Is there a case for a further co-ordinated global fiscal stimulus? Part 1

On September 16 and 17, the Earth Institute at Columbia University (well, at least it’s not called the Universe Institute) and the Asian Development Bank organised a conference at Columbia University on The Future of the Global Reserve System.  Papers were presented by the members of the Asian Development Bank’s International Monetary Advisory Group (IMAG), of which I am one (the other members are Prof. Jeffrey Sachs, Dr. Nirupam Bajpai, Dr. Maria Socorro G. Bautista, Prof. Barry Eichengreen, Dr. Masahiro Kawai, Prof. Felipe Larrain, Prof. Joseph Stiglitz, Prof. Charles Wyplosz, Dr. Yu Yongding).

The paper “Is there a case for a further co-ordinated global fiscal stimulus” is my take on the subject assigned to me for the New York conference: Are the coordinated stimulus plans working and are they effective? Should we continue with fiscal stimulus? Are there other approaches to aggregate demand management?”

I will publish the paper in this blog in two or three installments, as I revise the initial draft.  Installment one follows below.

Introduction

For further internationally co-ordinated expansionary fiscal policy measures to be desirable today, a number of conditions must be satisfied.

First, there must be idle resources – involuntary unemployment of labour and unwanted excess capacity.  Output and employment must be demand-constrained.

Second, there must be no more effective way of stimulating demand, say through expansionary monetary policy.

Third, expansionary fiscal policy must not drive up interest rates, either by raising the risk-free real interest rate or by raising the sovereign default risk premium, to such an extent that the fiscal stimulus is emasculated through financial crowding out.

Fourth, at given interest rates, the expansionary fiscal policy measures are not neutralised by direct crowding out (the displacement of private spending by public spending or of public dissaving by private saving at given present and future interest rates, prices and activity levels).  Such direct crowding out can occur in the case of tax cuts (strictly speaking, cuts in lump-sum taxes matched by future increases in lump-sum taxes of equal present discounted value) because of Ricardian equivalence/debt neutrality.  In economies with very highly indebted households, debt neutrality can occur when taxes on households are cut, because of what I shall call “Minsky equivalence” (see Minsky (2008)).  Increases in public spending on real goods and services (“exhaustive” public spending) can fail to boost aggregate demand because of a high degree of substitutability (in the utility functions or the production technology) between private consumption and investment on the one hand and public consumption and investment on the other.

Fifth, there must be cross-border externalities from expansionary fiscal policies that cause decentralised, uncoordinated national fiscal expansions to be suboptimal.

This paper will consider these issues in turn.  After reaching some fairly discouraging conclusions on the scope for further conventional expansionary fiscal policy now, unless there are significant political realignments in fiscally challenged nations that support coalitions in favour of significant future fiscal tightening through tax increases or public spending cuts, I briefly outline some unconventional fiscal/financial policies that may be effective in their own right and may help to enhance the effectiveness of conventional expansionary fiscal policy.  Collectively, they can be characterised as the equitization of debt – household mortgage debt, bank debt and public debt.

(I) Idle resources

The first of these conditions, the existence of idle resources reflecting deficient effective demand, is clearly satisfied in the industrial world: at the time of writing (July 2009), unemployment throughout the industrial world is high and rising.  Capital utilisation rates are low and falling.  Chart 1 below summarises the state of the G7 economies, indicating high and rising spare capacity; even though GDP is now growing again almost everywhere, it is growing more slowly than potential output, so unemployment will be rising in much of the overdeveloped world for the rest of this year and much of 2010.  In some countries, including the UK and US, unemployment could still be rising in 2011.

Idle resources in the G7
Idle resources in the G7


The picture is rather better, although subject to considerable country-specific variation, among the emerging markets and developing countries, however.  In the aggregate, emerging markets and developing countries now account for around 50 percent of global GDP and trade.  China’s growth rate is reported to be well above 8 percent again, at an annual rate and India is growing fast and never even experienced much of a slowdown.  Brazil is recovering briskly from a sharp decline in industrial production at the end of 2008.  Other emerging markets and developing countries such as Indonesia also appear to have experienced only minor growth pauses.  Individually none of these countries, not even China, is large enough to act as a global locomotive.  Collectively, however, they are a powerful force for global recovery.

The strength of the emerging markets is, not uniform, however.  Even among the BRICs, Russia demonstrates, with a more than 10 per cent year-on-year decline in GDP, the vulnerability of some emerging markets to the interruption of global financial intermediation, the decline in commodity prices and the collapse of world trade.  Perhaps the BRICs should be re-labelled the BICs.

(II) Monetary policy

As regards alternative ways to fiscal policy of boosting demand, monetary policy in the overdeveloped world seems to have been pushed about as far as is possible, unless governments undertake the (minor) institutional and technical reforms to permit them to set short nominal policy rates at negative values (see Buiter (2009)).  Exchange rate changes (deliberate devaluations/revaluations in managed exchange rate regimes or endogenous responses to policy actions or other exogenous shocks) are globally zero-sum as regards their effect on demand – they redistribute demand between currency areas.

This does not necessarily mean that exchange rate changes cannot be welcomed by all parties involved, but this will only be the case if the country with the depreciating currency has deficient aggregate demand while the country experiencing currency appreciation has excess demand.  Even better would be the configuration of the country with the depreciating currency experiencing both deficient aggregate demand and an excessive trade deficit while the country with the appreciating currency experiences both excess demand and an excessive trade surplus.  Within the set of advanced industrial countries, such lucky pairings cannot be found, but between the group of industrial countries and the group of emerging markets and developing countries, there may be some partial fits.

In what follows I will look at monetary policy as working mainly through the official policy rate, through expectations of the future official policy rate and through quantitative easing, credit easing and enhanced credit support.

The official policy rates in much of the industrial world are very close to zero.  The Federal Funds target is a zone between 0.00 and 0.25 percent; the Bank of Japan’s target for the uncollateralized overnight call rate is 0.10 percent; the Bank of England’s Bank Rate is 0.50 percent; the European Central Bank’s Main refinancing operations (fixed rate) at 1.00 percent.  There is, of course, no good reason why all four official policy rates should not be at zero.  In Japan and the US this would not make much of a difference, but 50 basis points in the UK and 100 basis points in the Euro Area are low-hanging fruit that should have been harvested already.  If the Swedish Riksbank can have a negative interest rate for commercial bank deposits with the central bank, why can’t the Fed, the ECB and the Bank of England?[1] And don’t tell me the Fed cannot do it for fear of money market funds ‘breaking the buck’.  That is a buck that needs to be broken occasionally, if money market funds offer higher expected returns than deposit accounts.

When challenged on their failure to lower the official policy rate to its lowest possible level, the authorities tend to offer a technical-operational story as to why a zero official policy rate would be awkward/difficult/impossible to implement.  With the official policy rate at zero, deposit rates on commercial bank reserves with the central bank would probably have to be negative.  Interest rates on some private deposits might also be negative.

Modestly negative interest rates on private deposits with commercial banks, as for commercial bank deposits with the central bank, pose no problem at all.  With the high carry costs on currency, we are most unlikely to see commercial banks holding their reserves in currency notes yielding a zero interest rate rather than in deposits with the central bank yielding even minus one percent or less.  There is only a (slightly less than) zero lower bound on the nominal interest rate because the authorities maintain a constant exchange rate (at unity) between bank deposits with the central bank and currency.  As shown in Buiter (2009), by putting deposits with the central bank at an appropriate forward premium to currency, even very negative interest rates on deposits don’t present an operational problem for the central bank, despite the presence of zero nominal interest-yielding cash. Abolishing currency or taxing currency has the same effect of eliminating any lower bound on the nominal interest rate.

Conventional monetary policy is not exhausted even if the official policy rate is at its lower bound (zero, say).  Conventional monetary policy is not exhausted until risk-free nominal interest rates at all maturities are at their lower bounds (zero).  Risk-free here means free of default risk.  The official policy rate is a short-term risk-free rate, often a target for the overnight rate.

The monetary policy authorities can influence longer-maturity risk-free interest rates either by committing themselves to a given sequence of future (short-term) official policy rates, by lending and borrowing over longer maturities at the risk-free rate of interest and/or by buying and selling longer-maturity risk-free financial instruments.  Lending by the central bank is only risk-free if it is secured by risk-free collateral, because no private counterparty is free of default risk.

All leading central banks now accept as collateral in repos, at the discount window and for any of their wide range of ad-hoc facilities created for the crisis, private collateral.  The ECB, for instance, did a very large-scale repos operation on June 24, 2009, when it lent €442 billion for a one-year maturity at 1.00 percent against its usual wide range of eligible collateral.  Although central banks steadfastly refuse to provide the information required to value the illiquid collateral they have accepted (and continue to accept), we cannot be certain that the rate of return to the central bank on these operations includes an appropriate risk premium rewarding them for the private credit risk they are taking on.  Even the profits recently reported by the Fed and the ECB on some of their liquidity operations don’t provide sufficient information to determine whether these central banks have been handing out ex-ante quasi-fiscal subsidies to their commercial bank counterparties during the past two years.

With neither the borrowing banks nor most of the collateral offered free of default risk, the ECB’s massive operation on June 24, 2009, was probably aimed at more objectives than just influencing the one-year risk-free rate.  The operation is likely to have involved a quasi-fiscal subsidy to the participating banks and may have encouraged banks to lend more at somewhat longer maturities, where private borrowing and lending rates are likely to still include a material liquidity premium.  Nevertheless, the uncapped (fixed-rate) repo did also provide a strong hint, almost a commitment, that the official policy rate would not be raised from its present level during the coming 12 months.

Quantitative easing – expanding base money in circulation (mainly bank reserves with the central bank) by purchasing government securities – does not appear to have had a noticeable and persistent effect on risk-free or on private market rates in the US, the UK or Japan.  Credit easing – outright purchases of private securities by the central bank, which can either be monetised or sterilised – is achieving little in the US or in the UK, although it has not been pushed very hard yet in the UK, where total Bank of England purchases of private securities amounted to less than £ 3 billion on July 16, 2009.  The Bank of England’s cumulative acquisition of Gilts through the Asset Purchase Facility on that same date was almost £114 billion.[2] In August 2009, the Bank of England increased the limit on the size of its Asset Purchase Programme by £50 bn to £175bn.  Almost all of this will take the form of Bank of England purchases of UK Treasury debt.  With UK annual GDP at £1.4 trillion, the Asset Purchase Programme amounts to about 12.5% of annual GDP, about the same magnitude as the public sector financial deficit for the current fiscal year.  Exiting from this extremely Treasury-friendly Asset Purchase Programme could well create material tensions between the Bank of England and HM Treasury.

In the Euro Area, the announcement by the ECB that it would purchase up to € 60 bn worth of covered bonds was followed up with just € 2.5 bn worth of actual purchases (as of 23 July 2009) thus far.[3] It appears, however, to have prompted a large increase in private issuance of and investment in covered bonds.

Enhanced credit support in the Euro Area – providing collateralised loans on demand at maturities up to a year at the official policy rate – is not working either, or not with enough speed and impact to expedite a cyclical recovery. None of these policies appear to materially improve the ability and willingness of banks to lend to the non-financial sectors.  They have had some positive impact on the corporate bond markets (which allow corporates and ultimate savers to bypass the banking system), but not enough to prevent a sharp contraction in total credit extended to non-financial corporates and households. It is not surprising that this should be so, once we reflect on the nature of these policy actions and on the conditions under which they are taking place.

In a nutshell: quantitative easing (QE), credit easing (CE), and enhanced credit support (ECS) are useful when the problem facing the economy is funding illiquidity or market illiquidity.  They are useless when the dominant concern of banks and their counterparties is the threat of insolvency, unless these special central bank operations contain a significant amount of quasi-fiscal subsidy, in which case they can be helpful in recapitalising the banking sector.  This appears to be the case, especially in the US, although a precise estimate of the magnitude of the quasi-fiscal subsidies bestowed by the Federal reserve system on the US banking system (and on part of the non-US cross-border banking system too) is not yet available.

Today, liquidity is ample, even excessive: commercial banks either re-deposit funds obtained from the central bank with the central bank or use it to purchase government debt.  Little of these funds appears to be finding its way towards the non-financial private sector.  Capital is scarce. Capital is scarce first and foremost in the banking sector. A panoply of central bank and government financial interventions and support measures have ensured, at least for the time being, the survival of most of the remaining cross-border banks. It has not done enough to get them lending again on any scale to the household and non-financial enterprise sector. In addition, while the fiscal authorities are prompting banks to raise more capital and have injected public capital in the weakest systemically important bank,s and while central banks are injecting liquidity into the economy on a scale never seen before, regulators and supervisors are often forcing banks to act pro-cyclically, by building up their liquid assets now and by aggressively deleveraging now.

That banks have ample or even surplus liquidity is apparent from the divergent behaviour of the stock of bank reserves with the central bank, which is increasing fast, and the broad money stock held outside the financial sector (for these purposes, the non-bank financial sector is just the off-balance sheet segment of the banking sector and should be consolidated with it). As pointed out by Ben Broadbent of Goldman Sachs, the increase in M4 outside the financial sector in the UK has recently been much smaller than the growth of commercial bank reserves with the bank of England. Similar patterns exist in the US and in the Euro Area.

To be sure, it is not just the supply of credit that has contracted sharply.  As the economy weakened, the demand for credit has engaged the supply of credit in a race to the bottom, but there can be no doubt that many firms and households are credit-constrained, and cannot find external finance either from the banks or from the capital markets.  Only the larger enterprises, and among these only those with a good credit track record, have access to the capital markets, even in good times. Small and medium-sized firms and new firms without a credit track record cannot go the markets.  Households, of course, cannot raise funds from the markets directly at all. So with zombie banks and highly selective access to the corporate bond markets, we are set for a slow and anaemic, restricted-credit recovery in much of the overdeveloped world.


[1] Since July 8, 2009 Sweden’s Riksbank has its official policy rate, the repo rate, at 0.25 percent and the deposit rate at -0.25 percent.

[2] The amount invested by the Bank of England in Commercial Paper was £1.8 billion, in Corporate Bonds £ 0.9 billion and in Gilts £113.9 billion.  Source: Bank of England http://www.bankofengland.co.uk/markets/apf/.

[3] Source: ECB http://www.ecb.int/mopo/implement/omo/html/index.en.html.

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