Banks upgrade to the next generation CDO pricing tools

Banks are installing CDO pricing systems from vendors such as DataSynapse, Sophis and CDO2

London, 1 November 2005

More banks are turning to vendors to install systems for pricing and risk-managing collateralised debt obligations (CDOs) as vendors upgrade the functionality of these tools. In September, French banks BNP Paribas and Ixis Corporate and Investment Bank deployed DataSynapse's grid computing solution in their London and Paris offices respectively. Grid computing helps banks price CDOs by cutting the time it takes to run risk reports on them to a fraction of the time previously taken.

In the same month, vendor Sophis released a new version of its buy-side platform for valuing CDO and credit default swap (CDS) indices. The platform, Value 3.0, includes for the first time direct links to data for implied correlation on single names, and full base correlation. "Before this version, users had to calibrate the market correlation externally from the system," explains Jean-Noel Dordain, head of research at Sophis in Paris.

And vendor CDO2 has upgraded its CDOSheet pricing and risk management tool to support new trade types. This includes new functionality to calculate the trigger risk on leveraged super-senior trades; price path-dependent trades such as step-up subordination trades; and price bespoke cashflow trades, such as coupons linked to inflation or different levels of subordination. Commerzbank's market risk group in New York installed CDOSheet over the summer.

Vendors say these tools are helping to solve a very real danger currently facing CDO investors and managers: that they may be unaware of the default risk of their portfolios, and the hedges they would need to place to manage that risk. Understanding correlation risk has also become crucial following the behaviour of the CDS indices after the downgrades of autos firms GM and Ford in May.

Solving these problems is also a potential revenue generator for banks as it will enable them to take on more trading of these products, and take them to market quicker. As computer power evolves, "today's exotic becomes tomorrow's commodity," says Willy Ross, managing director of DataSynapse for Europe, the Middle East and Africa, based in London. He says DataSynapse's technology enabled one unnamed bank to trade CDOs-squared (CDOs of CDOs) within 10 weeks of coming into contact with them. This means banks can bring products such as CDOs to market earlier, "which ultimately means a great difference to profits", says Ross. Another bank reports that without the speed provided by grid computing technology, it would have been unable to move to trading CDOs-cubed.

Meanwhile, Dordain at Sophis believes the market is only six months away from developing a standard correlation risk model. This would represent the first step towards the creation of correlation swaps, he says, which would enable traders to trade correlation risk directly.

The value of such a move is amply demonstrated by the effects of the GM and Ford downgrades, says Sophis. "If the market had been able to directly trade correlation risk, CDO investors could have traded out of their positions simply," says a spokesperson for the firm.

Standard & Poor's also released a new tool in September for monitoring and structuring synthetic CDOs. The new version of CDS Accelerator (version 2.2) includes European asset-backed security assets and five-year CDS spread data for the first time. This is a "timely response to the fact that synthetic CDOs are increasingly referencing new assets", says Perry Inglis, head of European CDOs at Standard & Poor's structured finance group.