S&P Futures, MS CDS and MS Bonds

Posted by on Oct 5, 2011 in Uncategorized | No Comments

What do MS CDS and S&P Futures have in common?  Everything AND Nothing.

MS CDS and ES (E-mini S&P Futures) are clearly correlated.  As MS CDS tightens, S&P futures rally, and vice versa.  That is pretty clear.  They are also the two most talked about things all day long lately.  That is where the differences become blatantly obvious.

ES trades from 6 pm on Sunday until 4:15 Friday virtually continuously.  It is global and the price is constantly known, and you can see the depth of the market at any given time.  ES has been trading about 3.1 million contracts a day.  Each contract represents an exposure of $56,112.50.  That is a notional volume of about $175 billion.  Yesterday, ES hit a low of 1068 and a high of 1119, about a 4.5% range.  With volumes of $175 billion and huge swings, there must have been massive demand for counterparty protection?  As far as I know, CDS on CME does not trade.  I’m not sure anyone has ever asked for CME CDS and it isn’t in the DTCC list of top 1000 Reference Entities.

Now look at MS CDS?  No one even knows how much traded yesterday.  No one knows what the volumes were on the way up or on the way down.  I am guessing that about 50-100 trades occurred in the entire market yesterday.  It is just a guess from a couple of volume numbers I did get.  Given the volatility, a bunch of the trades were probably for $5 million instead of the standard $10 million.  At the low end of the range, 50 trades at $5 million, volumes would have been $250 million.  At the other end it could have been as high as $1 billion.  Realistically the number was probably right around $500 million.  Yesterday, in one of the busiest days ever, MS stock had volumes of 76 million shares.  Assuming the average trade price was 12.50 that is a volume of $950 million.  So for all the talk about how “illiquid” or “thinly traded” the CDS market is, the volumes are meaningful compared to stock volumes.  It would be nice to know the exact numbers, and it would probably be helpful, but that is something the regulators have decided isn’t necessary for functioning markets.  Until the regulators get their act together, we will have to just work on best guesses, and yesterday the volumes were probably about 3:1 stocks vs CDS, which I don’t think that is a particularly abnormal ratio.

For all the complaints about how “such a low volume market” like CDS can push the stocks, it is worth comparing to the bonds.  It looks like MS has about $187 billion of debt outstanding.  According to “TRACE” data, there were 201 trades totaling $687 million.  Not bad until you examine how TRACE reporting works.  The most active bond was the MS 5.5% of 2021, the benchmark 10 year.  If a dealer buys from a client, sells to another dealer in the street through an intra-dealer broker, who finds a client to sell to, that counts as 4 trades.  So one real trade, where the bond goes from one client to another generates 4 trades, each with a little bit of P&L if all is working correctly for the banks.  Since every buy and sell is accounted no matter what, and the interdealer trades add even more, the right volume number is somewhere between 50% and 25% of the stated volume.  That gets real volumes to around $300 million as a good estimate.  At the lower end of the range of estimated CDS volumes.  But at least with the bonds, you could download all the data and filter out the “D” trades, which are dealer to dealer, and adjust for double counting of “B” and “S” trades.  Again, why isn’t that available for CDS?

TRACE even lets you see what price bonds traded at.  If anything, it only confirms that bonds were moving similarly to CDS and that they have similar volumes.  Why have regulators been so insistent on making some transparency in the bond markets available but have done nothing to push CDS into the public domain in spite of the obvious impact on individual stocks and bonds, and even broader markets?  I don’t know.  It cannot be that they believe that CDS shouldn’t be available, and it shouldn’t be because it is too complicated – it isn’t much more complicated than stock or treasury futures, all of which are extremely liquid and transparent.

The other big talking point is that CDS is being driven by investors hedging “counterparty” risk.   That I am willing to believe plays a role in this whole move.  I have written repeatedly that the financials and counterparty risk are extremely correlated and can create negative feedback loops.  As CDS widens, clients who have bought protection are in the money.  If the spread widening that they are profiting from is name specific and non systematic, then they don’t really care much about the counterparty risk since the bank they are facing is fine.  They do care when they are profiting from a move wider in financial spreads, and the counterparty that owes them more money by the day is seeing its spread widen as well.  That creates the need to hedge the counterparty risk.  Either that, or accept collateral from the bank.  The collateral idea doesn’t work because most hedge funds aren’t set up to manage collateral, and banks are extremely reluctant to provide collateral to a hedge fund.  So we are left with the status quo, that as bank spreads widen, systematic risk fears increase.  Investors start talking about the possibility of a daisy chain of defaults.  Clients rush to hedge counterparty risk.  Traders short banks in expectation that someone will come in to hedge counterparty risk.  The banks have fought clearing and exchanges, so it really does serve them right.  So greedy to keep the market opaque and keep boutiques away from this revenue stream that they have once again set up the potential for bear runs.  After Bear Stearns this should have been fixed, but wasn’t.  After Lehman it should have been fixed, but wasn’t.  If we make it through this crisis, it likely won’t be fixed.  For the safety of the system, and to reduce systematic risk, and to eliminate inane talk of daisy chain counterparty defaults, these trades need to be cleared.  It works for stock futures which are more volatile, it should work for CDS.  The ironic thing is that the best dealers would figure out how to make money in the new system.  It wouldn’t exactly be the same way they make money today, but they would adapt and be successful.

I have to admit that I am sick of listening to talk about MS CDS when so much of the conversation addresses issues like volumes, liquidity, transparency, depth, counterparty risk, etc., when all of those issues could be, and should have been, addressed by regulators.  The focus should be on whether or not there is value in MS credit at these prices/spreads not whether the prices/spreads are merely an illusion.  I suspect that if we had all the same transparency that exists for stocks, MS CDS and bond spreads would be exactly the same as they are now, but at least we could be focused on the real problems and issues at MS.