I am doing some research on the credit index vs single name arbitrage (dynamics under different market condition, feasibility of different replication approaches etc). Since index-single name arbitrage is a pretty capital intensive exercise (esp if using full replication technique), I think only the dealers may have the resources to run that. I have not worked for a dealer before. There are a few issues that I would like to have comments on:1) First and foremost, are the dealers actively involved in running index-single name arbitrage? I can see the deal spread mismatching between single names (traded in all-running) and the indices (traded with deal spread) could create a big operation issue. 2) CDX12 has a deal spread of 100bp, so as the the US investment grade single names after the big bang. For the CDX deal spread side, is it something "permanent" or just happen to be 100bp via dealer pool?3) Each on-the-run index has a life of 6 months (in the case of 5Y index, the maturity is 5.25Y at day 1 and become 4.75Y when replaced by the next on-the-run index). On the single name side, people tend to stick with the IMM dates (in the case of 5Y CDS, the maturity of CDS at inception would range from 5 to 5.25Y). In other words, you cannot even find CDS with the same maturity as the on-the-run index for half of the time? How do people deal with it?4) What's the typical ticket size? For single names, I heard that you would probably go for 5M each if you want to have decent bid-ask spread. For a 125M index, it would be 625M. Is it common to print something that big?

have not done this since 2006, so I may be 'old skool' now. Regarding (3), the issue comes down to putting a value on an off the run single name cds. What is the mtm of this 5yr cds u traded 1.69 yrs ago? Same problem as valuing 'off the run' IR swap: u need a curve building method to interpolate. So the first step in setting up this 'arb' is to create the bestest possible single name curves (1,3,5,7,10yr...whatever available) for each index member.

From what I see a lot of dealers have index vs. single name position, but few of them like having it. Most of these positions are held by correlations desks who hedged deltas from bespokes with an index. They hate the position because of the MTM vol that resulted from a massive basis widening a few months ago (especially in IG). A number of dealers were actually offering contracts where a client could have bought index protection and sold single name protection with restructuring clause, deal spread and maturity matching an index exactly. This trade was as close to free money as I've seen in credit recently, but it did let the banks get the basis off their books.

Assuming we can get buy CDS with matching maturity/ deal spread, I have computed the recent index skew for CDX9 (mid June). It is now less than 5 bps. Not much room at this moment. But who knows if it would widen out in the future. StructCred: thanks for your market color. Do you mind telling us who might offer CDS/index basis contracts (or leave me with a PM if you don't want that bit to go public)? I have not seen much from the correlation runs I happened to be in.

Afraid I can't recall the names on this trade. I never looked at trading this arb myself, so always treated it as a curiousity. Also, for matched term packaged arbs, pickup was never large, you maybe could pickup 10bps on IG when skew was trading at -40.

recent skew in IG12 IG12 close Skew --------------------------------- 6/30/09 131½ -17 6/29/09 135 -15 6/26/09 139½ -12½ 6/25/09 138½ -13½ 6/24/09 140½ -12½ 6/23/09 145½ -9 6/22/09 146 -7 6/19/09 141½ -6½ 6/18/09 143½ -6½

QuoteOriginally posted by: StructCredAfraid I can't recall the names on this trade. I never looked at trading this arb myself, so always treated it as a curiousity. Also, for matched term packaged arbs, pickup was never large, you maybe could pickup 10bps on IG when skew was trading at -40.My initial guess is they would split the loot and offer something like 25bps early this year. If it was the case, we might have some ground to think about the relative value between this and say long super-senior... In this case it is more like an "academic" exercise that I need to go thru to fulfil management's curiosity. Personally, I am quite sceptical about most yield pickup ideas...

The skew in -40s was never on matched contracts. The -40 included the MR/NR basis, which was chunky on it's own. At -25 on fully matched contracts I may have cared on this trade PA