Jul 22

Default correlation is the weakest link in credit risk transfer markets (BIS working paper)

by David Harper, CFA, FRM, CIPM

FRM |

There is some insightful material (and a nice overview of CDO technology) in Darrell Duffie's BIS working paper on credit risk transfer (CRT) and Mohamed El-Erian's keen remarks on the same. They go deeper than the usual riff on credit ratings; both seem to agree that default correlation is a central character, and the weakest link, in structured finance of the future.

Duffie's recap of the mixed, sometimes counterintuive impact of default correlation on tranches is helpful (emphasis mine):

"Default correlation across a pool of loans that constitute the collateral of a CDO can have a significant impact on the risks and market value of individual CDO tranches. As default correlation increases, so does the volatility of the cash flows from the collateral pool. At one extreme, if defaults are perfectly correlated, then either all loans default or none default. At the other extreme, if defaults are independent across a relatively large homogeneous collateral pool, according to the law of large numbers, the default rate would be close to the average expected default rate. A senior CDO tranche is effectively short a call option on the cash-flow performance of the underlying collateral pool. The market value of a senior tranche therefore decreases with risk neutral default correlation. The value of the equity piece, which resembles a call option on the collateral pool cash flows, increases with default correlation. Optionality does not have a clear effect on the valuation of intermediate tranches, however. Each of the intermediate tranches has given up an option to the tranches with less priority and taken an option from the tranches above it. The overcollateralisation of a tranche is the principal amount of debt below it. With sufficient overcollateralisation, the option given to the lower tranches dominates, but it is the other way around for sufficiently low levels of overcollateralisation."

In a few paragraphs, Mohamed El-Erian is fabulous on (i) shift in emphasis from traditional credit characteristics to correlation, (ii) the new frontier of everywhere basis risk, and (iii) the impossible job of supervisors:

"By influencing the fundamentals of balance sheet management by dominant players, the proliferation of CRT technology generates efficiency gains for pre-existing market activities. It also places enormous pressure on banks to shift to greater reliance on an "originate and distribute" model. As a result, the detailed evaluation and structuring of individual loans gradually gives way to the mass production of composite products, causing the emphasis in risk assessment to move from the credit characteristics of individual borrowers to the extent of correlation within the composite products being originated, warehoused and distributed…
This shift entails an interesting challenge for banks, since it forces them to abandon the two extremes of the hedging spectrum – no hedging and perfect hedging – for the world of imperfect hedging with significant basis risk…
These enthusiastic new buyers – state pension funds, insurance companies, etc – do not fall within the purview of supervisors and regulators possessing, in both absolute and relative terms, the technical sophistication required to understand CRT instruments. It is not that they are not regulated but that they are regulated by entities that have had little exposure to CRT technology…We are in the midst of a large- scale migration of risk out of the strongest regulatory/supervisory regimes to regimes that historically have lacked the necessary sophistication. The consequences for systemic risk are heightened by the growing probability of some type of political backlash that could lead to an ad hoc regulatory response."

Comments

  1. Be the first to leave a comment!

Leave a Comment