This rush for ever higher capital requirements, while well-motivated, may be counterproductive. To see why, we need to look at how large financial institutions fail.

How big banks become stressed

There are two different modes of financial institution stress:

  • Solvency becomes an issue if an institution suffers losses that are comparable to the institution’s equity.  Thus if a bank is 20 to 1 leveraged, and it suffers a loss of 2% of its assets, half of its equity is wiped out, and stress is highly likely.
  • Liquidity is an issue for most institutions because they have shorter term liabilities than assets, and hence they need to keep borrowing in order to carry on. If the market withdraws funding, then they need either to access a lender of last resort (such as the central bank) or else fail.

It is important to note that liquidity is not necessarily related to solvency; a solvent firm can suffer from liquidity stress. Indeed, liquidity stress usually comes first in a crisis.  Large financial institutions do not often fail because they are insolvent, but rather because they are illiquid. As an illustration, consider three examples from the 2007+ crisis:

  • Bear Stearns failed and a hasty sale was arranged.  The purchaser, JPMorgan, paid Bear Stearns’ stock holders about $10 per share indicating that it was solvent at the point of failure.
  • The Valukas report into the failure of Lehman makes it clear that the immediate cause of failure was liquidity, saying that the firm ‘no  longer had sufficient  liquidity to fund  its daily operations’ and by the evening of September 12, its ‘only hope to survive was federal [liquidity] assistance or a merger.’
  • Similarly the FSA report into RBS gives a central role to liquidity risk, showing how the banks’ ability to borrow deteriorated until it was forced to apply to the Bank of England for emergency liquidity support.

Lending decisions

Big banks fail, then, because short term lenders refuse to let them continue to borrow. (The traditional form of this is the run on deposits whereby depositors withdraw the funds that they have lent to a bank.  Modern deposit insurance schemes have somewhat mitigated the risk of this, but instead we see stress manifested in wholesale uninsured borrowings such as those in the repo and commercial paper markets.) It might be thought that this is odd: after all, if a firm is solvent why won’t people lend to it? There are two interlinked answers to this.  The first is that solvency is a matter of debate. While a firm might have persuaded its auditors that it has a positive net worth, investors may take a different view.  Some assets are hard to value because they are illiquid; others are contestable because they rely on an assessment of the future. In particular loan loss provisions for a bank are highly sensitive to assumptions about upcoming defaults, and the banking system has a lamentable track record of being proved to have been conservative in their assessments. Therefore an investor may be unwilling to lend money to a bank even if it appears to be solvent.

The second issue is that if solvent banks can fail for liquidity reasons, a potential investor must assess whether other investors will lend to the bank. Thus what can matter is not fundamentals but rather market confidence: rationally I should only lend to a firm that I think others will have confidence in. Banking is in this sense a confidence trick.  Confidence breeds lending which allows the institution to carry on and thus justifies itself, while a failure of confidence can lead to failure.

Capital requirements

Financial institutions have to adhere to their capital requirements at all times. Therefore if a bank is set, say, a capital ratio of 10%, it must always have that amount of capital.  Required amounts of capital cannot be used to absorb losses as using them for that purpose would leave the bank badly-capitalised. It is only capital above the regulatory minimum that is available to absorb losses.

Regulatory forbearance does not help here because of the funding markets. If supervisors say in the good times that a bank must have a 10% capital ratio, it will do no good if they later change their minds and say 7% will do. Investors will carry on demanding 10% if the bank wants to borrow.

In this light, the dramatic rise in capital requirements since the crisis is destabilising.  Clearly a bank which has more capital can absorb more losses if all other things are equal. But a bank which has more capital because regulators and investors demand it can’t necessarily absorb more losses before suffering stress.

Policy implications

A consideration of the central role of confidence and liquidity stress in bank failure suggests a rather different policy from that adopted since the crisis by leading financial institution policy makers:

  • Banks need high levels of capital above the regulatory minimum, as only these are available to absorb losses.
  • A key test of a regulatory and accounting regime should be that if a bank is officially well capitalised then it can borrow.  Various pre-crisis regimes, such as Basel II and the SEC’s consolidated supervised entity program, failed this test.  It is not clear that Basel III will do any better, not least due to the lack of confidence investors now have in complex models based capital models, whatever the headline capital ratio.  Recent calls for much simpler capital calculations, such as those from Haldane & Madouros and Hoenig, highlight the importance of simple, easily understood, easily checked requirements.
  • Other measures to retain confidence in institutions that are solvent are important.  Improved disclosure on valuation policy, better accounting standards (a loan loss provisioning framework for banks that investors can have confidence in would be a good start), more challenging and independent audits, and better supervision of valuation practices all have important parts to play.
  • Market signals of falling confidence are important early warning signals for supervisors. If, for instance, a bank’s price/book ratio falls too far, that bank’s supervisor should consider intervening to restore confidence.

This post is a based on D. Murphy, Maintaining Confidence, Special Paper 216, LSE Financial Markets Group Paper Series (December 2012).

Note: This article gives the views of the author, and not the position of the British Politics and Policy blog, nor of the London School of Economics. Please read our comments policy before posting.

About the Author

David Murphy is the principal of rivast consulting, a leading risk management and regulatory consulting practice.  He blogs on regulatory capital, derivatives and risk management at

Print Friendly, PDF & Email

About the author

Blog Admin

Posted In: Economy and Society

Leave a Reply

Your email address will not be published. Required fields are marked *

Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported
This work by British Politics and Policy at LSE is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported.