Taking the CVA route to valuing CLNs

Request a Copy

credit-linked notes can't be said to have risk-free collateral any more. In this article Dmitry Pugachevsky, Director of Research, Quantifi, suggests pricing them using techniques developed for bank counterparty risk.

One of the challenges in investing in credit linked notes (CLN) is the shortage of high quality debt for funding. This article explores a new type of trade - CLNs with risky collateral. It highlights that by taking into account all possible risk, including uncertainty of market value at early redemption, one can calculate values and sensitivities of such products, and trade them consistently as traditional CLNs.

In a CLN, a special purpose vehicle (SPV) enters into an unfunded credit derivative with a dealer and buys highly-rated “collateral” to receive funding. The collateral was presumed riskless. Recently though, analysis of collateral risk of default has become an important part of pricing and managing CLNs.

Dmitry also outlines how any type of credit derivative can be structured as a funded note supported by collateral that can default before deal maturity. In the case of collateral default, underlying credit derivative will be unwound at market price, which can lead to additional 10 e. For example, a funded credit default swap collateralised by a risky bond will be terminated when either the CDS reference or the collateral defaults.

At first glance, this deal looks similar to a funded first to default (FTD). However, further detailed analysis is shows that in order to price funded CD with risky collateral one has to estimate two additional terms: the value of the collateral at CDS default and the value of CDS at collateral default. This is similar to evaluating credit value adjustment (CVA), which also requires pricing credit derivatives (although floored at 0) at default of a counterparty. It is important to emphasise that these two terms cannot be priced in a regular "credit copula" framework and require a more sophisticated CVA-like methodology implemented a Monte Carlo simulation or a semi-analytic model.

This paper shows that the difference between funded CDS and funded FTD is mostly due to the shape of credit curves and the correlation between default events.


Request a Copy