Learning the Lessons of Financial Crises:
The Roles of the Public and Private Sectors
First Deputy Managing Director
International Monetary Fund
Emerging Market Traders' Association Annual Meeting
New York, December 9, 1999
Ladies and Gentlemen.
It is a great pleasure to be with you here today. Addressing the Emerging Market Traders'
Association makes a change from speaking to gatherings of central bankers, finance ministry
officials, or financial supervisors. On those occasions I explain how the IMF labors tirelessly to
reduce economic and financial volatility. And the audience seems to appreciate our efforts.
But here I am not so sure. For most of you, volatility—to be sure, predictable
volatility—is your daily bread and butter, not a source of distress. But the dramatic events
of the last two-and-a-half years surely demonstrate that you can have too much of a good thing.
The economic instability associated with the recent wave of financial crises, and the large
publicly financed support packages they demanded, have left policymakers determined to reform
the architecture of the international financial system.
One widely held conclusion is that the efficient operation of the international system requires
more private sector involvement in the prevention and resolution of financial crises. I am well
aware that this view has been causing concern to some in the private sector—and this was
well reflected in the thoughtful article in the fourth quarter EMTA Bulletin, "Is
Burden-Sharing Being Pushed Too Far?" But we surely have a common interest in many
aspects of reform in this area, as well as other parts of the international financial architecture.
For one thing, we all have an interest in the development of fair, efficient and predictable
trading arrangements for emerging market securities. This will help integrate emerging market
countries more deeply and more durably into the global economy, to the benefit of all. You have
already played a valuable role in this process, by helping to create the Emerging Markets
Clearing Corporation. Similarly, all of us want to avoid situations in which the relationship
between creditors and borrowers breaks down completely in difficult times. As the Russian
crisis last year demonstrated all too clearly, the consequences of such breakdowns may be
systemic in nature, bad for almost everyone involved, and with far too many countries and people
I will begin today by discussing the health of the emerging market economies in the wake of
their recent traumas. The bottom line is that they are in encouragingly good shape, but that we
are far from what used to be normal conditions. I will then turn to the lessons of recent events
for the architecture of the international financial system, beginning with the importance of further
measures to help prevent financial crises. Finally, I will turn to the controversial question of
crisis resolution, to bailing-out and bailing-in.
2. The Road to Recovery
First, the global economy. It is hard to believe that it is just over a year since there were
serious and well-justified fears of a looming world recession, one that seemed to be
spreading—from Russia to Latin America to New York. Fortunately, the doomsday
scenario did not come to pass. Policymakers responded decisively. The Federal Reserve cut
interest rates and other central banks followed. Congress finally supported expansion of the
Fund's capital base. Japan announced further fiscal stimulus and stepped up the pace of banking
reform. And the IMF assembled a support package for Brazil, that helped stabilize markets at a
time of unprecedented global financial fragility. When Brazil was forced to devalue in January
this year, the effects in Brazil and the market fallout were far less adverse than they would have
been a few months earlier.
In recent months activity in the emerging market nations has been stronger than we and
others expected. Take Asia first. Korea is expected to record economic growth this year of 9 per
cent. Thailand has been growing at an annualized rate of between 3 and 4 per cent. Growth has
resumed in Indonesia, where macroeconomic stabilization took longest to achieve and where
market sentiment has been undermined in recent months by the Bank Bali scandal and violence
in East Timor.
The big economies in Latin America have also outperformed expectations. With the possible
exception of troubled Ecuador, the entire region seems to be pulling out of recession. In
February the Brazilian economy was expected to shrink by 4 per cent this year. Now it looks as
though growth will be slightly positive. Mexico has put in a robust performance in the third
quarter. Argentina's deep downturn seems to have bottomed out. And Chile—whose
former Finance Minister Eduardo Aninat will next week join the IMF as a Deputy Managing
Director—is undergoing an impressive turnaround, helped by an aggressive easing of
monetary policy and a weaker currency.
These improvements are welcome indeed. But some are not yet totally secure. The outturn
depends on a host of factors. I will touch on three.
First, the health of the industrialized economies. Activity in the US remains robust and the
Fed's recent tightening of monetary policy should help ensure that this remarkable upswing
continues at a sustainable pace. Europe's biggest economies are also perking up, as the
unexpectedly strong factory orders and labor market data from Germany illustrated earlier this
By contrast, the third quarter drop in Japanese GDP comes as a disappointment, although in
part it reflects upward revisions to output in Q2. But the latest data underline the need for
macroeconomic policies, especially monetary policy, to remain expansionary—indeed to
become more so—until the recovery in Japan becomes firmly rooted.
Second, policies in the emerging market countries. The resumption of growth in Asia
reflects expansionary macroeconomic policies and an encouraging start to ambitious structural
reforms, as well as the recovery of the Japanese economy and electronics exports. But these
achievements should not be used as an excuse to slacken the pace of financial sector reform and
corporate restructuring. If growth is to be sustained, and long-term growth rates in the Asian
miracle economies are to return to close to their previous levels, policymakers in these countries
will need to maintain the momentum of reform efforts, even as the direct role of the international
financial institutions in their countries is scaled down.
Third, access to finance in the emerging market countries. International fund raising by the
emerging markets dropped sharply in the third quarter. At $33bn, inflows during the quarter
were barely up on the immediate aftermath of the Russian default. Capital flows to Asia were
stable, but Latin America saw big declines in both bond issues and loan commitments. The data
show that financing picked up in October, though that is to some extent a statistical illusion,
reflecting the voluntary Brady debt exchanges. The pace in November was down from October,
and Y2K concerns suggest an imminent upsurge is unlikely.
In the secondary market, emerging market bond spreads have recovered considerably since
their peak of around 1700 basis points after the Russian crisis. The declining spreads have made
these markets more attractive to many. But at over 900 basis points today, spreads remain
uncomfortably high. No-one wants a return to indiscriminate and ultimately destabilizing capital
inflows. But some improvement in financing conditions would certainly help strengthen these
still fragile upturns. That is another reason why the issue of private sector involvement in crisis
financing is so critical.
These factors will all be important to the outlook for emerging market economies in the short
term. In the longer term, it is also vital that the international community restore the momentum
of trade liberalization, which has been the engine of global growth during the post-war period.
The recent events in Seattle were profoundly worrying. We must hope and work for progress
here before too long.
3. Crisis Prevention
Let me turn now, very briefly, to the lessons recent events have taught us. We will
undoubtedly talk a great deal about crisis resolution and what this means for the role of the
private sector. But the first priority for the Fund and the emerging market nations should be
crisis prevention. Among other things, this means:
- promoting sound economic and regulatory policies;
- avoiding exchange regimes that are vulnerable to attack;
- ensuring effective prudential regulation of financial systems;
- opening capital accounts in an orderly fashion;
- making debt structures more robust to external shocks;
- ensuring that implicit or explicit guarantees do not shield the private sector from the
consequences of unduly risky behavior; and
- improving transparency and the provision of information to the public and to investors.
This list is familiar. What may be less familiar is the intensified work the international
community is pursuing along these lines. The IMF has stepped up its surveillance of economic
policies and prospects in our member countries. For some time we have been paying greater
attention to capital account and financial sector questions, the sustainability of exchange rate
systems, plus debt and reserve management practices. We are also looking more at vulnerability
analyses, the spillover effects from national policies, inter-country comparisons, and regional
developments. We have also taken steps that will encourage countries to act more decisively on
the policy advice they get.
The remarkable improvements in IMF transparency and in the information we provide are
visible on our website. We are also encouraging countries to make more information available to
the private sector, not only so that investors are in a position to take more soundly-based
decisions, but also through greater transparency to encourage good policies by governments. For
example, our Special Data Dissemination Standard (SDDS) now demands better coverage of
international reserves. Countries are also being encouraged to provide more information on the
level, structure and composition of their external debt.
International standards have also been agreed for transparency in a number of other policy
areas. These include monetary and financial policies, fiscal policy and banking supervision. The
IMF is also working together with the World Bank to identify and help remedy weaknesses in
financial sectors. These failings were central to most of the recent crisis episodes. Pilot schemes
have been under way for some time to pull together information on the observance of these
standards into transparency reports, or as they are known in their most recent incarnation,
ROSCs—Reports on the Observation of Standards and Codes.
As a result of these changes, much more information—more timely and of better
quality—is now becoming available on national economic and structural policies. But the
incentives for good economic policy provided by greater surveillance are not just the
responsibility of the official sector. If this information is to help prevent crises, investors must
make good use of it. We know you will do that as you become more familiar with the
4. Crisis Resolution and the Role of the Private Sector
Improved crisis prevention can help make the global financial system a safer place. But
countries will still get into trouble from time to time, as a result of poor policies, bad luck in the
form of internal or external shocks, or a combination of these factors. The international
community is ready to help when this happens, provided countries take measures to deal with
their problems. As our Articles of Agreement explain, IMF lending is there to provide our
members "with opportunity to correct maladjustments in their balance of payments without
resorting to measures destructive of national or international prosperity". The lending is to
take place "under adequate safeguards", which is interpreted as meaning subject to
conditionality on the policies the country will pursue.
A strong policy response by the country, together with official lending, should provide the
basis for stabilization and a return to growth. But typically that is not enough, and normal
growth can only be restored when private capital begins to return.
In most cases capital inflows resume spontaneously through the catalytic effect of policy
adjustments and structural reforms supported by the Fund and the rest of the official sector. This
remains the case today—in most IMF programs, it is business as usual. That could be seen
for instance in the economic program with Colombia that was announced in the same week as the
Ecuadoran default took place.
But there have been, and there will be, occasions on which the catalytic effect of the policy
program is not enough to finance the necessary economic adjustment. If a country does not have
a reasonable chance of regaining spontaneous market access without excessively large public
lending, then it may be necessary to take extra steps to secure private sector involvement. There
are two reasons for this:
First, there is the question of moral hazard. You are all familiar with the argument: by
sheltering private creditors from the consequences of previous lending decisions, officially
financed bail-outs can encourage imprudent lending in the future. This makes financial crises
more likely to happen and more severe when they do. So we should not offer any promise that
countries will always be able to service all their external debts.
I would like to amplify a bit on the topic of moral hazard. It is possible that official lending,
unless accompanied by the right incentives, including private sector involvement, can create the
possibility of moral hazard. But I do not want to push the argument too far. For example, some
critics claim that the Mexican rescue package in 1995 caused the Asian crisis two-and-a-half
years later, by encouraging investors to buy risky Asian assets, confident that the international
community would shield them from losses if things went wrong.
But the evidence does not support this theory. Investments in Asia did not flow
predominantly into the sorts of assets that were most likely to benefit from an IMF program.
Holders of government debt were the main beneficiaries of the Mexican package, yet holdings of
Asian government debt increased relatively modestly in the months that followed. One might
also have expected greater lending to Asian banks, which might have been thought likely to
enjoy protection in the event of a crisis. But again this was not the case.
It seems highly implausible that Asia was a moral hazard play. It is more likely that lenders
could not imagine that the miracle economies of the region could get into such serious trouble
after decades of outstanding economic performance. But moral hazard exists, and for a clear
example of the dangers that it poses, look no further than Russia. Many investors thought Russia
was too big to fail. They were wrong, and the consequences for them and for other countries
Let me return now to the second reason private sector involvement is needed: the IMF does
not have enough money to ensure that countries can always service their debts. As a practical
matter, our quota resources are limited, as they should be. And it is crystal clear from the
reaction to the crises of this decade that the public sector as a whole is not willing to provide the
resources that would be necessary to enable any country that gets into trouble to continue
servicing its debt in full.
We are of course familiar with the counter-arguments on private sector involvement in the
resolution of external financing crises. As EMTA argued in September, with respect to
sovereign debt: "While burden-sharing by the private sector is acceptable in principle,
forced rescheduling of bonds will drive investors away from the emerging markets and
effectively deprive countries of much-needed access to the bond markets".
In working towards a solution to the difficult and sensitive question of how best to secure
private sector involvement when this is necessary, we are learning from experience. At the
moment private sector involvement is at the center of programs with four of our member
countries: Ukraine, Pakistan, Romania and Ecuador.
Let me touch briefly on the circumstances of the four countries currently in the
spotlight—and I should emphasize that the circumstances of these countries differ
- With a relatively modest level of overall debt, Romania faced a liquidity
problem meeting repayments on two bonds due in May and June this year. After consulting with
their financial advisers, the authorities said they were confident of obtaining $600m in new
money, covering 80 per cent of the maturing amounts. In the event, the Romanians managed to
raise only a small proportion of the amount required, but our Board agreed to support—with
less financing than originally envisaged—a program under which they promised to raise
$470m for the first review. Even that money has proved hard to raise at a reasonable price in
current markets. With other aspects of the program proceeding well, in particular with foreign
exchange reserves much higher than expected in the program, the Executive Board has again
indicated that it could go ahead with much less private sector financing.
- Ukraine also faced a liquidity problem, arising from a highly bunched debt
service profile. The government negotiated with three major groups of creditors to secure
conversion of short-term instruments into longer-term ones. Based on the results of past
restructuring and the country's immediate balance of payments needs, it seemed prudent to seek
refinancing of 80 per cent of the maturing amounts. With Ukraine's reserve position suggesting
little immediate threat of default, the agreement proved difficult to reach. It has also left even
steeper repayment peaks in 2000 and 2001. The authorities have indicated that they now favor a
more comprehensive approach.
- In October Ecuador became the first country to default on Brady bonds since
their inception in 1990. Investors unwilling to accept the government's swap offer assembled the
necessary 25 per cent of bondholders to vote for acceleration. This case illustrates the problems
that arise from a severe liquidity crisis, delays in reaching agreement with the IMF on adjustment
measures, and delays in their implementation in a very difficult political situation, and a highly
complex debt structure. A consultative group has now been set up, which is realistic in
encompassing a wide range of bondholders.
- Like Ecuador, Pakistan faced insolvency rather than illiquidity. The Paris Club
agreed in February to reschedule debt service payments due over the next three years, on
condition that private creditors—including bondholders—accepted comparable terms.
An offer to exchange three bonds maturing between now and 2002 closed on Monday and more
than 90 per cent of holders accepted. This case raises several issues, including the definition of
comparable treatment, how to coordinate a diffuse range of creditors and what role the official
sector should play in the negotiations.
We are in the process of drawing lessons from this experience. We are aware that watching
our approach evolve through these cases may be frustrating. Your position paper recognized the
need for a case-by-case approach, but wanted it "guided by clearer and more consistent
principles". We too hope that we will be able to develop a consistent set of general
At this point two general principles are clear. First, when a country that has not managed its
policies and its external borrowing well, and requires large scale public financing to continue
servicing its debt, gets into trouble, the private sector will have to contribute to the resolution of
its financial difficulties. In some cases, those corresponding in a sense to insolvency, this would
involve debt restructuring.
And second, to the maximum extent possible, the official sector must encourage countries to
honor contracts, and to service their debts. The international financial system cannot operate
efficiently if default ever becomes anything other than a very painful option, a last and
much-regretted resort. Nothing that has happened so far provides any incentive to countries to
default; we have seen no country that wants to default, no country that has not gone to
extraordinary lengths to avoid default, and no country that having defaulted does not wish it
could have done otherwise. That is as it should be.
Beyond those general principles, lie mainly questions. First, at the organizational level: How
should the debtor and the creditors interact—via creditor councils, or some other way?
What should be the role of the official sector in the discussions between the debtor and its
creditors? And at the analytic level: Does the crisis appear to be one of long-term insolvency or
short-term illiquidity? What is the type and structure of the debt? How should the debt be
restructured if that is necessary? How do you judge when treatment is comparable between
different creditors? And, from the viewpoint of the official sector, what consequences might a
bail-in imply for other countries?
As we in the public and private sectors draw the lessons of these and other recent
cases—and this evening's discussion is part of the attempt to do that—we should at the
same time press ahead with pre-emptive measures that would make restructuring more orderly,
bearing in mind the principle that contracts should be honored.
First, we should encourage better communication between countries and their creditors when
times are good. Private investors who exchange views regularly with national authorities are
better informed investors, more likely to be there for the long haul. These consultation
mechanisms have shown their worth in Latin America during times of market stress.
Second, and I come now to a topic on which both EMTA and the IIF have expressed strong
views, we need to find ways to encourage the adoption of contract clauses that will make
restructuring easier when that becomes essential. For instance, emerging market nations could
adopt British-style bond contracts, with sharing, majority-voting, minimum-legal threshold,
nonacceleration and collective-representation clauses. This idea was first proposed in an
important report of the G-10 Deputies in 1996, in the wake of the Mexican crisis. It was only
subsequently that many discovered that such bonds already exist, in the form of British-style
Critics claim that features like collective-action clauses will push up the cost of capital for
emerging market economies. There have been times, for instance in the spring of 1997, when
spreads on many bonds have been too low. But that is not presently the situation.
However, let me raise the possibility that including such clauses in bond contracts could
reduce spreads. The argument, which I first heard from a market participant, is that there
is now great uncertainty about how any debt servicing difficulties would be resolved—and
everyone knows that debt servicing difficulties can arise. The inclusion of renegotiation clauses
would remove part of that uncertainty, and could thereby reduce spreads.
There has been some empirical work on this question. A preliminary study of spreads on
existing US and British-style bonds by Eichengreen and Mody suggests that including
negotiation-friendly clauses reduces spreads for the more credit-worthy borrowers and increases
them for the less credit-worthy. This is a more subtle outcome than the one I expected, and it is
of course preliminary; however it could be interpreted as suggesting that the inclusion of such
clauses makes markets more efficient than they would be without them.
In any case, these results suggest that this is a topic that deserves far more serious
consideration on both the demand and supply sides of the emerging markets than it has received
We have emerged from the dramatic events of the last two-and-a-half years with plenty of
lessons, both for crisis prevention and crisis resolution. We must now ensure that the conclusions
we have already drawn are translated into action, and that we work as fast and hard as we can to
draw lessons in other areas, especially private sector involvement.
But crises will never be abolished entirely. When countries do get into difficulties, the IMF is
available to provide short-term assistance, under appropriate safeguards. In the great majority of
cases, an IMF program of normal size should be sufficient to restore access to private capital
spontaneously or reasonably soon.
Under certain circumstances, however, it will also be necessary to bail in the private sector,
sometimes by restructuring external debts. This will never be, and never should be, done except
in extreme circumstances. But to recognize that possibility, and to take it into account both in
pricing securities and in the design of contracts, will make for more efficient markets, and
thereby for a more prosperous world.
IMF EXTERNAL RELATIONS DEPARTMENT