Central banks can't determine how much leverage is out there

After the Flood:
How Central Banks Fret
About Failures
Once Liquidity Dries Up

By John Plender
Financial Times, London
Tuesday, January 30, 2007

http://www.ft.com/cms/s/539c92d0-b006-11db-94ab-0000779e2340.html

In September 1998 Bill McDonough, the then president of the Federal Reserve Bank of New York, corralled representatives of 14 leading banks into the Fed's offices at 33 Liberty Street in Manhattan's financial district and urged them to bail out the ailing Long-Term Capital Management hedge fund. It was a classic central banker's response to a potential systemic crisis.

"Gentle pressure" is the euphemism often employed to describe such central bank bullying to persuade competing banks to collaborate in the common interest. The interesting question, in the light of huge structural upheavals in financial markets since 1998, is whether the nature of systemic risk has changed and whether a central bank could pull off the same trick today.

In the period between the break-up of the Bretton Woods semi-fixed exchange rate system in the early 1970s and the near-collapse of LTCM in 1998, financial crises were frequent. Yet for the best part of a decade an eerie stability has prevailed. Big financial institutions have collapsed, notably Refco, the derivatives dealer, and the Amaranth hedge fund. Yet neither initiated a systemic shock, even though Amaranth's $6bn losses were greater than those of LTCM.

Many private sector bankers believe that the newer markets in credit default swaps, which investors use as insurance against corporate default, and collateralised debt obligations, packages of debt instruments used to back the issue of new securities, are inherently stabilising. That is because they spread risk more widely around the system. At the same time technology, which facilitates trading in complex new financial instruments, serves to make markets more efficient.

This, together with a big surge in global liquidity, has contributed to a dramatic decline in financial institutions' concern about risk to the point where some companies are issuing securities at a zero or negative risk premium. The risk premium is the additional return over the return on risk-free government bonds that investors normally require as a reward for taking risk.

The credit euphoria in the markets, which has caused the yields of riskier bonds to move closer to the risk-free bond yield, is partly driven by the prime brokerage divisions of investment banks competing ferociously for hedge fund business. They have loosened lending standards and margin requirements relating to the amount of collateral they require to support a given amount of hedge fund debt.

Even central bankers, traditionally cautious about the consequences of financial innovation, see some advantages in the new world of high-octane derivatives trading. Tim Geithner, president of the New York Fed, points out that past crises would cause less damage today if they were to recur because of the greater dispersion of credit risk, the improvements in risk management, the size of the capital cushions maintained by banks and the improvements in many parts of the payment and settlements infrastructure.

That said, neither he nor any other leading central banker believes that we are witnessing the end of volatility or the demise of the credit cycle, though some youthful bankers in the private sector are prepared to argue that case.

According to Gerald Corrigan, a former president of the New York Fed who is now a partner in Goldman Sachs, there is a virtual consensus among leading practitioners and central bankers that "the statistical probability of a major financial shock with systemic features has got lower over time". But there is also agreement that another major shock is likely and that the potential damage could be greater. Mr Corrigan gives three reasons for this increased toxicity: speed, complexity and tighter linkages across institutions and markets, as the system has become more integrated thanks to financial innovation.

"The trouble," he adds, "is that we do not have the capacity to anticipate the timing and triggers of such a shock -- every now and then stuff happens. And if we could anticipate the timing and triggers, the shocks wouldn't happen."

There is no shortage of potential accidents, ranging from an over-abrupt unwinding of global financial imbalances to a dollar collapse. A particular concern, raised at the World Economic Forum at Davos by Jean-Claude Trichet, president of the European Central Bank, is the likelihood that credit spreads - the gap between the yield on risky bonds and the risk-free rate - could widen sharply if perceptions of risk change, inflicting large losses on traders. The collapse of a hedge fund or bank might then cause widespread disruption in the markets.

In the euphoria that has accompanied the explosive growth of credit derivatives and collateralised debt instruments, there is not just a possibility that risk is being seriously underpriced. Much trading in credit derivatives assumes that liquidity -- the ready availability of funds -- will remain when any adjustment in credit markets takes place. Liquidity permits traders to close positions rapidly when risks and potential losses are escalating.

Christopher Whalen of Institutional Risk Analytics, a consultancy, argues that, given the lower risk premiums in credit markets, it may no longer be prudent to assume credit default swap contracts will be liquid when the adjustment comes. In other words, traders may be unable to escape from positions where losses are ballooning because nobody will be willing to deal. He notes that a hedge fund that sells insurance protection against default may depend indirectly upon another under-regulated hedge fund having the resources to meet that guarantee.

Maintaining confidence in counterparties, adds Mr. Whalen, is absolutely required for the game to continue; and the stability of the entire credit derivatives market rests on the notion that hedge funds will somehow have access to sufficient liquidity to meet their obligations. For some, that looks a dangerously optimistic assumption.

Jim O'Neill, head of global economic research at Goldman Sachs, recently remarked that "liquidity is there until it is not -- that is the reality of modern markets." The liquidity glut, he thinks, could reverse at any time. So much for what some claim is a secular increase in liquidity.

Optimists downplay the risk to the system of the potentially problematic credit derivatives, which are still only 7 percent of estimated total notional over-the-counter (that is, unquoted) derivatives contracts. Yet the New York Fed's Tim Geithner emphasises that despite this underwhelming percentage, credit risk in the OTC derivatives market is large relative to more traditional forms of credit and is also quite large relative to the capital cushions and earnings of the major banks and investment banks.

He adds that these exposures are harder to measure because investments in credit derivatives contain "embedded leverage" where one's exposure to profit or loss is multiplied many times compared to the same investment in the underlying conventional security.

The problem for central bankers is that "embedded leverage" has expanded phenomenally and does not appear on balance sheets, so it is impossible to quantify embedded leverage across the financial system.

In other words, no one can be sure how much capital to set aside as insurance against these leveraged bets going wrong. While risk management techniques have improved, they remain flawed in fundamental respects.

It is widely acknowledged, for example, that mathematical models of risk, which are used to stress-test derivatives, give too much weight to the low volatility of recent times. In other words, they use the recent past as a guide to predicting the future. In financial markets this is the one sense in which history is bunk, since financial shocks have a habit of coming from unexpected quarters.

These risk models can ignore the potential occurrence of very low-probability scenarios with potentially extreme outcomes, in which one big loss can wipe out several years of positive returns. Statistically driven models and risk metrics are poor at capturing these low-probability financial blow-outs. If stress-testing does throw up an outcome that looks scary, people in financial institutions tend to declare the result "unrealistic" because a conservative assessment of risk would put them at a competitive disadvantage to more "realistic" competitors.

Academics such as Harry Kat of the Cass Business School at the City University in London have produced evidence that many hedge funds are, in fact, pursuing trading strategies that can be relied on to produce positive returns most of the time as compensation for a very rare negative return. They are encouraged to do this by a fee structure that does not require the fund managers to pay back their earlier profit share to investors if an extreme event strikes and wipes out the fund.

At the same time, big financial institutions have no incentive to incorporate the potential costs and risks to the system of their own collapse in their market pricing. They prefer others to incur the costs of providing the "public good" of financial stability, while under-insuring against the risk of failure and under-investing in systems to enhance financial stability. So central banks and governments pick up the tab in the event of a systemic collapse.

Considerable work has been done by banking authorities and private sector institutions to address these problems, notably through the work of the Counterparty Risk Management Policy Group II headed by Gerald Corrigan. He characterises the objective as being to strengthen the shock-absorbers of the global financial system. The group's recommendations were aimed primarily at the private sector, ranging from strengthening corporate governance to improvements in transaction processing.

Meantime, US and European financial watchdogs launched a probe before Christmas into lending to hedge funds and margin practices. This involves looking at risk-management in individual firms and telling them where they stand in relation to best practice, but without necessarily being prescriptive.

In essence, the goal of the authorities in dealing with potential shocks is damage-control and containment. As far as the co-ordination of bailouts is concerned, persuading bankers that they have a collective interest in rescuing competing financial institutions has never been easy, since most central banks have no legal powers to enforce such action. In a very different environment from that of 1998, it is a moot point whether a rescue would work when an institution deemed too big to fail finds itself in trouble.

Sir John Gieve, deputy governor of the Bank of England, has publicly questioned whether it would now be possible to put a failing firm's bankers into a room and persuade them to do their stuff. He points out that firms nowadays often do not know who holds their shares and debt, and many investors are looking to take the hit and get out as quickly as possible. Others add that some banks' proprietary trading desks might have short positions in a failing firm as well as outstanding loans, which could dilute their interest in joining a rescue.

Yet this is not something on which all central bankers agree. Tim Geithner acknowledges the difficulties of putting the lending banks in a room, but points out that we are now several decades into the securitisation of bank loans and dispersion of credit risk and there is no general increase in bankruptcies or decline in average recovery rates, though he adds that there are many other factors that may help explain this.

As for the conflicting interests within banks in relation to a failing firm, he adds that structurally, the banks have long positions in credit overall. It is worth noting too, that the New York Fed also has a big advantage in lender-of-last-resort operations relative to many European countries, including the UK, in that monetary policy decision-making and banking supervision are in the same institution, which minimises problems of communication and co-ordination.

Whoever is right, the one certainty is that lightning will eventually strike. The systemic crisis could arise in a conventional corner of the markets. But given the novelty, opacity and complexity of derivatives trading, and challenges that central banks face in trying to understand the risks involved, there is a high chance that the lightning will go there.

* * *

Help Keep GATA Going

GATA is a civil rights and educational organization based in the United States and tax-exempt under the U.S. Internal Revenue Code. Its e-mail dispatches are free, and you can subscribe at www.GATA.org. GATA is grateful for financial contributions, which are federally tax-deductible in the United States.