A quarterly magazine of the IMF

March 2009, Volume 46, Number 1

Viewpoint

The Shape of Things to Come
Brad Setser

National decisions, not international summits, will remake the global financial system

After Asia's financial crisis, the world's leading economies launched a major effort to remake the international financial system. Ten years later, they decided to try again. The 1998 effort to revise the world's "financial architecture" followed a crisis that had originated in the unwinding of the external deficits in the emerging world—deficits that were for a time willingly financed by banks and private investors in the world's wealthy economies. The second effort will follow a systemic financial crisis that started in the United States, spread to European banks that had borrowed dollars to buy U.S. securities, and then infected most of the world economy.

A downturn in U.S. home prices that led to large losses at the large banks and broker-dealers triggered the current crisis. But the household deficit of the United States, the United Kingdom, and many euro area economies couldn't have been financed for as long and at as low a rate without an unprecedented increase in the assets of the emerging world's central banks and sovereign funds. Private investors were never that keen on financing large deficits in the slow-growing United States; they wanted to finance the fast-growing emerging world.

The 1998-99 effort never quite lived up to its name: "architecture" suggested building new institutions or at least remodeling existing institutions for international economic and financial cooperation. That clearly didn't happen. What emerged instead was a host of suggestions to help emerging economies reduce their vulnerability to sudden swings in capital flows—along with new "restructuring" clauses in international sovereign bonds governed by New York law and new IMF lending facilities designed to help countries facing crises stemming from sudden swings in capital flows.

At the same time, the global financial system that emerged from the last crisis was fundamentally different from the financial system that existed before the crisis. A world where unprecedented growth in the foreign assets of emerging economies' central banks helps finance a large U.S. current account deficit at low rates is not the same as a world where private investors in the United States finance deficits in the emerging world. Neither the IMF nor the G-7 changed all that much. But the world around them did.

Decisions, decisions

The most important lesson from the past is that the international financial system is defined more by the decisions key countries make during and after a crisis than by carefully chosen communiqué language. The architecture for responding to "capital account"crises emerged from the U.S. decision to lend large sums to Mexico when it couldn't refinance its dollar-denominated bonds in 1995, the IMF's subsequent decision to take the lead in providing large-scale financing in the Asian crisis, and the conditions the IMF attached to its loans to Asia. The G-7's Koln Communiqué—which explicitly tried to lay out the G-7's vision for the financial architecture—didn't define the world's exchange rate regime. That emerged from the collapse of Argentina's currency board, the success of Brazil's managed float, the persistence of currency boards in many Eastern European economies, the Gulf 's ongoing dollar peg, and—above all—China's decision to maintain its link to the dollar even after the dollar started to depreciate in 2002. The regulatory regime was defined as much by the decision not to rein in the shadow financial system—the largely unregulated or lightly regulated institutions that came to play much the same role in the economy as banks—as by the work of the Financial Stability Forum.

Three additional lessons from the 1998 architecture debate—and subsequent debates on the global framework for preventing and managing international financial crises—are worth remembering.

Getting the right group of countries around the table doesn't guarantee results. Real change happened when there was a broad consensus among relevant countries on the nature of the needed reforms. The most obvious example is that advanced and emerging economies agreed that emerging economies could reduce their vulnerability to crises by holding more reserves and replacing debt denominated in foreign currencies with debt denominated in local currency. The governments of most emerging economies proved far more able to finance themselves with debt denominated in their own currency than many expected 10 years ago—helped by a growing sense that most emerging market currencies were undervalued. Don't doubt the magnitude of this change. A world where Russia's government enters a global downturn with $10 billion in foreign currency reserves and $140 billion in foreign currency debt is quite different from a world where Russia's government enters a downturn with $600 billion in reserves and only $35 billion in external-currency-denominated debt. Russia's current vulnerabilities are real, but they aren't found on the government's external balance sheet. The balance sheets of the governments of other large emerging economies—Brazil, for example—also look nothing like they did in 1998.

In other areas, real consensus proved elusive. Getting the right countries around the table was rarely the main problem. The biggest difficulty was that the key countries didn't agree—and saw no need to reach agreement absent the pressure of a crisis. The G-7, for example, was arguably the right group to discuss when the IMF should lend large sums to emerging economies. The G-7, however, was not willing to abandon the option of providing countries facing a run on their currencies—or countries unable to roll over their maturing external debts—with large amounts of front-loaded financing nor willing to recognize that the IMF's old access limits were no longer the norm for major emerging economies. Real differences over the desirability of continuing to allow emerging economies to borrow large quantities of reserves to help manage large swings in capital flows were papered over; hard decisions were left for the next crisis.

The G-7 made an effort to broaden the discussion of exchange rate regimes to include the emerging world: even in the 1990s, it was clear that the global economy couldn't be reduced to the United States, Western Europe, and Japan. But broadening the dialogue to include the emerging world—through forums such as the Group of 22 and then the Group of 20 (G-20)—didn't bring coherence to the world's exchange rate regimes. The dollar floated against the euro and—most of the time—the yen. The pound sterling and the Canadian dollar floated against the currencies of their larger neighbors. But the enormous acceleration in the growth of emerging market reserves from 2002 onward belied any notion that the monetary financial system was defined by independent, inflation-targeting central banks that let their currencies float against each other to preserve their monetary autonomy.

Don't ignore tough problems . . . One of the key sources of pressure on the foreign exchange reserves of major emerging economies in the 1997-98 crisis was a reduction in cross-border bank lending. Only in Argentina's crisis were payments on international sovereign bonds a major drain on the country's balance of payments—and even in Argentina, a domestic bank run was a bigger source of capital outflows. Yet the most animated debates that emerged from these crises focused on removing legal impediments to the restructuring of international sovereign bonds. Limiting outflows associated with short-term bank credit—or, for that matter, monitoring the risks associated with a large rise in bank lending to the world's emerging economies—didn't get comparable attention.

But it should have. Policymakers assumed that emerging economies would finance themselves through the sale of traded securities rather than by borrowing directly from the world's banks. But at the end of the second quarter of 2008, cross-border bank lending to the emerging world (largely to private banks and firms) totaled $1.2 trillion, a sum that easily exceeds the outstanding stock of international sovereign bonds. The roll-off of cross-border bank exposure will prove to be a larger source of pressure on emerging economies in the current crisis than maturing international sovereign bonds. Ignoring a difficult problem doesn't necessarily mean it will go away.

. . . but also challenge your assumptions. It is always easier to highlight the reemergence of old vulnerabilities than to imagine new risks. It is striking that the "architecture" debate focused almost entirely on the risk that a financial crisis in a single emerging market economy could spill over to other emerging economies and then to the global economy. The risk that financial trouble in an advanced economy might prove far more destabilizing to the emerging world than financial trouble in another emerging economy was never seriously considered.

The G-7 put pressure on the IMF to look more closely at balance sheet, not just fiscal, vulnerabilities in emerging economies. No comparable push was made to evaluate whether the U.S. and European financial sectors were too exposed to the household sector. Yet emerging economies' complaints about the procedural inequities of IMF surveillance miss an even more important point: there is little evidence that more intensive IMF surveillance would have made a difference. In 2007, the IMF extolled the "highly innovative" role of U.S. financial markets in "supporting capital inflows," arguing that the U.S. edge in creating complex financial products helped pull in the funds needed to sustain large U.S. external deficits. Its Article IV report noted that "core commercial and investment banks are in a sound financial position, and systemic risks remain low," though it did concede that "financial innovation has complicated risk assessment at a time of higher risk taking and deteriorating lending standards in some sectors." We all make mistakes, but that wasn't exactly a Roubiniesque warning of building risks.

With the benefit of hindsight, it is striking how many key issues of the past few years were left off the architecture agenda. Ways to help emerging economies manage volatile capital flows were discussed, but not ways to help manage volatile commodity prices. The modalities of restructuring sovereign bonds were discussed, but not the challenges of restructuring mortgage-backed securities—or mortgages denominated in a foreign currency. The need to limit the IMF's lending to emerging economies was discussed endlessly, while the risks of excessive self-insurance were ignored. And there certainly was no discussion of how advanced economies should manage the risks associated with increased demand for their debt from emerging economies looking to raise their stock of reserve assets—including the risk that the emerging economies' desire for reserves might distort financial markets in the advanced economies and mask the impact of large fiscal and household deficits.

Hopeful signs

Will the current effort to remake the international financial system succeed?

This crisis is still in its early stages. Past experience suggests that national decisions, often made under extreme distress, will do more to define the shape of the world's future financial system than international summitry.

But it is still important to try to forge a global consensus on the kinds of changes that are needed in the international financial system. Some signs are encouraging. Key emerging economies have been brought to the table in a new way. The G-7 lives on—but the annual G-20 leaders summit looks likely to attract far more attention than the G-7's annual summer retreat.

The crisis itself has already remade the architecture of the U.S., U.K., and European financial systems—with governments playing a far larger role in financial intermediation than in the past. It also looks certain to produce large changes in the regulatory structure.

Unless too-large-to-fail institutions are broken up, it would be a mistake to rely on credit markets to discipline them. In the short run, containing the current crisis has to take precedence over avoiding the next one—and right now forcing institutions to hold more capital would be self-defeating. Too much leverage has given way to too rapid deleveraging. Over time, though, large institutions need to hold larger buffers of capital and liquidity.

This is work that the world's mature economies have to do themselves. But emerging economies should insist that they do this job well. The emerging world's interest in well-regulated institutions at the core of the global financial system is clear: Lehman's collapse generated a bigger funding crisis for many emerging economies than Russia's default.

Less hopeful signs

Better regulation isn't all that is needed to help create a stronger basis for global growth. Three additional issues stand out:

• coordination of macroeconomic stimulus,

• the evolution of the world's exchange rate regime, and

• strengthening global institutions for crisis lending.

And in each area, it isn't yet clear that there is a real consensus for change. At the peak of the housing boom in the United States and Europe, large household deficits in some key mature economies offset the emerging world's surplus. But that boom could last only as long as the already heavily indebted household sector took on more debt. When households gave out, governments had to step in with large stimulus packages. But—as Martin Wolf of the Financial Times tirelessly points out—there is a risk that overindebted households will lead directly to overindebted governments.

The obvious conclusion is that countries with lots of debt and large deficits shouldn't do all the heavy lifting to support global growth. The big-surplus countries also need to do their part—and not just rely on the spillover from large stimulus packages in the deficit countries. Otherwise the next systemic crisis could easily come from loss of confidence in the public sector balance sheet of an advanced economy with a large external deficit.

A second issue—exchange rates—wasn't even mentioned in the communiqué that emerged from the first meeting of the G-20 leaders. That isn't encouraging; it suggests unwillingness to discuss the key issues facing the world economy. The Fed's decision to cut U.S. interest rates after the dot-com crisis could have led to a weak dollar and a boom in exports, not a boom in residential investment. But key countries followed the dollar down, limiting the dollar's overall depreciation. The end result of the combined depreciation of the dollar and the renminbi against the euro was China's larger trade surplus and more Chinese financing of the United States—not a smaller U.S. trade deficit. U.S. regulators looked the other way as households took on large amounts of debt—and key financial institutions kept profits up as margins fell by levering up (that is, by borrowing to buy more assets). But it is hard to see how the vulnerabilities in the household sector of the United States could have been allowed to build for so long absent large inflows from the world's central banks.

The integration of the large emerging markets into the world economy is bound to be complicated if their exchange rate regimes differ dramatically from the exchange rate regimes of the world's other large economies, especially if the currencies of countries with large external surpluses are tied to the currency of the country with the world's largest external deficit. The IMF has concluded that reduction of the world's imbalances likely implies depreciation of the currency of the deficit country—and it isn't clear whether the currencies of the big-surplus countries should fall too. It also isn't obvious that most oil exporters should continue to peg to the dollar. Too often the dollar has gone down when oil has gone up and gone up when oil has gone down.

Finally, the IMF lacks sufficient resources to stabilize the global financial system—or to provide a large enough pool of shared reserves to offer a real alternative to national self-insurance (or large bilateral swap lines). Many emerging economies are likely to conclude that maintaining their own financial stability in the face of huge swings in exports, commodity prices, and capital flows requires almost unimaginably large reserves. Emerging economies with trillion-dollar GDPs and $200 billion in reserves now aren't sure that the IMF's $200 billion ($350 billion counting the IMF's credit lines, including a new one from Japan) is enough. Russia started the crisis with close to $600 billion—and that won't be enough if its reserves continue to fall by $100 billion a quarter. Recent swings in global capital flows have been extreme—with record net (annual) private inflow to the emerging world turning into large net outflows in the span of a single quarter. Swings in commodity prices—and associated export revenues—have been no less extreme. A $200 billion Fund is too small to be relevant for any major emerging economy.

Tomorrow's world

Today's crisis is hitting all parts of the global economy hard. Countries that relied on private capital inflows to cover external deficits are suffering from a sudden stop in capital flows comparable in magnitude to the sudden stop that marked the 1997-98 crisis. Countries with current account surpluses that didn't rely on ongoing capital inflows are not faring any better: commodity importers are struggling alongside commodity exporters amid an extraordinary contraction in global trade. Something clearly has gone wrong. The need for a more robust financial system—one less prone to finance excessive deficits in mature and emerging economies alike and also less prone to sudden reversals and sharp crises—should not be in doubt.

Anticipating future risks is hard. But the scorching experience of the past few months ought to push all countries to try harder—and think carefully about the kind of world economy they want to see emerge from the current crisis. One suggestion: don't rely on large external deficits in the United States to drive demand growth globally, or solely on U.S. deficits to supply the world with large quantities of reserve assets. That hasn't worked.