March 26, 2009

Notional Value of Derivatives

Posted by JSpur at March 26, 2009 4:55 PM

I was in the audience at the Spring meeting of the Pension Real Estate Association in Washington, DC, when Tim Ryan, CEO of the Securities Industry and Financial Markets Association, addressed us about the current situation in the financial markets and the governmental response. He was modestly positive that the new Public-Private Investment Partnership and TALF programs will work, so the general message was reassuring.

In response to one question from the audience, he said that the book value of derivative positions tend to be the "notional value" of those positions. This set up a little alarm bell in my head, so when I got back to Houston I looked this up here.

As I understand it from this explanation, the $1.6 trillion book value of AIG's derivatives overstates substantially its total credit risk exposure, which is some small fraction of that number- so my Tuesday morning post about the amount at risk if the book is mismanaged by new and less competent staff clearly exaggerated the potential size of the risk. Actually to calculate the risk would require a level of understanding of each position that only those involved in the trades could possibly have.

As to this flawed analysis, apologies, and mea culpa. I was erroneously applying the accounting we use in commercial real estate to derivatives. My bad.

I think the main point does remain the same, however- as seen by the mad scramble in the AIG Paris office today to replace the two guys who resigned there, thus potentially triggering defaults in their derivatives contracts. The question of whether the taxpayers will be called upon to prop this company up to the tune of even more billions rides in large measure on whether the current book gets unwound as competently as possible. Whether the replacements Banque AIG finds for its two resigning executives will be good enough to satisfy the French banking regulators and thus avoid a default, and who knows what additional damage to this company, is a risk we, as 80% owners of the company, wouldn't be running but for all this whole bonus anger.


It's even trickier than that. For interest rate or fx derivatives, the loss will usually be much less than the notional amount. For example, an interest rate swap on 10MM notional will only swing in value by a few hundred k, usually, although I suppose that if interest rates moved substantially enough then it could be much greater than that. I don't see why CRE is different, if you are trying to price an option to buy a building then the notional would be the price of the building but the economic value of the option would be much lower.

Also, in terms of CDS, usually you can lose up to the entire face value of the contract, it just depends on the recovery rate. Usually people price in a 40% recovery rate, I believe, but the Lehman recovery was something like 9%. This would be more appropriate to the AIG situation.

So the answer is even more, it depends.

Posted by: Andy | March 27, 2009 1:12 AM

Andy, I was actually going a slightly different (but still erroneous) direction with my thinking on this, which is that if you have commercial real estate assets on your books at a value, that value is the amount of cash money for which you genuinely expect you could sell those assets, especially in these days when we in private equity real estate are being called upon to mark everything to market per FASB 157. That is clearly not the case with interest and fx derivatives, but more true as to CDS, it would seem.

Since most of AIG's book is apparently CDS, your comments are supportive of the argument I've been making. It most certainly does depend, and in this case the "it" is the end-of-the-day body-slamming that will be inflicted upon the taxpayers, as the none-too-proud 80% owners of AIG, and the importance of managing AIG going forward so that that trauma is minimized.

Posted by: JSpur | March 27, 2009 9:57 AM


Posted by: Klmjiyhx | July 15, 2009 10:13 PM
Post a comment