Alan S Blinder. Foreign Affairs. Volume 78, Issue 5. September/October 1999.
The economic conflagrations that lit up the world throughout the last half decade sent a very clear message: There are fatal flaws in the global financial architecture. The Bretton Woods system was designed for a very different world. The IMF, part schoolmarm and part firefighter, no longer plays either role well. Too often, it ignores the real victims and makes crises worse. The system must be redrawn to stabilize markets and head off panics before they spin out of control. Herewith a simple, eight-point plan for such reforms that uses existing institutions and respects current notions of national sovereignty.
Back To Bretton Woods
Financial crises once made most people’s eyes glaze over; they were subjects of intense interest to only a limited clientele, many of whom wore green eyeshades. Not any longer. The topic has unfortunately acquired a mass audience in the second half of the 1990s. Stunning currency collapses in Mexico (1995), southeast Asia (1997), Russia (1998), and Brazil (1999) have pushed the subject to the front page. Financial conflagrations have become too frequent, too devastating, and too contagious to be ignored.
As the World Bank’s chief economist Joseph Stiglitz has put it, when so many cars run off the road, you start wondering whether the road itself might be the problem. And indeed, many questions are now being raised about the global financial architecture. Much of the discussion centers around the concept of “moral hazard,” an awkward phrase that economists borrowed years ago from the world of insurance. In the financial context, it means that people (or banks, or governments) who are shielded from the consequences of their actions may take imprudent risks-hoping they will be bailed out if things go wrong.
But there is a vastly more important hazard of much greater moral urgency: the fact that financial crises afflict literally hundreds of millions of innocent bystanders who play no part in the speculative excesses but nonetheless suffer when the bubbles burst. The present global financial system manifestly fails to protect these poor people from extreme hazards. This failure is the chief reason to seek reform. Those who bet wrongly in financial markets should suffer losses. And borrowers should repay their debts, even when they are onerous. But citizens who take no part in the game should be shielded from the consequences of financial collapse to the maximum extent possible. This is plainly not happening now. How did we get into such an awful mess?
The story starts back in 1944, when the major Allied nations met at Bretton Woods, New Hampshire, to design a mechanism for restoring the shattered world economy to health. They created a new international financial architecture based on fixed exchange rates and the convertibility of the U.S. dollar into gold. And they established a new multilateral institution, the International Monetary Fund, to police the system.
The demise of fixed exchange rates in the 1970s robbed the IMF of its principal raison d’etre. But the fund morphed its mission several times and now has evolved into a global advice-and-rescue squad-one part wealthy benefactor, one part stern schoolmarm, and one part global firefighter. It lectures countries on economic orthodoxy, proffers financing in return for approved behavior, and rides dramatically to the rescue when countries fall prey to financial crises.
The fund has long had its critics, and their criticisms have grown louder and more frequent in the 1990s. Since the Mexican crisis erupted at the close of 1994, the IMF’s schoolmarmish advice and firefighting work have been attacked as wrong-headed. Some prominent voices-among them former Secretary of State and Treasury George P. Shultz’s-have even called for the fund’s abolition. That raises a key question: Have all the recent financial travails happened despite, or because of, the IMF? There now seems to be a widely shared consensus-even within the fund-that the current design is old and creaky and needs to be brought up to date. Unfortunately, the consensus ends right there.
How can we make the international financial architecture designed in 1944 work better in the fast-paced, high-octane world of 1999? The eight-point plan below focuses on broad principles of design rather than on the detailed bricks and mortar but not because the latter are unimportant. On the contrary, both God and the Devil (to merge the two cliches) will ultimately be found in the details. However, meaningful debates over numerous important issues must wait until we first agree on basic principles to govern the new financial system. The physics must precede the engineering.
The Art of the Possible
A logical way to begin is by setting the goals of international financial reform. What do we want to accomplish? We cannot expect to design an international financial system that totally eliminates risk. Manias, panics, and speculative excesses are inherent in free-market capitalism. Speculative markets have always been subject to alternating waves of greed and fear; they go to extremes, exhibit herdlike behavior, and probably always will. All this is part of the price we pay for vibrant, highly productive capitalism-a necessary blemish on an economic system that has produced a dazzling record of achievement.
But we should recognize the blemish for what it is. The current system breeds too many crises that are too severe. Much of the nonWestern world has suffered through one or another such malady in the past few years. What the global economy needs now is a way to minimize the frequency and intensity of crises. Moreover, the present system lets financial epidemics blossom into global pandemics. Our new financial architecture should have sturdier defenses against contagion. Perhaps most critical of all, but least mentioned, the world badly needs a system that protects innocents from the financial hurricanes that sometimes swirl around them.
But how? Grandiose schemes involving elaborate international institutions tend to get nowhere fast. Regardless of its merits, no world central bank will be created in my lifetime. Even establishing a global financial super-regulator, a far more modest step, would require an implausible surrender of national sovereignty. And an international bankruptcy court would require coordination among nations that have very different bankruptcy laws. Better to stick with more modest plans that require little or no institution-building. The world’s poor cannot wait for grand edifices to be built.
The plan outlined here can be accomplished through two channels: reform of the IMF, and changes in national practices that countries can implement on their own (indeed, some are already being implemented). The recommendations divide neatly into two categories, corresponding roughly to the IMF’s roles as schoolmarm and firefighter.
Four Ounces of Prevention
The first four recommendations focus on policies designed to build a financial system with fewer vulnerabilities-including both policies that sovereign nations should adopt on their own, and advice the IMF should dispense to its client states. As will be seen, parts of the schoolmarm’s catechism need to change.
Principle 1: Don’t fix your exchange rate. This point seems the most fundamental, but was curiously de-emphasized until former Treasury Secretary Robert E. Rubin more or less endorsed it at an April 1999 IMF meeting. All the international financial crises of the 1990s, like most of the crises of the 1970s and 1980s, have shared a common element: a fixed exchange rate that crumbled under speculative attack. Indeed, in the cases of Mexico, southeast Asia, Russia, and Brazil, the defense of an exchange rate pegged at an untenable level was at the very heart of the crises. The lesson seems painfully obvious: Fixed exchange rates, being hazardous to a country’s economic health, should be avoided. This is not an ideological position founded on religious devotion to free markets. It is, rather, based on pragmatic observation of many painful experiences. Nor do I insist on purity-“dirty” or managed floats are just fine. Indeed, even some versions of exchange-rate fixity may make sense under the right circumstances. But floating rates ought to be considered the norm by the IMF of 1999, just as fixed rates were in 1944.
I do not believe in a “one size fits all” exchange-rate policy. Particular countries at particular times may have good reasons to peg their exchange rates to a hard currency. For example, Brazil’s crawling peg and Argentina’s currency board (which guarantees the convertibility of pesos into dollars) helped rid those nations of hyperinflation in the 1990s. Hong Kong’s currency board helped restore confidence that was badly shaken in the 1980s by the prospect of Chinese rule. Some small countries dominated by a big neighbor may deem it either foolhardy or hopeless to have an independent exchange rate. And fixed rates may be necessary for other reasons; remember the European Exchange Rate Mechanism, a stepping stone to monetary union. But these should be viewed as exceptions to a general rule. As global financial markets grow bigger and more fluid, the viability of fixed exchange rates that are not literally locked in place forever (as in a monetary union) diminishes accordingly. Market forces are simply too powerful relative to the resources at central banks’ command. In this important respect, we need to recognize that the world of 1999 is quite different from the world of 1944. Just as fixed exchange rates were the linchpin of the financial architecture designed at Bretton Woods, floating rates should be the accepted norm in the new financial architecture.
Any nation that decides to peg its exchange rate will be well advised to have an exit strategy-and to use it at a propitious moment before speculators take aim. Forced devaluations can be disastrous. Think how much easier and smoother the necessary adjustments would have been if the Mexican peso, the Indonesian rupiah, and the Brazilian real had floated down gradually over a period of months or years rather than dropping abruptly. As George Soros (among others) has observed, with no fixed exchange rate for a target, speculators have nothing to shoot at.
Principle 2: Borrow less in foreign currency. A second common element in the financial crises of the 1990s-with the possible exception of Brazil’s-has been an excessive amount of borrowing in foreign currency, especially for the short term. This extremely risky practice should be actively discouraged by both the IMF and national governments.
When combined with an allegedly fixed exchange rate, such borrowing makes for an especially volatile-and sometimes toxic-brew, a “fixed exchange rate bubble.” It works like this: banks and corporations in emerging markets, where local interest rates are comparatively high, find the lure of lower U.S. (or German, or Japanese) rates irresistible. So they borrow in, say, dollars and then either lend in local currency or invest in local assets.
That is precisely what the Mexican government did in 1994, and what many southeast Asian banks and businesses did throughout the 1990s. When the fixed exchange rate collapsed, they found themselves buried under an avalanche of unpayable debt. Worse yet, millions of innocent victims fell with them as the economies crumbled under the weight of all that debt.
Critics will object that borrowers in emerging markets cannot afford to pay expensive local-currency interest rates. They must borrow at lower rates in dollars, some claim, or else development will be retarded. I disagree-on at least three grounds.
First, in the case of southeast Asia (though not Mexico or Russia) much of the foreign borrowing was not necessary for development. Each of the Asian “tigers” generated an enormous volume of domestic saving relative to the size of its economy. The influx of foreign capital seeking quick returns just “boomed the boom,” leading to overcapacity and to speculative bubbles in equity and real estate markets.
Second, the risk premium for borrowing in home currency may not be all that onerous if a country’s fundamentals are reasonable, a liquid market is created, and the exchange rate regime does not present market participants with a one-way bet by signaling to speculators that the currency can only go down. For example, markets have recently been charging South Africa only roughly three percentage points more to borrow in rand rather than in dollars-not an outlandish premium.
Third, and most fundamental, pretending that borrowing a U.S. dollar was just another name for borrowing, say, 25 Thai baht or 2,500 Indonesian rupiah was a cruel hoax that led to excessive borrowing and inappropriate allocation of risk. By borrowing dollars and lending baht (or making loans collateralized by baht assets), Thai banks and finance companies took on much more risk than those bargain–basement dollar interest rates indicated. The higher baht interest rates, which included a market-determined premium for exchange-rate risk, indicated risk more accurately. Had Thai companies been forced to pay those higher rates, they would likely have borrowed much less-which would have been a good thing. As it was, when the exchange rate finally tumbled, millions of innocent Thai citizens were left holding the bag. Similar scenarios played out in Indonesia, South Korea, and Malaysia.
The point is that borrowing in dollars does not eliminate foreign exchange risk-it just forces the wrong people to bear it. Wouldn’t it have been better if, say, hedge funds and international banks had lent the money to Thailand in baht, rather than in dollars, thereby assuming the exchange rate risk themselves? They would have charged a market-determined fee for doing so, of course. But that would have better reflected the actual risks rather than covering them up. Of course, no one can force financial market participants to lend to emerging markets in local currencies. Thus this principle constitutes a policy recommendation only in the sense that national and international supervisory agencies (like the IMF and the Bank for International Settlements) should cast a jaundiced eye on unhedged foreign-currency borrowing-and place handicaps, like lower supervisory ratings and higher capital charges, on banks that do so. National governments should also shift their sovereign borrowing toward home currency, rather than dollars.
Principle 3: Don’t rush to open capital markets. Unfettered international capital mobility is not the best system for all countries. I do not recommend the cessation of global capital flows, nor emulation of Malaysia’s heavy-handed approach to restricting them. And I certainly do not favor protectionism in the financial service industries. Chile sets a fine moderate example by being open to foreign banks while imposing prudential controls on capital inflows rather than outflows. Other countries may find different ways, tailored to their own unique circumstances, to slow down the flow of international hot money. But the hard-core Washington consensus-which holds that international capital mobility is a blessing, full stop-needs to be tempered by a little common sense.
This finally seems to be happening. But only two years ago, the IMF was well on its way toward requiring full capital account convertibility of all its member states. During the Asian crises, the IMF saw open capital accounts as part of the solution, rather than part of the problem. I found that attitude badly misguided, and it pains me to admit that the U.S. government was a primary pusher of this bad advice. The fund has now backed away from its former position, but not far enough.
The problem is one of proper sequencing. Too many emergingmarket countries rush to open up their capital markets too soon—before, for example, they have proper supervisory structures in place. Full openness to international capital flows is a fine long-range goal. But the IMF should not encourage, much less require, premature liberalization. Nations should not adopt America’s bad financial habits until they are rich enough to afford them. Malaysia and Indonesia clearly were not.
Principle 4: Follow sound macroeconomic and financial policies. I have saved the platitudes for last-not because I disagree with them, but because they have been said so many times before. Most contemporary suggestions for international financial reform focus on urging countries to follow “sound” fiscal and monetary policies, to avoid large current– account imbalances, to develop strong and competent financial regulatory and supervisory structures, and to adopt various codes of good conduct in such areas as transparency, accounting standards, and bankruptcy law.
Most of the items on this list are uncontroversial in principle. Who, after all, favors unsound macroeconomic policies, opaqueness, and incompetent regulation? But things become highly contentious once you get down to specifics. I will therefore simply endorse the general approach-noting only two things. First, I would assign top priority to two of the most boring topics on the standard list: bank supervision and accounting standards. Second, although transparency-which is the current rage-is all to the good, no one should expect it to accomplish very much in the way of crisis prevention. Bubbles form and burst even in extremely transparent markets like the New York Stock Exchange.
Redesigning the Fire-and-Rescue Squad
No matter how sound the new policies, accidents will happen and financial crises will occur. When they do, the IMF must remain the world’s designated fire-and-rescue squad. I see no need to reassign the task and certainly would not want the IMF abolished. If that were done, the world would soon reinvent a similar institution from scratch. But the IMF’s modus operandi needs significant change. Critics charge that the fund has sometimes aggravated crises rather than mitigating them. Although exaggerated, such criticisms contain important elements of truth. So what should IMF crisis-management teams do differently?
Principle 5 Austerity is not always the right medicine. One common criticism holds that no matter what the problem, austerity always seems to be part of the IMF’S proposed solution. As a former central banker, I well understand that fiscal and monetary orthodoxy have their place. But that place is not every place. For example, tighter budgets and slower money growth helped much of Latin America in the 1980s, when irresponsibly large budget deficits financed by money creation had fueled inflation. But the fund was wrong to prescribe austerity in southeast Asia in 1997, when neither budget deficits nor high inflation were part of the problem. Indeed, the IMF itself has all but acknowledged this error.
The standard rationale for imposing fiscal and monetary stringency is, of course, to defend a fixed exchange rate-just as was done under the old Bretton Woods system and the gold standard before that. The IMF regularly answers critics of its tight monetary policies by scoffing at the notion that lower interest rates will help defend a currency from speculative attack. They won’t, of course, but that is precisely the point. If currencies were allowed to float, there would be no pegged rates to defend, and hence much less need for crushingly high interest rates. Furthermore, killing a nation’s economy hardly seems the best way to bolster confidence in its currency. It didn’t work very well in Asia.
It is high time the IMF recognizes that the global macroeconomic situation in the late 1990s differs fundamentally from conditions in the 1970s and early 1980s. In particular, inflation-which was the bane of the 1970s and early 1980s, and the other rationale for austerity-is no longer a problem. Instead, a worldwide shortage of aggregate demand has emerged as the world’s premier macroeconomic malady. Programs that force austerity everywhere aggravate this problem rather than ameliorating it. In a world with floating exchange rates and low inflation, fiscal and monetary austerity ought to be prescribed far less. Not never, just less frequently than is the IMF’s current habit.
Principle 6: Devote more resources to protecting innocent bystanders. As I stated at the outset, the new financial architecture needs to give greater weight to developing and strengthening the social safety nets that shield innocent bystanders from the fallout of financial crises. This idea is not alien to the IMF’s way of thinking. But neither is it central. The fund pays inadequate attention to the protection of innocents-compared, say, to the protection of creditors who may have made ill-considered loans.
The IMF, we are assured, always strives to preserve social spending. But let’s think realistically about what happens to a country in crisis. Mexico, Brazil, or any of several Asian countries can serve as examples. The government sees its tax receipts falling due to a recession. The interest rates it must pay on its outstanding debt soar. In addition, it will probably have to shoulder the budgetary burden of a major banking bailout. Now the IMF team arrives, demanding that the overall budget deficit be cut. How, in these circumstances, can the country increase transfer payments to its poor and its unemployed? The answer is obvious: It probably cannot, and so the needy suffer from budget austerity.
A reformed IMF, working in conjunction with the World Bank and regional development banks, should ensure that foreign creditors are not bailed out while local populations drown.
Principle 7: Agree on some procedures for orderly debt settlement. Financial crises typically mean that some beleaguered entities-be they governments or private businesses-cannot pay all their debts. When this occurs, an obvious question arises: Who will get paid, and how much? In the aftermath of the 1994-95 Mexican crisis, the G-10 deputies issued a report that suggested, among other things, that some sort of orderly workout procedure for resolving conflicting claims might be preferable to the current system, in which each creditor grabs whatever he or she can get. The underlying idea was the same one that motivates domestic bankruptcy laws: Assets lose value in a mad scramble, so a more orderly procedure for apportioning losses might leave both creditors (as a class) and debtors better off. Furthermore, the IMF does not have the financial means to ride to the rescue time and again.
Unfortunately, the private sector was singularly uninterested in any such proposal in 1996, and the idea died. But I was glad to see one of last fall’s reports by the G-22 (a committee of nations working to strengthen the international financial system) revive this suggestion-with a greater sense of urgency. By now, the world has experienced several additional financial crises, each with its own disorderly scramble for an inadequate pile of assets. One might hope that these unhappy experiences, plus the realization that the IMF has limited funds, would have convinced some private creditors that orderly workout arrangements are preferable to the status quo. But, alas, this does not appear to have happened.
There are many ways to solve this problem. Collective action clauses in bond contracts-which, for example, might allow bondholders’ committees to make decisions about restructuring-are one. The clever proposal by Willem Buiter (a member of Britain’s Monetary Policy Committee) and Anne Sibert (an economist at Birbeck College in London) for mandatory rollover of debt, but at a penalty rate, is another. The private sector must be intimately involved in designing this aspect of the new architecture. Debt workout provisions should not be dictated by governments against the wishes of private investors, lest credit markets dry up. But national governments and the IMF can and should push things along. For example, G-7 nations could lead the way by incorporating collective action clauses in their own sovereign bonds.
Principle 8. Prevention is better than cure. Recent economic history amply demonstrates that, once a country has been devastated by a financial crisis, there are no good options. The economic and social costs become enormous; it is devilishly difficult to restore growth. It is therefore imperative to find a way to deter speculative attacks on countries with sound fundamentals.
Under the traditional approach, rich countries and the IMF come to the rescue only after a poorer country has collapsed in financial ruin. Thus much of the house burns down before the fire brigade arrives. But the financial world of 1999 moves to a much faster beat than did the world of 1944, or even 1973. So the new global architecture should offer protection to countries that follow sound policies, before they collapse. One good way to do so would be to put enough international money on the table in foreign reserves to scare speculators away. If the idea worked flawlessly-which, of course, it will not-no money would ever be spent. That is the essential idea behind the IMF’S newly instituted and experimental contingent credit lines (CCLS).
Unfortunately, implementing this idea effectively is extremely difficult. Among the thorny issues to resolve are, what criteria will be used to decide which countries prequalify for international assistance and which do not? What happens to the countries that fail to make the grade-are they just left to the financial wolves? And what happens when a country that was once prequalified slips and falls below the standards? Wouldn’t defrocking it invite speculative attack?
Although many details need to be worked out, I offer the following two-part answer to these questions. First, the principles discussed earlier-including sound macroeconomic and financial policies, a flexible exchange rate (or very good reasons to fix it), and little borrowing in foreign currency-are a good place to start compiling a list of IMF “dos” and “don’ts.” Indeed, the IMF’S recently published guidelines for the CCLS embody such a list already.
Second, prequalification should not be treated as an either-or question, as the fund has been doing. Finer gradations need to be made. The most worrisome aspect of the prequalification idea is that markets might react violently when a country crosses the line from being prequalified for IMF assistance to being unqualified. But no such bright line exists in well-functioning credit markets. When they operate smoothly, financial markets gradually impose higher interest rates and stricter terms as a borrower’s creditworthiness deteriorates. Only when panic sets in-which, unfortunately, seems to happen all too frequently in emerging markets-do markets resort to all-or-nothing evaluations. The IMF’S prequalification plan should emulate orderly markets, not panicky ones.
How can this be done? As part of its regular surveillance procedures, the IMF could classify member nations into several categories-much as rating agencies do now. The ratings would be based on the sorts of criteria discussed earlier. In case of a contagious crisis, countries in the highest class would have unquestioned access to sizable amounts of IMF lending, with few if any conditions, at near-market interest rates. The next highest class would face less credit availability, more conditions, and higher rates-and so on down to the lowest-rated class, which would have no guarantee of support. They would have to apply to the IMF in duress, as at present.
Although there are difficulties with such a scheme, perfection is not required. Since a workable prequalification system might prevent some crises and mitigate others, the potential gains from getting one in operation are large enough that the details-difficult as they may be-should not be allowed to stand in the way. The billions of innocent victims of future crises, many of them desperately poor, should not be asked to wait until the perfect design is found.
Eight is Enough
It would be conceit to claim that all eight principles of my suggested financial architecture must be held together as a package. In truth, it is possible to mix and match them-and the first two (floating exchange rates and less foreign-currency borrowing) are far more important than the rest.
But the list has a certain internal integrity: one principle supports and reinforces another. For example, borrowing in foreign currency would be both less alluring and less dangerous if exchange rates were allowed to float. Less-open capital markets in developing nations and a workable IMF prequalification plan should reduce the incidence of crises. Floating rates and the IMF reforms suggested here would reduce the need for fiscal and monetary austerity-and that would, in turn, ease the plight of the poor and unemployed, reducing the strain on social safety nets. Orderly workout procedures and more social spending would limit the economic devastation when crises hit. And so on.
None of the eight principles is revolutionary; you have probably read all of them (or variants) before. But together they constitute an agenda that is ambitious yet eminently achievable. And if instituted, they will make the financial world a far safer place-helping not just market players, but the billions of people who unknowingly depend on them and suffer from their failures.