A damp squib from the G-7 in Washington DC

The G-7 have done the absolute minimum required not to have their Washington DC meeting on Friday, October 1o characterized as an abject failure.   So let’s just call it a disappointment and a missed opportunity. There was, unfortunately too much of the “we will use all available tools” verbiage (the G-7 variant of the British “we will do whatever it takes”) and not enough commitment to concrete, specific schemes and firm actions that leverage international cooperation to address the global financial crisis.

Each country is now supposed to outline its commitment to take specific action to support liquidity in key financial markets (such as the interbank markets, markets for commercial paper, for asset-backed commercial paper and for mortgage-backed securities), bank recapitalisation, funding for the financial system and deposit guarantees.

Even if the G-7 had agreed on a substantive international plan of action, the G-7 would not have been the right forum to guarantee the effectiveness of a plan to stabilise the situation sufficiently to prevent a global financial crisis from becoming a global economic crisis rather  than just a global recession.  Effectiveness requires that the countries that belong to the G-13 (those members of the G20 that are not also in the G-7) be on board also.

For those whose knowledge of G-{N, N=1, 2, …, ∞} is bounded from above (I always have to look them up myself), the G-7 member countries are the USA, Japan, Canada, the UK, France, Germany and Italy; the members of the G-20 are the finance ministers and central bank governors of 19 countries (the G-7 countries plus Argentina, Australia, Brazil, China, India, Indonesia, Mexico, Russia, Saudi Arabia, South Africa, South Korea and Turkey) plus the European Union.  While the G-20 is too large a group for meaningful deliberation and discussion, at least it is not flawed by errors of omission.  With a bit of luck we will in due course replace the current G-7/G-8, which is flawed both by errors of omission and errors of commission, with a new G-8, consisting of the USA, the EU, Japan, China, India, Brazil, Russia and Saudi Arabia, which includes all political-economic entities that have global systemic significance and which will meet regularly to address global economic and financial issues,

How much international coordination is actually required?

For fighting the financial crisis, international co-ordination and co-operation are required in eight areas

(1) Guarantees for interbank lending.

Without international agreements on which national fiscal authority guarantees unsecured or secured lending between banks domiciled in different jusrisdictions, only lending between banks in the same country can be guaranteed.  That would greatly reduce the effectiveness of guarantees for interbank lending.

Instead of Treasury guarantees for interbank lending, there could be an alternative arrangement (equivalent from an economic point of view) in which a publicly guaranteed clearing house is created, guaranteed by the fiscal authorities of the countries whose banks participate in the clearing arrangement (according to some sharing rule), that would act as the counterparty for all unsecured (or for all secured and unsecured) interbank lending.

A special case of this is where the central banks of the countries/jurisdictions of the participating banks play the role of counterparty in interbank lending. So instead of UK bank A borrowing from US bank B, UK Bank A would borrow from the Bank of England (unsecured), the Bank of England would borrow that same amount from the Fed and the Fed would borrow that same amount from the lending bank, US bank B.  Alternatively, British Bank A could borrow directly (unsecured) from the Fed and US bank B would lend that same amount to the Fed. The last interesting alternative would be where US bank B would lend directly to the Bank of England, which would lend the same amount to British Bank A.  The risk exposures of these arrangements are, of course, quite different, and a clear set of internationl rules has to be worked out promptly for this to work.

(2) Government guarantees for other bank liabilities

Many governments have guaranteed some bank liabilities, be they deposits or bank debt held by other banks or by non-bank institutions or individuals.  Whether such guarantees are applied to new liabilities only (as I would prefer) or are applied to both existing and new liabilities,  their effectiveness would be much reduced if they applied only to liabilities issued to persons or to institutions domiciled in the same national jurisdiction as the bank issuing the liability.

(3) Recapitalisation of banks with significant cross-border activities.

The economic interest a nation has in ensuring the survival of a domestic bank (or subsidiary of a foreign bank) may not be well-aligned with the official national domicile of the bank.  The Dutch, Belgian and Luxembourguese authorities negotiated minority capital injections in the national subsidiaries of the Fortis Group (that deal subsequently got unstuck because the Dutch withdrew from it).  The Belgian, French and Luxembourguese authorities negotiated another 3-country recapitalisation plan for Dexia.  There will have to be many more of these cooperative multi-country fiscal bailouts, and not only for small countries with large banks.  A template for fiscal burden sharing for rescues of banks with major cross-border signatures would be very helpful.

(4) Fiscal bail-outs of countries whose systemically important banks have a solvency gap that exceed the government’s fiscal capacity

The problem of systemically important banks or other financial instutions whose need for external resources beyond what can be obtained privately exceeds the fiscal capacity of its government,  is not necessarily confined to small countries with large banking sectors – the ‘Iceland problem’.  There are some quite large countries that, although their banking sectors are not terribly large in relation to their GDP, nevertheless could find themselves at risk of not being able to support their systemically important banks because their governments are in bad fiscal shape to begin with.  I hope that, although it is not possible to plan for this openly,  contingency planning for the fiscal rescue of insolvent neighbours is taking place in the basements of the ministries of finance of countries with deep pockets.

(5) Mandatory debt-to-equity conversions

I hope that, before this crisis is over, holders of debt of banks and other highly leveraged institutions will have seen outcomes other than (1) a full guarantee of their debt and (2) default of the issuer followed by a lenghty and uncertain extraction of  X < 100 cents on the $/€/£ through the bankruptcy process.  An across-the-board partial debt-to-equity conversion, in inverse order of seniority, would help reduce leverage without the need for public sector capital.  It would also be wonderful from a moral hazard perspective.

But without international agreement on the types of debt instrument involved and the fraction of each type of debt instrument to be converted, such a mandatory debt-to-equity conversion would be a mess, with great opportunities for cross-border evasion.

(6) Avoiding a moral hazard race to the bottom

Unless their are binding international agreements on the kind of guarantees extended by governments to financial institutions and creditors in their jurisdictions, we will see a replay of the moral hazard explosion that followed the Irish decision to guarantee all liabilities of the six largests majority Irish-owned banks.  The race to the bottom that would result could end up with every creditor guaranteed and the tax payer screwed royally everywhere. Not only would this be outrageously unfair, it would create terrible incentives for future reckless lending.  The IMF would be the obvious organisation to monitor and police any common norms adopted.

(7) Agree common access rules and common methods for valuing illiquid assets in different national TARP-like structures

TARP-like  toxic asset dumps will be essential for gaining insight into the true state of the balance sheets of banks and other highly leveraged institutions.  They are also key to an orderly cleaning up of banks’ balance sheets (through mark-to-market accounting rules) and to a resumption of issuance in securitisation markets and markets for (a limited subset of rather simple and transparent) structured products.  There should be common rules for who gets access to the toxic asset dumps managed by different nations (e.g. subsidiaries of foreign banks).  There also should be common methods for valuing illiquid assets, lest the same asset gets valued in different ways at different national TARPs, putting excessive strain on the public finances of the country offering the highest valuation.

(8) Agree to adhere rigorously to mark-to-market accounting and reporting principles and agree on common rules for relaxing regulatory requirements attached to marked-to-market valuations

It has taken decades to agree on mark-to-market (or ‘fair value’) accounting and reporting.  It would be the height of insanity to throw that achievement out of the window because regulators cannot apply common sense in their application of capital ratios, liquidity ratios and other regulatory requirements that are influenced by marked-to-market  asset valuations.  The information provided even by asset prices set in illiquid markets is bound to be better than what the managements of financially troubled companies would come up with, which would be lies and damned lies.  But the regulatory consequences of valuation distortions introduced by mark-to-market accounting and reporting in the presence of illiquid asset markets can be adjusted, in a uniform, internationally coordinated manner.

After the storm

To support the real economy once the worst of the financial crisis subsides, further cuts in official policy rates are likely to be required.  Doing these rate cuts cooperatively probably enhances their effectiveness in the short run because it boosts confidence.  I would not waste interest rate cuts while the current financial hurricane is blowing.

Expansionary fiscal policy (debt-financed or money-financed tax cuts or public spending increases) are likely to be especially effective in a credit crunch, because more firms and households are apt  to be liquidity-constrained.  To gain the greatest bang-per-buck, personal income tax cuts or transfer payments should be targeted at those most likely to be liquidity constrained; this includes households with temporarily low incomes (the temporarily unemployed) and with rising age-earnings profiles (yuppies).  Others with high marginal propensities to consume include the old. This is not the time to worry about the Golden Rule, the Sustainable Investment Rule and the SGP.  Once the crisis and the recession that will follow it are over, there will have to be a quite steep increase in the tax burden or cut in public spending as a share of GDP.  Let’s see whether whoever is in office then has the guts to act countercyclically during good times.

There is a risk that, with every nation in the world busily reflating the economy come 2009, we end up with an aggregate fiscal and monetary impulse that is way to big.  No-one has any idea, after all, as to how great the damage to the real economy is that has been caused by the financial conflagration so far, or how much more damage will be caused before financial markets normalise.  Here too is a role for the IMF – its systemic surveillance role.

Other helpful hints for the G-20

Other key lessons the G-20 should make use of when they put together their national and internationally cooperative rescue packages include the following

(1)  Do’nt do an Irish: don’t guarantee all liabilities of the banking system (with nothing in return)

Governments elsewhere should not emulate the Irish example and guarantee all the liabilities of the banking system.  This would be silly for two reasons:

(1) It does too much.  To get new lending going again requires only a guarantee of new lending and of the issuance of new liabilities.  The existing stock can be left without guarantees.  The UK scheme for guaranteeing (up to an estimated value of £250 bn but with more available should it be needed) gets it right.  It only guarantees newly issued debt instruments.

(2) It gives no upside to the tax payer.  The only thing the tax payer gets out of the Irish scheme (other than the prevention of the possible collapse of the banking system, which can be achieved more cheaply) are the fees charged for the guarantees.  Governments are always very vague about this.  Words like ‘commercially priced’ and ‘no subsidy from the tax payer’ abound, but seldom if ever any specifics are forthcoming.  Even today, for instance, we don’t know the terms charged by Bank of England and the UK Treasury for Northern Rock’s use of the Liquidity Support Facility starting in September 2007.

If a government is going to guarantee all bank liabilities it should insist on some serious upside for the tax payer.  I believe a fair quid-pro-quo (and one which would at least give the right incentives for future equity investors) would be for the government to nationalise any bank whose liabilities it guarantees comprehensively.  You might as well own the assets if you are going to be responsible for all the liabilities.  I don’t think that any compensation for the shareholders is required – if the government needs to guarantee all the liabilities of the bank, it obvious is not a viable going concern.  Mor cautions (wimpish) governments could insist on equity stakes (with upside) of less than 100 percent.

(2) Address the market failures directly

Even the UK plan, which was otherwise not too bad and certainly best-of-breed, contained a number of key deficiencies.  The interbank markets are disfunctional.  Banks don’t lend to each other unsecured for longer than overnight and even overnight there is little direct interbank lending, with banks preferring to keep large amounts of overnight reserves with the central bank.

This matters  for two reasons.  First, banks that cannot lend to and borrow from other banks are likely to have to curtail their lending to non-bank customers pretty soon.  Second, for historical reasons that are hard to undo swiftly, many loans to the private sector are priced off Libor (especially 3-month Libor).  When the G-7 monetary authorities cut their official policy rates by 50 basis points and Libor rises, you know that interbank distrust has taken on pathological proportions.

So any plan of action that makes sense should address the interbank lending drought directly.  The central bank could interpose itself between the banks in the interbank market, acting as universal counterparty (the international dimensions of this were discussed earlier).  At 3 months maturity it could, for instance, offer to lend (unsecured) at the 3-month OIS rate plus 50 basis points and to accept deposits at the 3-month OIS rate minus 50 basis points.  Central banks don’t like lending unsecured, but these are interesting times, and even central bankers have to grow up.   If the Fed can buy commercial paper and asset-backed commercial paper from the banks, then all central banks can lend unsecured to banks in the interbank market.  Alternatively, the Treasury could guarantee interbank lending, for a fee. In any case, the Treasury would have to indemnify the central bank for any losses it incurs by acting as counterparty of last resort in the interbank market.

(3) Don’t forget about moral hazard

The best – really the only – time to make progress on imposing proper incentives for avoiding future excessive and inappropriate risk taking is during a crisis when the banks and other financial institutions that need to be reined in, are on the ropes.  When the economy is healthy  and the banks and other financial institutions are strong, their lobbying power and political influence make it virtually impossible to impose any constraints that their CEOs don’t want to see imposed.

The French minister of finance, Christine Lagarde is quite wrong when she says “Moral hazard has to be dealt with later … Maintaining the functioning of our markets is the top priority”.  She is wrong for two reasons.  First, she believes that addressing moral hazard now would undermine actions to keep the markets functioning properly.  That is incorrect as a general statement.  Putting out fires and taking care of moral hazard can be complementary objectives.  Second, she has the typical politician’s binary view of choices: either nothing or everything.  The notion that there may be a rather less abrupt trade-off (somewhat more stability today at the expense of a somewhat greater likelihood of somewhat greater stability in the future), does not appear to occur to her.

Let me illustrate.  A bank needs recapitalisation.  It cannot raise the money privately.  The government could just give it the money – no questions asked.  Or it could provide the money in exchange for preference shares.  Or it could provide the money in exchange for ordinary shares, or for preference shares convertible into ordinary shares over some future period.  Or it could provide the money in exchange for some combination of preference and ordinary shares and a seat on the board.  Or it could ask for the previous package as well as requiring a partial debt-to-equity-conversion.  And/or it could impose restrictions on dividends and share repurchases and on executive remuneration.

Another example.  The guarantees for bank borrowing (existing and/or new) and for  other bank liabilities (existing and/or new, including deposits of various kinds) could be provided for free.  Or they could be priced in such a way that the authorities are not expected to make a loss for the tax payer.

The nadir of a financial crisis is clearly the ideal time to address moral hazard in an intelligent way.

(4) Apply measures to all highly leveraged institutions, not just banks

Many institutions not classified as banks engage in the same activities as banks.  Many institutions are highly leveraged and too systemically important (too large, to interconnected or too politically connected) to fail.  Common suppport should be given to and common constraints should be imposed on all these highly leveraged institutions (HLIs).  This includes hedge funds, money market mutual funds, other investment funds, structured investment vehicles (SIVs), private equity funds, insurance companies like AIG with high embedded leverage through exposure to derivatives, and non-financial corporates like GE and GM (through its exposure to GMac).  Focusing only on banks can destroy the level playing field between banks and other institutions with which it competes.

(5) Don’t make recapitalisation voluntary and subject to delay

Another major flaw in the UK package is that it does not compel all banks (let alon all HLIs) to achieve certain minimal capital ratios (and, I would argue, maximal leverage ratios) immediately.  The confidence collapse is now. The capital is needed now.  In practice, since banks cannot attract new capital from private sources overnight, this would mean that the government would recapitalise all banks.  I will just make a brief comments on the near-instantaneous capital injection by the HSBC group into its UK subsidiary, achieved by taking the money out of another member of the group.  It worries me that (a) HSBC’s management appear not to be taking the need for additional capital too seriously, based on this financial legerdemain, and (b) that HSBC may be able to get away with this vivid example of taking something out of one pocket and putting it into another.

Giving the banks days or even weeks and months to raise the required additional capital from private sources would be to put the banks’ desire for continued independence ahead of the demands of systemic stability.  Confidence crises move with the speed of electronic data transmission.

Once the required capital ratios and leverage ratios have been achieved through government capital injections, the banks will be able to attract private capital to repay the government capital, subject to any minimum public shareholdings the state wishes to keep for governance reasons or to give the tax payers some more upside.

(6) Reduce leverage at least in part through mandatory debt-to-equity conversions applied to all HLIs

This ‘Chapter 11 lite’ for the entire range of systemically important highly leveraged institutions should be applied immediately and preferably on uniform terms by all of the G-20.  It could be restricted to all debt incurred before a certain cut-off date.  I don’t think it is necessary to extinguish the existing equity first.  Existing shareholders will get diluted by the exercise.  From a moral hazard point of view is is a perfect antidote to the blanket guarantees for existing debt that some benighted governments have seen fit to hand out.

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

Maverecon: a guide

Comment: To comment, please register with FT.com, which you can do for free here. Please also read our comments policy here.
Contact: You can write to Willem by using the email addresses shown on his website.
Time: UK time is shown on posts.
Follow: Links to the blog's Twitter and RSS feeds are at the top of the page. You can also read Maverecon on your mobile device, by going to www.ft.com/maverecon