Via Peter Tchir of TF Market Advisors,
With CDX and credit indices being such a topic of conversation, I took a look at the 1 month changes as of May 12th. I selected U.S. and European Credit Indices that had NET position changes of $1 billion during that 4 week period. I also included some with smaller changes where it made sense to me as either part of “normal” roll flows or the now legendary “whale” trade.
Starting with “on the run” indices
Looking at IG, HY, MAIN (EUROPE), and FINS, you see a very typical trend. The NET is increasing on the “on the run” series and decreasing in the older or “off the run” series. That is typical as investors, speculators, hedgers, spiv monkeys, funds, etc., want to remain in the more liquid on the run series.
The overall reduction in HY and XOVER is interesting. Also, even in financials, the riskier sub index experienced a net decrease. I’m not sure what it means. Complacency? Increased volatility forcing smaller position sizes? JPM cutting HY short and shorting IG18 against IG9?
Looking at this, other than not being able to explain the net reductions in the riskiest indices, the rest of the positioning looks like traditional role activity.
This data is a bit more interesting, especially in light of all the “whale” questions.
It looks like positions in European indices got reduced pretty dramatically. On series 9 you saw large reductions in the net exposure on both the tranche and the index side. This is indicative of correlation desks reducing positions as they would have both sides of the trade. Since the net is relatively close, it was either some senior (low leverage) tranches being unwound, or some real risk reduction in outright tranches.
Across some old series, the reduction in Net exposure are across the board (and in series 15). The total for those 4 tranches is a meaningful reduction. There was less of a move in the untranched indices, which seems to mean that these were outright tranches, or someone was hedging more with single names or more liquid indices. A bit weird, but could have been some serious risk reduction.
IG9 tranche net actually increased in the period, though outright index dropped off. Is that a sign that it was hard to get out of tranches? IG9 with that special place in everyone’s heart, does seem strange.
IG8 and IG7 both had serious amounts of notional reductions in the tranched and untranched. The ratio of tranche to index in them is very different with IG7 looking like it must have been some very risky tranches (big ratio).
IG15 and IG6 both had decent reductions. Not sure what if anything it means.
HY10, one that JPM was allegedly short had a reduction as well. Assuming this was the “5 year” tranche, it matures in June 2013 (consistent with story that JPM bought a lot of jump to default risk back in the fall of last year).
There has been so much talk about IG9 and the basis, and how much JPM must have lost since the announcement based on eyeballing IG9 or the basis, but the story is far more complex than that.
HY10 is currently at 102. That is lower than it was before the JPM announcement when it was trading at 102.5. At the end of April, when it was pretty clear that JPM was aware of the problems in the CIO, this was at 102.75, so it may have gained back a lot of the losses. It is pretty unclear the exact date the $2 billion was as of. Probably marked closer to the conference call than month end, but not sure exactly. This index was pretty stable the entire month, so while IG was getting wider one of their shorts wasn’t moving (and was bleeding carry daily). That is consistent with them losing money according to guesses on position, but they have down well since then.
I am getting some additional price history on tranches. That is a painful exercise, but to put it in perspective, they are volatile and pricing is completely dependent on where you are. The 3-6% tranche of IG9 10yr was recently 36% up front + 500 running. The “super senior” was 54 bps running. Assuming the whale was at lower end, we have seen some serious volatility in that. While IG9 10 year was tightening the entire first quarter, it is possible the tranche was underperforming. They may have made some money, but not much. Then when the index started widening, the tranches moved in step. So that by the time the index itself was tighter on the year, the naked tranches could have been wider on the year. It is that break down of correlation that could have occurred that would have explained the losses. So more important than how IG9 is moving, is how the tranches in IG9 are moving and if the “delta” equivalent remains the same.
Guessing what they have is almost impossible, but any guess that looks at IG9 and assumes the worst is probably more wrong than right.
In any case, all these products need to be moved to an exchange. Look at the huge differential between the gross and the net? That would go down. Yes, banks would have to unwind offsetting trades, but who cares? Banks would have to post collateral, possibly on longs and shorts, but again who cares? The IG9 tranche might be the worst case of regulatory misconduct ever. It has $800 billion of tranches outstanding. $150 billion of NET outstanding, dwarfing most companies, yet, you cannot get a breakdown of how risky the tranches are, and finding a price if you aren’t in the market is somewhere between difficult and really annoying. The scale and lack of transparency have to end. The beauty is that it wouldn’t take long to get indices on an exchange if anyone had the will.