What the Chancellor did next

Corporate Insolvency 101

Corporate bankruptcy or insolvency (I will use the two interchangeably – pace the lawyers) occurs when a borrower cannot meet his contractual debt obligations. Without a legal and regulatory framework for bankruptcy law, a creditor has two main legal remedies in the north Atlantic area: when the loan is secured, or the debt collateralised, the creditor can seize the assets that serve as collateral. With unsecured debt, a creditor can request the courts to sell some of the debtor’s assets.

Much of what follows is cribbed from the work of Oliver Hart, now at Harvard, who together with John Hardman Moore, of the LSE and the University of Edinburgh, has done much to rescue thinking about and designing insolvency regimes from the lawyers by focusing on efficiency issues See e.g. Oliver D. Hart “Different approaches to bankruptcy”, in Governance, Equity and Global Markets, Proceedings of the Annual Bank Conference on Development Economics in Europe, June 21-23, 1999 (Paris: La Documentation Française, 2000), also available as National Bureau of Economic Research Working Paper No. W7921, September 2000).

There is a need for legal/regulatory insolvency regimes because the alternative, having complete debt contracts that specify exactly what is to be done in every conceivable and inconceivable contingency is not feasible. With inherently incomplete debt contracts, when there are many creditors who cannot coordinate their actions perfectly, and the debtor’s liabilities exceed the value of his assets, there will be a ‘run on the debtor’; individually rational attempts by creditors to recover their debts as swiftly as possible may lead to a fire sale of the firm’s assets, and possibly to a premature or unnecessary break-up of the firm as a going concern, with a loss of value for all creditors and shareholders. It is therefore in the collective interest of creditors to realise the debtor’s assets in an orderly manner, via a bankruptcy procedure.

Economic approaches to bankruptcy stress three key desirable properties of bankruptcy regimes.

A good insolvency regime/bankruptcy procedure has three properties, that may have to be traded off for each other

(1) it should deliver an ex post efficient outcome, that is, it should maximize the total value of the resources to be distributed between the creditors, the debtor and other legal stakeholders, like the tax authorities and the workers.

(2) It should preserve ex-ante efficiency, that is should preserve the ‘bonding role of debt’ by penalising managers and shareholders adequately when bankruptcy occurs. It should preserve the incentives for servicing debt in the future.

(3) It should preserve the absolute priority of claims (secured creditors first, then senior unsecured and junior unsecured, shareholders last). The only qualification is that some positive value may have to be reserved for the shareholders. Without this, a management that serves the shareholders’ interest be incentivised to ‘go for broke’ – to avoid bankruptcy at all cost.

Why banks need a special insolvency regime

What has been said so far holds true for a ball bearings manufacturing company in Coventry and for a bank. Is there anything special about banks that would require them to have a special insolvency regime? Insolvency regimes for non-financial corporations are designed to achieve an orderly debt workout, a restructuring of the assets and liabilities of the firm. It does this by protecting the firm (its shareholders, other stakeholders and management) against uncoordinated creditor actions that would violate Property 1 – maximising the size of the pie; it allows the firm to get on with its main business. When credit is the main business of the firm, however, as it is for banks, protection from creditors puts a bit of a dampener on the firm’s ability to get on with its business.

The UK has no special insolvency regime for banks. This, as was pointed out by the Governor of the Bank of England, Mervyn King during his first statement to the House of Commons Treasury Committee, just after Northern Rock hit the wall, was one of the reasons the Tripartite Arrangement, involving the Treasury, the FSA and the Bank of England, for dealing with financial crises could not take effective preventive action at an early stage of the crisis – for instance at the time Northern Rock first requested special financial assistance from the Bank.

A normal feature of any bankruptcy procedure is something akin to a stand-still. The debtor is protected by temporarily ‘freezing’ the claims of the creditors, e.g. stopping them from exercising pre-payment options, or requiring them to temporarily extend (‘roll over’) loans that mature. Depositors are one class of (unsecured) creditors. In the UK, without a bank-specific insolvency regime, this means that as soon as an administrator is appointed for a bank that enters an insolvency procedure, all its deposits, including retail deposits, are frozen. Since up to that moment deposits are withdrawable on demand, on a first-come-first served basis, depositors do a runner as soon as there is any risk of a bank administrator being appointed. This arrangement, in the words of Mervyn King, invites a bank run.

Thus there is a case for a special insolvency regime for banks (and indeed for other financial institutions with (normally) highly liquid, short-term liabilities, that does not rely on the assumption that banks are uniquely important. When there is financial stability, banks are no more valuable or important than a ball bearings manufacturer with the same value added. They are just different. However, when banks collapse in a disorderly manner, they can, because of their strategic location in the centre of the spider’s web of financial intermediation, do more damage than is indicated by the loss of their value added. Their collapse can trigger further bankruptcies elsewhere in the economy, both in the real and in the financial sectors. Default and bankruptcy are not just a Coasian reassignment of property rights. They are processes that destroy real resources. In the presence of such real resource costs of default and bankruptcy, the pecuniary externalities that are the immediate manifestation of bank default give rise to real efficiency losses. This too provides a reason for taking bankruptcy arrangements for banks seriously.

The Chancellor’s FT Interview of January 3, 2008

The Chancellor has now announced he will create a special insolvency regime for banks, with primary legislation to that effect being introduced in May 2008. It is, in principle, a good idea – an idea in fact, whose time did come quite some time ago already. This legislation will be part of a wider set of legal and regulatory changes involving the Financial Services Authority, the Bank of England and the Treasury, to deal with banks that are in financial trouble, even if no formal insolvency procedures ever get initiated.

Mr. Darling was not very specific about the number of key aspects of the legislation and regulatory changes he proposes, but the following is clear.

A. The Tripartite Arrangement will be able to intervene earlier and more effectively than under current law when an individual bank gets into trouble.

B. The regulator (the FSA) will get the powers wielded in the US by the Federal Deposit Insurance Corporation, to appoint someone (an administrator, say) who has the power to restructure a bank that has got into trouble. This administrator would, to all intents and purposes control and run the bank. The Chancellor, referring to the FDIC, called this “… effectively takes over … it can run a bank or it can order it to make changes and so on”. Existing management, if it stayed around at all, would be in a subordinate position to the FSA-appointed administrator, and the shareholders would have no decision-making power. The administrator can ensure “…that the bit that has got in trouble can be separated from the other bits that are perfectly healthy”. This presumably means that the administrator can sell of some, or even all of the bank’s assets. Yo could even sell the bank as a going concern to new owners (possibly via the mechanism discussed in point B).

C. Given the wide powers of the FSA-appointed administrator, there really is no further need for formal nationalisation of banks in trouble. The administrator can, presumably, order the bank to issue additional equity to be bought by the state, if, for some reason, formal state ownership were to be required.

D. The administrator will be able to “seize and protect” the deposits of a bank in trouble. The administrator or the FSA will be able to decide to transfer the retail deposits of the troubled bank to another institution.

E. The administrator’s mandate will presumably be different from and wider than that outlined in points (1), (2) and (3) for a ‘conventional’ insolvency procedure, because systemic financial stability considerations would have to play a role. The ‘stakeholders’ include all those affected by systemic financial stability, both inside and outside the financial sector.

F. To achieve (A), the trigger or triggers prompting an intervention by the FSA will be very specific and clear. In the Chancellor’s words “…you can’t have a situation, …, where you turn up at the doors of a perfectly healthy bank and say look we are going to reorganise it. The trigger points that I have in mind is that once a bank is in financial difficulties, once a bank is using lender of last resort facilities, people know what the rules of the game are at that point.”

Note that the lender of last resort (LOLR) facilities referred to in the previous paragraph are not the use of the discount window (the overnight standing (collateralised) lending facility, which is open to all eligible institutions, on demand, against eligible collateral, at a rate 100 basis points about the official policy rate). It refers to special emergency facilities like the Liquidity Support Facility created for Northern Rock.

It is also important that the Chancellor referred to using LOLR facilities as an obvious trigger, not requesting or discussing the use of LOLR facilities. That is important, because if the mere opening of discussions between a troubled bank and the FSA were to be a sufficient reason for triggering a de facto transfer of control of the bank to an FSA-appointed administrator, banks would avoid approaching the FSA like the plague. This could mean that opportunities for exiting the troubled state without having the FSA take charge might be wasted.

G. Deposit protection levels for retail depositors will be raised and will not be subject to a ‘deductible’ up to the new ceiling. My guess is that the new limit will not be much higher than £50.000. This would be roughly the same level of cover as that provided in the US and would, if it were implemented today, cover about 98% of UK retail depositors (although no doubt as smaller percentage of UK retail deposits).

H. The pay-out of the protected deposits will be much faster than it is under current arrangements (which apparently can be up to six months). At most two working days would be a reasonable target.

I. The FSA will be explicitly instructed, in a rewritten Memorandum of Understanding for the Tripartite Arrangement, to monitor the liquidity of the banks it supervisors with the kind of care and attention it has hitherto reserved for the capital position of the banks.

J. The Bank of England will also be given an explicit mandate to monitor the liquidity positions of individual banks. Because of (I) and (J), it will be necessary to clarify “… the line at which the FSA supervisory interest stops and Bank’s interest in the system in general (begins)…..needs to be clarified …”. One thing is clear: just as you cannot write a complete contingent contract, you cannot write a complete contingent MOU. The Bank and the FSA will have to be in constant close touch to minimize the risk of misunderstanding and of things falling into the cracks between the two institutions.

K. The Chancellor will be in charge of the Tripartite Arrangement. Well, he always was. The references to COBRA are a bit of macho spin. Even for the semi-botched rescue of Northern Rock, it was always clear that no public money could be put as risk without the explicit permission of the Chancellor. This does, however, raise some issues that need further clarification.

Remaining Open Questions:

OK, so the Chancellor is in charge of the Tripartite Arrangement for dealing with financial stability issues. What does this mean?

A. The Treasury and the Bank of England

Ai. Does this mean the Chancellor can instruct the Bank of England to change its official policy rate, if he deems this to be necessary for financial stability? I assume that the Chancellor could only do this by invoking the Treasury Reserve Powers, which permit the rate setting power of the MPC to be temporarily repatriated by the Treasury.
Aii. Does this mean the Chancellor can give the Bank of England binding instructions about its collateral policy, both at the standing lending facility and in its repo operations? I assume not.
Aiii. Does this mean that the Chancellor can give the Bank of England binding instructions about any other aspect of its discount window operations or of its open market operations?
Aiv. Can the Chancellor instruct the Bank of England to make loans or provide financial resources through any other means or mechanism to any individual bank or banks? Can the Chancellor do so if the Treasury guarantees the loans or other resources made available by the Bank of England to individual banks on the instructions of the Chancellor?

B. The Treasury and the FSA

Bi. Assume the trigger or triggers have been written down. Who pulls the trigger, the Chairman and/or the CEO of the FSA or the Chancellor? This is especially important if, as I suspect, the trigger(s) will be principles-based rather than rules-based. The only trigger explicitly discussed is a troubled bank using LOLR facilities. Presumably, since the FSA has no budgetary resources of its own, the decision to let a bank use LOLR facilities will have required the authorisation of the Treasury, so there is no potential for conflict here. If there are other triggers, it must be absolutely clear where the decision to pull the trigger is taken. I assume this will always be the Treasury, on the advice of the FSA and the Bank, but without either the Bank or the FSA (or the two jointly) having any power of veto.
Bii. Does the FSA appoint the administrator, or is there a Treasury veto?
Biii. Does the Treasury approve only the general decision to provide an LOLR facility to a bank, or does it get involved in the details of size, conditionality, terms etc?
Biv. If LOLR funds are provided to an individual bank, does this money come through the balance sheet of the FSA? Will it instead be routed through the Bank of England (as with Northern Rock), or will it come straight from the Treasury. If it is the Bank of England that provides the money, does it have a veto? In my view the only clear solution is an uncapped and open-ended overdraft facility (credit line) of the FSA with the Bank of England, guaranteed by the Treasury.
Bv. Where does the FSA get the financial resources to ‘seize and protect’ the deposits of a troubled bank?
Bvi. Will the shareholders of the troubled bank retain any decision making power once an administrator has been appointed? Can the administrator sell the business as a going concern?
Bvii. Will there remain any legal recourse for shareholders and creditors against the actions taken by the administrator?
Bviii. Which institutions are covered by the new FSA mandate? Only commercial banks and other licensed deposit-taking institutions or other financial institutions as well?


While the devil in financial stability arrangements is always in the fine detail that determines the actual structure of the incentives faced by all the parties involved, I would have to say: so far so good.

The division of labour between the Bank and the FSA is clarified. The Bank is responsible for systemic stability and is concerned about systemic liquidity, that is, the liquidity of key financial markets and instruments. It should know the liquidity positions of key individual banks and financial institutions, but it is not responsible for LORL operations for individual troubled banks. At most it will act as a conduit for financial support authorized by the Treasury and administered, monitored and supervised by the FSA. The Bank can now play a largely passive role in the individual LOLR operations, leaving the active decisions to the Treasury and the FSA.

The best part of the solution is that, with the threat of a run by retail depositors removed, we are at last in the UK getting to the point where it will be possible for the authorities to allow a bank to fail when that bank deserves to fail. The only systemic risk left is that a fire-sale of an insolvent bank or of its assets could unduly depress the value of illiquid assets and threaten systemic instability. However, provided the Bank of England does its job and keeps key financial markets liquid, even a fire-sale of illiquid assets should not have material systemic consequences. Under these conditions, not only small banks like Northern Rock ought to be unable to extort financial support from the state, even the largest banks are no longer too big to fail. That will be the true solution to some of the moral hazard problems we find in the banking system. It won’t solve all incentive problems, however. The US has for decades had the kind of insolvency regime for banks that the Chancellor now proposes. Yet in that same US, the Chairman and CEO of a not-so-small bank that under his leadership sailed far too close to the edge of the precipice, is sacked with a golden parachute of $100 million or so. Only a radical overhaul of corporate governance can remedy the profound distortion of incentives that produces such obscene outcomes.

Once the systemic issues have been addressed, the role of the FSA-appointed administrator is in fact the same as that of a regular administrator in a corporate insolvency. You will have to strike the best possible balance between the three objectives listed at the beginning of this blog: maximise the size of the cake; make sure management and shareholders pay the price for failure; and assure the absolute priority of claims of all stakeholders, including the state, both as tax collector and as lender.

The main risk I see is that there will be an excessive emphasis on the private provision of liquidity through socially inefficiently high private holdings of inherently liquid assets. Banks are supposed to borrow short, in markets that are normally liquid, and to lend long and invest in illiquid assets. If they are required to hold inherently liquid assets (like Treasury bonds) in amounts sufficient to tide them over even when their normally liquid sources of credit have, exceptionally, become illiquid, there will be too little term and liquidity transformation. It is the job of the Bank of England to act as market maker of last resort when key financial markets for normally liquid financial instruments dry up because of a systemic loss of confidence and trust. Liquidity is a pubic good whose private provision is possible but socially inefficient. The role model here is not the US, but the ECB, which no doubt would be happy to provide the Bank of England with a copy of its handbook on liquidity management in the interbank markets.

Maverecon: Willem Buiter

Willem Buiter's blog ran until December 2009. This blog is no longer active but it remains open as an archive.

Professor of European Political Economy, London School of Economics and Political Science; former chief economist of the EBRD, former external member of the MPC; adviser to international organisations, governments, central banks and private financial institutions.

Willem Buiter's website