Lessons from Northern Rock: How to handle failure

Willem Buiter 05 March 2008

a

A

In my previous column, I examined proposals for preventing financial crises in the UK Treasury Committee Report The Run on the Rock. Here, I look at mechanisms that might reduce the impact of failing banks, provide appropriate deposit insurance, and coordinate the three institutions responsible for financial stability.

Dealing effectively with failing banks

The authorities are effectively proposing to put in place the kind of legal and regulatory arrangements currently found in the United States and a number of other countries. A special resolution regime (SRR) would be created, led by a new authority (I shall call it the special resolution regime authority or SRRA, not to be confused with SSRI, lest we get some very depressed bankers), who could take control of a troubled bank before it hit the normal insolvency buffers – inability to service its debt. The assets of the pre-failing bank, or any of its activities and business, could be transferred to one or more healthy banks or some other third party; a ‘bridge bank’ could be created to allow the SRRA to take control of all or part of a bank or of its assets and liabilities; a ‘restructuring officer’ could be appointed by the SSRA to carry out the resolution; and finally, if the judgement is reached that pre-insolvency resolution is not feasible, a special bank insolvency procedure could be invoked to facilitate swift and efficient payment of insured depositors. Public ownership of all or part of a bank as a last resort is also part of the package. The Treasury document refers to it as temporary public ownership, but unless this means that a fixed time table has to be provided, the word ‘temporary’ only indicates hope or intent and is not operational.

The government proposes that the FSA would be the SRRA, and I agree with that. It should not be the Bank of England (because the job of the SRRA is too political) or the Treasury (because the Treasury is too political for the job of the SRRA). A new separate entity would be possible, but further balkanisation of the responsibility for financial stability in the UK would seem undesirable (anyone really wants a Quadripartite Arrangement?).

The key issue is the specification of the circumstances under which the SRRA would be able to impose the SRR on a bank. What will be the threshold conditions or triggers (quantitative or qualitative) that would cause the SSRA to compel a bank to enter the SRR? If the threshold is set too low, competition is distorted. If the threshold is set too high, there may be risk of systemic instability. Of course, with adequate deposit insurance and an appropriate bank insolvency procedure, contagion effects and other systemically destabilising manifestations of panic ought not to happen. Even the failure of a large bank should not be of greater public interest than the failure of a ball-bearings manufacturer in Coventry with equal value added.

The Treasury believes the decision on whether and when a bank should be ordered into the SRR should be based on a regulatory judgment exercised by the FSA after consultation with the Bank of England and the Treasury. Provided it is clear that the ultimate decision lies with the FSA, I would agree with this proposal.

Deposit insurance

I believe that the new deposit insurance arrangements should be located in the same institution that has the SRRA, that is, with the FSA. The existing Financial Services Compensation Scheme should either be moved into the FSA or wound up. In its current form, it is useless.

As regards the limits of the insured amount, the current UK figure of £35,000 (since October 1, 2007, the idiotic run-inducing 10% deductible after the first £2000 has been abolished) appears to be in the middle of the 19-country pack reported in the Treasury document. Eyeballing the charts, it looks as though about 97% of all retail deposit accounts hold less than £35,000. At the same time, the top 3% of deposit accounts hold about 50% of total deposits in the UK. This means that an increase in the limit would raise the value of the deposits covered by significantly more than it would raise the number of depositors covered. I cannot see a strong case for raising the limit, and no case for raising it above £50,000. What matters is the speed with which insured deposits can be paid out should a bank get into trouble.

Strengthening the Bank of England

It is apparent that the Bank of England, since it became operationally independent for monetary policy, and lost banking supervision in 1997, has done a much better job as regards its monetary policy mandate of price stability than it has as regards its financial stability mandate. There has been really only one serious test of the UK’s Tripartite Arrangement for financial stability between the Bank, the FSA and the Treasury. It failed the test. Much of the blame lies with current and past Treasuries and with the FSA, but the Bank contributed to the problems through its mismanagement of market liquidity. The Treasury Report does not address this issue at all.

It is key that the Bank of England should follow the example of the ECB and extend its list of eligible collateral at the standing lending facility and in open market operations to include routinely private securities, including asset-backed securities. It should also extend the maturity of its standing lending facility loans from overnight to up to one month, taking a leaf from the Fed this time. Finally, it should extend the list of eligible counterparties at the standing lending facility and in its repo operations to include not just banks and similar deposit-taking institutions. Currently, open market operations are open to non-Cash Ratio Deposit-paying banks, building societies and securities dealers that are active intermediaries in the sterling markets. Access to the standing facilities is restricted to participants in the Bank of England’s Reserves scheme and a few others. Both OMOs and standing facilities should be accessible to all financial institutions regulated in a manner approved of by the Bank.

While in a first-best world, the Bank would not be the active player in LoLR operations, it will always be involved in funding liquidity matters through its standing facilities. It is therefore key that the use of the standard lending facility be de-stigmatised. This can be achieved by abolishing the unbelievably complex operational procedures for setting the Official Policy Rate or Bank Rate (official policy sets the target for the overnight unsecured sterling interbank rate) and managing short-term liquidity.

The current framework has three main elements: rather plain-vanilla standing facilities and OMOs and a mysterious and pointless reserves-averaging scheme (from the Bank’s Redbook):

Reserves-averaging scheme. UK banks and building societies that are members of the scheme undertake to hold target balances (reserves) at the Bank on average over maintenance periods running from one MPC decision date until the next. If a member’s average balance is within a range around their target, the balance is remunerated at the official Bank Rate.’

The reserves-averaging scheme should go. There should be no reserve requirement at all. The Bank should stand ready to repo (against eligible collateral) or reverse repo any amount at any time at the Official Policy Rate. That, after all, is what it means to set the Official Policy Rate. Anything else is an attempt to set both price and quantity – and is doomed to failure.

Commercial banks would therefore be borrowing from the Bank of England all the time, as a matter of routine, and no stigma would be attached to such operations. This would also keep the overnight interbank rate closer to the Official Policy Rate than it is under current procedures, decoupling the Monetary Policy Committee’s interest rate decision from the liquidity policy not managed by the MPC but by the Bank’s Executive.The Bank still could retain its standing lending facility by accepting a wider range of assets as collateral at the standing lending facility than it accepts in repos to peg the Official Policy Rate.

In its open market operations, the Bank should act as market maker of last resort, by standing ready to purchase, at a properly conservative/punitive price, normally liquid assets that have become illiquid through a systemic flight to quality and liquidity caused by fear, panic and other contagion effects. As for the securities acceptable for rediscounting at the standing lending facility, there should be a positive list of securities (including private securities and indeed private ABS) that are acceptable as collateral by the Bank. This would help concentrate the minds of (the supervisors of) those maniacal financial engineers generating ever more complex and opaque financial structures, which would be unlikely to figure on the list of eligigble collateral.

What becomes of the Tripartite Arrangement?

It is obvious that, whenever taxpayers’ money is put at risk, the Treasury must be consulted and should have a veto over the operation. The Treasury document makes this clear. The Treasury is also ‘in charge of’ the whole arrangement, although it appears obvious that there are certain things it cannot instruct the two other parties to do without risking damaging resignations. I doubt whether it could give the Bank instructions on its collateral policy, OMOs and standing facilities operations. In my view it ought not to be able to do so. It is also unclear as to whether the Treasury expects to be in a position to instruct the SRRA (that is, the FSA) to invoke or not to invoke the SRR for a particular bank. I would hope it would not be able to do so. What the role of the Treasury would be in the decision to invoke the new bank insolvency procedure remains unclear. Obviously, nationalisation could only be authorised by the Treasury.

In the proposals of the Treasury, the FSA continues to be the regulator and supervisor of the banking sector (and of most other financial institutions). It remains responsible for the default risk (solvency), the funding liquidity of the institutions it supervises and other risks, including operational and reputational risk. It will lead the SRR and act as the SRRA. I assume it would also be responsible for the management of the deposit insurance scheme, although the Treasury document is not clear on this. The Bank of England does get its nose into the tent for most of these activities and responsibilities, however. To my mind this further troubles the allocation of responsibility and authority.

The financial cost of the deposit insurance scheme can only be borne by the participating institutions (either through pre-funding or ex-post funding) if the banking sector trouble causing the scheme to be called upon for a pay-out is a ‘local’ problem affecting only a minority of the banks. When there is a systemic bank run (or bank default), only the Treasury can credibly meet the insurance claims. This should be recognised. Any serious deposit insurance scheme represents a contingent claim on the Treasury.

The Bank of England remains responsible for market liquidity, both in normal times and, under disorderly market conditions, by acting as market maker of last resort. It is involved in funding liquidity through the (on demand against the proper collateral) standing lending facility. The Treasury Report (and even more strongly the Treasury Committee Report) favours an enhanced role of the Bank of England in the LoLR process. The Treasury Report wants the Bank to spend time and resources becoming and remaining informed of the liquidity situations of the individual UK banks. This clearly would also require it to be aware of the solvency-related aspects of the balance sheet and operations of individual banks. The Bank and the FSA would effectively become joint supervisors with shared responsibility for funding liquidity and solvency. I doubt whether such an arrangement would work well.

As far as I can tell, the Treasury Committee wants all of banking supervision and regulation to be returned to the Bank of England, with the FSA taken completely out of the game. A new Deputy Governor and Head of Financial Stability would take the lead in all financial stability matters, and could even order the FSA around.

It is clear that the Treasury Committee’s proposal would put strains on the Bank of England’s independence in monetary policy. The Committee therefore raises the possibility that the new Deputy Governor/Financial Stability Czar might not be a member of the MPC. I still cannot see it. What would be the authority relationship between the new Deputy Governor/Financial Stability Czar and his/her notional boss, the Governor? If the Bank of England is to be put in charge of (the operational end of) Financial Stability, better not to appoint a new Deputy Governor but to give the job to the Governor and to take MPC out of the Bank of England. The Governor of the Bank would, under this model, not necessarily be the Chair of the MPC or even a member of it.

A different solution

Rather than putting money and individual bank-specific information together in the same institution by making the Bank of England responsible for banking supervision again, I would move in the opposite direction. The lender of last resort (which would not be the Bank of England although the lender of last resort, if it is not the Bank of England should have an open-ended uncapped credit line or overdraft facility with the Bank of England, guaranteed by the Treasury), should be the SRRA, that is, the FSA. It would make liquidity available to a troubled bank that could no longer fund itself in the interbank markets, the repo markets or at the standing lending facility. The collateral that would be accepted, the terms on which it would be accepted, and the other terms and conditions attached to LoLR funds would be decided by the SRRA (the FSA) on a case-by-case basis.

The current Tripartite arrangement is sketched in Figure 1. The Treasury Committee’s proposal is in Figure 2, the Treasury’s proposal in Figure 3 and my own proposal (for a minimalist central bank) in Figure 4. Finally, Figure 5 shows how, under my proposed arrangement, a potentially troubled bank would be handled.

With effective deposit insurance and a sensible insolvency regime for banks, all proposals share the feature that it could, at last, become conceivable that a non-trivially small bank in the UK might fail. That would be the best guarantor of greater future financial stability.

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

a

A

Topics:  Financial markets

Tags:  financial stability, Northern Rock, bank failure, deposit insurance, Bank of England

Special Economic Adviser of Citigroup and CEPR Research Fellow

Events

CEPR Policy Research