Memorandum submitted by Professor Marcus
Miller, Department of Economics and Centre for the Study of Regionalisation
and Globalisation, University of Warwick
Professor Miller is grateful to Santayan
Ghosal for his comments
In his current book on globalisation, George
Soros (2002) argues that it is the rapid evolution of capital
markets that is the major issue. He identifies four key steps
in the recent historical process of the globalisation of these
(i) the progressive removal of capital controls
following the ending of the Bretton Woods system in 1973;
(ii) the sharp rise in offshore financing
after the first oil price in 1973 (as petro dollars were recycled
as countries borrowed to buy oil);
(iii) the acceleration of capital movements
in the 1980s under the free-market deregulatory administrations
of Ronald Reagan and Mrs Thatcher; and
(iv) the collapse of Communism in 1990, after
which financial markets became truly global.
While this process has liberated innovative
and entrepreneurial abilities and accelerated global economic
growth, it has a negative side. For, as Soros puts it, financial
markets can be "unstable" or prone to crisis. The
well-known phenomenon of "bank runs" is consistent with
so too was the excessive rate of company liquidations seen in
nineteenth century London, and in America during the Great Depression.
These runs and excess liquidations are handled at the national
level by the institutions which have evolved for the purpose,
including the Central Bank, Federal Deposit Insurance Corporation
(FDIC), Bankruptcy Courts, etc as summarised in Table 1 in the
What about global capital markets, where crisis
may occur for two main reasons: (i) problems of creditor co-ordination
exemplified by a bank run (see for example Sachs 1995) and (ii)
those of debtor's moral hazard characteristic of sovereigns who
misuse funds raised by issuing debt on terms of sovereign immunity.
Because banks who lent to emerging markets took large losses under
the Brady plan of the late 1980s, emerging market finance has
since then been mainly in the form of New York bonds which are
very difficult to "restructure". This may be useful
as a check on debtors "moral hazard", but it poses serious
problems of creditor co-ordination when debt service capacity
is impaired by events largely outside the debtors control.
The rapid growth of global markets has outpaced
the competence of global institutions to handle them and the latter
have been forced to play "catch up". (Table 2 in the
Annex provides an overview). But there have been successive crises:
Mexico 1994-95: East Asia 1997-98: Brazil 1999: Turkey and Argentina
in 2001and now possibly Ecuador and Brazilagain.
These market failures not only imply some resource misallocation:
much more, they represent shocking setbacks in the growth prospects
for many promising emerging markets, with serious implications
for income distribution and living standards of their citizens,
To achieve greater financial stability without
undermining incentives for efficient allocation of capital is
surely one of the principal economic challenges posed by globalisation.
(Another is achieving the redistribution objectives set by the
Millennium Development goals; Birdsall and Williamson, 2002.)
As noted by the UK Executive Director to the IMF in his answers
to the Treasury Committee, this involves measures for both crisis
prevention and crisis resolution.
2. CRISIS PREVENTION
The agenda for crisis prevention was summarised
by John Williamson (2000) as including:
(a) macroeconomic discipline;
(b) standards and codes for the financial
sectorand incentives for emerging market countries to implement
(c) a CCLwith an automatic right to
draw by countries satisfying certain standards;
(d) active encouragement by the IMF for the
use of price-related measures to limit capital inflows in inappropriate
forms and excessive quantities; and
(e) a switch in the form of lending from
short-term loans to instruments that effectively place much of
the risk on the lender.
In this context it is worth noting that sustainable
ratios of Debt-to-GDP relevant for emerging markets may be substantially
lower than those for the OECD countries. Since interest rates
paid by emerging market governments are both higher and more variable
than those paid by industrial country governments (and since borrowing
costs tend to increase sharply with the Debt-to-GDP ratio), Stanley
Fischer (2001, Chapter 2) concluded that the upper limit of 60
per cent of GDP "that seems to have gradually gained status
as a norm" is too high for an emerging market country for
whom "ratios nearer 30 per cent are much safer". (Brazil
currently provides a case in point: with Government Debt close
to 60 per cent of GDP and much of it short-dated, interest rates
rising over 18 per cent pose serious problems of sustainability).
But it is clear from the above list of measures
that things have moved on a long way from the days when IMF advice
could beand wascaricatured by the phrase "It's
Mostly Fiscal"! The East Asian Financial crisis in 1997-98
marked a decisive changefor the simple reason that most
of the countries involved did not initially have significant fiscal
problems, as measured by Deficits and Debt relative to GDP,
while their poorly regulated financial systems and their short-term
liquid liabilities in dollars left them dangerously vulnerable
to capital outflows.
As was confirmed by the representatives from
the Treasury in their evidence to the Committee, the IMF has backed
away from its advocacy of rapid capital account liberalisation
prior to the Asian crisis in favour of a more carefully sequenced
approach. As John Williamson (2000) has put it, however, "the
question now is whether it should go further and actively urge
members to implement controls when capital inflows are excessive".
He continues: "The precedent here is Chile
in the 1990s and its encaje (also Colombia and Malaysia)
. . ., Chilean-style capital inflow taxes can provide a useful
instrument to emerging markets suffering actively urge its members
to use when threatened by excessive inflows."
3. CRISIS MANAGEMENT
As the Treasury document notes "the IMF
has an important role resolving financial crises, but the availability
of official resources is limited in relation to private capital
flows. This increases the importance of securing the involvement
of private sector creditors in crisis resolution. The way crises
are resolved today may have important implications for the behaviour
of the public and private sectors in the future."
(a) IMF as an International Lender of last
Two new facilities have been designed to provide
credit in crisis conditions, subject to prequalification or higher
interest rates to avoid overuse. First is the Contingency Credit
Line (CCL) introduced in April 1999 to provide precautionary credit
facilities for countries hit by financial contagionconditional
on sound macro policies, adherence to financial standards, and
good relations with private creditors. Although in September 2000
interest charges and commitment fees were reduced, there are still
no takers, probably because of the "signalling problem"if
you ask for the credit line investors fear the worst. Second is
the introduction during the Asian Crisis of Supplementary Reserve
Facility (SRF) to offer larger sums for shorter periods at higher
interest rates than traditional IMF lending.
It is widely agreed that easy availability of
liquidity in the form of "bail outs" can lead to moral
hazard on the part of both creditors and debtors and undermine
the proper functioning of capital markets. In Ghosal and Miller
(2002) we argue that bail-outs do not necessarily solve the underlying
causes of a sovereign debt crisis. In fact bail-outs may end-up
increasing the probability of a sovereign default.
(b) Collective action clauses, Sovereign
Debt Restructuring Mechanism (SDRM), and a Sovereign Debt Forum
Both self-organising creditors (ie sovereign
debt contracts with collective action clauses) and international
bankruptcy procedure are ways of creating commitment devices which
prevent and resolve sovereign debt crisis. Recent proposals for
the former by John Taylor of the US Treasury and the SDRM of Anne
Krueger of the IMF are discussed in some detail in Miller (2002),
already circulated to the Committee. Which will work better?
Collective action clauses have the advantage
that they can more easily be implemented: after all, they have
been incorporated in London bonds since the late nineteenth century.
But there are two major problems, referred to as those of Transition
and Aggregation. Because most existing sovereign debt instruments
have been issued under NY law, they do not contain such clauses:
how are they to be changed? Some have suggested that bond swaps
be used to solve this problem of Transition.
Perhaps more serious is the fact that collective
action clauses operate within one class of debt, while a sovereign
debtor typically has many types of debt in issue, eg Argentina
has more than 80. How can agreements to restructure be reached
across the different classes of debt? JP Morgan Chase and Company
have suggested a two-stage bond swap could in principle be used
to handle this Aggregation problem. But this has not been put
to the test.
The Sovereign Debt Restructuring Mechanism advocated
by M Krueger of the IMF is meant to solve both these problems.
Her proposals require international agreement to major institutional
change, however, and will surely imply giving more power to the
IMF and less to Wall Street. The stand taken by the United States
is crucial as it has a blocking vote on the necessary Amendments
to the IMF's articles (which require an 85 per cent super majority
vote and the USA alone has 17 per cent of the votes). But the
signals given so far are unclear. On the one hand, Mr O'Neil apparently
encouraged the IMF to explore the avenue of institutional change;
on the other hand, John Taylor of the US Treasury countered this
initiative with proposal for collective action clauses. It appears
that the latter is more "market-friendly": an open letter
signed by the IIF (Institute for International Finance) the EMTA
(Emerging Market Trading Association) and the EMCA (Emerging Market
Creditors Association) has come out with a clear preference for
Despite the apparent dissonance, however, it
may be that the approaches taken by the IMF and the US Treasury
are complementary rather than contradictoryboth being elements
of what is referred to as a two track approach. The first track
is to proceed with contractual changes, while keeping the second
trackstatutory interventionas a live option. It
could well be that the threat of statutory change gives lawyers
and the markets the incentive they need to write ingenious contracts
for creditor co-ordination. Should the contracts solution fail,
further institutional change would be necessary.
As interim measure, it has been proposed that
a Sovereign Debt Forum (SDF) be established to facilitate restructuring.
Its supporters argue firstly that bondholders realise institutional
change is required; and second that an informal procedure, where
those who have money at stake make the deal, may be more effective
than the legal procedures proposed by Krueger. Richard Gitlin
(2002), for example, has formulated principles to guide a Forum
along these lines, principles which include changes to bond contracts
(to include collective action clauses with super majority voting
and provision for the appointment of Trustees to promote restructuring).
Although the Paris and London Clubs offer interesting precedents
for such an approach, it is widely believed that bargaining goes
better "in the shadow of the law".
In the absence of decisive measures to improve
the performance of global capital markets (such as a credible
bankruptcy procedures, which takes account of both moral hazard
of the sovereign debt and creditor co-ordination) capital controls
may be necessary to prevent further crises. In the words of the
Hippocratic oath, Dani Rodrik (1998) has
warned that "where knowledge is limited the rule for policymakers
should be, first, do no harm"; and, in the light of the East
Asian financial crisis, he concluded there is a compelling case
for maintaining controls or taxes on short-term borrowing. Inflow
controls along Chilean lines are not too controversial. But outflow
controls may also be necessary in times of crisis as Anne Krueger's
It is believed for example that Argentina experienced enormous
capital outflows while the overvalued peso was still pegged at
one to one with the dollar and wealthy citizens were free ship
dollars across the Rio de la Plata into private bank accounts,
out of the sight and out of reach of the sovereign state. It seems
clear that the benefits of capital mobility in these circumstances
accrue to a small and privileged elite while the costs of the
crises are borne by the whole population. As Tobin warned in 1999,
these activities can threaten the viability of the currency peg
and the divergence between private and social costs can justify
public intervention. 
Bartholomew E, E Stern, A Liuzzi (2002) "Two
Step Sovereign Debt Restructuring" New York: JP Morgan Chase
Birdsall N and J Williamson (2002) "Delivering
on Debt Relief", Washington DC: Centre for Global Development
Fischer S (2001) "The International Financial
System: Crises and Reform", Lionel Robbins Lecture, London
School of Economics, October.
Ghosal S, Miller M (2002) "Co-ordination
Failure, Moral Hazard and Sovereign Bankruptcy Procedures",
Mimco, University of Warwick.
Gitlin R (2002) "A Proposal: Sovereign
Debt Forum", Mimeo, Bingham Dana LLP, New York, March.
HM Treasury (2001) "Responding to the Challenges
of Globalisation; the UK and the IMF 2001", London : HM Treasury.
Miller M (2002) "Sovereign debt Restructuring:
New Articles, New Contractsor no Change?" International
Economics Policy Briefs, no PB02-03, Washington DC: IIE.
Rodrik D (1998) "Who needs capital convertibility?"
in Peter Kenen (ed) "Should the IMF pursue capitalaccount
convertibility?, Essays in International Finance, no 207,
Rogoff K (1999) "International Institutions
for Reducing Global Financial Instability", Journal of
Economic Perspectives, Fall.
Sachs J (1995), "Do we need an International
Lender of Last Resort?" Mimeo.
Soros G (2000) George Soros on Globalisation
NY: Public Affairs.
Stiglitz J (1999) "Must Financial Crises
be this frequent and this painful?" in P-R Agenor et al.
The Asian Financial Crisis, Cambridge, Cambridge University Press.
Robin J (1999) "Financial Globalisation:
Can national currencies survive?" Cowles Foundation Paper
No 985, Yale University.
Williamson J (1999) "What should the Bank
think about the Washington Consensus?", Policy Analyses in
International Economics, Washington DC: IIE.
Williamson J (2000) "Modernising the International
Financial Architecture: Big Outstanding Issues" www.iie.com/papers/williamson0900-lhtm,
Washington DC: IIE.
SUMMARY OF DOMESTIC PROCEDURES AND INTERNATIONAL
As a background to the present debate on the
international financial architecture, it may be useful to summarise
institutional mechanisms devised to handle domestic liquidity
crisis, such as bank runs and creditor grab races. Table 1 lists
mechanisms used for banks and non-financial co-operations inside
a nation state. These include the lender of last resort role for
the Central Bank as recommended by Walter Bagehot, bank deposit
insurance and regulation as practised in the US by the Federal
Deposit Insurance Co-operation, and the role of the bankruptcy
BANK RUNS AND CREDITOR GRAB RACES
|For whom||What action?
||Supply Liquidity||Co-ordinate a Rollover
||Suspend Convertibility||Provide Insurance
||Impose Regulation||Supervise Workout
|Who does it? ||
||Central Bank (CB)|
||Debt equity swap
Table 2 uses the same framework to see how such actions are
taken at the global level and by whom.
HANDLING SOVEREIGN LIQUIDITY CRISIS AND KEN ROGOFF'S LIST
OF "8 WAYS TO SAVE THE WORLD"!
|What action? ||
||Who is to do it|
|Supply Liquidity||Co-ordinate a Rollover
||Suspend Convert'y||Provide Insurance
||Impose Regulation||Supervise Workout
||Rogoff's List (JEP 1999)
||(1) "Deep Pockets ILLOR"|
||G7 IMF?||(2) "Crisis Manager"
|(3) Capital controls|
||(4) Global FDIC|
||(5) Global Financial Regulator|
|(6) Transparency and Regulation|
||Debt Restructuring||Paris and London Clubs IBC
(8) Rogoff's planan ex ante debt equity swap
KEY: The headings in the
last column refer to the various points made in Rogoff's article.
The concept Soros uses to characterise this instability is "reflexivity".
It seems to correspond, broadly speaking, to the economist's notion
of multiple equilibria. Back
Perhaps the position taken in the Treasury Report on the UK and
the IMF 2001 is too. It notes that "Greater capital mobility
has the potential to raise welfare . . . However, undertaking
liberalisation too quickly, before the necessary conditions are
in place, can jeopardise stability". Paragraph 3.22. Back
Which corresponds fairly closely to the conditions for capital
account liberalisation described by the UK Executive Director
to his evidence to this Committee. Back
There were nevertheless significant contingent liabilities-to
bail out the banking system for example-which have added substantially
to the ex-post debt figures. Some have argued that these off-balance
sheet liabilities should be included in the ex-ante Debt figures:
but this poses the risk of labelling as fiscal events whose origins
lie elsewhere. Back
Also cited as a good example of well-sequenced capital account
liberalisation by the Treasury witness to the Committee. Back
In his contribution to essays on the topic "Should the IMF
pursue Capital Account Convertibility". Back
The Treasury Document on the IMF and UK 2001 notes that "We
should be prepared to support a country that must impose temporary
capital controls, as part of as an orderly process of crisis resolution".
This presumably includes outflow controls. Back
"The Central Bank, committed to honour the peg and to maintain
the country's terms of trade, has to protect its reserves. It
cannot be indifferent to the claims on those reserves negotiated
by private parties, domestic and foreign, who ignore the social
costs" Tobin (1999). Back