Valuation models: The value of number crunching

COPYING AND DISTRIBUTING ARE PROHIBITED WITHOUT PERMISSION OF THE PUBLISHER: SContreras@Euromoney.com

By:
Louise Bowman
Published on:

New valuation models have underscored the need for accurate mark-to-market pricing for credit derivatives.

Two new credit derivatives valuation models were launched last month, an illustration of the challenge investors face in accurately pricing the risk embedded in these instruments. But the two platforms approach the market in contrasting ways and are targeting very different types of investors. Rating agency Fitch Ratings has launched Risk Analytics Platform for Credit Derivatives (RAP CD), which aims to drive transparency in market risk for single-tranche synthetic CDOs. Data provider Markit Group is using its proprietary data to provide independent portfolio valuations across a range of vanilla and exotic OTC derivative instruments.

Kim Slawek, group managing director at Fitch Ratings, describes its new product as a natural evolution for the rating agency. But it marks a significant departure in that it focuses on market risk, not credit risk. “Mark-to-market risk and understanding volatility are the biggest barriers to investing in synthetic CDOs,” she says. “We want to bring as much transparency to the market risk side as we have to the credit risk side. There is a lot of market risk embedded in credit and this is a far more efficient way to get intelligence on the drivers for CDO pricing.”

RAP CD will produce a mark-to-model price and a market risk report, which will outline the drivers of the valuation and what is driving correlation. It is a web-based open platform and will incorporate a number of different pricing models, sensitivity and stress testing. Market data can either be supplied by the user or from other sources, such as Bloomberg. Two recent Fitch acquisitions will provide pricing modelling (from Algorithmics, which Fitch acquired in January 2005) and CDS pricing, which will come from Valuspread, bought by Fitch in August 2005. Two further third-party CDO pricing modelling firms will be involved.

The first release of the platform will encompass just single-tranche CDOs; portfolio analytics may be incorporated in the future. But mapping a bespoke portfolio to an index portfolio to extract the correlation will be a challenge – particularly for very complex trades such as leveraged super seniors and CDO-squareds.

Slawek explains that RAP CD is targeted at fixed-income investors who are new to the sector and are buying synthetic CDOs or CDS for the first time. But Markit is aiming for the far more sophisticated investor as well. Its new service, Markit Portfolio Valuations, is based on proprietary data that it receives from more than 60 industry market makers. It is this that the company believes “sets Markit apart from alternative services” as its prices span over one million data points collected daily. Markit Portfolio Valuations provides a post-trade calculation of the gross asset value of a portfolio of trades and has been test run with 20 Markit clients throughout 2005. “What is important is accuracy of pricing,” emphasizes Tim Barker, executive vice-president and head of valuations at Markit. “In the CDO market, we have an accurate and complete view in terms of individual names and correlation skew between tranches.” Markit claims to have a unique view on second order inputs (such as option volatility) as well as detailed correlation, which ensures accuracy. “Implied option volatilities differ whether you are deep in the money, at the money or out of the money,” says Barker. “If you work at a hedge fund or a bank you are trading the relative value of option strikes and it is no good to price these off a flat volatility – you need the volatility levels for different strikes.”

ACCURATE OPTION STRIKES DEPEND ON VOLATILITY
Cap volatility surface
Source: Markit Group


Although there is little disagreement that there is a need for more accurate CDS pricing tools, that need differs sharply across the market. “There is a big discrepancy among investors as to the sophistication of pricing tools,” observes Olivier Renault, head of European structured credit strategy at Citigroup. “The big hedge funds have super-sophisticated pricing tools and they invest a lot in this area.” Barker at Markit agrees; “Hedge funds challenge us down to the last basis point on pricing,” he says. But industry estimates suggest that only about 20% of hedge funds have exposure to harder to value assets. “The vast majority of hedge fund managers are trading long/short equity and not a lot else,” Barker observes.

The majority of investors don’t have pricing models and have to rely on other providers, which Slawek at Fitch sees as the market opportunity. “Real money investment managers are now forced to buy complex credit,” she says. “Buyers are beholden to the sell side to give them pricing. The correlation shakeout brought the complexity of modelling to the fore and reinforced the need for new tools.”

Whether or not occasional buyers of complex credit can justify investing in pricing models is open to question. “A small savings bank or pension fund that occasionally ventures into the market will tend to look just at the rating of the tranche and get the marks from their dealers,” says Renault. “It is probably not worth investing a lot of money to get a pricing tool to mark those small and safe positions to market. The cost of the model would erode the already low returns these triple-A tranches offer.” Slawek insists, however, that the need to mark the book on a regular basis will override any concerns over the cost of a model. And more third-party models will certainly be useful for fund administrators, which often need counterparty prices and can find it very difficult to get consistency of information.