Following the near-collapse of Bear Stearns, even trades conducted with interbank dealers can no longer be considered risk-free. With so much of the derivatives market in the hands of a few dealers, what would happen if a major counterparty were to go bust? What are banks doing to manage this risk? By Duncan Wood Taking an aspirin is an easy way to treat a headache. The potentially unpleasant side-effects - heartburn, indigestion, nausea - are rarely considered. Similarly, a bank suffering the pain of excess credit exposure often gulps down a credit default swap (CDS) without reading the warning on the packet: credit derivatives can produce painful, unanticipated build-ups of counterparty risk. That warning could be ignored with impunity while CDS counterparties - predominantly other banks - were raking in vast profits. That's not the case any more. Dealers have reported billions of dollars in writedowns on their subprime mortgage exposures and leveraged loan warehouses. Concerned about potential losses at other dealers, banks have been reluctant to lend to each other, causing interbank lending rates to soar and credit lines to be pulled - a factor that contributed to the near-collapse of Bear Stearns in March. Now that a major dealer has suffered a liquidity event, triggering intervention by the Federal Reserve and an eventual acquisition by JP Morgan, counterparty credit risk has been pushed firmly up the risk management agenda. A growing number of participants believe the market should break up its unwieldy chains of bilateral trades - created when one player sells protection and immediately obtains a hedge from another dealer, who does exactly the same with a third counterparty - and instead create a central counterparty clearing house, to provide more stability and clarity. In a commentary published at the start of April to promote his new book, famed investor George Soros argues that the market needs "a clearing house or exchange with a sound capital structure and strict margin requirements to which all existing and future contracts would have to be submitted". The establishment of a central clearing house would hopefully clear up one of the perceived weaknesses of the credit derivatives market. Each dealer in the market has no idea what kind of exposures other players might have accumulated, says Aaron Brown, a risk manager at Greenwich, Connecticut-based hedge fund AQR Capital Management. And while any individual trade between two dealers is routinely collateralised, the collapse of a counterparty could still result in a level of uncertainty that the market as a whole would not be able to tolerate, he adds. "If a major dealer goes under, it might turn out they had a net profit on their CDS book, but it could take months to work that out and the whole market could be in trouble while that happens," says Brown. "But if we could clean up all the market's offsetting trades - take each contract, trace it all the way through and cut out all the people in the middle so it became a direct contract between two end-users - that would dramatically reduce the volume of CDSs and the credit risk associated with them. I'm not saying that's a practical solution, but it's mathematically where you'd be looking to end up." Cutting margins That wouldn't go down well with the market-makers that dominate CDS trading. According to a survey of the credit derivatives market published by Fitch Ratings in 2007, the top 10 dealers account for 89% of the total notional volume bought and sold (see table on page 54). Those institutions would have to accept both a smaller market and thinner profit margins, says Don van Deventer, chairman and chief executive at credit risk software vendor Kamakura in Hawaii. "Dealers as a group are not enthusiastic about CDSs becoming more standardised - that narrows margins," he says. They may not have any choice in the matter. Soros says one of the factors that forced the Federal Reserve to support the Bear Stearns rescue was concern about what would happen to the rest of the industry while the securities firm - one of the top 10 CDS dealers, according to Fitch - tried to work out the net position of its trades and its counterparties scrambled to replace the protection on which the bank had defaulted. Speaking at the International Swaps and Derivatives Association's (Isda) annual general meeting in Vienna on April 17, Patrick Parkinson, deputy director in the division of research and statistics at the board of governors of the Federal Reserve System, admitted Bear Stearns' swaps book was one cause for concern. But he said the main reason for the Fed's support had been the dealer's inability to raise financing, and worries that other securities firms with similar business models would also be hit by a liquidity crisis if the central bank did not take action. Nonetheless, if the Fed wants to avoid a repeat, it may now decide it has to force the market to accept change, adds Soros. Even some bankers agree. "Moving to an exchange - or to a market built on a clearing house, at least - is inevitably the way forward," says Emmanuel Ramambason, London-based head of fixed-income counterparty risk management at BNP Paribas. "The idea of going through an exchange is not widely accepted at the moment because people see it as something that may reduce their OTC business margins. But Soros floating the idea could prompt regulators to make it a recommendation. I think we'll see momentum building for the market to organise itself differently." That is already starting to happen. A group of major credit derivatives dealers is working to set up a clearing house for OTC credit derivatives in response to growing concerns about counterparty credit risk. The initiative is being developed with the Clearing Corporation, a Chicago-based company jointly owned by 11 dealers, three interdealer brokers, Frankfurt-based exchange Eurex, London-based data vendor Markit and New York-based trading platform Creditex. The service will initially be open to credit derivatives indexes, but is likely to be extended to single-name CDSs. Reducing risk "We feel confident this is viable, and market participants are spending a lot of time pursuing it," says Athanassios Diplas, chief risk officer and deputy chief operating officer of global credit trading at Deutsche Bank in New York, speaking at the Isda conference. "This will mean we will be able to take a lot of risk out of the system, and people will not have to worry about one dealer going under and causing a shock to the market." Admission criteria will be strict, he adds, noting that only well-capitalised institutions will be able to participate in the new venture. "All these issues have to be dealt with carefully," says Diplas. "We are not going to jump into something unless we are very confident it will work." Isda itself, which until now has provided most of the mechanisms through which counterparty risk in the CDS market is managed - such as netting rules and collateral agreements - is reluctant to accept that anyone other than the big dealers and end-users will determine what happens next. But the association accepts that regulatory pressure could create a new dynamic. "Unless you're going to have a controlled market, the only determinants of how a market develops will be the market participant," says Bob Pickel, chief executive of Isda in New York. "But if the major regulators decided to intervene in the market, we'd be involved in that. I'm not sure it's the best way for the market to develop, but we'll certainly engage in that discussion." David Richardson, a London-based independent risk consultant, has some idea of the political wrangling involved in changing the way a market operates. When he worked at Standard Chartered in the mid-1990s, he was the bank's representative on the board of a venture called the Exchange Clearing House (Echo). Superficially, Echo resembled the kind of mechanism now being proposed for the CDS market - a central counterparty that reduced webs of foreign exchange transactions to single net payments and also took on the settlement and counterparty risk. The venture was fairly successful - 30 banks had signed up as members within three years of its launch in 1996 - but the big US banks supported a rival project. Eventually, both were taken over by London-based CLS Group, a forex settlement system. Richardson says Echo's users were enthusiastic - research carried out at the time showed that up to 95% of the risk was removed from currency trades using the scheme. He argues a similar approach could make sense for the CDS market - but only if regulators decide to twist the industry's arm. "People in banks won't do things that involve spending money and effort unless regulators make them do it," he adds. So far, regulators in Washington have made little mention of the need for a central clearing house. Both the Financial Stability Forum, which published a report on April 12, and the President's Working Group on Financial Markets, which released its own investigation into the causes of the subprime crisis on March 13, have highlighted the need to improve OTC derivatives infrastructure. Yet it is the dealers themselves that have taken the lead on the clearing house. Simultaneous default So if an exchange or a central counterparty may be the solution, what exactly is the problem? In theory, what a buyer of protection worries about is not the default of the underlying reference entity (because the default swap provides a hedge), nor the default of their counterparty (because the protection could be moved to another counterparty), but the simultaneous default of both, says AQR's Brown. Until recently, that was considered a minute risk, but times have changed. Spreads have blown out massively for big bank counterparties and, more importantly, there's also a real possibility the default of a reference entity might actually be the trigger for a collapse of a counterparty. Prior to the past year's spread widening, around 5% of an individual trade's value might have been calculated as being at risk from a joint default, says Brown. Today, he puts it at 25%. "It's just a quantitative fact that the importance of counterparty concentration has gone up, although it's actually off from its peak," he says. "I was more worried a month ago - but not a lot more. Things are calming down, but there's still a lot of scary stuff out there." It also doesn't require an actual default to make life unpleasant, says Barrie Wilkinson, a partner in the finance and risk practice at management consultancy Oliver Wyman in London. He outlines a scenario where a bank lends EUR1 billion to a company, then hedges the full amount with a CDS - the lender no longer has exposure to the company, as long as its counterparty can be relied on to pay. This counterparty exposure is measured by marking the CDS position to market. "If spreads move so that the CDS contract is in-the-money for the company's lender, it gets marked-to-market as a profit," says Wilkinson. "That profit is then the extent of the exposure to the CDS counterparty." In some cases, mark-to-market movements would be significant enough that the lender would transact a second CDS contract to offset the counterparty risk on the first trade. "Banks have bought all these CDSs to manage single-name risk, but now they've got a build-up of counterparty exposure to the big CDS dealers. To manage that counterparty risk, they'll then buy another layer of CDSs on top of that, but that just creates more counterparty exposure to more banks. It's quite difficult to get rid of it completely," says Wilkinson. It was a variant of this scenario that was behind much of the latest round of banking industry writedowns. Dealers had bought protection on large chunks of structured credit exposure from monoline insurers, and when the uncollateralised trades moved heavily in the dealers' favour, they were left with big mark-to-market gains - or, to put it another way, mushrooming exposure to their monoline counterparties (see Asia Risk March 2008, pages 14-17). Many attempted to mitigate that by taking out a layer of protection against the monolines themselves. But when dealers rushed to cover themselves against a potential default, they actually made it more likely, says the chief risk officer of one US monoline. "When all the players in the market go looking for protection at the same time, it pushes the spreads way out - and they're still out there," he says. "Now, some of that may be speculative activity, but the vast majority of it was the result of risk managers deciding they had to reduce their counterparty risk. That had a direct impact on the monolines' business." A more traditional way to manage CDS counterparty exposure is to cap the amount of mark-to-market risk a dealer will tolerate and put collateral agreements in place to ensure exposure above the cap is offset. But even this is problematic, in part because wildly gyrating credit spreads mean the CDS business is consuming more collateral than it used to. That's been exacerbated by a recent change in market practice. In the past, only protection sellers were expected to post initial margin, says AQR's Brown; today, protection buyers have to do it too. Separately, the collateral itself has to be vetted to ensure it is not too closely correlated with either the CDS reference entity or the counterparty. "We don't want Turkish bonds collateralising Turkish banks that are selling us protection on Turkish sovereign risk," says BNP Paribas' Ramambason. Dealers say these measures would prove effective if a major counterparty defaulted. That may be true - or it may not. In February, analysts at Barclays Capital estimated the default of a counterparty with a $2 trillion book of credit derivatives would result in losses of between $36 billion and $47 billion, purely as a result of the risk repricing that would follow the event. However a default played out, it would certainly take time for the market to find its feet - and this is where Bear Stearns comes into the picture. Timothy Geithner, president and chief executive of the Federal Reserve Bank of New York, didn't mention credit derivatives on April 3, when he told the US Senate how and why the securities firm had to be rescued. But he did say a default "would have precipitated a rush by Bear's counterparties to liquidate the collateral they held against those positions and to attempt to replicate those positions in already very fragile markets. A sudden, disorderly failure of Bear would have brought with it unpredictable but severe consequences for the functioning of the broader financial system." The head of credit portfolio management at one US bank puts it rather more bluntly: "Believe me - if Bear Stearns had gone down, Lehman Brothers would have been dead in 24 hours, and Merrill Lynch would have gone down another 24 hours after that. It's unbelievable we got so close to Bear collapsing. It would have been a catastrophic error by the Fed." Remaining alert Of course, now the Fed has shown how concerned it is about the prospect of a major dealer collapsing, risk managers may feel they can relax somewhat - surely, any other sizeable player would also get a Fed-backed bail-out? "I tend to agree the Fed couldn't allow a key derivatives market counterparty to collapse." says the credit portfolio manager. "But the Fed makes mistakes like everybody else, so we're still insisting on our limits, we're still managing the counterparty risk and we're still being very, very careful." The Bear Stearns CDS portfolio is now set to be absorbed by JP Morgan, which by some estimates would make the bank a counterparty to almost 25% of the credit derivatives market's outstanding trades. That's a problem, says the credit portfolio manager. "Everyone's looking at JP Morgan and saying 'well, I had a lot of exposure to them before they were JP Morgan plus Bear - how am I going to manage this?'," he says. "So you're looking at your mark-to-market caps and you're looking at your collateralisation agreements, but there's still a problem. We're becoming more concentrated at a shrinking number of good guys." Despite the growing concentration, he refuses to countenance the alternative - spreading the bank's business around - because he argues it's better to deal with a small number of trusted dealers than to diversify but end up taking on counterparty exposure to untested players. In his view, JP Morgan remains a solid counterparty, despite the additional counterparty concentration. AQR's Brown agrees. "If Bear Stearns had been absorbed by someone less well respected, there might have been bigger concerns," he says. Rather than seeking diversification, CDS market participants have generally looked to cut off counterparties with which they do marginal amounts of business, adds Brown. "With credit being so prominent, you really want to focus on people you have a strong relationship with," he says. Still, the CDS market faces a series of overlapping problems: the risk of simultaneous default is higher, the number of reliable counterparties is shrinking, and the whole apparatus is so unwieldy that taxpayers' money had to be used to prevent one dealer from defaulting for fear of a knock-on effect. Could a clearing house address all these issues? AQR's Brown thinks so. "It would need to be more complicated than the kind of clearing house you see in equity markets, but it's doable," he says. "And if somebody went into business to do it, they could immediately get people to recognise that it's superior to the existing system," he says....
Start a FREE trial or subscribe to continue reading:
Start a 4 week free trial
Try Risk.net's premium content for a limited period. Register now for your FREE trial to one of our leading brands.
*not available to previous trialists or subscribers.
Log In or Subscribe Now
Subscribe to Risk.net Business now to access all our premium news & features content for 1 year.
Pay by Credit Card for immediate access.