◆Paradise lost
經濟學人雜誌 2008年5月18日

Banks are bound to fail from time to time. But does the fallout have to be so painful?

WILSON ERVIN, the chief risk officer at Credit Suisse, a large Swiss bank, cannot pinpoint the precise moment he knew something was up: “This was not like Paul on the road to Damascus.” But signs of the gathering subprime storm in America started to trigger alarms in late 2006. Data from the bank's trading desks and from mortgage servicers showed that conditions in the subprime market were worsening, and the bank decided to cut back on its exposures. At the same time Credit Suisse's proprietary risk model, designed to simulate the effect of crises, signalled a problem with the amount of risk-adjusted capital absorbed by its portfolio of leveraged loans. It duly started hedging its exposure to these assets as well.

Mr Ervin could not have guessed at the sheer scale of what was coming. For nine months now, banks have been in a panic: hoarding cash, nervous of weaknesses in their own balance-sheets and even more nervous of their counterparties. More damaging still, money-market funds have steered clear of banks as well. The drying-up of liquidity not only created havoc in the backrooms of the financial system. It also wrecked the front door, thanks to the dramatic collapse of Bear Stearns, an 85-year-old Wall Street investment bank that was bought for a song by JPMorgan Chase in March. The Federal Reserve offered emergency funding to the investment banks for the first time since the 1930s, and there were bank bail-outs in Britain and Germany too.

The economic effects are set to be just as striking. According to a study of previous crises by Carmen Reinhart of the University of Maryland and Ken Rogoff of Harvard, banking blow-outs lop an average of two percentage points off output growth per person. The worst crises reduce growth by five percentage points from their peak, and it takes more than three years for growth to regain pre-crisis levels. With so much at stake if the banks mess up, regulators and politicians are now asking fundamental questions. Should banks be allowed to take on as much debt? Can they be trusted to make their own assessment of the risks they run? Bankers themselves accept the need for change. “We've totally lost our credibility,” says one senior European banker.

To regain trust, banks will not need to be totally bomb-proof. Safe banks are easy enough to create: just push up their capital requirements to 90% of assets, force them to have secured funding for three years or tell them they can invest only in Treasury bonds. But that would severely compromise their ability to provide credit, so a more realistic approach is needed. This special report will ask how banks should be run and regulated so that the next time boom turns to bust the outcome will be less miserable for all concerned.

If the crisis were simply about the creditworthiness of underlying assets, that question would be simpler to answer. The problem has been as much about confidence as about money. Modern financial systems contain a mass of amplifiers that multiply the impact of both losses and gains, creating huge uncertainty.

Standard & Poor's, one of the big credit-rating agencies, has estimated that financial institutions' total write-downs on subprime-asset-backed securities will reach $285 billion, more than $150 billion of which has already been disclosed. Yet less than half that total comes from projected losses on the underlying mortgages. The rest is down to those amplifiers.

One is the use of derivatives to create exposures to assets without actually having to own them. For example, those infamous collateralised debt obligations (CDOs) contained synthetic exposures to subprime-asset-backed securities worth a whopping $75 billion. The value of loans being written does not set a ceiling on the amount of losses they can generate. The boss of one big investment bank says he would like to see much more certainty around the clearing and settlement of credit-default swaps, a market with an insanely large notional value of $62 trillion: “The number of outstanding claims greatly exceeds the number of bonds. It's very murky at the moment.”

A second amplifier is the application of fair-value accounting, which requires many institutions to mark the value of assets to current market prices. That price can overshoot both on the way up and on the way down, particularly when buyers are thin on the ground and sellers are distressed. When downward price movements can themselves trigger the need to unwind investments, further depressing prices, they soon become self-reinforcing.

A third amplifier is counterparty risk, the effect of one institution getting into trouble on those it deals with. The decision by the Fed to offer emergency liquidity to Bear Stearns and to facilitate its acquisition by JPMorgan Chase had less to do with the size of Bear's balance-sheet than with its central role in markets for credit-default and interest-rate swaps.

Trying to model the impact of counterparty risk is horribly challenging, says Stuart Gulliver, head of HSBC's wholesale-banking arm. First-order effects are easier to think through: a ratings downgrade of a “monoline” bond insurer cuts the value of the insurance policy it has written. But what about the second-order effect, the cost of replacing that same policy with another insurer in a spooked market?

The biggest amplifier of all, though, is excessive leverage. According to Koos Timmermans, the chief risk officer at ING, a big Dutch institution, three types of leverage helped propel the boom and have now accentuated the bust. First, many banks and other financial institutions loaded up on debt in order to increase their returns on equity when asset prices were rising (see chart 1). The leverage ratio at Bear Stearns rose from 26.0 in 2005 (meaning that total assets were 26 times the value of shareholders' equity) to 32.8 in 2007.

Second, financial institutions were exposed to product leverage via complex instruments, such as CDOs, which needed only a slight deterioration in the value of underlying assets for losses to escalate rapidly. And third, they overindulged in liquidity leverage, using structured investment vehicles (SIVs) or relying too much on wholesale markets to exploit the difference between borrowing cheap short-term money and investing in higher-yielding long-term assets. The combined effect was that falls in asset values cut deep into equity and triggered margin calls from lenders. The drying-up of liquidity had an immediate impact because debt was being rolled over so frequently.

That is not to suggest that the credit crunch is solely the responsibility of the banks, or that all of them are to blame. Banks come in all shapes and sizes, large and small, conservative and risk-hungry. Alfredo Saenz, the chief executive of Santander, a Spanish retail giant, recalls attending a round-table of European bank bosses during the good times at which all the executives were asked about their strategic vision. Most of them talked about securitisation and derivatives, but when it was Mr Saenz's turn, he touted old-fashioned efficiency. He did not get any questions. “There were 'clever' banks and 'stupid' banks,” he says. “We were considered one of the stupid ones.” No longer.

Beyond the banks, a host of other institutions must take some of the blame for the credit crunch. The credit-rating agencies had rose-tinted expectations about default rates for subprime mortgages. The monolines took the ill-fated decision to start insuring structured credit. Unregulated entities issued many of the dodgiest mortgages in America.

And no explanation of the boom can ignore the wall of money, much of it from Asia and oil-producing countries, that was looking for high returns in a world of low interest rates. “It is indisputable that the global glut of liquidity played a role in the 'reach for yield' phenomenon and that this reach for yield led to strong demand for and supply of complex structured products,” says Gerald Corrigan, a partner at Goldman Sachs and an eminence grise of the financial world.

Many blame the central banks: tougher monetary policy would have encouraged investors to steer towards more liquid products. Others blame the investors themselves, many of whom relied on AAA ratings without questioning why they were delivering such high yields.

Wheel-greasers

Still, the banks have been the principal actors in this drama, as victims as well as villains. The S&P 500 financials index has lost more than 20% of its value since August, and many individual institutions have fared far worse. Analysts have been forced to keep ratcheting down their forecasts. “The downside will be longer than anyone expects,” says David Hendler of CreditSights, a research firm. “There is so much leverage to be unwound.”

According to research by Morgan Stanley and Oliver Wyman, investment banks will be more severely affected by this crisis than by any other period of turmoil for at least 20 years. By the end of March the crunch had already wiped out nearly six quarters of the industry's profits, thanks to write-downs and lower revenues. Huw van Steenis of Morgan Stanley reckons that the final toll could be almost two-and-a-half years of lost profits (see chart 2).

Other industries have gone through similarly turbulent times: airlines in the wake of the terrorist attacks on September 11th 2001, technology firms when the dotcom bubble burst. Even within the financial sector the banks are not the only ones currently suffering: hedge funds, insurers, asset managers and private-equity firms have been hit too. But banks are special.

The first reason for that is the inherent fragility of their business model. Bear Stearns, an institution with a long record of surviving crises, was brought to its knees in a matter of days as clients and counterparties withdrew funding. Even the strongest bank cannot survive a severe loss of confidence, because the money it owes can usually be called in more quickly than the money it is owed. HBOS, a big British bank with a healthy funding profile, watched its shares plummet on a single day in March as short-sellers fanned rumours that it was in trouble. It survived, but the confidence trick on which banking depends—persuading depositors and creditors that they can get their money back when they want—was suddenly laid bare.

The second reason why banks are special is that they do lots of business with each other. In most industries the demise of a competitor is welcomed by rival firms. In banking the collapse of one institution sends a ripple of fear through all the others. The sight of customers queuing last September to withdraw their money from Northern Rock, a British bank, sparked fears that other runs would follow.

The third and most important reason is the role that banks play as the wheel-greasers of the economy, allocating and underwriting flows of credit to allow capital to be used as productively as possible. That process has now gone into reverse. Banks have seen their capital bases shrink as write-downs have eaten into equity and off-balance-sheet assets have been reabsorbed. Now they need to restore their capital ratios to health to satisfy regulators and to reassure customers and investors.

For some, that has meant tapping new sources of capital, often sovereign-wealth funds. For most, it has meant reducing the size of their balance-sheets by selling off assets or by cutting back their lending. Quantifying the impact of this tightening is hard, but one calculation presented by a quartet of economists at America's Monetary Policy Forum in February suggested that if American financial institutions were to end up losing $200 billion, credit to households and companies would contract by a whopping $910 billion. That equates to a drop in real GDP growth of 1.3 percentage points in the following year. If the banks suffer, we all do.