Seeking Alpha

David Harper, CFA

 
View as an RSS Feed
View David Harper, CFA's Comments BY TICKER:
Latest  |  Highest rated
  • What's The True Risk Of Credit Default Swaps? [View article]
    Is a short position a derivative? (I didn't think so but i could be wrong). I thought we were talking about derivatives, which are generally bilateral contracts. I do think your point about supply is great; but protection buyers (i.e., short the bonds) are in an analogous (not identical) position, they need to find seller, they can't keep buying without increasing their cost (the CDS spread). It's really hard to figure the implication of growing gross notion in zero-sum market (except, to me, the systemic counterparty chains, that i do think correlates with size of gross notional).

    I agree with your negatives with respect to CDS, though. Thanks for the conversation (I will be mindful of my first sentences from now on!)
    Jan 20, 2012. 03:55 PM | 1 Like Like |Link to Comment
  • What's The True Risk Of Credit Default Swaps? [View article]
    The only disagreement i have with your last post is: a hedge doesn't require owning the underlying. It only requires long (short) some instrument with common factor exposure to the short (long) hedge instrument.

    Many useful hedges in practice are technically cross hedges; the textbook example is an airline hedging fuel costs with heating oil, but (eg) a bond portfolio can duration hedge and typically does with derivatives that have no reference to the underlying bonds (derivatives parse risk while they create dangerously unfunded contingent liabilities), or just use sovereign CDS to hedge out the sovereign component, with derivatives that do not reference the same underlyings as held by the portfolio. So, an argument against the very broad set of hedging practices does not imply the need for insurable interest. Hedging the bond you own is a very particular case of hedging.

    I still don't see, anywhere, why the CDS deserves special treatment, as a credit derivative, among several others, in a broader class of derivatives.
    (it is not to suggest your insurance measures would not lead to better safety. It looks like they would)

    I do agree, clearly, making them liquid by exchange trading will make them vulnerable to pricing manipulations ("naked CDS buyers could do likewise by calling a company's credit into question by whatever means necessary"). Absolutely. Although, that's manipulation of the M2M price, and yes he can profit (like any other market), but the CDS payout is triggered by a credit event, and his manipulation may lead to temporary mispricing, but it's unclear how he can influence the ultimately binary credit event. In the meantime (except for feedback on the company's funding cost, which i concide), those trades are zero-sum gain/loss between two counterparties. So, I agree with the (dangerous?) feedback on market price but don't see how it feedsback on probability of default (credit).

    I need to get back to work. My first/last current view on this is similar, I think, to Carlos: the logical solution would seem to be in accurately and fully margining or collateralizing the exposure of the CDS seller. Regardless of insurance status, or even OTC/exchange, if the exposures are fully funded such that a key counterparty can't fail, at least the systemic threat of counterparty chain failure is addressed. The rest of it, seems to me, to be valid concerns but that can be levied against the general class of derivatives.
    Jan 20, 2012. 03:11 PM | 1 Like Like |Link to Comment
  • What's The True Risk Of Credit Default Swaps? [View article]
    LOL, well-played and deserved "Meh." Apologies for the first paragraph, I guess I can't edit it out or i would.

    The key difference between a drop in stock price (short sale beneficiary) and credit event (long CDS beneficiary) is the corporate, and even more so sovereign events, are painstakingly well-informed decisions by a group of (basically) insiders (or, really, the decision makers on a credit event are tighter than even insiders, insiders is too broad). Market risk is not credit risk. A short can libel and do all sorts of immediately dangerous things. A hedge fund that takes a long CDS in company/sovereign can promulgate rumors, or whater, until he's blue in the face. But he's stuck on the outside. Unless he gets invited to the meetings as a hostile, meeting that include a lot of lawyers, where event decisions get made (corporate restructuring? coupon default? these are not decisions made in public markets based on rumors), even his illegal means are far more restricted than the equities short.

    In practice, I can agree with you that they tend to lack a "pro" side. But it's unclear how they should be isolated from the whole derivatives category. Any of the instruments for speculation, are also used by some as genuine hedges. Somebody's speculation is another's hedge. On this, i agree with you in practice, but disagree in theory.

    I do agree with your argument-question: "why would the means have to be legal?"

    the weakness is the reductio ad absurdum: if the argument is that (i) that have no "pro" side (which is very debatable) and (ii) illegal means can be employed, then I fail to see, on close examination, how they differ from many OTC contracts, which have obvious speculative uses (and therefore, by definition!, valid hedging applications) where one counterparty is motivated on the financial loss of an asset, and where we can imagine illegal means to produce that gain.
    Jan 20, 2012. 01:01 PM | 2 Likes Like |Link to Comment
  • What's The True Risk Of Credit Default Swaps? [View article]
    Meh. We've all read this argument a million times, can we add some depth or perspective to it? The life insurance analogy (the hitman appears!), although copied/pasted endlessly still doesn't help me.

    As Carlos says, market participants can already short financial assets routinely; Einhorn shorts Green Mountain without an "insurable interest." Are we worried he's going to break into their headquarters and break things? Only to a point: he doesn't have a board seat; probably isn't a claimant in the capital structure. Or Bill Gross shorting U.S. Treasuries, any less visible conflicts there? A naked CDS protection buyer, to my knowledge, doesn't sit on the board or have a legal claim, like bond and debt holders, on the firm's capital structure.

    Can you explain to me, how in the real world, a naked CDS protection buyer, via LEGAL means, can sneak past the syndicate of sophisticated legal capital structure claimants, all of their lawyers and corporate arrangements (hypnosis?) and trigger or even influence a credit event?

    I would be much more persuaded on this point if somebody can point me to a single example of where a protection buyer influenced default (unlike your life insurance analogy, where literature gives me plenty of murder mystery examples!)

    Please include, in your answer, an awareness that the long CDS (who benefits from the default) is matched by a short CDS who is equally motivated in the other direction. In a naked CDS, you have two counterparties on opposite sides of the trade, neither with any obvious legal (is that correct?) means to influence default. If good so far, it reminds me of ... just about every other other derivative in existence: they are all bilateral bets, where one of the counterparties wins on a drop in value of a financial asset.

    (There are issues, but they more realistic; e.g., as the CDS market appears to influence bond spreads, net protection buying can widen the basis. And, I think you can make an argument than naked CDS can exacerbate a companies funding costs ...)

    No, the more real issue is margin/collateral: if the protection seller were fully collateralized, there couldn't be counterparty chain failures. But to fully collateralize them would probably make them too expensive.
    Jan 20, 2012. 12:04 PM | 2 Likes Like |Link to Comment
  • What's The True Risk Of Credit Default Swaps? [View article]
    Hi Carlos, Yes, agreed, excellent questions to which I do not have the answer.

    I think the contrast to futures margin is spot-on: with a CDS, there is "jump-to-default" risk. Basically, if you are short protection, you are posting collateral on a currently priced (mark-to-market) position in the CDS but this tends to be far less than the notional, and in sudden default, will be less than the obligation. The notional, after all, is real in the sense that it represents the maximum payout (under 0 collateral). So, historically at least (I am not current), the margin/collateral has been the issue.

    I agree we would still have question of evaluating the numbers, the question of "we don't know who the players are, and how solvent they truly are" is EXACTLY why, imo, we can't know the total risk (as the net notional understates in the case of even one insolvency). Thanks!
    Jan 20, 2012. 11:41 AM | 2 Likes Like |Link to Comment
  • What's The True Risk Of Credit Default Swaps? [View article]
    Hi Carlos,

    The firms don't really get into offsetting trades immediately, it's not quite as silly as you suggest. Although it is silly in other ways.

    CDS are OTC (i.e., illiquid) bilateral contracts, and time is a dimension.

    If you sell me protection today, say $100 notional, as we go forward in time, just like with futures or any position, you may want to later reduce/unwind your position. But unlike a futures contract you don't have an exchange to give you immediate liquidity; you have a somewhat illiquid long-term contractual obligation. So, probably you enter a new trade in the other direction; buying protection (+$100 gross notional = $200 gross, net notional = 0). It's like selling stock, but the OTC market structure sort of makes you put on a new trade to sell. Cumulatively, you are a paradox, but it's only because you changed your mind ;)

    What's the risk? Mathematically, the net notional is conservative for the zero recovery but it contains an devious assumption: that all the counterparties in this extensive netting chain do honor their contracts. (and if we assume they do, you can see why the gross notional indeed vastly overstates the risk). The rub comes into play because, under a worst case scenario, a single net protection seller fails and this ripples thru the chain. In this case, as the net notional does assume all counterparties honor their obligation, the risk can quickly exceed the net notional.
    Jan 20, 2012. 12:40 AM | 2 Likes Like |Link to Comment
  • Fascinating Data For Credit Default Swaps [View article]
    I used "firm" as a proxy for "counterparty family" to keep the example simple, to try and help you understand the difference between netting and recovery. Why firms? To hedge (netting is a counterparty risk mitigant; e.g., basel capital relief). I can't speak to how a financial reporter generated a number, do i have this correctly, that "some [unsourced?] analysts speculate."

    No, it's not really as complex as a CDO, a synthetic CDO would be a basket of CDS.

    I have not researched it deeply, but just based on the DTCC pdf above, I see no reason to prima facie trust the net notional numbers. But my skepticism would be based on the key implied assumptions (e.g., what is the basis for aggregating "families;" do these correspond to actual contractual/isda in-place netting) not on a recovery assumption (as the recovery assumption appears to be the most conservative possible; i.e., 0%)
    Jan 19, 2012. 12:37 PM | Likes Like |Link to Comment
  • Fascinating Data For Credit Default Swaps [View article]
    Say I (working at my firm A) buy protection on Greece, $100 notional CDS, from you (working at your firm B).

    Also, your colleague (firm B with you) buys protection from my colleague, $150 notional (working at firm A with me).

    Two trades = $250 gross notional

    But, if Greece defaults, my firm (i.e., "counterparty family") has a net notional exposure of $50. Or, i guess per DTCC method, my "family" would have -$50 net notional exposure. But this is still a maximum assuming 0 recovery.

    I totally agree with your insurance example, the actual transfer will be $50 notional * (1 - recovery rate).

    So there is firstly a huge set of assumptions around the netting that produces net notional, but note that it is assumption about how trades are aggregated (which hopefully matches the actually contractual netting!)

    From the your pdf but i think the illustration shows this is a netting that does not include a recovery assumption:

    “Net Notional Values” with respect to any single reference entity is the sum of the net protection bought by net buyers (or equivalently net protection sold by net sellers). The aggregate net notional data provided is calculated based on counterparty family. A counterparty family will typically include all of the accounts of a particular asset manager or corporate affiliates rolled up to the holding company level. Aggregate net notional data reported is the sum of net protection bought (or equivalently sold) across all counterparty families. Net notional values are displayed only for single reference entities (single names and indices) and therefore only provided for tables 6, 7, 14 and 15. Net notional positions generally represent the maximum possible net funds transfers between net sellers of protection and net buyers of protection that could be required upon the occurrence of a credit event relating to particular reference entities. (Actual net funds transfers are dependent on the recovery rate for the underlying bonds or other debt instruments.)
    Jan 18, 2012. 10:16 PM | 1 Like Like |Link to Comment
  • Fascinating Data For Credit Default Swaps [View article]
    I think you are (understandably) confusing recovery with netting. The net notional implicitly assumes 0% recovery; i.e., as if the contract paid the entire notional. With 0% recovery, there is still netting within counterparty families. So i think the parenthetical about recovery means: assuming netting, these are max fund transfers but transfers will be less as/if recoveries are higher than 0. Put another way, i don't see how recoveries can be inferred from the data.
    Jan 18, 2012. 06:33 PM | Likes Like |Link to Comment
  • Wall Street Breakfast: Must-Know News [View article]
    Rachael, thank for the breakfast must-know, it's a must-read
    May 5, 2010. 04:52 PM | 1 Like Like |Link to Comment
  • Review of a Classic: Quality of Earnings, by Thornton O'Glove [View article]
    Marc is correct: it it dubious to suggest that cashflows always trump accruals. Some research finds otherwise (see "Is Cash Flow King in Valuations? CFA Institute, FAJ Journal. Their Answer: "Contrary to the common perception that operating cash flows are better than accounting earnings at explaining equity valuations, recent studies suggest that valuations derived from industry multiples based on reported earnings are closer to traded prices than those based on reported operating cash flows").

    In addition to Marc's points, a couple of examples: capex (lumpy) and financial instruments that change value without immediate cash flow implications
    Feb 6, 2010. 04:33 PM | Likes Like |Link to Comment
  • My Problem with Cuomo Going After Ratings Agencies [View article]
    Weak argument. The issue is whether the work product and methodology justified the fees, or if not, constituted a professional negligence. This can't be decided at the superficial level of your argument: clearly, at some hypothetical extreme, taking fees for inappropriate ratings would be actionable. The problem is that, in regard to structured products, the rating agencies *clearly* applied dubious model practices to insubstantial datasets - I don't know the legal implications, but it's worse than you make it sound. David Harper, bionicturtle.com
    Jul 21, 2009. 07:48 PM | Likes Like |Link to Comment
  • My Problem with Cuomo Going After Ratings Agencies [View article]
    Weak argument. The issue is whether the work product and methodology justified the fees, or if not, constituted a professional negligence. This can't be decided at the superficial level of your argument: clearly, at some hypothetical extreme, taking fees for inappropriate ratings would be actionable. The problem is that, in regard to structured products, the rating agencies *clearly* applied dubious model practices to insubstantial datasets - I don't know the legal implications, but it's worse than you make it sound. David Harper, bionicturtle.com
    Jul 21, 2009. 07:47 PM | Likes Like |Link to Comment
  • My Problem with Cuomo Going After Ratings Agencies [View article]
    Weak argument. The issue is whether the work product and methodology justified the fees, or if not, constituted a professional negligence. This can't be decided at the superficial level of your argument: clearly, at some hypothetical extreme, taking fees for inappropriate ratings would be actionable. The problem is that, in regard to structured products, the rating agencies *clearly* applied dubious model practices to insubstantial datasets - I don't know the legal implications, but it's worse than you make it sound. David Harper, bionicturtle.com
    Jul 21, 2009. 07:47 PM | Likes Like |Link to Comment
  • My Problem with Cuomo Going After Ratings Agencies [View article]
    Weak argument. The issue is whether the work product and methodology justified the fees, or if not, constituted a professional negligence. This can't be decided at the superficial level of your argument: clearly, at some hypothetical extreme, taking fees for inappropriate ratings would be actionable. The problem is that, in regard to structured products, the rating agencies *clearly* applied dubious model practices to insubstantial datasets - I don't know the legal implications, but it's worse than you make it sound. David Harper, bionicturtle.com
    Jul 21, 2009. 07:47 PM | Likes Like |Link to Comment
COMMENTS STATS
29 Comments
12 Likes