America's banking crisis
Worse than Japan?
Feb 16th 2009 | WASHINGTON, DC
From The Economist print edition
Look carefully and the answer could be yes
Illustration by Claudio Munoz
THERE is one consolation for the depressing instability of finance. History offers a rich array of banking crises from which policymakers can draw lessons—and against which today’s rescue plans can be judged. According to an IMF database, there have been 124 “systemic” banking crises since 1970—episodes in which bad debts soared across the economy and much of the banking sector was insolvent.
Most of those were in the developing world. But the list also includes half a dozen rich-country crashes, from Japan’s slump after its property bubble burst in the late 1980s, to the Nordic bank crises in the early 1990s. All involved deep recessions, required massive government intervention to clean up bust banks, and led to big increases in public debt as economies shrank while government spending soared. But the speed of recovery differed dramatically; Japan endured a decade of economic stagnation, whereas South Korea returned to growth within two years of its 1997 banking disaster.
Received wisdom holds that policy choices determined the pace of recovery. Sweden rebounded quickly because it acted fast: removing dud assets from banks’ balance-sheets, recapitalising weak banks and nationalising where necessary. Japan stalled for a decade because it took years to recognise the scale of its mess. In his first presidential news conference on February 9th, Barack Obama pointed to Japan as an example of what not to do. Its “lost decade”, he argued, was the consequence of “not act[ing] boldly or swiftly enough”.
There is truth to that analysis. Japan’s central bank took too long to fight deflation; its fiscal stimulus was cut off too quickly with an ill-judged tax increase in 1997; and it did not begin to clean up and recapitalise its banks until 1998, almost a decade after the bubble had burst. But the history of bank failures suggests that Japan’s slump was not only the result of policy errors. Its problems were deeper-rooted than those in countries that recovered more quickly. Today’s mess in America is as big as Japan’s—and in some ways harder to fix.
Double bubble
This crisis, like most others in rich countries, emerged from a property bubble and a credit boom. The scale of the bubble—a doubling of house prices in five years—was about as big in America’s ten largest cities as it was in Japan’s metropolises. But nationwide, house prices rose further in America and Britain than they did in Japan (see first chart). So did commercial-property prices. In absolute terms, the credit boom on top of the housing bubble was unparalleled. In America private-sector debt soared from $22 trillion in 2000 (or the equivalent of 222% of GDP) to $41 trillion (294% of GDP) in 2007 (see second chart).
Judged by standard measures of banking distress, such as the amount of non-performing loans, America’s troubles are probably worse than those in any developed-country crash bar Japan’s. According to the IMF, non-performing loans in Sweden reached 13% of GDP at the peak of the crisis. In Japan they hit 35% of GDP. A recent estimate by Goldman Sachs suggests that American banks held some $5.7 trillion-worth of loans in “troubled” categories, such as subprime mortgages and commercial property. That is equivalent to almost 40% of GDP.
As the world’s biggest debtor, America headed into this bust in a very different position from Japan, a creditor nation rich in domestic savings. Nonetheless the macroeconomic trends in America look more like those that existed in Japan than other crisis-hit countries. Most big banking failures, from Sweden’s to South Korea’s, were created or worsened by a currency crash. Tumbling exchange rates raised the real burden of foreign-currency loans, forced policymakers to keep interest rates high and pushed up the fiscal costs of bank rescues. (Because of the rupiah’s collapse, for instance, the clean up of Indonesia’s failed banks in 1998 cost more than 50% of GDP.) But, by boosting exports, a weaker currency also offered a route to recovery. In Japan, by contrast, the yen stayed strong as the economy slumped, deflation set in and the banking problems grew.
In some ways America’s macroeconomic environment is even trickier than Japan’s. America may have a big current-account deficit, but the dollar has strengthened in recent months. America’s reliance on foreign funding means the risk of a currency crash cannot be ruled out, however. That, in turn, places constraints on the pace at which policymakers can pile up public debt. And even if the dollar were to tumble, the global nature of the recession might mean it would yield few benefits.
That is a big change. The most concentrated recent episode of systemic banking crises was the Asian financial crash in the late 1990s. The regional downturn was dramatic as firms slashed investment and paid back debt. But buoyed by strong demand in the rest of the world, particularly America, these countries’ exports soared, allowing a quick recovery. Even Japan eventually built its economic recovery on the back of booming exports. Today demand is falling rapidly across the globe and most big developed economies face simultaneous banking crises. With demand weak everywhere, the familiar route to recovery is blocked.
Beyond the banks
Administratively, today’s crisis is far more complex than it was in countries where the clean-ups are presently being praised. In Sweden’s highly concentrated banking system, one firm—Nordbanken—accounted for a quarter of all loans. The government dealt with a big part of the problem by taking over two banks. America’s finance industry is more diffuse. Even after a wave of government-induced consolidation, there are at least a dozen systemically important commercial banks.
More important, Sweden’s much-praised bad banks, into which the government shovelled troubled loans, dealt with straightforward credit backed by clear collateral. Even then the success was unusual. According to the IMF, asset-management companies were set up in 60% of banking crises, but were generally “ineffective”. That seems more likely today when the complexities of securitisation have left “toxic assets” that range from pools of car loans to fiendishly complex collateralised-debt obligations, which are much harder to unravel, value and manage.
What is worse, today’s bust is not just about banking. America faces twin financial crashes (as, to a lesser degree, do other Anglo-Saxon countries): one in the regulated banking sector and a simultaneous collapse of the “shadow banking system”, the universe of hedge funds and investment banks responsible for much of the recent securitisation boom as well as for the sharp rise in financial leverage.
As a result, standard measures of banking distress, such as the level of non-performing loans, understate the contractionary pressure. So far most of the credit collapse in America has come from the demise of securitisation. In 2007, for instance, $668 billion of non-traditional mortgages were securitised. Last year that figure dropped to $40 billion. Rapid deleveraging outside traditional banks also means that cleaning up banks’ balance-sheets may not break the spiral that is driving down asset prices and stalling financial markets. As the lower chart shows, financial-sector debt was the fastest-growing component of private-sector debt in recent years. Many of those excesses are being unwound at warp speed.
Illustration by Claudio Munoz
A final difference between today’s bust and most other big banking crises is the importance of household debt. Historically, serious banking busts have mainly involved overborrowing by firms. In Japan, for instance, corporate borrowing soared in the 1980s against the collateral of rising share and property prices.
Today, however, household profligacy, which underpins much of the other debt, has been the problem. After the dotcom bust, American firms held back. Virtually all the rise in non-financial debt since 2000 was among households, as Americans tapped into the rising equity in their homes. Although troubled business debts, such as commercial property, are rising, households are the worst hit.
That has important implications. Household balance-sheets are more difficult to restructure than corporate ones, which involve far fewer people. Politically, the process raises questions of fairness. How far, for instance, should taxpayers bail out reckless homeowners who bought mortgages they could not afford? On the other hand, the economic dislocation from unwinding a household-debt binge may be less disruptive than restructuring swathes of firms. As Anil Kashyap of the University of Chicago points out, one reason Japan was so loth to acknowledge the depths of its banking problems was the knowledge that a banking clean-up would require a large-scale restructuring of Japanese firms which, in turn, would throw many people out of work. Restructuring household debts may be political dynamite, but it would not require a wholesale remaking of corporate America.
Nonetheless, the rebuilding of American households’ balance-sheets is likely to force a reliance on government demand that is bigger and longer-lasting than many now imagine. In the aftermath of Japan’s bubble, firms spent more than a decade paying down debt and rebuilding their balance-sheets. This sharp rise in corporate saving was countered by a drop in the savings rate of Japanese households and, most importantly, by a huge—and persistent—increase in budget deficits.
A similar dynamic will surely play out in America’s over-indebted households. With their assets worth less and credit tight, people will be forced to save much more than they used to. The household saving rate has risen to 3.6% of disposable income after being negative in 2007. For much of the post-war period it was around 8%, and in the short-term it could easily exceed that. But, whereas dis-saving by Japanese households countered the corporate balance-sheet adjustment, American firms are unlikely to invest more while consumers are in a funk. Propping up demand may therefore require more persistent, and sustained, budget deficits than in Japan.
Add all this together and the ease with which American policymakers dismiss Japan’s experience is probably misplaced. Japan’s outcome—a decade in which growth averaged 1% a year and gross government debt rose by 80 percentage points of GDP—was not one to be proud of. But given the magnitude of today’s mess, it may soon seem not that bad after all.
Alıntı
Yer imleri