Here are two case studies of Co-relation trades & blowing up.
The common theme in both of these trades: an allergic reaction to be negative carry or paying out premiums.
1) London Whale (2012)
Bruno Iskill initiated a curve trade on the CDX.NA.IG.9 Credit Index. This index represents the CDS spreads of 125 US corporate bonds. This particular Index was chosen because it had the most liquidity since it was initiated in 2007 and a lot of structured products launched in that timeframe had elements of CDX.NA.IG.9 embedded in those deals. The rumored trade initiated approximately in Jan 2011 was:
- Selling protection on the 10-year contract of the CDX.NA.IG.9 (that matures in December 2017)
- Buying three times the notional amount on the 5-year tenor (that matures in December 2012).
For a visualization of the curve- click here .
The proportions were supposed to be such that the overall trade was CV/DV01 neutral which means that regardless of the movement of the overall credit yields that trade would not have lost money IF the change in spreads happened in the same quantities across the curve (i.e. for both the 5 yr & 10 yr). The trade was supposed to be also “rebalanced regularly” as prices change so that the proportions between the 5yr and 10yr cotinued to be CV/DV01 neutral. The trade could also make money in a somewhat bearish environment if the shorter maturity (5Yr- 2012) CDS spreads moved up more than the longer maturity (10yr-2017) which in effect would have “flattened the curve”. They initiated this curve trade instead of buying the CDS protection outright because it would have been money out of thier pocket and a negative carry trade. The curve trade achieves a short credit bearish view without money out of the pocket.
There were two problems with the trade:
1) JPM recently updated thier risk and VAR models which got the math wrong on the hedge proportions for the 5yr & 10yr contracts that they were supposed to buy & sell.
2) They sold 10 yr protection in such large quantities that they themselves became the market (See this chart for an increase in the market). In effect, that created a skew i.e. a difference in value between the Index and the 125 underlying bonds (see this chart for the skew).
2) Howie Hubler / Morgan Stanley Subprime Losses (2007/2008)
Investors would buy credit protection on mortgage bonds they thought were likely to fail, happily paying small insurance premiums now in return for what they hoped — rightly — would be a massive payday when the bonds started defaulting.
The problem was that no one liked paying those premiums. Lewis details the enormous fights that one manager, Michael Burry, had with his investors: they hated the fact that lots of money was flowing out of his fund, in insurance premiums, and nothing was coming in. So other people tried to put together trades where they could make enough money in one part of the mortgage market to pay the insurance premiums elsewhere. Lewis writes about the most disastrous such trade, put on by Morgan Stanley’s Howie Hubler:
The crown jewel of their elaborate trading positions was the $2 billion in bespoke credit default swaps Hubler felt certain would one day very soon yield $2 billion in pure profits. The pools of mortgages loans were just about to experience their first losses, and the moment they did, Hubler would be paid in full.
There was, however, a niggling problem: The running premiums on these insurance contracts ate into the short-term returns of Howie’s group. “The group was supposed to make two billion dollars a year,” said one member. “And we had this credit default swap position that was costing us two hundred million dollars.” To offset the running cost, Hubler decided to sell some credit default swaps on triple-A-rated subprime CDOs, and take in some premiums of his own. The problem was that the premiums on the supposedly far less risky triple-A-rated CDOs were only one-tenth of the premiums on the triple-Bs, and so to take in the same amount of money as he was paying out, he’d need to sell credit default swaps in roughly ten times the amount he already owned.
This turned out, by the time the dust settled, to be without a doubt the single worst trade that any investment bank has ever entered into in the history of the world: it ended up losing Morgan Stanley $9 billion. It was a long-short trade: Hubler was long triple-A CDOs and short triple-B CDOs, and he ended up losing vastly more on the long side than he could possibly make on the short side.