Musings on Credit Default Swaps

As readers may know, I view the credit default swaps market with more than a bit of skepticism. I can point to cases where it has caused harm:

1. Bagholders. Dealers claim that CDS are really not bad at all because they haven’t been taking risk, oh no, they hedge their position with offsetting swaps.

So let’s assume they are right. The dealer community is a closed system with no net exposure. But you have the buyers of contracts (hedgers and punters). Someone somewhere has to be assuming the risk.

We used to know who that someone was: AIG, and on CDOs, Ambac and MBIA (their policies were structured more like traditional insurance but was also sometimes called credit default swaps). But other bagholders have included public investors who bought products with enhanced yields (typically synthetic CDOs) but where they’d actually lose money if defaults on the referenced bonds went above a trigger level. And guess what? Those products typically contained names of real turkeys like AIG, the auto companies, Lehman. They were also peddled heavily overseas to chumps that would only know these were famous American companies.

2. Distortions in cash bond pricing. The theory is that CDS make markets more efficient, but in reality, they can distort the price of new issues. That means they are damaging the real economy by making costs higher than they “ought” to be. From March 2008:

The problem started in the second half of last year when subprime mortgage delinquencies started to rise, causing investors to retreat from complex instruments such as synthetic collateralized debt obligations, or packages of credit-default swaps that became hard to value. The swaps are contracts based on bonds and used to speculate on a company’s ability to repay debt.

As values of CDOs began to fall, banks that had sold swaps underlying the securities started to buy indexes based on them instead, a method of hedging their losses on portions of the CDOs they owned. The purchases are driving the cost of the contracts higher, raising the perception that company bonds tied to the swaps are suddenly riskier and leading investors to demand higher yields throughout the corporate debt market.

The Financial Times reported that this was continuing in May.

3. Incentives to push companies into bankruptcy. One worry, that seems to be coming to pass, is that CDS are leading investors to be indifferent to a bankruptcy, and in cases to push for it. Since they own a CDS, they’ll get their payoff, while negotiating a restructuring takes time and money. Why bother? But negotiations can keep companies out of BK, and are also necessary for Chapter 11 to succeed. And if a restructurings fail, more job losses result. This too is a toll on the real economy.

I have a fourth concern, which is harder to prove, and I’d welcome reader feedback. CDS also create an incentive for lenders and investors to skip or do a cursory job on credit research. Why bother if the market give you a price? The reason that this is particularly bad is that it is the lender that can do the best examination, by virtue of direct interaction with the borrower and being able to obtain non-public information. Having the original lender be able to lay off risk presumably had lead them to cut back on their credit function. Is this correct? If so, what evidence can you point to? When did this start and how pronounced has it become?

Yet another argument when the bank stocks were taking a beating was that speculators pushing up CDS spreads were much bigger culprits than short sellers, but I am not clear as to the mechanism.

I am also wondering if the same thing has happened on the investor side, that credit research isn’t what it used to be.

I am similarly skeptical of the purported benefits. More liquidity does not appear to have produced less volatility, which is the theoretical reason why liquidity is a boon. I also don’t buy the price discovery argument. In the Volcker era, when bond trading was under stress, I cannot once recall anyone complaining about price “discovery” being a problem in corporate bonds. Please. Even then, there were enough liquid issues to easily grid prices for those that didn’t trade much (bonds trade by attribute, duration, rating/credit quality, presence or absence of call features, etc,).

Another bone of contention (although this is a more specific complaint) is that when the industry argues against moving to exchanges, they argue that they need to do OTC trades so they can “customize”. Yet I have never read a single example of why they need customization (as in when and how it serves investor needs, with specific examples) and how much the investor would lose by buying off the rack products. The lack of specificity in any of these claims makes me believe the main reason is to create complex products to hive risk off onto bagholders, or to charge extra for an added feature that really does not do the investor a meaningful amount of extra good.

If you disagree with that view, you need to provide examples with realistic pricing. Assertions merely support my suspicion.

A final query: does anyone have a sense of CDS economics to dealers, as in how important a profit source this is to the big banks? I know it varies over the cycle, but an order of magnitude idea would be helpful

Print Friendly
Tweet about this on TwitterDigg thisShare on Reddit0Share on StumbleUpon0Share on Facebook0Share on LinkedIn0Share on Google+0Buffer this pageEmail this to someone

17 comments

  1. Anonymous Jones

    I simply would be floored to find out that certain CDS markets do not decrease the amount of credit research. There’s a reason why people know the acronym IBGYBG, and even then, the problem is more insidious than than IBGYBG (there are always the Madoffs of the world who keep a scheme going even when they are massively behind the 8 ball). There are risk seekers in this world. I’m not convinced that the braintrust behind AIGFP was simply into scamming as much as they were risk-seekers. Look, there is a legion of Option-ARM borrowers who attest to the risk-seeking nature of humans. A risk-seeker in the CDS market does not need to do any credit research. He needs only to make an incredibly leveraged bet; if it works out, he is rich. If it doesn’t, well, he’s judgment proof anyway (and debtors prisons have gone the way of the Dodo). This is why Rocky Balboa (ante-Apollo Creed) was employed in the art of breaking people’s thumbs. The breaking of thumbs is the disincentive to punters to make bets they can’t cover. This is standard practice in gambling (even the ‘legal’ sorts you find in Vegas); of course, even extreme physical pain is insufficient disincentive to determined punters, but it’s certainly better than nothing.

  2. Timothy

    Yves,
    I think the bankruptcy issue is a moot point. Pushing companies into Chap11 is not a bad thing. In the old days, restructuring is done during Chap11. Negative basis trades (which are arbitrage trades) are just reflecting this idea. However, the idea of stealth restructuring (ie making debt holders bear the cost while equity guys benefit) is a perversion of the system. If a company has problems that pushes it to Chap 11, so be it. The short termism of the markets have let to stealth restructuring as the rule rather than the exception. As a credit investor, I take offence to it. A company that is run well will never run the risk of this happening. And companies that deserve to be bankrupt SHOULD BE allowed to be bankrupted.

    The second note is credit creation has expanded to a large extent due to the credit derivative market. I am not saying this is right or wrong. But the ability to distribute risk is greatly enhanced by the CDS market. Of course, like any effective tool, it can be abused. The lack of good regulation governing credit derivatives was the main reason why things got out of hand. Think of the S&L crisis. Lending was rampant and without control, which led to the crisis, and subsequent regulations capped future blowouts. The same is happening now. Except it is just a hell of a lot bigger. The unholy trimvate of bad regulators, greed and over-reliance on incompetent rating agencies let to this mess we are in. Do not blame the tools. Blame the authorities. Anyone in the industry could have told you the risks involved.

    As for distortions, well, when times are good, every company achieved extremely tight pricing, because of the CDS market. Greed and total dis-regard for risk let to that. Imporper regulations let to excess leverage on the derivatives. And the result was a credit bubble. Leveraged synthetic CDOs, CDO-squared, counterparty risk were all misregulated. Regulators were supposed to keep greed in check. Thats the point of having them around. Guess what? They didn't do their job. Combine that with excessively cheap money and we have a molotov cocktail mix.

    All I am saying is that tools are just that. Like a stick of dynamite, it can be used to do good or evil. Don't blame the tools for the folly of people.

  3. bobo bobo

    yves: Yet another argument when the bank stocks were taking a beating was that speculators pushing up CDS spreads were much bigger culprits than short sellers, but I am not clear as to the mechanism.

    ——–

    The idea isn’t too complicated. A hedge fund shorts stock in a company, and then buys CDS on the company’s debt in order to cause a spike in the company’s implied risk of default, causing a reduction in the stock price. The reason this works is that the CDS market isn’t too liquid, so one can spike credit spreads without spending too much money buying CDS.

  4. bobo bobo

    timothy: I think the bankruptcy issue is a moot point. Pushing companies into Chap11 is not a bad thing. In the old days, restructuring is done during Chap11. Negative basis trades (which are arbitrage trades) are just reflecting this idea.

    ———–

    The problems with CDS and bankruptcy is that if a creditor is hedged they exercise their voting rights in bankruptcy to maximize their overall position (ie, position in debt and CDS), which may encourage them to torpedo ANY plan of reorganization. The bankruptcy rules need to change to deny creditors who are hedge voting rights, or to give them to the counterparty that takes on the credit risk of the underlying position. Separating voting rights from economics defeats the bankruptcy code’s policies of helping viable companies restructure so they can get a fresh start.

    So for example, you buy a very big position in credit protection, and you buy a much smaller position in company debt, just large enough to let you veto the debtor on the underlying instrument’s plan of reorganization. You can make more money from your big CDS position if the company liquidates than you make on your small debt position if the company successfully restructures.

  5. x2KgqGo1o4PT5WNr4mv.5z79X40-

    How can I come out ahead by buying a bond plus a CDS instead of simply buying a different bond with lower risk of default? I shouldn’t be able to break even as the CDS writer has to make a profit.

    Once you Google “AIG” and “side letters” things make a lot more sense.

  6. Timothy

    Bobo,

    That is actually not true. Reorganisation done under Chap11 is after the event. Once the company goes into Chap11, the arb trade is crytallised. At that point, the risk taker can go take over and restructure the company. In the case of a debt restructuring, it also counts as ‘an event of default” which will also trigger the cds. What it prevents is stealth restructuring. For example, when GM threatens bond holders with a debt exchange or bankruptcy. Creditors have become a bullied lot, even though they are senior to equity guys.

    So, if a company wants to reorganise, the most equitable way is through Chap 11. And debtholder/creditors must realise this.

  7. bobo bobo

    timothy: That is actually not true. Reorganisation done under Chap11 is after the event. Once the company goes into Chap11, the arb trade is crytallised.

    ———-

    If a CDS only allows physical settlement, and a creditor can drive down the price of debt by vetoing a plan of reorganization, it looks like doing that would be a way for a creditor to enhance their CDS claim by letting them buy cheap debt to settle a CDS trade.

    Ignoring bankruptcy voting to enhance a payout under a CDS, an out of court workout might be preferable to creditors. You have the same issue with bankruptcy filings. If a creditor buys just a big enough portion of a company’s debt to cause a credit event, and a much bigger position in CDS, the creditor might vote for a default in order to make money on the CDS even if unhedged creditors think that is a bad idea.

  8. bb

    Yves,
    as W.Buiter pointed on the subject, CDSs are not hedges, they are not insurance on bonds. for an insurance to be regarded as such, the insured should have insurable interest: he cannot be better off in the case of an insured event. this certainly does not apply to a market that is multiples of the ‘insured’ bonds. why would otherwise people purchase insurance for 10x their risk?
    and besides that, anyone who follows credit spreads knows that they track more closely the equity than the bonds of the issuer. why? because you can trade CDSs with other broker/dealers at 0 margin and with a very small margin with the smaller fish. compare that to an equity position that requires real money: 1/2 cash the other 1/2 borrowed with interest. as well, if you want to place a pure equity bet, it can take some time and you would not know the price, even the ballpark number, if you want to get a 100m exposure for example. you can always set off the CDS exposure later in the futures or equities or even another CDS contract.
    the CDS game adds no value to the market: it kills volatility in normal markets and excaberates it in tense ones.

  9. Timothy

    Bobo,
    In a restructuring event (at least outside the US), a CDS event has occurred. In the US, if it involves all bond issued, it is also an event of default (at least pre the Big Bang in April this year). From experience, buying bonds to create a credit event is a dodgy thing at best. For one, getting cheap bonds is not easy. And getting it at a decent arb level is even harder. As for out-of-court workouts, so far, I have only encountered companies who wish to restructure debt without restructuring other liabilities or for that matter, allowing equity dilution. That is not reorganisation, it is blackmail and bullying. A creditor is supposed to rank senior to equity and subordinated stakeholders but without a proper court organised reorganisation, most creditors are generally treated unfairly. From where I am sitting (in Asia), I see creditors taking a harder line to weak corporates because of this reason.

    Taking the view of an unhedged creditor, the main reason they do not want to go through bankruptcy proceedings is two-fold: (1) they are not allowed to hold defaulted assets, or (2) they are not willing to go through the time consuming process. Both are very bad reasons. There are a group of investors who will look at these assets : vulture funds. They ware willing to do the work to get the profits and they too will push for bankruptcy to get the best structure for recovery. Without the bankruptcy card, companies are generally not willing do reorganisations that hurt in the short term but benefit in the long. Just look at the GM fiasco.

  10. Snarky Librul

    yves – As a former banker where we did use CDS as hedges I’m not sure that it is true having the capacity to hedge weakened our credit underwriting per se. What CDS were more likely to be used for was to enable us to take a larger piece of a corporate loan than we might otherwise. So absent CDS we may have lent company X $25m, but with CDS we could lend company X $75m then hedge say $50m. The theory at least was by lending the $75m we’d get more ancillary biz from the company to make up the cost of the hedge…sometimes that was true, sometimes it wasnt.

    So to me CDS added alot of lending capacity which I suppose in and of itself is a good thing. Whether that expanded capacity was used wisely is a different question.

  11. Jacob

    On your fourth point, there has been a decline in credit research generally, although I think that has more to do with banks cutting back their research departments rather than CDS incentives.

    I can see your point about CDS removing a research incentive for some investors, but doesn’t it also create one? The very speculators in CDS we are so quick to decry (rather than the people who use them “legitimately” to hedge some exposure) have a very strong financial incentive to do the underlying credit research. The theory that the swap price has discovered anything at all requires those participants to do their research.

    I’m not sure whether the net effect is more or less research. Maybe the research is not being done by the same parties, but it most assuredly is still being done by someone. As long as there is a way to profit from more accurately understanding the quality of a firm’s credit, that’s not likely to change.

  12. Greg

    Customized OTC trades create information asymmetries that can be far more lucrative than exchange trades. Unfortunately, you can find such asymmetries behind just about any service. Some folks exploit the asymmetries. Others find ways to profit from reducing the asymmetries.

    It really comes down to the ethics of those who hold the valuable information. If the counter-party doesn’t have the information to protect his own self-interests, under what obligation is the other party to provide it?

    In cases such as residential mortgage lending, the counter-party may lack the mathematical or other skills needed to understand and analyze a complex instrument. To what extent can information asymmetries be tolerated while creating the “meeting of the minds” necessary to contract?

  13. JJ

    it is a two-way street with customized products in which each side benefits: the sell-side can increase the profit margin by charging more and the buyer can justify a MTM which differs from true value

  14. Valentine Michael Smith

    Hi.

    Yves, you say: “CDS also create an incentive for lenders and investors to skip or do a cursory job on credit research.”

    The capacity for offloading risk is almost by definition an mandate to disregard risk. We can all see what happened in the mortgage business when loans (good or bad – it didn’t matter any more!) were instantly bundled up and sold off.

    Going through the motions of risk-assessment at some point becomes a waste of time, and makes you uncompetitive with those who interpret the rules at their lowest common denominator.

    Yves, you then say: ” …it is the lender that can do the best examination, by virtue of direct interaction with the borrower and being able to obtain non-public information. Having the original lender be able to lay off risk presumably had lead them to cut back on their credit function.”

    Maybe I’m not sophisticated enough to see the differences, but this seems to me to parallel the logical steps in the subprime mortgage loan process.

    As soon as people realized what could be done with the tools available — and all legal! — there was a race to the bottom. Its like they had a contest to see who could create the worst loans. The term “Toxic Waste” was not created by the press after the fact, don’t forget. It was invented by the very people making and packaging and selling them, at the time.

    So what’s going to be different in human nature when it’s a CDS?

  15. john c. halasz

    Yves:

    One point you seem to have overlooked is the capacity to synthesized fictive mirror bonds without any real underlying principal/actual real investment. One can apparently do the same using interest rate swaps, though it’s not obvious to an outsider like me how that would be done. But what real economic purpose would fictive mirror bonds serve? We know that some synthetic CDOs served to “arbitrage” regulatory capital requirements, but others seem to have been created purely for speculative purposes, to generate fees and payment flows in excess of the available supply of underlying bonds or loans, which would sheerly be a symptom of an excess supply of credit and leverage.

  16. Yves Smith

    john,

    I forgot that, and it’s important point.

    To the extent that CDS were used to create synthetics, one might argue it impacted the cash bond market. In theory arbitrage should lead to the prices being the same in cash bonds. But with CDS being name specific, I’m not sure that argument holds.

    What would help is understanding the real economy flows. My understanding is you could create synthetics with IR swaps via using govvies, so you’d still have some real economy benefit (ie, more gov’e bond purchases, lower risk free rate). If these synthetics were purely synthetic, that means no benefit to the real economy.

  17. charles

    Yves,

    I disagree completely with the “Distortions in cash bond pricing” argument for the simple reason than CDS spread are not the same than bond spread. CDS pricing includes the cost of borrowing (or lending) the underlying securities on a long term basis (I.e.the maturity of the swap) and a complex contingent interest rate option because the underlying can be delivered against par . What people interpret as a discrepancy is simply this implicit cost.

    Long term asset borrowing agreement always involve a dissociation between the economic interest and the voting rights that are associated with the asset. Asset lenders (or synthetically, naked CDS writers or naked puts writers) are compensated for precisely that. It is a "picking pennies in front of a steamroller" business : lending shares or bonds brings your a few bps per annum quite regularly, but sometimes leaves you stuck with a lousy economic exposure. You can't have your cake and eat it.

    Having a "yield curve" not only on currencies but also on assets is very useful information from an economical standpoint (see for instance the commodity market) and increases liquidity. It is even critical when there are rules imposed on sensitive economic agents (such as banks) that limits their ability to enter into maturity transformation (not only in currencies but also on assets).

    Finally, "failed restructuring => more job losses => toll on the real economy" is an assertion that support my suspicion. I don’t even think that it can even be refuted or proved.
    On the other hand, it is politically more expedient to say that, for instance, GM is bankrupted by nasty bankers with their toxic derivatives rather than by its inept organization that is unable to sell cars above costs…
    The truth is that “innocent” economic agents that loose their job because their company has been “unfairly” put into bankruptcy are not that innocent, Cf. Dick Fuld.

    Felix Salmon addresses a similar topic in his blog and I posted here a comment that complements this one

Comments are closed.