Thursday, September 13, 2007

Message from Mervyn

Yesterday, the argument for rate cuts to save the US from impending recessionary disaster of an unspecified nature was aired by it's most learned and esteemed advocate, Martin Feldstein, who repeated his narrowcast doom and gloom speech from the recent Jackson Hole conference in the mass market pages of the WSJ.

m* strongly disagreed with this view and held out the example of BOE's Mervyn King as a counterweight to the luminaries calling for easing. In typical fashion, King's voice during this period has thus far been conspicuous by it's absence, until today when King broke the silence with his first comments weeks into the current crisis in the money markets and structured debt markets. In a statement prepared for regularly scheduled testimony to the Treasury next week, King sends an eloquent, reasoned message to the markets about the hazards of short term, emotionally driven thinking and the responsibility of a central banker. The short of it is captured by headline-of-the-day award winner and frequent source Macro Man with "No Soup for You!"


TURMOIL IN FINANCIAL MARKETS: WHAT CAN CENTRAL BANKS DO?

Paper submitted to the Treasury Committee

Mervyn King, Governor of the Bank of England

The recent turmoil in financial markets has increased uncertainty in the world economy.

Problems that surfaced first in the US sub-prime mortgage market are now visible more
widely. All those involved, whether banks, other financial institutions, regulators or
central banks, need to analyse carefully the causes of the recent turmoil and think through
the long-term consequences of any actions if they are to respond appropriately.

So in this note I shall try to answer two questions. First, what are the immediate causes

of the recent turmoil and what are its implications? Second, what can and should central
banks do to alleviate the problems? In due course we shall all have time for a more
detailed examination of this episode and the lessons to be learned from it.

1. What are the immediate causes of the recent turmoil and what are its implications?


Since the beginning of August, there have been sharp movements in financial markets:

prices of loans, and assets backed by loans, have fallen; prices of government securities
have risen; and interest rates on inter-bank lending have risen.

Rising default rates on sub-prime mortgages in the United States were the trigger for the

recent financial market turmoil. It is important, though, to put recent events in
perspective. The world economy has been strong for the past five years. Our own
economy has been growing at a steady rate for a considerable period. There are major
problems in the US housing market to which the authorities there are responding with
both macro and micro measures. But the losses from defaults so far remain small relative
to the capital of the banking system.

None of this is meant to say we should be complacent. But the source of the problems

lies not in the state of the world economy, but in a mis-pricing of risk in the financial
system. And it is on that set of issues that we need to focus to determine the remedies,
both short-term to address the current problems and long-term to prevent a recurrence.

Why have developments in one part of the US mortgage market proved so important for a

wide range of financial markets? Sub-prime mortgages are one type of loan that banks
have parcelled together into securities backed by the cash flows from those loans – a
process known as securitisation. Those securities have been sold by banks to investors.
They have also been sold to investment vehicles, many of which have been established
by the banks themselves. Many of these vehicles have financed their purchases by
issuing short-term commercial paper.

Securitisation of loans has separated the information held by loan originators from those

exposed to the risk of default – investors in asset-backed securities or commercial paper.
The unexpected losses sustained on assets backed by US sub-prime mortgages have
highlighted the potential costs to investors of uncertainty about the types of loans
underlying the assets they purchase. So for the time being the markets in these
instruments have either closed or become very illiquid. Vehicles financed by short-term
commercial paper are holding assets which can no longer be traded in liquid markets.
They now find that they have borrowed short to lend long – normally thought of as a
function of banks.

As a result of this maturity mismatch, vehicles set up by banks and others are now

finding it extremely difficult to obtain funding through asset-backed commercial paper.
The markets are now withdrawing short-term funding from such vehicles, a process not
unlike a bank run. Many investment vehicles have been forced to shorten the maturity of
their commercial paper, making their borrowing even more short-term and their maturity
mismatch even greater. Other vehicles have been unable to issue at all. For example,
since the beginning of August the value of asset backed commercial paper outstanding in
the US has fallen by almost 20%.

Some investment vehicles will need to be wound up. In many cases, however, the

sponsoring bank will have written a backup line to the vehicle, guaranteeing its funding.
Many of the securitised loans may now be re-priced, restructured or taken back by the
banks. A process is starting that will expand the balance sheet of the banking system.
But how far that process will go is hard to tell.

The vehicles can be taken back onto banks’ balance sheets. Banks as a whole are well

capitalised and should be able to do this. Moreover, the funds that were directed to assetbacked securities and commercial paper will now be available elsewhere. In the end, that funding will come back to the banking system, although between banks the distribution
will differ. So the adjustment period may be awkward and, during it, banks are placing a
premium on holding assets which can quickly be turned into cash.

The increase in demand for liquid assets during the adjustment period is one reason why,

in all the major economies, yields on liquid assets like government securities have fallen.
It also helps to explain why the compensation needed for banks to lend to other banks
over periods longer than overnight has risen and why the volume of inter-bank lending
has been increasingly concentrated at shorter maturities. Since the beginning of August,
the spread between interest rates for 3-month inter-bank lending and central bank interest
rates expected over that period has risen in all the major economies. At present, the
average spread is 110 basis points in sterling and 90 basis points in dollars. This is the
natural economic result of a change in the preferences of banks over the composition of
the assets they wish to hold on their balance sheets.

In addition, banks have raised their demand for reserves at central banks. Banks settle

payments with each other using central bank money and they hold reserves at the Bank of
England to manage their daily payment needs. Conditions in financial markets have
made their payment needs less predictable. As a result, banks have wanted to hold more
reserves. They have tried to fund those reserves by borrowing overnight from other
banks. Over the past month, interest rates on secured overnight borrowing have averaged
5.91% – 16 basis points above Bank Rate. That spread was wider than usual – since the
introduction of the current money market regime, it has averaged 3 basis points.

These changes in the distribution of assets across the financial sector, and banks’

preferences over different assets during the adjustment period, are likely to have
consequences for the wider economy through the interest rates for borrowing and lending
faced by households and companies. It is too soon, however, to quantify the impact on
the economy as a whole.

In the short term, some corporate loan rates will rise in line with inter-bank rates. Banks

that are unable to sell pools of loans that they had securitised, or who need to support offbalance sheet vehicles, may cut back on new lending. But banks whose potential funding
liabilities to vehicles or conduits are small as a proportion of their balance sheet may be
able to exploit profitable lending opportunities, which may not be as open to those banks
which are now hoarding liquidity. So there may be a redirection of borrowing from
within the banking system. This is part of a normal market adjustment.

Funds that had been invested in asset-backed commercial paper issued by vehicles and

conduits will find their way back to the financial system, perhaps directly through bank
deposits or indirectly via the corporate sector by purchases of corporate debt. It is
notable that yields on investment-grade corporate bonds are unchanged since the
beginning of August. And companies, including some financial institutions, have been
able to issue long-term debt.

Nevertheless, there has since mid-July been a widespread reassessment of the

compensation investors seek for bearing risk. Equity prices have fallen in all major
economies. Most of that adjustment took place in July – before the turmoil in credit
markets. The FTSE All-Share today is 6% below its level at the beginning of July. As
this re-pricing of risk passes through to borrowers, the supply of credit faced by
households and companies may tighten somewhat.

In summary, the turmoil in financial markets since the beginning of August stems from a

reluctance by investors to purchase financial instruments backed by loans. Liquidity in
asset-backed markets has dried up and a process of re-intermediation has begun, in which
banks move some way back towards their traditional role taking deposits and lending
them. That process is likely to be temporary but it may not be smooth. During that
process, demand for liquidity by the banking system has increased, leading to a
substantial rise in inter-bank rates.

2. What can and should central banks do to alleviate these problems?


Three distinct policy instruments can be deployed by central banks: interest rates, money

market operations, and other general liquidity support operations.

First, what role should monetary policy play in the present situation? The answer is to

protect the public from the consequences of the recent turmoil by continuing to maintain
economic stability. That is done by setting interest rates in order to meet the 2% target
for inflation. Interest rates are a flexible tool and can be adjusted quickly when
necessary. If, in the wake of a shock to the financial system, the terms on which the
financial system extends credit to the private sector become less favourable, then
borrowing and overall demand would weaken. Other things being equal, that would
lower the inflation outlook. Of course, other things are not equal. When the Monetary
Policy Committee meets each month it reviews all the evidence on the outlook on
inflation before reaching a judgment. The August Inflation Report implied that some
slowdown from recent strong rates of economic growth was needed to meet the inflation
target. The new element introduced by the recent turmoil is that effective borrowing
rates facing households and companies will rise somewhat. So, as we said in the August
Report, the Committee is monitoring credit conditions intensively. It is too soon to tell
how persistent and how large any change in credit conditions for household and corporate
borrowers will prove to be. A new Bank of England Credit Conditions Survey will be
available to the MPC at its next meeting.

Second, the central bank is responsible for the smooth functioning of the payment system

among banks – the short-term money markets and what is known as the high value
payment system. Central banks discharge that responsibility by providing reserves that
enable banks to settle among themselves. In the reform of our money market operations
a year ago, we made very clear, and this is a unique feature of the British system, that the
banking system as a whole will get the reserves that it itself requests. Each month, at the
beginning of what is known as the maintenance period, running from one MPC meeting
to the next, banks set their own reserve targets. They are not imposed. We then supply
the reserves that the banking system as a whole requests. The objective is to allow banks
to deal with their own day-to-day liquidity needs and, by supplying in aggregate the
banks’ demand for reserves, to keep the overnight interest rate close to Bank Rate set by
the Monetary Policy Committee. If any individual bank has misjudged its reserves target
and finds that on any day, due to unusually large payment flows, it needs additional
liquidity, then that is supplied against eligible collateral at a penalty rate. There is
automatic and guaranteed access to the standing facility in return for eligible collateral
and a penalty rate of 1% above Bank Rate. It should be clear that because standing
facilities are available at the borrower’s discretion and against eligible collateral, they are
quite distinct from what is known in other financial centres as “emergency liquidity
assistance”, and under the UK tripartite framework as lender of last resort arrangements,
where the central bank decides that there is a policy objective in lending to one or more
institutions. Reflecting these different aims, the collateral required is different.

The interest rate for secured overnight borrowing was, in August, unusually high relative

to Bank Rate, indicating that banks’ aggregate demand for central bank reserves had risen
since they set their reserves targets. For the current maintenance period, which began on
6 September, the reserves banks raised their target levels of reserves by 6%. That larger
quantity of reserves was supplied by the Bank of England in its open market operation on
6 September.

As expected, some pressure on interest rates for overnight borrowing was relieved. Last

week, we announced that, during the current maintenance period, we will make available
to banks additional reserves, up to 25% of the reserves target, if the secured overnight
rate remains higher than usual relative to Bank Rate. The reason for this is that there are
grounds for suspecting that banks may, at the start of the current maintenance period,
have underestimated their demand for reserves, and the additional reserves will help to
bring the overnight rate into line with Bank Rate. We will announce the terms of this
week’s operation on Thursday. Provision of central bank reserves, in exchange for highquality
collateral, cannot be expected to narrow the spreads between anticipated policy
rates and the rates at which commercial banks can borrow from each other at longer
maturities, and has not done so elsewhere.

So, third, is there a case for the provision of additional central bank liquidity against a

wider range of collateral and over longer periods in order to reduce market interest rates
at longer maturities? This is the most difficult issue facing central banks at present and
requires a balancing act between two different considerations. On the one hand, the
provision of greater short-term liquidity against illiquid collateral might ease the process
of taking the assets of vehicles back onto bank balance sheets and so reduce term market
interest rates. But, on the other hand, the provision of such liquidity support undermines
the efficient pricing of risk by providing ex post insurance for risky behaviour. That
encourages excessive risk-taking, and sows the seeds of a future financial crisis. So
central banks cannot sensibly entertain such operations merely to restore the status quo
ante. Rather, there must be strong grounds for believing that the absence of ex post
insurance would lead to economic costs on a scale sufficient to ignore the moral hazard in
the future. In this event, such operations would seek to ensure that the financial system
continues to function effectively.

As we move along a difficult adjustment path there are three reasons for being careful

about where to tread. First, the hoarding of liquidity is a finite process. When any
transfers of the assets of vehicles back onto banks’ balance sheets are complete, the
demand for additional liquidity, and the associated rise in LIBOR spreads, will fall back.
The fragility of sentiment at present means that the system is vulnerable to further shocks
and it is important to monitor financial conditions extremely closely. But the banking
system as a whole is strong enough to withstand the impact of taking onto the balance
sheet the assets of conduits and other vehicles.

Second, the private sector will gradually re-establish valuations of most asset backed

securities, thus allowing liquidity in those markets to build up. Indeed, there are market
incentives to speed up the process both of taking assets back onto balance sheets and to
re-open markets in securities that have closed. Already there are tentative steps in this
direction which will allow the price discovery process to restart. Strong institutions have
incentives to reveal their positions to obtain better access to funding. Some are tapping
long-term paper. And there are opportunities to make money for those who can assess
and value instruments and eventually repackage and reissue them. Difficult and timeconsuming though that process may be, it will also slowly reduce that part of the rise in
market rates which reflects counterparty risk.

Third, the moral hazard inherent in the provision of ex post insurance to institutions that

have engaged in risky or reckless lending is no abstract concept. The risks of the
potential maturity transformation undertaken by off-balance sheet vehicles were not fully
priced. The increase in maturity transformation implied by a change in the effective
liquidity in the markets for asset-backed securities was identified as a risk by a wide
range of official publications, including the Bank of England’s Financial Stability
Report, over several years. If central banks underwrite any maturity transformation that
threatens to damage the economy as a whole, it encourages the view that as long as a
bank takes the same sort of risks that other banks are taking then it is more likely that
their liquidity problems will be insured ex post by the central bank. The provision of
large liquidity facilities penalises those financial institutions that sat out the dance,
encourages herd behaviour and increases the intensity of future crises.

In addition, central banks, in their traditional lender of last resort (LOLR) role, can lend

“against good collateral at a penalty rate” to an individual bank facing temporary
liquidity problems, but that is otherwise regarded as solvent. The rationale would be that
the failure of such a bank would lead to serious economic damage, including to the
customers of the bank. The moral hazard of an increase in risk-taking resulting from the
provision of LOLR lending is reduced by making liquidity available only at a penalty
rate. Such operations in this country are covered by the tripartite arrangements set out in
the MOU between the Treasury, Financial Services Authority and the Bank of England.
Because they are made to individual institutions, they are flexible with respect to type of
collateral and term of the facility. LOLR operations remain in the armoury of all central
banks.

Conclusions

The path ahead is uncertain. There are strong private incentives to market players to

recognise early and transparently their exposures to off-balance sheet entities and to
accelerate the re-pricing of asset-backed securities. Policy actions must be supportive of
this process. Injections of liquidity in normal money market operations against high
quality collateral are unlikely by themselves to bring down the LIBOR spreads that
reflect a need for banks collectively to finance the expansion of their balance sheets. To
do that, general injections of liquidity against a wider range of collateral would be
necessary. But unless they were made available at an appropriate penalty rate, they
would encourage in future the very risk-taking that has led us to where we are. All
central banks are aware that there are circumstances in which action might be necessary
to prevent a major shock to the system as a whole. Balancing these considerations will
pose considerable challenges, and in present circumstances judging that balance is
something we do almost daily.

The key objectives remain, first, the continuous pursuit of the inflation target to maintain

economic stability and, second, ensuring that the financial system continues to function
effectively, including the proper pricing of risk. If risk continues to be under-priced, the
next period of turmoil will be on an even bigger scale. The current turmoil, which has at
its heart the earlier under-pricing of risk, has disturbed the unusual serenity of recent
years, but, managed properly, it should not threaten our long-run economic stability.

2 comments:

HeraclitusO said...

It's obviously a bit late to worry about moral hazard. With confidence in the whole system at stake, even smaller fry can't be allowed to fail. And this has been clear for a couple of decades now... so there is no discipline enforced by depositors.

Of course in part this raises questions about whether it was really such a smart idea to remove regulation of banking sector from the Bank's responsibilities, but it is hard to imagine a set of regulatory institutions that leads to the right decisions being made at the right time - the regulators just don't have the knowledge even if one were to consider the political dimension not to play a role.

So is it time to revisit once again the idea of forcing banks to issue subordinated debt, and using spreads to trigger an automatic independent review of that institution's soundness? Of course, credit markets aren't always as forward looking as one might hope, but at least the asymmetric payoff might lead to better alignment of incentives of management and the ultimate providers of capital.

m* said...

limbicpua, thanks for your comment.

Presumably you are referring to the blow to confidence in UK banks with the Northern Rock crisis. while m* has general praise for the BOE, it appears that the FSA dropped the ball on this one allowing a deposit taking institution to lever itself quite heavily and with short term capital to boot. At the same time, the incomplete depositors insurance scheme in place in the UK (until the Treasurer's alteration this week) was in confidence terms only marginally better than no scheme at all, hence a polite bank run but a bank run nonetheless.

m* does not believe it is yet too late to worry about moral hazard or that the confidence in the whole system is yet at risk - though folks like Bernard Connolly and Marc Faber, and owners of gold would certainly agree with you.

Before the Fed decision, 39% of respondents to a WSJ online poll (well, ok - point taken) were against a rate cut and the addition to moral hazard it would produce.

http://online.wsj.com/public/article/SB119007144524930441.html

The idea of using market based metrics to trigger an institutional review of bank soundness is interesting, but unlikely to be practical given the timing issues between market-based decision making and a government review.

Increased transparency of leverage, duration matching, and stricter standards for off-balance sheet dealing would be an improvement.

Many of these problems stem from the conflict between regulatory standards for soundness and the profit motive of shareholder driven managements. In a perverse twist, one could make a fairly good argument that Basel II is actually a step towards "free banking" by legitimizing (with the exception of the off-balance sheet stuff)the risk assets on bank balance sheets.

Perhaps there is a market for an independent bank rating agency utilizing increased disclosure (ie: not the equity or CDS market) that would be accessible to the retail consumer.