Risk Management - September 17, 2007

Interview with Dominic O’Kane

Dominic O'Kane

As an authority on credit risk and credit modelling, what is your overall view of the subprime crisis?

Dominic O’Kane: A widening in credit spreads was inevitable after the long period of low interest rates and the tight credit spreads we had over the last 5 years. However the way it happened was new. Previous credit crises have been a result of sovereign default as in 1997, or corporate default as in 2002. This time it was the mortgage sector which triggered this widening. Why has it happened? In my view, the biggest factor is the long period of growth we have seen in US house prices in the period 1996-2005. I think that this caused lenders to think that mortgage lending was a one-way bet – if the borrower defaults, house price inflation will always ensure that the loan balance will be fully repaid. However, once this upward trend was reversed by interest rates rises, there was an erosion of lending standards which has now become apparent. Also, the whole process of mortgage securitisation, which is generally a good thing, has actually created layers of complexity which have made the credit analysis of individual deals impossible for all but the most sophisticated investor. Investors have therefore been very dependent on the credit rating provided by the credit rating agencies.

Do you think there are any initial risk management lessons to be learned? Do you think that the current techniques for evaluating extreme risks, like the VaR or BVaR, can account for the risk of illiquidity and its interaction with credit risk?

Dominic O’Kane: Yes. I think we learned that structured products need to be analysed carefully, and in doing so we need to think about the co-dependence of risk factors, in this case, most notably house price inflation and interest rates. Regarding extreme risks, I think that the choice of risk measure is secondary to the choice of model, although I do prefer measures which are coherent. What needs to be captured is co-dependence and that can be done using a number of techniques including tail-dependent copulas. Rather than try to model liquidity, which is very hard, I think it’s better to model the risks that led to the loss of liquidity in the first place!

The ratings agencies were heavily criticised during the subprime crisis. Do you think that this criticism is justified?

Dominic O’Kane: I think some of it is. Clearly some of their modelling assumptions for these deals, especially the ones with high concentrations of sub-prime loans, now seem overly-optimistic and they were slow to correct them. For example both Moody’s and Fitch downgraded a large number of deals and revised up their US subprime default probabilities in July 2007. What is not clear is why they waited so long since the increase in US mortgage delinquencies was already known in early 2007. Some of their estimates also failed to take into account the poor quality of documentation on these loans.

In times like this, questions also arise regarding the perceived conflict of interest which arises in the rating of structured transactions. Since the investment banks who issue a deal get to choose which agencies should rate a deal, and since the rating agency gets paid by the investment bank for the rating, the whole setup does, in my view, discourage a rating agency from breaking ranks to take a much more negative view on a deal than others. The question is whether this perceived conflict of interest really does translate into more favourable ratings. It’s not clear, especially as the rating agencies are also motivated to protect their reputations. However sometimes even a perceived conflict of interest can do damage since it can create suspicions and undermine confidence.

You are currently finalising a book on credit risk. Do you think that the subprime crisis will lead to new research on credit risk?

Dominic O’Kane: Yes, I have almost completed a book on the pricing of credit derivatives which will be published early next year by Wiley Finance. In it I set out the main models used in the market to price mainly credit derivatives on corporate and sovereign credit. These models capture credit risk in its many facets i.e. probability of default, loss given default, spread volatility, default correlation and contagion, and the negative correlation between default rates and recovery rates.

However, the credit risk of sub-prime mortgages introduces another facet of credit risk. That is the link between interest rates, house price inflation, and loss severity. When we think about it, we realise that there is a negative feedback effect here – if interest rates go up, monthly payments go up on adjustable rate mortgages, borrowers fail to make payments, houses are reclaimed and sold for less than their initial value, house prices fall, loan to value ratios increase, mortgages rates go up and so on … It is this negative feedback loop which has caused the crisis we see now and it is a scenario which was underweighted in most models. A process like this can only be halted by some regulatory intervention.

What were the reasons that led you to join EDHEC as an affiliated professor?

Dominic O’Kane: Well let me start by saying that I have always enjoyed doing research and teaching. I have a PhD in theoretical physics from Oxford and was a post-doctoral research fellow at Imperial College in London before I gave it up to work in the city. After working for Salomon Brothers for 3 years, I moved to Lehman Brothers, where I ended up running a team of 20 PhD quants with the responsibility for building the pricing and risk models for the fixed income business. Therefore, you can see, my career since leaving academia has always been one of research in a financial setting. I also did a certain amount of teaching at workshops and also on the Oxford University Master of Finance programme.

I joined EDHEC because after moving to France, I was looking for a place where I could continue my research work. However, I wanted a place which values the experience that I bring and EDHEC certainly does that. It is also keen to take a view on many of the issues affecting the financial markets today, and I think that its independence gives it an influential voice.

What are your research projects for the coming months?

Dominic O’Kane: First I need to finish my book on credit derivative pricing. After that I will be continuing some work on the modelling of European and US sub-prime mortgages, which I have already begun.

About Dominic O’Kane

Dominic O'Kane is an affiliated professor with EDHEC Business School. He was previously a Managing Director at Lehman Brothers where he headed the Fixed Income Quantitative Research team, covering the pricing and risk models used across credit, interest rates, FX and commodity derivatives. He was at Lehman for over 7 years. Previously he spent 3 years at Salomon Brothers. He has a doctorate in theoretical physics from the University of Oxford.

URL for this document:

Hyperlinks in this document: