## Mervyn King: Banking and the Bank of England
Speech by Mr Mervyn King, Governor of the Bank of England, to the British Bankers Association, London, Tuesday 10 June 2008.
* * *
After a decade or more of economic stability, we are now facing a period of rising inflation and falling economic growth. Part of the reason for this change of economic weather is that we are passing through the most prolonged period of financial turmoil that most of us can remember. Whether, as the IMF has argued, it is the worst period of financial stress since the 1930s is too early to judge. After all, the crisis is not yet over.
Since last August there has been a great deal of soul-searching about what we all might have done differently, both before and after 9 August when the crisis hit. And in circumstances where livelihoods are at risk it was inevitable that the 'blame game' was played. Unfortunately, although such a game helped to fill newspaper columns, it generated more heat than light. So today I want to step back.
Financial crises have been a regular, and disturbing, feature of our and other developed economies for as long any one can remember. They are as frequent now as in the nineteenth century, when Henry Thornton and Walter Bagehot were developing their ideas about how central banks could prevent instability in the banking system. Are these repeated crises the inevitable result of human nature at work in financial markets, with greed and fear alternating as sentiment swings from irrational optimism to irrational pessimism? Or is it the failure over many years to find the right incentive structure for banking markets? Or is it a mixture of both?
I don't pretend to be able to offer a complete answer to those questions. But I do want to focus on the relationship between the Bank of England and the banks because it helps to explain why crises occur and what we can do about them.
Let me start by saying that, although I do think the role of the Bank needs to be enhanced, it would not be right for the Bank to take back banking supervision from the FSA, as some have suggested. We should remember that the period before 1997, when FSA and the 'tripartite arrangements' were set up, was not seen at the time as a golden age of official oversight of banks. Before 1997 the relationship between the Treasury and the Bank of England in respect of banking supervision was often fraught. Remember too that the experience which shaped the design of the 1997 settlement was that of idiosyncratic problems in Johnson Matthey, BCCI and Barings, not a system-wide crisis affecting the liquidity of the banking system as a whole. As a result, the tripartite Memorandum of Understanding which underpinned that settlement was not so much faulty as incomplete. Since the onset of the crisis last summer, the way in which the tripartite arrangements work in practice has evolved considerably.
The concept of a golden age in the relationship between the Bank and the banks is also far from the truth. Our relationship was described by David Kynaston, in his magisterial history of the City, as 'surprisingly prickly'. In the 1920s the Chairman of the Midland Bank, Reginald McKenna, derided the Governor, Montagu Norman, as 'an intellectual without an intellect', and in turn Norman said about McKenna, 'some people cannot avoid the limelight'. Since the crisis broke last August we too have had our differences, although more recently the close cooperation between the Bank of England and six of the largest banks in working through the details of our Special Liquidity Scheme was a model of effective collaboration.
But perhaps we should not be surprised that sometimes our relationship is 'prickly' because our interests do not always coincide. Banks are, entirely legitimately, concerned primarily with their commercial interest. The Bank of England is concerned with the interests of the UK economy as a whole. The two sets of interests are not always the same.
BIS Review 75/2008
Where they may clash is over the use of our balance sheet. Companies which have misjudged risks or their business model should generally be allowed to fail to encourage prudent behaviour by others. But the failure of a bank can have an impact that goes well beyond the importance of that bank alone - so-called systemic risk. That is why central banks have sometimes acted as 'lender of last resort'. It can then be difficult for the authorities to make a credible commitment that in future banks will be allowed to fail. And that can in turn encourage banks to take greater risks - both in maturity transformation and lending - enabling them to earn higher profits. Because the authorities will face an incentive to step in to rescue banks tomorrow, there is an incentive for banks to take more risk today.
There is a parallel here with monetary policy. In their decisions today, people take into account how interest rates will be set tomorrow, just as banks when taking risks anticipate how central banks are likely to behave. In monetary policy we now have in place a framework which makes the commitment to low inflation credible by delegating the responsibility for achieving it to an independent central bank.
We need now to develop an equally strong framework for financial stability. It will be more complex than for monetary policy where there is a single instrument - Bank Rate - which can be delegated to a single decision-making body - the Monetary Policy Committee. Financial stability requires cooperation among at least three sets of decisionmakers: the supervisor, the central bank and the finance ministry. The existing tripartite framework needs to develop so that the authorities are able to reduce systemic risk not just by effective management of crises but by creating incentives for financial institutions to avoid excessive risks.
Some of the elements of such a framework I described in my evidence to the Treasury Committee on 20 September last year. In line with international experience, they included a special resolution regime for failing banks, changes to deposit insurance to reduce the incentives for bank runs, and a proper regime for the regulation of bank liquidity.
At the heart of the case for a special resolution regime is the need to find a way to allow banks to fail in an orderly manner. As the Treasury Committee said in its report The Run on the Rock , 'Banks should be allowed to 'fail' so as to preserve market discipline on financial institutions'. It is clear from the responses to the Consultation Document that there is widespread acceptance of the basic case for a resolution regime. But that is tempered by concerns about the details of its design. I want to highlight three key features that will be central to the success of the new regime.
First, responsibility for resolving a failing bank should be delegated to a resolution authority which should have a wide range of instruments available to help resolve a failing bank. Which body assumes the role of the resolution authority is less important than the granting of those powers to some body.
Second, the circumstances in which the regime would be triggered should be spelt out as clearly as possible. It is unrealistic to imagine that entry to the regime could be triggered by reference to quantitative triggers alone. Some discretion will be necessary. But the triggers should contain some quantifiable measures for both capital and liquidity, and there should be transparency in advance about their formulation to constrain the authorities' discretion. Moreover, looking ahead, although the regulator should have the leading role in triggering the regime, no regulator, however effectively managed, can ever fully insulate itself against the risk of forbearance which might delay a necessary resolution. That is why the Bank believes that the resolution authority, wherever it is located, should have a powerful voice in decisions on the use of the trigger. Moreover, any resolution authority will have a natural interest in the timing of a decision to place a bank under its control.
Third, a clear framework for accountability should be established to give confidence that decisions relating to the resolution regime are exercised in line with the objectives for the regime set out in legislation.
BIS Review 75/2008
Turning to deposit insurance, I understand the concerns that many of you have voiced about the reforms discussed in the tripartite Consultation Document. I know that you accept that to avoid bank runs there will need to be 100% insurance of deposits - up to a limit. That limit should not be too high. And in order to limit risk-taking by banks there is also a strong case for relating the payments made by different banks to finance the insurance scheme to the risks that they will fail. And in that context, some element of pre-funding is natural and is used in many countries, including those with banking systems similar to our own, such as Canada and Sweden. A degree of pre-funding is one of those ideas that is bound to be unpopular before the fund is called upon, but seems decidedly wise after the event as it lessens the burden on the banking system in a time of stress.
On the regulation of bank liquidity, the FSA has already issued a discussion paper, and I support strongly the case for reform of liquidity regulation made this morning by Sir Callum McCarthy. I very much hope that an international initiative to include liquidity in the Basel capital regime, in which the Bank is playing a leading role, will come to fruition. For our own part we, along with other central banks, have been reviewing the way in which our market operations can alleviate stress in the banking system.
The primary objective of our money market operations is to ensure that the interest rate on overnight lending in the market is close to the Bank Rate set by the Monetary Policy Committee. That constrains the net amount of central bank money we lend to the system. But we are also concerned with illiquidity in particular markets when it threatens the stability of the system as a whole. It is this concern that motivates the choice of assets we lend against in both normal and stressed conditions.
A loss of liquidity in markets on which the banking system as a whole relies can magnify the mismatch between the maturity of assets and liabilities on banks' balance sheets. A banking system in which large volumes of funding need to be renewed very frequently is a fragile one, vulnerable each day to a shock stemming from rumours and loss of confidence. And it is in those circumstances that a backstop is needed. The reason for it is not to support individual institutions. It is to avoid an unnecessary loss of confidence in the system as a whole. That backstop is central bank funding when private markets important to the stability of the system have temporarily closed.
But central banks are not there to support liquidity in all markets at all times. They are not there to create a liquid market in all financial instruments that traders or some rocket scientist have dreamt up. Nor are they there to maintain indefinitely a market when changes in information or risk appetites make it more difficult to sustain a private market, as in some asset-backed security markets recently.
It was concerns about confidence in the banking system as a whole that explain why the major central banks acted together in December, which in our case meant extending the range of collateral accepted in our regular auctions of 3-month lending. Auctions like this can play a useful role but, as we saw with Bear Stearns, a more continuously available facility may sometimes be needed to avoid an immediate loss of confidence. That is why, in April, when confidence was particularly fragile, we introduced a backstop that is available every day - the Special Liquidity Scheme.
As I said in September, in providing such a backstop, we face a balancing act, between avoiding a major shock to the system and encouraging future reliance on the same cheap but risky funding sources. The Special Liquidity Scheme aims to do this by exchanging liquid for illiquid assets that existed only before the end of last year.
We intend to learn from the experience of the Scheme to put in place a liquidity facility that works in all seasons - both 'normal' and 'stressed'. It will be part of a set of reforms to our Red Book, to be announced later this year. Any such facility will need to meet two challenges: it will need the right pricing structure and it will need to overcome the 'stigma' problem that has affected access to all central banks during the current crisis.
The pricing structure will be key to striking the right balance. It should vary according to the type of collateral which banks provide, as indeed it does to some extent now. Access to central bank money or asset swaps should, as far as possible, encourage banks to manage liquidity risk prudently. Without the correct pricing structure, the incentives to encourage future risk taking are obvious and potentially large.
Valuation margins - or haircuts - are used to protect our balance sheet. By themselves, they aren't sufficient to create a general pricing structure. The cost imposed by a margin depends primarily on the difference between the costs of secured and unsecured funding, and that will vary across banks and over time. So the key is in the interest rates charged by the central bank.
The second challenge is to overcome the 'stigma' problem that has plagued the use of central bank facilities throughout the crisis. Commercial banks have feared that, by accessing funding from the central bank, they would acquire a 'black spot' in the market. They have been reluctant to come to their central bank unless in good and extensive company.
These are challenges that all major central banks are grappling with. They will not be easy to overcome. But with our international colleagues we are aiming to devise a framework that does so. The prize will be an integrated framework, covering all our lending to the banking system and one that I hope, by making clear the terms on which liquidity will be provided in times of stress, will focus the minds of banks long before those stresses emerge.
During my second term as Governor I would like to see us establish a framework for financial stability that is on as sound a footing as the one we have successfully established for monetary stability. The measures I have described would go a considerable way to achieving that. But just as no framework for monetary policy can prevent shocks from hitting the UK economy, we need to think carefully about why financial crises have been both persistent and damaging.
That leads me to one final thought on which I want to end. For reasons of historical accident we have created over many decades a financial system in which the incentives to monitor risk-taking have been sharply reduced. It is often said that the role of a central bank is to take the punch bowl away just as the party is getting going. That approach has served us well in monetary policy. But all those efforts will come to naught if the opposite applies to the financial sector. If banks feel they must keep on dancing while the music is playing and that at the end of the party the central bank will make sure everyone gets home safely, then over time the parties will become wilder and wilder.
That might not matter were the consequences limited to the party-goers. But they are not. When the party ends, some innocent bystanders may lose their homes altogether. Moreover, the party-goers aren't just deposit-taking banks. A wide range of financial institutions, including investment banks, monoline insurers, even hedge funds, have the potential to cause significant damage to the rest of the economy in the wake of their demise. Are they all to be helped to get home safely when the party ends? If so, will they all be regulated in the same way as banks? If not, how can we limit the potential cost to the taxpayer? Add to this the problem that many of these institutions are global in scope, but the responsibility for both regulation and rescue remains firmly at national level, and you can see why policy-makers get headaches as bad as the party-goers'.
No individual bank, or indeed the authorities, can easily find a way out of this because their incentives are to continue in the same manner. There is, therefore, no point blaming anyone for the outcomes. But together we have to find an answer. Of course, excessively strict and detailed regulation would mean that there were no parties worth going to and the economy as a whole would suffer from a suppressed and inefficient financial sector. That is why requiring financial institutions to hold more capital to act as a shock absorber, while reducing the procyclical nature of existing capital requirements, may offer a balance between excessively burdensome regulation and excessive risk-taking.
BIS Review 75/2008
The proposals in the Banking Bill, and the other changes I have discussed today, will take us an important step forward in the creation of a framework for financial stability. But they will not be the final answer to a difficult, at times seemingly intractable, problem of how to change the incentives of both private and public actors to reduce the frequency and cost of financial crises. For that we will all need to do a lot more thinking. It is important that we do find the right solution. An efficient and thriving banking sector which can intermediate saving and investment, both domestically and internationally, is essential to our economic prosperity.