Are CDS a good thing?

By Felix Salmon
April 24, 2009

Kevin Drum asks whether I’m “still as bullish about credit default swaps as I have been in the past”. It’s a good question: after all, a large part of my speech to the regional bond dealers (more of the actual speech than of the notes) comprised me explaining that I was very wrong about a lot of this stuff.

On the subject of CDS specifically, yes I think I was entranced by the shiny-new-toy aspect of them. They were so elegant, they were such efficient ways of moving risk around the markets, that they had to be a good thing, right?

I’m still no fan of the CDS demonizers, who generally have no idea what they’re talking about: if they are remotely right, they’re right for entirely the wrong reasons. But I will admit that I did rather jump in an unwarranted manner to the conclusion that because the CDS critics were so very wrong, then the CDS market must have been largely benign.

The fact is that AIG could never have blown up if it wasn’t for CDS. UBS and Citigroup could never have suffered their billions of dollars in super-senior losses if it wasn’t for CDS. And therefore the CDS market was intimately involved in some of the most systemically-devastating events of the crisis.

That said, CDS had nothing whatsoever to do with the failure of Bear Stearns. The fact that CDS allowed you to effectively short a credit you didn’t own was not particularly harmful. The the CDS market was incredibly useful in terms of price discovery, and it remained impressively liquid even as virtually all other credit markets froze up altogether. Etc etc.

So right now I’m vacillating on this one. I’ve recently finished reading Gillian Tett’s new book, and although she tells a good story, I don’t think she really succeeds in making her case that the creation of the CDS market was the proximate cause which “unleashed a catastrophe” on the world.

On the other hand, I don’t think I can really add credit default swaps to the list of undeniably positive financial innovations, like ATM cards. CDS are powerful instruments, which can be — and were — misused, most egregiously by AIG. I’m not happy about that. But I’m not remotely ready to start banning them, or to start trying to require that any buyer of protection have an “insurable interest” before being allowed to do the deal.

Update: The comments, in general, are well worth reading. It’s true that in an enormous zero-sum market, you’re going to find situations where both buyers and sellers are booking profits, and that can’t be good. I’m also still pondering the excellent question asking why price discovery is a good thing. And less specifically but just as importantly, CDS encourage a certain kind of laziness on the part of the buy side — not just in bankruptcy situations, but more generally too. The easier it is to hedge your position dynamically, the less homework you’re likely to do before entering into that position. Which is a recipe for trades getting very crowded very quickly.


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see

I think there’s a learning curve associated with them, and they will do more net good after people have been through a business cycle with them.

I also happen to think CDOs are a good idea, especially in the long run (learning curve again), so some people might want to discount my opinions here.

I think the SIZE of the CDS market is evil.

CDSs are zero-sum insturments, which means in the long term, for every winner there is a loser.

Large scale zero sum instruments, of any stripe, are dangerous with wall street compensation schemes, because for some reason, both the buyer and seller as individual employees make out like bandits, but ONE of their two companies is going to get taken to the cleaners eventually.

Posted by Nicholas Weaver | Report as abusive

After mulling it over and trying to apportion “blame” to someone… anybody, over the CDS fiasco I arrived at the conclusion that the blame must be placed onto the policy issuer because he had little or no clue as to what kind of risk he was accepting, and certainly did not know how to price it. Car insurance, home, life, business insurance very seldom go bankrupt because of mis-pricing their policies.

The idea that “the product is so new that no one knows enough about it” just highlights the stupidity or ignorance of the executives in the business. Mind you if the intent is to make out like bandits on the personal lever then these perpetrators have obviously succeeded; but this has more to do with opportunity than with competence, experience, or a job well done.

H.F. Wolff

Posted by H.F. Wolff | Report as abusive

Felix, can you please explain (or yet better, give a link to something that does) _why_ price discovery (more specically, the prices that CDSs discover) is of social value?

I guess I have two questions.

First, in light of (what most people view as) a mystery of why sovereign debt CDS are priced where they are, and in general suspicions that many players used CDS coverage for essentially regulatory purposes rather than pure risk management, why should we believe the “discovered” prices are good?

But even if accurate (under some definition), I’m simply unable to work out for myself why it’s so useful to have price discovery. The closest thing I can come to is that if we believed CDS prices we’d have a “good” estimate for default probability. Ok … but WHY? If you think the chance of X defaulting is 3% but I think it is 5%, how important is it if you turn out to be more right?. After all, default is a yes or no proposition at the end, so isn’t this line of reaosning going to have to drag in various philosophies-of-probabilities to advance? What’s the probability of Ford, say, defaulting
in the next three years? The reality is presumably
that it’s nontrivial, say > 5%, but probably less than
50%. How would the world be actually served by someone
saying “We’ll, actually, my efficient CDS market shows
it to be 23.2127%”??? I mean, served in such a way that we should tolerate negatives of the CDS markets.

…As I say, I genuinely would like a pointer that discusses the value of price-discovery for CDS’s.


Posted by ax | Report as abusive

A new anti-CDS argument here, from Andrew Leonard is that creditors who have laid off their risk with CDS will benefit from bankruptcy relative to renegotiation. 7/cd_swaps_and_bankruptcy/index.html?sou rce=rss&aim=/tech/htww

I don’t believe price discovery was enhanced by CDS:

There was a huge demand for AAA rated paper after glass steagal being repealed in the US and the basel regulations in Europe. This was because regulators now allowed you to lever up on these AAA as good as cash assets. 40 to 1 in the case of basel requirements ( which being procyclical are simply an excuse to lever up in the good times ) So we had a bubble in bonds, and the answer was to simply increase the supply of AAA rated bonds thanks to CDS.

The liquidity of the CDS market in no way helped price discovery because the price we were discovering was not the default risk of the bonds. It was the price that the buyer was willing to pay simply to avoid capital requirements. Since the CDS market was unregulated the seller had no restrictions and thus was willing to sell at a price that did not compensate for the risk.

Now it looks like people are using CDS to attempt to price risk. However the product is not complete until it is on an exchange and under regulation. So I would keep vacillating until we find out what the CDS market ends up looking like.

Posted by Mark | Report as abusive

The CDS can be of value only if risk can be accurately assessed. I think the events of the last few years have demonstrated that there are enough people who are incapable of accurately assessing risk actually got that job. The whole idea of using CDSs to distribute risk depends on pricing them correctly, otherwise it turns into a roulette wheel.

Posted by KenG | Report as abusive

Asking whether CDS are a good thing is the wrong question. The right question is “How should CDS be regulated?”

You say that you are:
“not remotely ready to start banning them, or to start trying to require that any buyer of protection have an “insurable interest” before being allowed to do the deal.”

A clearinghouse may or may not reduce risk

( 009/04/21/tuesday-links-fail-to-reduce-r isk/),

but it will increase transparency – which can only be a good thing.

I don’t necessarily have a lot of sympathy for the large investment banks that were counterparties of AIG – they probably figured out that the protection that AIG was selling them was worthless: see this post by acredittrader:

“Did the banks realize the value of its protection held against AIG was zero? Of course they did – they aren’t as dumb as the media suggests. The reason they continued to pay the full market CDS offer (rather than a much lower level due to AIG’s massive wrong-wayness) to AIG was because they considered it a cost that allowed them to continue originating CDOs. If they could not offload super-senior risk to someone, their originating desks would be effectively shut down.”

I have more sympathy for the smaller counterparties of AIG (e.g. municipal governments) that were only relying on the info. from the ratings agencies. A clearinghouse may have been able to help them figure out what the game was.

Posted by cna | Report as abusive

Felix, your initial reaction to CDS is the correct one. They are a useful tool to hedge risk that, as with any such financial instrument, can result in losses if not underwritten well.

In a way, CDS are simply a variation of a personal guaranty, which creditors have used for centuries (both prudently and imprudently) to hedge risk. As with guaranties, CDS are a useful tool for hedging risk, so restricting them is not good policy. On the other hand, better understanding of the risks of CDS, as with the risk of guarantees, is always beneficial to the marketplace.

Felix, I think that CDS price discovery is a good thing; otherwise, you are relying on the bond rating agencies. If a CDS is trading upfront on a AAA company, which do you believe?

I suspect the market got too big and too complex because the players must have forgot about collateral. In the commodity markets, which is also a zero sum game, somebody explodes every two years or so, and sometimes spectacularly. But there are immense collateral requirements, partly mediated by exchanges, that prevent any systemic risk. It’s like Taleb’s first principle to prevent black swans: make sure that someone explodes every couple of years, and nothing systemically bad will happen.

Posted by macndub | Report as abusive

one of the pricing problems being that with the participation of specialized buyers and sellers, the same asset may be ‘worth’ a broad range of prices at any given time, depending who/what the buyers and sellers represent – perhaps backed up by their capacity to facilite the company’s bankruptcy or survival.

Posted by mtm | Report as abusive

I think CDSs can be a good thing, but I’ve never understood them in their current form.

More specifically, I can’t see why a seller would write such coverage on a *particular* security–unless, of course, it never had any intention of paying out. It creates a situation where some buyers (the holders of the insured securities) have more information about credit risk than those from whom they are buying protection. This strikes me as a textbook case of adverse selection.

However, I can see the value of a CDS market linked to a basket of similar securities, such as the ABX index. In other words, a buyer knows something about its own borrowers, but wants to hedge against risks outside of its control, such as a general decline in real estate prices. So it buys a security that pays out in cases of general default.

For this to work, nobody should be allowed to sell such protection without a determination that the seller has the resources to fulfill its obligation, and the party making that determination must face strong incentives to make a correct determination. Potentially this could be handled in the private sector, through an exchange that takes on liability in cases of nonperformance of one of its approved transactions.

Posted by Mike | Report as abusive

like any technology, it is dual use and can be purchased to fulfill constructive or destructive purposes. Much of the instrument’s ‘value’ depends on a specific owner’s intentions and capacity to utilize the instrument to achieve a certain goal.

Posted by mtm | Report as abusive

CDS are indeed a major cause of the problem. The fact is that CDS are supposed to act like insurance. But if they had been called insurance they would have been subject to tight regulation. So the instrument was called a Credit Default Swap. This way the instrument could be used without the burden of regulation. That tells me that execs wanted CDS to be used in ways other than intended. What other reason would there be for not treating CDS’ as insurance the way they should have been?

The so called geniuses on wall street wanted to be left to use their own judgment. And like a child with a loaded gun ignoring all safety precautions they blasted everyone. CDS should be regulated as the insurance instruments they are. And those executives and corporate owners that caused this mess should pay out of pocket and put the stolen money back. But this will never happen because just like any other country, you don’t have to follow the rules or fix anything you break if you have enough money. Berne Madoff was just one individual, and in the scope of the global economy we find ourselves in, his crime was just a drop in the bucket.

You’ll notice that AIG and other execs involved in this mess aren’t being held any where near as accountable as Mr Madoff. He’s just a scapegoat to distract us from the fact that the real criminals are still on the scene and still screwing with all of us.

There are several tins wrong with the CDS market as it is now constructed.

1) There can be substantially more CDS written against a companies debt than there is debt outstanding. This leads can lead to rampant speculation and adverse price movements.

2) There is no capital assigned against this insurance since it is off-balance sheet. Whe things go wrong there is no cushion.

3) iven the size of the markets, the counterparty risk can cause systemic risk in the marketplace.

4) Who controls the concentration of risk (ie. AIG.)

5) It is black or white risk. Most of the time the writer of the CDS collects the premium; but when things go wrong the writer doesn’t have the funds to pay the insurance (AIG, Fannie, Freddie, and Citi and numerous hedge funds that are undocumented).

In concept the ability to hedge credit risk is sound, but in practice it has been shown to be a dangerous systemic risk to the entire financial world.

Posted by hammer | Report as abusive

We all remember Napster, the original file sharing (aka music piracy) system, right?

This discussion of credit default swaps reminds me a lot about discussions I had with friends about Napster and its ilk. There was absolutely nothing wrong with the technology we agreed, and it was none of Napster’s fault that it was being used for bad purposes. We confidently decided that Napster should not be restricted because legal files could certainly be shared through it. During the dotcom bubble, I help found a startup aimed at peer-to-peer filesharing so I relished these justifications.

Of course the reality is that anonymous P2P filesharing was used almost entirely for piracy, and its kind have catastrophically harmed the music industry. I say this with regret because my own involvement a decade ago in such technology.

Felix, it is very intellectually satisfying to think of credit default swaps as a kind of technologically pure system that was contaminated by evil users but can be valuable in the hands of good users.

In reality, as we have seen, credit default swaps have had overwhelmingly destructive effects (hiding of massive risk, allowing almost unimaginable leverage, causing huge systemic instability, and allowing enormous fraud by major financial players against each other and ultimately against taxpayers) balanced by very few actual positive ones. I don’t care how many theoretical positive uses there are, Felix. These past few years have shown these instruments to be on the balance so harmful that they ought not to exist. We did just fine (better even) before CDS’s and we will do fine again if and when we do the right thing and get rid of them.

Posted by Daniel Hess | Report as abusive

Benny Acosta makes the important point that this is insurance, but in order to perpetrate the fraud of writing insurance not backed up by an ability to pay it was called something else.

It is stunning to me that folks still praise the merits of CDSs *after* they have reached their disasterous culmination. It’s like a friend I know who clings to the notion that communism is a fine ideology that just hasn’t been implemented properly. Well that’s just terrific. Hasn’t the world suffered enough?

Posted by Dan Hess | Report as abusive

Ax asks two fantastic questions:

1. Do CDS help illuminate the price of credit?
2. If so, why is such “price discovery” a societal good?

A month ago, when I was a vocal defender of CDS, I would have argued that because CDS are so liquid and so simple, they must contribute to price discovery; as to why that is a good thing, I would have touted reductions in asymmetric information.

I’ve since retreated into a cocoon to re-evaluate my views on CDS. Question 1 is critical and may well determine whether I emerge again as a CDS defender, detractor, or mere skeptic.

That said, I still think that CDS have garnered a mountain of undeserved criticism, much of it evident in some of the comments above.

Banning CDS’s because they enabled the current crisis would be akin to banning stocks because they enabled the 1929 crash. I think demanding that CDS defenders prove their social utility is misguided. They are financial instruments, and the ultimate measure of their worth should be whether or not they have financial utility. If they don’t, then they will cease to exist on their own, without the need for a blanket ban on their existence (were such a thing even practical). I do think there needs to be some changes to the way they were treated by accounting standards and how they were disclosed (or not disclosed) on company balance sheets, but that’s an accounting problem rather than a problem with the instrument itself. A CDS is little different in principle from other modern financial instruments — such as out-of-the-money put options, for instance — and banning them would be rather arbitrary and short-sighted.

Posted by Nate | Report as abusive

Probably the key problem with CDS — especially as AIG approached them — is the lack of a reliable DIVERSIFICATION strategy. The AIG execs probably figure, heck its just another line of INSURANCE. But writing insurance on property on the Gulf coast can be diversified by writing uncorrelated insurance on ranches in Montana. But credit markets are intrinsically much more linked than local weather ‘markets’.

Regulation of CDS should take account of the correlation risks inherent in writing a large net exposure (to say nothing of the various counterparty risks involved in the netting process itself).

Posted by Seth | Report as abusive

As analogies go, I’d think you might want to say “akin to banning margin debt because it enabled the 1929 crash.” Even then, spouting analogies with no explanation of why two things are analogous and/or why either restriction would be a bad thing is not too helpful.

Posted by Anonymous Jones | Report as abusive

Felix, how can you not be against naked CDS?

Would you like anyone being able to take out a fire insurance against your house? Having plenty of people paying a small fee and …. throwing a burning candle in your house?

I have read estimates of 80% of CDS’s being naked (no underlying bonds). Congress apparently knew what would happen, because the “Futures modernization act” (heavily lobbied by Greedscam, Rubin, Summers, Levitt et al) explained that all the gambling laws would not be applicable to (the gambling with) CDS.

How convenient that lobbying by wall street resulted in removal of the uptick rule (June/July 2007; a few months later the stock markets would peak): now they could buy CDS AND shorts of firms they knew were heavily loaded with subprime toxins, or gas guzzling SUV manufacturers, etc. And then they prevent an orderly restructuring as they get more profit by letting the firms go fully belly-up.

And what about the rumored side letters? Both parties knew there would not be any payment, but one party could bamboozle the sleep-at-the-wheel regulators.

Posted by carol | Report as abusive

A metaphor I’ve seen before about CDS. They are like buying insurance on someone else’s house so that you collect if it burns down. So wonderfully motivating, don’t you think?

Credit Default Swaps are financial crack, and I see that Felix is hooked. If you could put down the crackpipe for a minute (and there are very very few who, once hooked, can)and listen to that very rare commodity these days called REASON you can easily see that CDS is a piece of financial engineering that adds absolutely NOTHING to the economy. It’s brilliance is the sheer vertical market it creates in allowing financial institutions to skirt capital requirements in order to be able to participate in for what is now essentially a worldwide floating casino. Probably even more brilliant is how many crackheads it has created. The CDS market is not defensible under ANY circumstances but there are far too many glazed over eyes in smoked filled rooms full of drooling mouths thinking about how that new risk model is gonna make millions…..

Posted by Mbuna | Report as abusive

CDS was a less than perfect fix to a problem that is recurring in big companies balance sheet : the balkanisation of debt.
Big corporates should issue like governments : regularly, in a standardized way, and with a view of matching the duration of assets and the duration of liabilities. They don’t do it because funding short term is cheaper and cheating pari passu debt holders by pledging asset ex post by various means is a winning strategy. Banks encourage this situation because it preserves their turf and provide an easy justification of their own practice of maturity transformation on a great scale (an activity that a Nobel economies qualifies as counterfeiting money). In capitalization terms, the corporate bond market is of similar size than the equity market, yet, it is much more opaque. By comparison, CDS market was a hallmark of standardization and liquidity. This is why it was more efficient in information dissemination.

What has not been properly thought out and need a serious fix is the ramification of Credit Event Clause in Derivatives contract. I am not talking here only about CDS’s but ALL derivatives under ISDA. It is not the CDS’s written on Citigroup that causes systemic risk. The real source of systemic risk is the fact that all derivatives contract (fx, bonds, equities,commodities) would have to be unwound in a short amount of time together with their hedges. The only way to make derivatives safe and indeed useful for the economy is to tame them through exchange cleared contract,like financial futures, with robust enforcement of position limits. Exchange cleared CDSs, or even better, exchange cleared futures and options on liquid pari passu bonds have a place in a sound financial environment

Posted by Charles | Report as abusive

Nate –

“Banning CDS’s because they enabled the current crisis would be akin to banning stocks because they enabled the 1929 crash.”


Stock – A fractional ownership in a business. Businesses need owners.

CDS – The financial utility is quite clear to me. Warren Buffett explained it nicely (referring to derivatives generally) as did Nicholas Weaver above. Buyer and seller both book a nice profit today and go on their merry way, leaving losses for the future. This is great; lets meet again tomorrow and do it again!

No, Nate, we really do need to put our feet on solid ground and ask whether what we do serves a helpful purpose. The credo “first do no harm” doesn’t apply only to doctors but all professionals including financial ones. That is what it means to be a professional. Contrary to our recent experience, banking and finance as properly practiced can and ought to bring good benefit generally.

Wall Street has done a fine job convincing the world that it has no professional ethics. Questions of “what net benefit does instrument X bring” are now more relevant than ever.

Posted by Daniel Hess | Report as abusive

The CDS instrument creates the wrong incentives in bankruptcy. Normally, bankruptcy is a negotiated process by which debtholders are vying for a fair stake in a going concern. CDS put the wrong negotiators at the table. Instead, while the debtholders are negotiating, the CDS holders are really on the hook.

So, as a result, where workouts in a business would normally be found, now we are almost certain to always enter into bankruptcy. In addition, the prices in bankruptcy are likely to represent severe conflicts of interest and potentially fraud.

Posted by Doug | Report as abusive

I agree that the bankruptcy negotiation scenario is problematic. Perhaps that could be solved by a standardized CDS contract clause in which the bond-owner who buys protection transfers their bankruptcy negotiation rights to the protection seller.

Posted by Nate | Report as abusive

Charlie Munger has come out on Bloomberg favoring a ’100% ban’ of credit swaps.

“If I were the governor of the world, I would eliminate it entirely — 100 percent,” Munger said in a Bloomberg Television interview today. “That’s the best solution. It isn’t as though the economic world didn’t function quite well without it, and it isn’t as though what has happened has been so wonderfully desirable that we should logically want more of it.”

“The national policy that allowed the derivative markets to develop as they did was a stupid policy and we think the derivative markets as they evolved have done more public damage than public benefit,” Munger said. “That said, if they exist and they are legal and some opportunity therein is presented to us that we think makes sense to the shareholders of Berkshire, we would seize that opportunity.”

“The whole mass of incentives created is quite counterproductive,” Munger said. Buyers of the swaps get a “vested interest in the destruction of some business.”

Posted by Dan Hess | Report as abusive

“Soros Says Default Swaps Should Be Outlawed” 6/12/soros-says-default-swaps-should-be- outlawed/?ref=business

“The more I’ve heard about them, the more I’ve realized they’re truly toxic,” Mr. Soros said Friday, according to Reuters. Later, he added: “It’s like buying life insurance on someone else’s life, and owning a license to kill.”

Posted by Dan | Report as abusive