I’m not sure if it was the biggest quarterly loss of all time, but Merrill Lynch’s $16 billion loss in the fourth quarter of 2008 certainly ranks very high up there in the annals of investment-bank blowups. It happened after the bank had already been taken over by Bank of America (NYSE:BAC), and it was in the middle of the financial crisis, so it didn’t get nearly the amount of attention it deserved. But it was not simply a case of assets plunging in value. Instead, it was, in very large part, a basis trade blowup.
The basis trade is an arbitrage, basically. There are two different ways the market measures credit risk: by looking at credit spreads — the yield on a certain issuer’s bonds, relative to the risk-free rate — or by looking at CDS spreads, which are basically the same thing but set in the derivatives market rather than the cash bond market. Most of the time, CDS spreads and cash spreads are tightly coupled. But sometimes they’re not. And at Merrill, a huge part of that $16 billion loss was reportedly due to a bad basis bet: the basis on many credits became very large and very negative during the financial crisis.
This time around, the basis-trade disaster has happened at JPMorgan (NYSE:JPM), where the famous London Whale seems to have contrived to lose $2 billion on what was meant to be a hedging operation. And once again, although the details are still very murky, the culprit seems to be the CDS-cash basis.
I’ve been meaning to write a post about the CDS-cash basis for a few days now, which is why I happen to have this chart handy, showing the basis for various European banks as of Tuesday May 8.
These are very big numbers, for very big banks: UBS (NYSE:UBS) is at 75bp, Deutsche (NYSE:DB) is at 83bp, Natixis (OTCPK:NTXFY) is at 116bp, and IKB (OTC:IKBDF) is at a whopping 392bp. And this is just the banks — other corporates have seen similar price action. The cost of protection has gone up sharply, while the cash bonds are still trading at very low spreads.
Bruno Iksil, the London Whale, had a massive long position on corporate CDS in general, and the CDX.NA.IG.9 index in particular. He was selling protection, betting that credit spreads would go down, rather than up. The position was meant to be a hedge, although it’s a bit unclear how JPMorgan could have some massive short position in corporate debt that it was hedging against. In any case, CDS spreads went up — and credit spreads, in the cash market, didn’t.
Cue a $2 billion loss.
Rarely has a position been as widely publicized as Iksil’s, and I wouldn’t be at all surprised to learn that the credits with the highest basis were precisely the credits CDX.NA.IG.9 index. Whenever a trader has a large and known position, the market is almost certain to move violently against that trader — and that seems to be exactly what happened here. On the conference call, when asked what he should have been watching more closely, Dimon said “trading losses — and newspapers”. It wasn’t a joke. Once your positions become public knowledge, the market will smell blood.
Of course, this loss only goes to show how weak the Volcker Rule is: Dimon is adamant, and probably correct, in saying that Iksil’s bets were Volcker-compliant, despite the fact that they clearly violate the spirit of the rule. Now that we’ve entered election season, Congress isn’t going to step in to tighten things up — but maybe the SEC will pay more attention to Occupy’s letter, now. JPMorgan more or less invented risk management. If they can’t do it, no bank can. And no sensible regulator can ever trust the banks to self-regulate.