Created: 2000-03-05
Last Modified: 2000-03-05
Go to
Brad De Long's Home Page

Teaching | Writing | Career | Politics | Book Reviews | Information Economy | Economists | Multimedia | Students | Fine Print | Other | My Jobs

The IMF and Moral Hazard

J. Bradford DeLong



First, it seems to me that the moral hazard problem is not just a
matter of neoclassical theory, but as Robert Rubin himself
conceded a couple of years ago, a material phenomenon. So,
without an IMF bailout fund, there would be less risk-free
NY/London/Tokyo/Frankfurt portfolio capital flowing into the
emerging markets (we call this hot money in South Africa, as it's
come in very fast three times -- 3-12/95 after exchange controls
were dramatically relaxed, 1-6/98 and 3-11/99 -- but twice left even
faster, over periods of a few weeks on largely spurious grounds,
like a rumour Mandela was ill in 2/96, causing enormous real-
sector dislocations due to 30% currency collapses, rapid depletion
of forex reserves, massive interest rate increases, stock market
crashes, etc). So, if the portfolio-K inflow doesn't arrive because the
IMF isn't there to do bailouts, then we don't get the disastrous
outflow episodes.

My Comment:

Ummm... Yes we do. Back before World War II, back before there was an IMF, there were *lots* of disastrous outflow episodes. The absence of a moral hazard-creating IMF did not keep there from being a disastrous outflow episode in the U.S. in 1873, in the U.S. in 1893, in the U.S. in 1907, in Argentina in 1890, in Austria in 1931, in Germany in 1931, in Britain in 1931, and in a bunch of other places. To claim that financial markets will be "rational" and not overspeculate in the absence of a lender of last resort is just naive. Only those who have vast confidence in the rationality of financial markets make such confident assertions.

The live questions are: (a) How much in the way of additional capital flows are generated by confidence that there will be an international lender of last resort? (b) How much smaller are the financial panics that do come made because of the existence of an international lender of last resort? The Asian financial crisis was, a year before it happened, seen as roughly a 1% probability event (and the Mexican financial crisis was, a year before it happened, seen as roughly a 10% probability event). So I have to think that in the most recent financial crisis factor (a) was small: few who invested in East Asia in 1996 were thinking at all about the potential benefits of an IMF bailout in 1998. Claims about how long the East Asian depression would have lasted in the absence of the IMF are inherently speculative, but my use of analogies from the 1870s and 1890s (which are, I admit, extremely weak evidence) suggests that a lot of good was done by keeping East Asian banking systems liquid.


Second, the IMF should be shut down because their lead
bureaucrats retain enormous power, and are opposed to capital
controls which would halt the capital flight you worry about.

My Comment:

Say, rather, that they're scared that capital controls lead to large-scale corruption. And that they are attracted (as am I) by the idea that open capital markets (if they can be properly managed) allow for large-scale capital flows from Japan and Europe to the periphery that promise to cut as much as a decade off the time it takes to go through an industrial revolution.

Look: I like taxes on short-term capital inflows. I like the idea that developing countries should twist their relative price structures to make it extremely *expensive* to import foreign-luxuries and extremely *cheap* to import machine tools. But I am also fearful: the rhetoric behind India's anti-developmental-state policies of the past fifty years is very close to the rhetoric behind Korea's policies of state-led development. I don't think that the U.S. state has the power and competence to successfully use tariffs and capital controls as part of an industrial policy--and the U.S. state is, in broad perspective, one of the most autonomous in the world.

So I think that the neoliberal bet is the best one open. And I am pretty damn sure that getting rid of the IMF--without putting anything else in its place--moves us from a world in which developing countries have bad options when first-world (and third-world) investors pull the plug on their economies to a world in which developing countries have no options at all...


Brad DeLong


Sorry, I overstated my case and you're right, the moral hazard
argument is ahistorical. What I should have emphasised was that
this is not, at the end of the day, about institutions, but rather
about a cyclical phenomenon (global banking panics and crashes
during the 1820s, 1870s, 1930s and in recent/current years, each
following an overaccumulation buildup). Institutionally, JP Morgan
was the IMF and Fed rolled into one for a long time, early on this
century, but even he couldn't hold the line from 1929-33. So it's not
the institution that can or can't prevent the disastrous outflow
episodes, it's in the nature of the accumulation process. Perhaps
we'd agree to agree on this.

My Comment:

I think that central banking privately provided by plutocrats is a poor second best to government-provided central banking. And I'm more skeptical about J.P. Morgan-as-central-banker than many. But you have a good point.


What the IMF does so very well (and differently) today, however, is
police the outflows so that nearly all of it actually flows out, instead
of during the earlier periods when a good chunk of it got caught in
sovereign debt defaults (1/3 of all countries refused to pay, during
those periods).

My Comment:

True. But it also provides resources in times of crisis. And *if* the crisis is successfully surmounted those resources may ex post appear to have been extremely valuable. Stan Fischer likes to compare Mexico after 1982 and Mexico after 1994, and attribute a large part of the difference to the fact that sovereign debt defaults are extremely costly to both sides.


Two absolutely "live"
(and death) questions in emerging markets remain, why do we
need "additional capital flows" given what they entail, and what
degree of whoring is sufficient to generate confidence by portfolio
managers and the IMF cop?

My Comment:

Well, back at the end of the last century the U.S., Canada, Australia, Argentina, and some others were able to use their access to international capital markets to accelerate their economic development. It seems a shame to pass up one possible channel through which developing countries can rapidly add to their capital stocks. (Note, however, that both Harry Dexter White and John Maynard Keynes thought that international capital flows were too dangerous to be prohibited.)

The Context:

Wait a minute. As I understand it, the IMF directly ordered massive
East Asian banking system contraction (and hence foreclosures on
plenty of going concerns),

My Comment:

I had thought that the IMF sought the closure of too few East Asian banks, not too many. You do have to shut down insolvent (but not illiquid) institutions, after all. But I haven't looked at it closely enough to have an informed opinion...

As for the rest, you have a lot of good points. But alas! I have to go grade...


Brad DeLong


>I commend to you Stan Fischer's address on the IMF and the Asian Crisis,
>March 20, which discusses this and other aspects of the IMF's role. It can
>be found at
>Peter Temin
>Department of Economics

Two long selected excerpts from Stanley Fischer:

The charge that, by coming to the assistance of countries in crisis, the IMF creates moral hazard has been heard from all points of the political compass. The argument has two parts: first, that officials in member countries may take excessive risks because they know the IMF will be there to bail them out if they get into serious trouble; and second, that because the IMF will come to the rescue, investors do not appraise -- indeed do not even bother to appraise -- risks accurately, and are too willing to lend to countries with weak economies.
It would be far-fetched to think that policymakers embarking on a risky course of action do so because the IMF safety net will save them if things go badly.

All the evidence is many countries do their best to avoid going to the Fund. Nor have individual policymakers whose countries end in trouble generally survived politically. In this regard, Fund conditionality provides the right incentives for policymakers to do the right thing -- indeed, these incentives have been evident in the preemptive actions taken by some countries during the present crisis. These incentives may even be too strong, and I agree with Martin Feldstein that it would generally be better if countries were willing to come to the Fund sooner rather than later. But I do not believe countries should have too easy access to the Fund: the Fund should not be the lender of first resort; that is the role of the private markets.

The thornier issues arise on the side of investors. Economists tend to point to the problems of moral hazard and the inappropriate appraisal of risks; others are more concerned that some investors who should have paid a penalty -- and typically they refer to the banks --may be bailed out by Fund lending. These are two sides of the same coin: if investors are bailed our inappropriately, then they will be less careful than they should be in future.

First the facts. Most investors in the Asian crisis countries have taken very heavy losses. This applies to equity investors, and to many of those who have lent to corporations and banks. With stock markets and exchange rates plunging, foreign equity investors had by the end of 1997 lost nearly three quarters of the value of their equity holdings in some Asian markets -- though to be sure, those with the courage to hold on, have done better since the turn of the year. Many firms and financial institutions in these countries will unfortunately go bankrupt, and their foreign and domestic lenders will share in the losses.

Some short-term creditors, notably those lending in the inter-bank market, were protected for a while, in that policies aimed to ensure that these credits would continue to be rolled over. In the case of Korea, where bank exposure is largest, the creditor banks have now been bailed in, with the operation to roll over and lengthen their loans having been successfully completed earlier this week. Further, we should not exaggerate the extent to which banks have avoided damage in the Asian crisis: fourth-quarter earnings reports indicate that, overall, the Asian crisis has been costly for foreign commercial banks.

None of this is to deny the problem of moral hazard. It exists, and it has always to be borne in mind, and we need to find better ways of dealing with it. But surely investors will not conclude from this crisis that they need not worry about the risks of their lending because the IMF will come to their rescue. Investors have been hit hard. They should have been, for they lent unwisely. But there remains the question: if it was not mainly moral hazard that led to the unwise lending that underlies the Asian crisis, what was it? The answer is irrational exuberance.

Financial crises based on swings in investor confidence -- on irrational exuberance, and also on irrational depression, not really irrational in lacking some foundation in fact, but sometimes representing an excessive reaction -- far predate the creation of the IMF, and would not be avoided even if the IMF did not exist. This is not something to applaud. Rather we have to do everything we can to provide the information and incentives that will encourage rational investor behavior. We do need, as I will discuss shortly, to find better ways to bail in the private sector more systematically. But we cannot build a system on the assumption that crises will not happen. There will be times at which countries are faced by a massive reversal of capital flows and potentially devastating loss of investor confidence. Thus we need in the system the capacity to respond to crises that would otherwise force countries to take measures unduly "destructive of national or international prosperity".

The IMF is part of that system of response, to help countries when markets overreact. Here I would like briefly to discuss the role of IMF lending -- and I emphasize that the IMF lends money, and gets repaid, it does not give it away -- and the issue of bailouts on a more fundamental level.

When the IMF lends in a crisis, it helps moderate the recession that the country inevitably faces. That means that the residents of that country, its corporations, and some of the lenders to that country, do better than they otherwise would have. That is not in any meaningful sense a bailout, provided lending of this type can be sustained in future crises. Rather, if properly designed to avoid as far as possible creating the wrong incentives for the private sector, it represents rational lending -- not grants or handouts -- in conditions when markets appear to have overreacted.

To ensure that lending of this type can be sustained in future crises, we have to be surethat the required size of Fund loans does not keep rising, which means that in seeking to improve the architecture of the international system, we will have to find ways of discouraging unwise private lending -- that is to help ensure that risk is properly priced, and to limit the required scale of official lending, in part by finding ways of sharing the burden between the official and private sectors.

The alternative proposed by those who would abolish the IMF is to leave countries and their creditors to sort out the countrys inability to service its debts. That sounds simple, but it has rarely been so in practice. That is one reason that the IMF assisted the Asian crisis countries to avoid defaults or debt moratoria. In the absence of an accepted bankruptcy procedure for dealing with such cases, given that the debts involved generally involve both sovereign and private obligations, and given the free rider problem, the experience -- from the inter-War period and the 1980s -- is that workouts have been protracted, and that countries have been denied market access for a long time, at a significant cost to growth. By contrast, in the Mexican crisis of 1994-95, market access was lost for only a few months, and Mexico returned within a year to impressive growth assisted by its ability to tap the international capital markets. Similarly, in the present Asian crisis, it is quite likely that both Korea and Thailand will be back to the international markets within a few months. That surely bodes well for their recoveries, which it is reasonable to expect will begin later this year.

The second reason that the IMF tried to help countries avoid a standstill was the fear of contagion. We believed, and continue to believe, that a standstill by one country, at a time when markets were highly sensitive, would have spread to other countries and possibly other continents. That nearly happened in October, but due to prompt and courageous action by Brazil, did not.

Of course, we cannot know what would have happened had there been no official lending in the Asian crisis. But we do know that the crisis has been contained, and it is reasonable to believe that, deep and unfortunate as the crises in individual countries have been, growth in those economies can resume soon.


* * * * * * * * * * * *


Despite its other activities -- surveillance, information provision, and technical assistance -- the IMF is best known for its lending. The Fund operates much like a credit union, with countries placing deposits in the Fund, which are then available to loan to members who need to borrow and who meet the necessary conditions. Members' quotas in the Fund determine both the amount they have to subscribe, and their voting shares. The size of a member's quota reflects, but typically with a lag, the size of its economy and its role in the world economy. 4

Total quotas now amount to a bit under $200 billion. Countries have to pay in 25 percent of their quota (the so-called reserve tranche) in any of the five major currencies in the SDR; the remainder can be paid in the country's own currency.
This means that not all the quotas can be used for lending. Countries can have virtually automatic access to their reserve tranche, and the U.S. has drawn on its reserve tranche more than twenty times, most recently in defense of the dollar in 1978.

In September 1997 the members agreed to increase quotas by 45 percent, about $90 billion, with the United States' share of the increase amounting to nearly $14.5 billion. The Congress has before it at present both the Administration's request
for the quota increase, and a request for $3.5 billion for the United States contribution to the New Arrangements to Borrow (NAB). The NAB will allow the IMF to borrow from a group of 25 participants with strong economies in the event of a risk to the international monetary system. 5 It would thus provide backup financing that could be available if the Fund runs short of regular quota-based resources. The NAB doubled the resources available to the Fund under the
General Arrangements to Borrow established in the 1960s.

When a member in crisis approaches the Fund for a loan, the Fund seeks to negotiate an economic program to restore macroeconomic stability and lay the conditions for sustainable and equitable growth, paying careful regard to the social costs of adjustment. The decision whether to support the country will be taken by the Executive Board, based largely on the strength of the reform program the country is willing to undertake. The loan is typically tranched, paid out in installments, each conditional on the country's meeting the conditions to which it has agreed. These procedures, especially conditionality, constitute the adequate safeguards required by the Articles of Agreement.

The policies agreed in a Fund-supported program typically include fiscal and monetary policies, designed to restore viability to the balance of payments, help restore growth, and reduce inflation. Where appropriate, they also include
structural policies designed to remedy the problems that led to the need to borrow from the Fund. When a country's problem is purely balance of payments related, and can be expected to be reversed in a short time, the Fund loan will
typically cover policies for a year, with repayment starting after three years and concluding within five years. When the country's economic problems are more deep-seated and will take longer to deal with, the arrangement will last longer,
covering policies for up to three or four years. In these cases, the program will contain, along with monetary and fiscal policy changes, more structural measures, such as reform of the financial system, the pension system, labor markets, agriculture, and the energy sector. Such extended arrangements typically include reforms that will be financed during the period of the program by World Bank loans. Such is also the case with the financial sector and other structural reforms in Asian countries.

Despite the common usage, "IMF program", the Fund itself is careful to speak of a "Fund-supported program". Ideally the program should be that of the country, and one that its government is committed to carry out. Of course, in the loan
negotiations, the Fund will usually ask the government to do more than it initially wanted. But because a program is unlikely to succeed unless those who have agreed to it intend to carry it out, a key element in the evaluation of any
agreement is the degree of the government's commitment to the economic program which it has signed -- a conclusion which is reinforced by the recent Asian experience, in which the Korean and Thai financial markets both turned around
when new governments, strongly committed to carrying out the programs, came into office. The government's commitment may be difficult to judge, especially if it is divided, and if, as happens not rarely, the program is being used
by those who favor reform as a vehicle to implement changes that some of their colleagues oppose. Although a Fund-supported program is often seen in the press as the international community's way of imposing changes on a
country's economy, it is more often the international community's way of supporting a government or a group within the government that wants to bring about desirable economic reforms conducive to long-term growth.

But why then are programs so often unpopular? The main reason is that the Fund is typically called in only in a crisis, generally a result of the government's having been unwilling to take action earlier. If the medicine to cure the crisis had been tasty, the country would have taken it long ago. Rather the medicine will usually be unpleasant, in essence requiring the country to live within its means or undertake changes with short-term political costs. Probably the government knew what had to be done, but rather than take the reponsibility, finds it convenient to blame the Fund when it has to act. Similarly, when structural changes have to be made, the losses are often immediate and the gains some way off. Despite all this, there are countries where the Fund is popular, among them transition economies that have seen hyperinflation defeated and growth begin during Fund-supported programs.

The secrecy that until recently has often attended Fund-supported programs may well have contributed to their unpopularity. A public that does not know what is being done, nor why, is less likely to support measures that are difficult in the short-run but that promise longer-run benefits. Governments have often been reluctant to publish their agreements with the Fund, disliking to give the impression that their policies were in any way affected by outsiders. Recently, in the Korean, Thai and Indonesian programs, the government's Letter of Intent, its letter to the management of the Fund describing its program, has been published --another change welcomed by the management of the Fund...


>After reading your Mexico article, I wondered how you
>would respond to Greider's assertion, apparently drawn largely from Japanese
>economists, that most countries following the neo-liberal model are failing,
>and that rapidly developing countries have generally taken a Japanese
>approach (Greider emphasizing the importance of capital controls).

My Comment:

The most rapidly growing countries have taken a "Japanese" approach. Unfortunately, the *least* rapidly growing countries have taken a "Japanese" approach to...

If you have a government strong enough, independent of elites enough, and interested enough to undertake fundamental land reform to ensure a relatively egalitarian distribution of income; a bureaucracy *honest* enough to carry out the missions the central government assigns it (rather than using its leverage to extort bribes and reward its relatives); subsidy policies that promote industrialization while quickly withdrawing support from industries that are too poorly-led or poorly-positioned to generate significant exports; tariff policies that do not hinder attempts to boost exports by making it difficult for export industries to get the stuff that they need--if you have all these things, then you can follow the "Japanese" model of state-led development, and be very successful.

If not--well then, as Lant Pritchett of the World Bank puts it, there's nothing worse than state-led development carried out by an anti-developmental state. Restrictions on imports and exports--capital controls, import licenses, and so forth--become ways in which segments of the bureaucracy can reward the politically powerful or their relatives; subsidy policies prop up employment of factions that the government wishes to keep in its corner rather than assist industrialization; corruption spreads throughout the whole government; and the fact that the government is daily making thousands of decisions that alter the distribution of wealth means that the distribution of wealth gets more unequal because only the elites have access to the levers that the government uses to redistribute the stuff.

The economy stagnates.

Look at Wole Soyinka's _Nigeria:_Open_Sore_of_a_Continent_, at Robert Bates's _Markets_and_States_in_tropical_Africa_, or Carlos Diaz-Alejandro's _Essays_on_the_Economic_History_of_the_Argentine_Republic_ to get an idea of how catastrophically wrong the "Japanese" model of state-led development has gone in much of the third world (and in parts of the first world: Argentina in 1929 was a first world nation).

The "neo-liberal model" is, at some level, an admission of defeat: a bet that if you cut back on the government's role in the economy, you are likely to destroy more of the rent-seeking, corruption-causing, growth-retarding actions of the government than you are likely to hinder the pro-growth, pro-development programs (if there are any).

How this bet is going to turn out is still unclear. It looks like a good bet to me. But it would be far, far better to get a proper Japan-style developmental state up and running--neo-liberal policies are for countries where such a successful developmental state is simply not an option.


>Greider also predicts a coming global supply crisis similar in development
>to the Depression. Krugman rather glibly mocks him, saying there are no
>supply crises, only liquidity problems solved by monetary loosening.

My Comment:

Central banks do not always act properly: they could blow it, and we could have another great depression. But Krugman's right. We don't have--we never have--an "oversupply" problem: we have an "underdemand" problem.

Global productive capacity today is something like 60 times what it was a century ago; something like 200 times what it was two centuries ago; something like 300 times what it was three centuries ago.

There are always people talking about how the economy is "too productive", and how "technological unemployment" and "secular stagnation" because of over-productivity is inevitable. (In fact, if you read _Brave_New_World_, most of the social control exercised by the elite over the population is an attempt to artificially stimulate consumption to avoid such a "supply crisis."

After hearing "wolf" cried for two centuries, I understand where Paul is coming from. And I think that this part of Greider's argument simply cannot be taken seriously--because he doesn't make even a feeble attempt to say why this time the prophets of a supply crisis are correct.

Now corporate economists have something to worry about: high capacity in their industries and low demand means low prices in their industries--good for consumers, but not for the continued employment of corporate economists.

I think that there are actually two things to worry about--and Greider's book is not helpful for thinking about either one of them:

(I) So far, the past two generations have seen U.S. income inequality widen sharply, but increasing trade has played little if any role in the widening. What is going to happen in the next two generations, because globalization certainly has the potential to inflict massive changes on the U.S. income distribution?

(II) We have no reason to feel confident that the IMF and the world's central banks are up to the task of avoiding another Great Depression. Even a relatively small international financial panic like Mexico 1995 was contained by only the narrowest of margins. (No thanks to William Greider, I might add.) And it was contained over the objections of the government of Britain, the government of Germany, the Republican caucus in the House, the Democratic caucus in the Senate, and Banking Committee Chair Alfonse D'Amato.


Brad De Long


>No, they were significant. Significantly worse if I understand it
>correctly, in that the IMF will charge HIGHER rates and demand a
>SHORTER repayment period in the wake of the congressional deal.
>Enlighten me if I'm mistaken, please.

My Comment:

No, you are not mistaken. I know that I want (and Joe Stiglitz wants) a kinder, gentler IMF, willing to loan more money for longer periods of time at lower interest rates with less conditionality; and willing to broker debt-forgiveness deals.

That's not what we got. We got an IMF that is going to loan at higher interest rates in the future. I don't think *that* much damage was actually done--the IMF does have more resources, after all, and is able to loan more for longer terms--but some damage was done.

An IMF that is out of money is not a kinder, gentler IMF--willing to loan more money for longer periods of time at lower interest rates with less conditionality; and willing to broker debt-forgiveness deals. If you wanted a nastier, rougher IMF, you voted against recapitalization. Credit to the AFL-CIO for recognizing this.


>Does anyone (Doug?) have any information of what is actually going on in
>Indonesia. The official transmission belts (aka media) are notoriously
>unreliable, and they focus thier coverage along the lines of "angry mob
>looting the sacred cow of capitalism - private property." I think it was
>Noam Chomsky who graphically demonstrated the media bias on Indonesia
>reporting by actually measuring the volume of the Mossad Information Agency
>(aka New York Times)
[Note: why the antisemitism?] coverage and juxtaposing it with their coverage of the
>alleged "communist attrocities."

My Comment:

I think that the New York Times is still *way* *behind* on its coverage of Stalinist atrocities. Recall their lines during the 1930s on Stalin's Russia and during the 1970s on Mao's China...

Suharto. Tyrant seated on a throne of skulls. Killed 500,000 Communists, suspected Communists, ethnic Chinese-Indonesians who it was to someone's advantage to claim was a Communist, random ethnic Chinese-Indonesians, and so forth--plus another 200,000 dead in East Timor over the past two decades. Supported by the United States in one of our not-very-smart exercises in realpolitik on the grounds that someday Indonesia might be a useful ally against China should China turn expansionist and imperialist.

Suharto. Corrupt as Mobutu. The line in expatriate circles in Jakarta for decades was that his wife's name, Tien, was short for "Tien percent."

Suharto. Who has presided over a quadrupling of Indonesian real GNP per capita. Some of this was the result of the OPEC oil boom of the 1970s--but in other high-population OPEC producers like Venezuela and Nigeria, the political dynamic set in motion by the oil boom ended up making the country poorer than had it never had any oil at all. Suharto deserves some "objectively progressive" points for not following the path that has in Nigeria led to General Abacha.

Suharto. Who has presided over a *sharp* rise in income and wealth inequality--but even so, estimates I have seen suggest that the average working-class Indonesian today has about twice the material standard of living of his or her counterparts forty years ago (and the average upper-class Indonesian today has perhaps ten times the material standard of living of his or her counterparts forty years ago).

Suharto. Who lacked sufficient control over his country last year to fulfill the promises he had made about stopping the burning and deforestation that created the southeast asian smog of 1997.

The IMF. Which is trying to *loan* Indonesia $43 billion so that the sudden wave of domestic and international capital flight does not send the economy into a deep depression, but only into a shallow recession. The problem is that the IMF is not a grant but a loan-making organization, so that it wants its $43 billion back *with* *interest* in time for it to loan it out again to someone else facing a panic wave of speculative capital flight in two or three years when the next international financial crisis hits.

Thus as a result the IMF's policies--the conditions that it imposes on you if you want to borrow $43 billion--are crafted with an eye toward, first, making sure that the borrower country will repay its loan in two or three years. This means that the IMF wants to see, first, an export surplus against which loan principal and interest payments can be charged--and thus needs to see a fall in domestic demand (and a rise in domestic unemployment) and a fall in the exchange rate (and thus a sharp rise in the domestic-currency price of internationally-traded staple goods like rice) in order to diminish imports and boost exports and create the export surplus to repay the loan.

Only after it is sure that the recommended policies will be sufficient to guarantee repayment does the IMF turn its attention to trying to make sure that economic growth in the borrower country resumes as fast as possible.

If you ask Stanley Fischer why the IMF thinks that loaning $43 billion to Indonesia for two years would do any good--or would do more good than simply having Indonesia suspend payments and repudiate its debt--he will point to the contrast between Mexico after its 1982 debt crisis and Mexico after its 1994 exchange crisis. After the first suspension of payments and partial repudiation, it took seven years before GDP per worker reattained its old levels and began to rise. After the second IMF loan and Exchange Stabilization Fund loan rescue, it took only one year for Mexican economic growth to resume. The distributional consequences of both crises were bad--but I have seen nothing to suggest that the distributional consequences were worse the second time than the first.

I think we would live in a better world if the IMF worried less about getting its money back. I think an IMF with four times the resources willing to loan twice as much for twice as long with half of the conditions imposed on domestic economic policies would do a lot more good in allowing countries to cope with the aftermath of policy disasters or of international capital panics.

But that's not the way the political winds are blowing. Both Lauch Faircloth and Ralph Nader appear to want not a larger, kinder, gentler IMF but no IMF at all--in which case there is no one to loan $43 billion on any terms at all to developing countries that suddenly see their balances of payments go completely haywire because of international capital panics. Stanley Fischer and Michel Camdessus have bet the future of the IMF on successful resolution of this East Asian crisis. Indonesia's spiral suggests that their bet is not at all a sure thing.


>So if Bhagwati took you out of context, what is your position on capital

My Comment:

My position is massively confused and inconsistent...

I like the idea that the industrializing periphery can borrow on a large scale from the industrial core in order to cut perhaps a generation off of the time needed to go through the industrial revolution (and--perhaps more important in the very long run--the demographic transition).

I'm not very scared that the (successfully) industrializing parts of the periphery will wind up enserfed in a form of global debt peonage. Leland Stanford and Jay Gould did a wonderful job at making sure that British investors who loaned to American railroads wound up owning not high dividend-paying enterprises but instead overcapitalized shells on the point of bankruptcy. I think that this process will be repeated...

I am scared that (unsuccessfully) industrializing parts of the periphery may get into bad trouble as kleptocrats skim off the foreign loans to live on the Riviera and leave the country's taxpayers to repay. When Erich Honecker wanted to redistribute wealth to himself from the East German people, all he could get was a big house, a few imports, and a deer park. By contrast modern capital financial markets allow Suharto and family to skim off 3 billion? 10 billion? 20 billion? Enough to be worried about.

I do believe that *if* we are going to go the global-capital-mobility route, we need a kinder, gentler IMF: one that makes bigger loans at lower interest rates for longer periods requiring less conditionality when we have an episode like Europe '92, Mexico '94, or East Asia '97 when industrial core investors panic and flee. Thus I am depressed at seeing my options for change reduced to either Lauch Faircloth (who thinks that 30 million people in East Asia need to be thrown out of work because East Asian borrowers who overborrowed *must* be punished)or Ralph Nader (who thinks that 30 million people in East Asia need to be thrown out of work because New York lenders who "overlent" *must* be punished). There's a depressing symmetry here.

Here's a first-draft response to Bhagwati, by the way:


800 words...

I open my May/June _Foreign Affairs_ to discover myself pilloried in an article by Jagdish Bhagwati between Paul Krugman and Roger C. Altman (excellent company to be in, by the way: much better than I am used to) as a banner-waving proponent of international capital mobility, guilty of "assum[ing] that free capital mobility is enormously beneficial while simultaneously failing to evaluate its crisis-prone downside."

I rub my eyes in surprise. I had not thought of myself as a banner-waving proponent of international capital mobility. I wish that Jagdish Bhagwati's research assistants had shown him the sentence from my January 28 Los Angeles Times op-ed after the two he quotes (it reads: "But the free flow of financial capital is also giving us one major international financial crisis every two years"); or shown him my evaluation of the causes of the crisis a paragraph but one above where he quotes (it reads: "the sudden change in [market] opinion [toward East Asia] reflects not a cool judgment of changing fundamentals [of East Asian growth] but instead a sudden psychological victory of fear over greed").

If I am the the point man waving the banner, all I can say is that the ranks of the army of international capital mobility must be thin indeed.

But since I have apparently been elected, let me pick up the banner and wave it around a few times, for on this issue I am what Jagdish Bhagwati calls a "liberal"-- someone who believes that we should neither encourage governments to choke off international flows of saving and investment (as Bhagwati thinks), nor look with schadenfreude on and discourse on the long-run salutory effects of the great depressions caused by international financial panics; but instead try to have our cake and eat it too: to reap the benefits of international capital mobility, and to minimize the human costs of recurrent crises through appropriate and well-funded international central banking institutions and practices.

We should try to have our cake because the potential benefits of international capital mobility truly are mammoth. Between 1994 and 1996 some $200 billion of international capital flowed into Malaysia, the Philippines, South Korea, and Thailand. In all of these countries the private return on investment is high--higher than in the industrial core. In all of these countries the social return on investment is higher still: if the economic history of the past two centuries teaches us anything, it teaches us that investments in modern machine technologies are a very good if not the best way to upgrade the skills of the labor force and gain the organizational expertise necessary for high total factor productivity.

This inflow of capital to these four countries was worth at least $15 billion a year and perhaps as much as $40 billion a year in higher GDP to the receiving countries even after taking account of the interest, dividends, and capital gains owed to investors from abroad. Just as the flow of finance from the British core to the periphery in the late nineteenth century played an important role in producing the Australian and North American economies that have had the world's highest standard of living in the twentieth century, so the flow of finance from today's industrial core to the NIC periphery has every prospect of cutting a generation or so off of the time needed for East Asian workers and consumers to achieve industrial core levels of productivity and economic welfare.

Calculations of the effect of international capital mobility on economic
welfare are considerably more complicated and uncertain than
calculations of the effect on growth, but they carry the same message: the ability to attract international capital to boost development or cushion
the costs of macroeconomic policy mistakes can be very, very valuable.

We should try to eat our cake too because the costs of unmanaged international financial crises are horrific. Because of the Latin American debt crisis of 1982 the decade of the 1980s was lost to Latin American development--leaving the typical Latin American country between five and ten percent poorer at the beginning of the 1990s than it would have been in a counterfactual world in which borrowing from abroad had not financed oil imports and elite consumption in the late 1970s. The financial crisis of 1873 saw the share of the U.S. non-agricultural labor force employed in building railroads fall from perhaps eight to perhaps two percent. And international financial crises turned the global recession of 1929-1931 into the Great Depression, generating not only a decade of relative poverty but the rise of the Nazi regime and the fifty million dead from World War II in Europe.

If there were no reasonable prospect of successfully managing international financial crises, then I would agree with Professor Bhagwati: the risks of an 1873 or a 1982 or--worst of all--a 1933 would then significantly outweigh the benefits of capital mobility. But there is every reasonable prospect of successfully managing international financial crises. The much-larger-than-anyone-anticipated Mexican crisis of 1994-1995--successfully handled--saw Mexican economic growth resume after a single year of recession. The East Asian crisis of 1997 may not even generate an absolute recession: as of this writing it looks as though East Asian GDPs will not decline, but instead that growth will pause in 1998 and resume in 1999.

But successful handling of international financial crises requires political and economic skill. It requires rejecting the arguments of the Wall Street Journal's editorial page that East Asia "needs" a deep, prolonged recession with mass unemployment to punish entrepreneurs and banks in NICs who overborrowed. It requires rejecting the arguments of Ralph Nader that East Asia "needs" a deep, prolonged recession with mass unemployment to punish New York financiers who overlent. And it requires rejecting the arguments of Jagdish Bhagwati that international capital mobility--good enough to finance the industrialization of the NICs of Australia, Canada, and the U.S. a century ago--is too risky for the NICs of today.

On Thursday, January 16, a $3.5 billion electronic funds transfer reached the Federal Reserve Bank of New York: the government of Mexico repaying the last piece of the more than $13 billion that it borrowed from the United States during the peso crisis of the winter of 1995.

President Clinton saw nothing but sunshine: "The United States is being repaid more than three years ahead of schedule. We have earned more than half a billion dollars on our loan. Our exports to Mexico are at an all-time high and the Mexican economy is back on track... The Mexican economy grew by over four percent.... The exchange rate has stabilized. Inflation has been cut nearly in half. Close to one million new jobs have been restored to Mexico since the crisis bottomed out. And Mexico has regained the confidence of international investors. This is a remarkable turnaround."

Clinton praised his own policies, that he said had successfully protected "a strong and growing market for American products that supports 700,000 jobs here. We helped Mexico to sustain its program of democratic reform and economic growth. And we helped to give the Mexican people renewed hope for a more secure future."

Reading this, I winced. I did not wince because the U.S. policy of loaning Mexico lots of money at usurious interest rates during the winter 1995 peso crisis was a mistake--it was not a mistake, but the more-or-less successful carrying out of the standard governmental role of preserving the liquidity and functioning of the financial system during an irrational financial panic. I winced because, once again, President Clinton was overpromising.

There is a lot that is not sunshine in Mexico's economy, and in U.S. economic relations with Mexico. Very bad things are likely to happen to Mexico's economy, and in Mexico's political system, in the next few years. Whether the Mexican government's decision to open its economy to foreign competition and privatize state-owned industries will be a success is still not clear: it ain't over 'til it's over--and it ain't over.

When the next big setback to Mexican economic growth or to Mexican political democratization occurs the yahoos, who think that we should (i) ban trade with Mexico (ii) build a big electric fence along the border (iii) close our eyes and (iv) hope Mexico goes away, will be stronger. They can ask what happened to the "remarkable turnaround" and to the "sustained program of democratic reform and economic growth."

Moreover, Clinton was overselling the success of U.S. government policies as well. He talked as though the U.S. government deserved an A for its actions in the Mexican peso crisis. But in fact the U.S. and other governments deserve, at most, a C+. The Executive Branch and the Federal Reserve deserve better grades, but the Congress and the governments of other leading industrial nations deserve worse. To paper over the history by which government actions during the peso crisis fell short of the mark is to make it more likely that we will repeat the same mistakes--and more likely that the next such crisis will have a much more unhappy ending.

It is safe to say that virtually no economists saw the peso crisis of early 1995 coming.

Economists divide countries into two groups: those with governments that are solvent--that are balancing or could easily balance their spending and their revenues--and those with governments that are not solvent--that have no way to balance their budget, and next to no ability to raise taxes in the future to pay off what they might borrow today.

Those in the first group are vulnerable to small-scale speculative attacks on their currency, should international investors think that the exchange rate is too high. For example, the fall of 1992, for example, speculators made a lot of money by betting that the British pound's exchange rate was too high, and selling pounds back to the British government as fast as they could. When the cost of buying back all these pounds became too high the British government cracked, devalued its currency, and the exchange rate settled into a new value roughly twenty percent lower. Such a small-scale speculative attack is an embarrassment for a government but not an economic disaster.

Those in the second group are vulnerable to much worse crises: crises in which not only do speculators bet that the currency is going to lose value, but in which virtually all domestic and foreign investors decide that they need to pull their money out of the country immediately. Such a large-scale crisis sees the currency lose more than half its value in a matter of days, sees interest rates rise to levels that bankrupt businesses and banks, sees a spiral of high inflation take hold, and sees a depression. Such a large-scale flight of capital is an economic disaster for the country.

Before 1995 Mexico was in the first group of countries. Its government budget was balanced. International investors had ample confidence in its long-run development prospects, and funnelled large sums of investment capital into Mexico--in spite of the political risks associated with the corrupt and authoritarian ruling Institutional Revolutionary Party. It would violate the canons of financial market rationality if--without very bad news about Mexico's government finances--the same investors eager to pour money into Mexico in 1993 were to suddenly decide that they had to stampede out, no matter how much of their investment they were to lose.

So much the worse for the canons of financial market rationality.

The same investors who had been eager to stampede into Mexico in 1993 stampeded out in the winter of 1995. And economists--including me--were left dumbfounded at the spectacle of a complete currency collapse and liquidity panic in a country that (our models had told us) should have been vulnerable only to a small-scale devaluation. In a matter of weeks Mexico became a country that could not borrow to refinance its government debt from anyone in the private market, at any price.

The consequences for the Mexican economy? Think of an upper-middle-class American family with a rolling balance of -$5,000 on its credit cards, no other liquid assets, and no prospects for a home equity loan. Think what would happen to it if suddenly its credit card company cut off further purchases and demanded immediate repayment. That's what happened to the Mexican economy in the winter of 1995. Perhaps one in ten jobs in Mexico disappeared.

When financial markets fall into such a panic where no one will lend because no one else is lending, the right thing is for some very large authority--some government, central bank, or international organization--to act as a "lender of last resort": put its own money up or guarantee repayment in order to keep the flows of capital going until the panic passes. When the panic passes, people are happy to make the loans and hold the securities that they would not touch before--just as private sector investors are happy to lend to the Mexican government today.

So when the Mexican panic hit in the winter of 1995, the Federal Reserve and the Executive Branch came up with a plan to provide loan guarantees to calm and reassure the panicked market. The United States government would say that should for any reason the Mexican government not repay its debts, the United States government would step in. So President Clinton introduced and Chairman Greenspan backed a bill for Congress to authorize guarantees. Such a law should have slid through the Congress quickly: both the new Speaker of the House Newt Gingrich and the new Majority Leader of the Senate Robert Dole had been strong supporters of the policy of economic engagement with Mexico. Their political fortunes were at stake should the Mexican economy collapse into chaos as well.

But then Gingrich's chief lieutenant, Majority Leader Armey, started dragging his feet. Right-wing Republicans like Patrick Buchanan began calling any assistance to Mexico "Goldman-Sachsanomics." Ex-consumer advocate Ralph Nader demanded that Congress vote against the loan guarantees. One House Republican, Zach Wump, came out of a briefing by Chairman Greenspan on the rationale for assistance crowing that this was "an issue made for talk radio." California Senator Feinstein, whose constituents stood to suffer the most from a severe depression in Mexico, complained to Treasury Secretary Robert Rubin that she did not see why she should vote for it. Even the--usually staunchly internationalist--core of the establishment media, the New York Times and the Washington Post, ran incoherent op-ed pieces denouncing loan guarantees.

Gingrich and Dole were soon missing in action: if President Clinton wished to provide assistance to Mexico by loaning out the Treasury's Exchange Stabilization Fund (never mind that that Fund had never been used in such a way before), they would not stand in his way. But they were not going to spend any of their political capital rallying support.

Such rapid erosion of support was horrifying. The United States government had been undertaking lender-of-last-resort operations to calm markets in financial crises since at least the Gold Panic of 1870. When such operations have not been undertaken--as when the Federal Reserve failed in its mission at the end of the 1920s, at the start of the Great Depression--the consequences have been disastrous. Yet this one-hundred and twenty-five years of historical experience did not register on the minds of the Congress, even though the seekers of the loan guarantees were the head of the Democratic Party, President Clinton, and the senior Republican in Washington, Chairman Greenspan.

So when U.S. assistance did materialize in the Mexican panic of 1995, it came later than it should have because several weeks had been wasted in the search for Congressional approval. And it came on a smaller scale than it should have, because it was limited to the amounts the Executive Branch could claim it could commit without Congressional authorization.

It is true that late is better than never, and that some is better than none. But late is still late. And some is not as good as more: the support package as implemented probably turned the Mexican Great Depression of 1995-1997 into merely a recession, but a somewhat larger package might well have saved another million jobs in Mexico, and turned the depression we had into a mere recession.

Perhaps the World War II battle of Dunkirk is a good analogy. At the Battle of Dunkirk it looked as though the Germans were going to capture 300,000 British soldiers and all their tanks, artillery, vehicles, and equipment. Instead the 300,000 soldiers got away across the channel, and all the Germans captured was their equipment, vehicles, artillery, and tanks. The British House of Commons cheered. But Prime Minister Winston Churchill pointed out that "wars are not won by evacuations."

Dunkirk was a SNAFU and a disaster--but not as bad a disaster as it for a time looked like it was going to be.

And Mexico? Mexico's unemployment rate has come down several points since it peaked in July of 1995. Mexico's exports are booming. The long-term benefits from the economic policy reforms remain because tariffs and non-tariff barriers to imports have been slashed, restrictions on foreign investment have been lifted, and state-owned enterprises have been privatized.

But many of Mexico's banks are still in danger of bankruptcy. Mexican inflation is still double-digit. Whatever recovery from the nadir of the depression in 1995 has been accomplished has not trickled down to the streets of Mexico City.

Moreover, Mexico's progress toward democracy is not assured. The Institutional Revolutionary Party is corrupt to the core and not eager to hand over power. And the underpaid police are vulnerable to bribery financed by the high prices U.S. consumers pay for drugs moving north.

Mexico's destiny is its own to make. It is wide open. There is an apocryphal story that Zhou Enlai was once asked his opinion of the consequences of the eighteenth-century French Revolution: his answer was, "It is too soon to tell."

Mexico's destiny is probably a better one as a result of recent U.S. policies. NAFTA has probably increased the odds that successive Mexican governments will continue to dismantle the structures of government control and political influence that have kept Mexico's growth far below what it might have been. Support for the peso in the 1995 panic did keep Mexico's liquidity crisis an economic misfortune, as opposed to an economic disaster. President Clinton is entitled to be happy that he did the right thing.

But "the Mexican economy... back on track"? "A remarkable turnaround"? A sustained "program of democratic reform and economic growth"? Renewed hope for a more secure future"? Nope. The future remains uncertain, and the risks immense. It would be more accurate to say that the Mexican reformers--and their supporters in the U.S.--have not yet crapped out. It ain't over 'til it's over--and it 'aint over.

Brad De Long


>So if Stanley Fischer is so smart, why did the IMF respond to Asian
>deflation with austerity packages, and to Russian collapse with austerity
>packages? Why is Russia committed to running a budget surplus now?

My Comment:

I think that the answer to your questions is an expansion of my point: that "Stan Fischer is a lot smarter than Montagu Norman... but Lauch Faircloth is not."

I think that our policy toward Russia is insane. I think that having spent $4 trillion in the generation before 1990 buying weapons to defend ourselves from the Commies, that any sane U.S. government would have been willing to spend a tenth that amount--$400 billion--on a Marshall Plan for the former Soviet Union. I think that a hundred years from now historians will judge that the greatest crime committed by Ronald Reagan was that his tax-cut rhetoric robbed the U.S. government of the ability to do the right thing at the end of the Cold War.

As to why the IMF responded to Asian deflation with austerity packages, I wish that the IMF had loaned a lot more money to East Asia on easier terms. But I don't see how it could have done so, at least not without a mammoth injection of new capital. And that is nowhere on the horizon.

I asked my patron (and the former chair of my dissertation committee Lawrence Summers) this same question: why isn't the IMF loaning East Asian countries twice as much for twice as long at lower interest rates with less conditionality?

His answer had two parts--

(i) That I must remember Mexico in 1995: how the IMF and the U.S. Treasury had tried to do more, and had been pulled up short by others. (I do remember Newt Gingrich trying to make administration acceptance of the Balanced Budget Amendment a precondition for even bringing up aid to Mexico for a vote; I do remember the British and German Executive Directors of the IMF registering strong opposition to the Mexican loan package; I do remember Principal Deputy Managing Director Fischer stating at a San Francisco Federal Reserve conference that most of the 6% fall in Mexican real GDP in 1995-1996 could have been avoided had the peso support package been the size that the U.S. Treasury and the IMF had originally envisioned.)

(ii) That, as he politely put it, "neither your friends on the left nor your friends on the right want to see a larger, better-funded IMF. Where is the political coalition to support increasing the IMF's resources going to come from?"

The IMF is not a central bank: it cannot create large amounts of purchasing power to pull countries (or the world) out of global recession. It can provide temporary injections of liquidity to (somewhat) soften the trauma when the world economy (and its own government) have just creamed some developing economy. But a kinder, gentler IMF would need to be a better-funded IMF.

And at the moment we are caught between Lauch Faircloth (who doesn't want to see a kinder, gentler, refunded IMF because such an IMF would reduce the amount by which East Asian workers suffer), and Ralph Nader (who doesn't want to see a kinder, gentler, refunded IMF because such an IMF would reduce the amount by which New York-based investors suffer). Moreover, we have people at the head of the Bank of Japan and the European Central Bank who may be as clueless as Montagu Norman was...

>broadly, why has the IMF been pushing deflationary policies - not just
>fiscal contraction, but the encouraging 120 poor countries to compete for
>the privilege of exporting to 20 rich ones - on the whole damn world at
>least since Mexico's 1982 default?

Exports to the first world are the best way to get the purchasing power to buy the industrial capital goods that embody so much modern technology--and those countries that have exported and used the proceeds to boost their investment rates have done much, much better than other developing countries over the past half century.

Of course, there are those developing countries that have boosted exports and used the proceeds to pay for elite vacations in Davos. Those have not done so well...

And most developing countries have not gone for export promotion: export promotion (usually) requires a relatively low exchange rate. The elite is much happier with controls on imports that keep the non-elite from buying from abroad, with the high exchange rate controls on imports generate, and with the ample ability to purchase foreign luxuries that a high exchange rate allows...

I agree that if the developing world as a whole actually listened to what the IMF economists said in their Article IV consultations--that if all developing countries tried to follow export-led development strategies--we would see exactly how elastic first-world demand for third-world imports is. But there are powerful domestic political reasons why only East Asia and Southwestern Europe undertook the export-led road to growth in the second half of the twentieth century. Those reasons remain. There are many places in the world today where the state is not an executive committee for managing the affairs of the business class, and there is no reason to expect economic growth in such places...


Brad De Long

Sign up for Brad Delong's (general) mailing list

Go to related links...

Add a comment on this page...

Read other people's comments on this webpage

Professor of Economics J. Bradford DeLong, 601 Evans Hall, #3880
University of California at Berkeley
Berkeley, CA 94720-3880
(510) 643-4027 phone (510) 642-6615 fax

This document:

Search This Website