Some FACTS
Source: DTCC
- DTCC Posts 6-Month Market Activity Snapshot on CDS Market
- Default in Today’s Advanced Economies: Unnecessary, Undesirable, and Unlikely
- The Impact of Banks’ Cumulative Reserve Position on Federal Funds Rate Behavior
- A Wolf in Sheep’s Clothing
- TBA Trading and Liquidity in the Agency MBS Market
- Impact of High and Growing Government Debt on Economic Growth
- Financial Amplification of Foreign Exchange Risk Premia
- The Central-Bank Balance Sheet as an Instrument of Monetary Policy
- Exorbitant Privilege and Exorbitant Duty
- The Information Value of the Stress Test and Bank Opacity
- Price of Risk—Recent Evidence from Large Financials
- Shadow Banking
- Summer Doldrums
- Credit Default Swaps: What Are the Social Benefits and Costs?
- The International Role of the Euro
- Chronicle of Currency Collapses: Re-Examining the Effects on Output
- The Paradox of Toil
- Links
- What the Fed Did and Why
- Markets and Government Before, During and After the 2007-20xx Crisis
- Detecting and Interpreting Financial Stress in the Euro Area
- Are We Building the Foundations for the Next Crisis Already? The Case of Central Clearing
- Links
- Links
- OIS Discounting
- Multimarket Trading and the Cost of Debt: Evidence from Global Bonds
- Is Economics Coursework, or Majoring in Economics, Associated with Different Civic Behaviors?
- China’s High Saving Rate: Myth and Reality
- Oil Spill
- Eurozone €440 Bln SPV Aid Fund
- ECB Financial Accounts and ECB Financial Strength
- Central Bank Swap Networks
- DTCC Posts CDS Market Activity Snapshot
- ECB: Financial Stability Review (June 2010)
- Global CDO Issuance by Transaction Structure
- Links
- TBAC Minutes: Sovereign CDS and Swap Spreads
- Goldman Sachs Underwriting Market Shares in Subprime RMBS and CDOs
- Abacus for Dummies
- The World’s Safest and Riskiest Sovereign Debt (Update)
- Increased Sovereign Risk Behind Negative Swap Spreads
- Haircuts
- Markets, Religion, Community Size, and the Evolution of Fairness and Punishment
- Beyond the Dollar: Rethinking the International Monetary System
- Links
- Amazing Discovery
- ABCP Outstanding: Still Crashing
- Greek Government Bond Market
- Links
- Sterling Today
The customer/retail deals are most likely not listed with DTCC as the bank typically (as part of its ‘value added’) does the derivatives processing for its clients.
======
Well, it sounds like we need a call to the DTC. I understood that the major bank dealers registered all their trades/contracts, including those with clients, not just those with each other.
Also, in your numbers, if the dealer didn’t show the $10 billion short to its client base to the DTC, but had laid off $8 billion to other dealers, which is did show, it’s not clear how the DTC’s figure is “highly misleading”.
I could be wrong, but I think it is a stretch that exotic terms would double and treble the amounts. Of course it is possible, but … don’t most exotics usually only account for a small amount of the total trade? Is it different for CDS market?
Bi-lateral netting might not be important, as you point out, in a highly fragmented market. I would guess, however, that an 80/20 rule applies – that the 20 largest client players account for 80% of the trades.
The credit exposure to the hedge funds is real, but not missed/exaggerated by the DTC figures. Whatever lax attitudes about credit lines and collateral were extended in the past, that’s not what is likely going today, right? Stands to reason that some part of the deleveraging that jck talks about is about collateral calls for all types of contracts, including CDS. [Of course, that's an over-reaction, because the counter-party credit risk has been accurately priced all along, and the dealers are amply reserved for these risk, by having pocketed the bid-ask spreads along, in case of a rainy day ... yuk, yuk.]
ss:
The data on international banks, show that there has been a huge contraction in credit, of over $1.1 trillion, in the second quarter, to put things in perspective this is multiple times the contraction that occurred after the bursting of the dotcom bubble or the LTCM crisis. As I have said before, the CDS market has nothing to do with it, de-leveraging is all there is to it, and you don’t “cure” years of over-leverage in 3 days or in 3 weeks.
All I can say is the equity markets are tanking, banks and insurance companies are ponying up for Fed money, bailouts of sovereign nations by the IMF are happening in droves, stock markets are closing for trading regularly intraday — pretty much the kind of points emphasized over at the Roubini article.
Add in the unwinding of the carry trade and the flight to treasuries and you have too much for this little brain to handle. If it’s not the CDS market causing these problems, and I’m not wedded to the idea that it is, I’d love to hear a coherent explanation of what is the cause(s) of all this turmoil. We’ve discussed the credit market freeze potentially being the result of funds being parked at the Fed rather than lent. There is some disagreement about this and I saw an article this morning which put forth my initial impluse that the freeze is related to a repricing of risk and that no one wants to borrow at the current risk adjusted rate. Having said that, the rest of the explanation for the current meltdown eludes me.
Not to beat a dead horse, but I think the DTCC data is highly misleading and borders on being disingenous. Netting is not all that it appears to be at first blush and the systemic risk of a bank (or investment bank failing) is significant.
A bank (or investment bank) most likely does CDS transactions hundreds of ‘retail’ counterparties (hedge funds, mutual funds, corporates, insurance companies, wealthy individuals, smaller banks). These deals may be highly non-standard (forward starting, step-up, step-down, amortizing, mutual termination). The bank typically takes the net postion from these deals and then hedges with another dealer.
For example, a bank may have as many as 5000 CDS with 3000 different customers with LEH as the underlying. The net exposure to the bank might be short $10 billion in protection. The bank hedges this with a few deals with other dealers. It is important to note that the bank is the sole determinant of the ‘net’ $10 billion position. The customer/retail deals are most likely not listed with DTCC as the bank typically (as part of its ‘value added’) does the derivatives processing for its clients.
The bank also most likely doesn’t do a daily cash mark-to-market/margin on these deals with their customers. There probably is some kind of broad collateral agreement that covers all customer deals (loans, FX, interest rate derivs, credit derivs, etc.). This collateral most likely consists of illiquid assets such as real estate, personal guarantees, receivables/inventory, preferred shares and some liquid assets such deposits with the bank.
Given the non-standard nature of the customer deals and the lack of daily mark-to-market cash margining, the bank assumes considerable basis and credit risk (for which they no doubt charge a hefty sum). In the event the bank (investment bank) fails, there are two major risks:
1) The bank’s ‘netting’ is wrong and they are unhedged
2) The bank’s customers fail and they are unhedged
Right now, the bank’s balance sheet and mark-to-model masks/covers these risks. So the DTCC can say that the net exposure is quite small and nothing to see here folks go on about your business.
In reality, a bank/investment bank failure exposes these risks which is why the Tsy/Fed are anxious to prop up the GS and MS of this world
Amicus:
Bi-lateral netting applies to all signatories of the ISDA master agreement, that is everybody trading CDS and other derivatives subject to that agreement, so there is bi-lateral netting between dealers, and between customers and dealers.
oh, I totally concur that the DTC figure is the best starting place for most estimates.
What I wanted to target is the assumption (of doomsters?) that, because the DTC doesn’t have everything in the registry that the net amount *necessarily* goes up.
In practice, it probably does, given what we know about some of the risk-takers presumably not on the list, but that’s no reason to jump to the worst possible figure or outlandish estimates, either.
Do you know if the ISDA protocol(s) require bi-lateral netting for all participants? It’s fairly clear that it’s done by the major institutions. I’m not sure how prime brokerages are handling that among their client base, though.
Amicus:
A very large number of contracts are registered with DTCC, including those written before 2006, through a process they call “backloading”. Nobody has a better estimate of the numbers than the DTCC. Everything else is a “pulled off a hat” number with zero credibility.
Can I chime in?
I grow weary of reading fantastic estimates from the Nouriel Roubini website.
The fact is, that no one knows the total net settlement figure.
Indeed, the fact that a large portion of contracts may not be registered, for any given name, does not imply that the net settlement figure will be larger than what the DTC has estimated.
We can, of course, guess and even make informed guesses. But Nouriel and others are not carefully qualifying the “doom” numbers in that way (that I’ve seen at least – and I don’t read or hear everything, obviously…).
I’m sorry, but I find his approach to estimation and his publicity of it … irresponsible, _almost_ to the extreme, and I’m not afraid to say so …
/end of chime in
interested reader:
You don’t bother me at all, but you certainly are confused and I feel your pain ;_)
Everyone knows that DTCC is a clearing house, BUT THEY DON’T CLEAR CDS, why do you think the Fed and others are pushing for a CDS clearing house?
Answer: ’cause there isn’t one.
What they do is fancy back-office work and payment calculations, they do not hold the collateral/margins behind CDS trades.
The ISDA piece doesn’t say they clear CDS, it says “all of the major global credit default swap dealers have registered in the Warehouse the vast majority all contracts executed among each other.” Including via “backloading”, most contracts written before TIW was launched.
Don’t panic, things are bad, many HF will go bust but not because of the CDS market.
http://www.dtcc.com/news/press/releases/2008/dtcc_processes_lehman_cds.php
“About DTCC
DTCC, through its subsidiaries, provides clearance, settlement and information services for equities, corporate and municipal bonds, government and mortgage-backed securities, money market instruments and over-the-counter derivatives.”
“all of the major global credit default swap dealers have registered in the Warehouse the vast majority all contracts executed among each other.”
“One of the many central servicing functions of the Trade Information Warehouse is to calculate payments due on registered contracts.”
“At the time of the bankruptcy of Lehman Brothers Inc., approximately $72 billion in credit default swaps written on Lehman Brothers were registered in the Warehouse.”
ISDA CEO Robert Pickel says in his Oct 21 statement that the notional outstanding was $400bn.
http://www.isda.org/press/press102108.html
I apologize again but we must set the record straight on this. When I think the whole chain through I can’t help feeling dismayed at the real pain this stuff is causing. Thanks for bearing with me, I won’t bother again.
interested reader:
I am afraid some of your statements are not correct.
First, the DTCC doesn’t clear dealers CDS trades, it merely services them.
Second, and more importantly the DTCC doesn’t just interfaces with dealers but also with “buy-side” firms and these firms have their trades registered and processed by the DTCC or more precisely by the Trade Information Warehouse.
If you want to know more:
http://www.dtcc.com/downloads/brochures/derivserv/DerivSERV%20Brochure.pdf
Sorry, I got carried away. I’m just trying to make sense of it.
Again, the DTCC clears mainly interdealer positions that are meant to be hedged (and even if they weren’t, dealers have Fed liquidity access anyway.) In the case of Lehman, $70bn out of $400bn notional contracts were registered, cleared and settled smoothly through DTCC (not a surprise, given largely hedged interdealer books and Fed liquidity.) The remaining $330bn notional is unregistered credit risk held by net credit protection sellers such as insurers, monolines, CDOs, CPDOs, hedge funds that are either unregulated or undercapitalized or both for the risk they are holding. Predictably, insurers and monolines are standing in line for a slice of TARP. Predictably, structured finance SPV that need to unwind trades with Lehman, WaMu, Icelanding banks as reference (which could take months) are driving high-yield and investment-grade CDS spreads beyond fundamental values and their own NAV below trigger values. Predictably, hedge funds that sold protection on Lehman thinking that Lehman would never fail or that invested in SPV tranches are folding in record numbers. The predicted market fallout from the Lehman settlement did not happen on October 21 but 3 days later but happen it did and we’re not finished yet. The DTCC itself is the non-event.
http://www.rgemonitor.com/economonitor-monitor/254133/post_mortem__lehman_cds_settlement_360bn_or_6bn
I was playing with some numbers from the WAMU auction.
It seems to have been more “well ordered” than the Lehman auction. Among the interesting factoids:
-The top players in the bidding phase seems to vary (more than I expected). Deutche was biggest for Wamu, but dead last for Lehman bonds. Goldman was at 22% of the “winning bids” in the Lehman auction, but just 10% of the WAMU auction.
-In WAMU, no one bid over the mid-market price during the auction phase.
-The bid-ask for mid-market seems to be set a 2 pts, by convention…
-For a net open to sell in WAMU, the wtd avg bid, among “winning” bids, averaged about 1.9 and 2.3 points below the indicated bids in the first phase and the mid-market determined by the protocol for all auction participants. They averaged *less* for the Lehman auction (1.2-0.53)…
-The final price is greatly influenced by the “largest” bidders. The distribution of all bids for WAMU suggests that there is “interest” at higher prices (perhaps client interest that must be “filled”), a gap, and then another set of bids. The size of that smallish gap greatly influences, it appears, the “final price”.
Coincidentally or not, the Lehman settlement date corresponded to the lowest recent s/t LIBOR rates …