Arnold Kling  

Why Credit Default Swaps?

Devil Says "Sin Less"... Seminar in Portland...

I have argued that credit default swaps are fatally flawed because there is no natural seller. One commenter on this post asked, then,

Has Arnold gone into detail on why CDSs exist? I understand his point that they have no natural buyer, and thus may be a result of some weird incentives in the system somewhere. But which ones? What regulation causes this anamoly?

I should go into more detail.

A B-rated bond has higher default risk than a AAA-rated bond. There are lots of institutions that could hold AAA-rated or AA-rated bonds but which are precluded from holding B-rated bonds. However, if it were not for regulation, those institutions might be able to do better with B-rated bonds than with AAA-rated bonds. Suppose that a B-rated bond has a default probability of .01, and a AAA-rated bond has a default probability of .0001 (note: those are strictly hypothetical figures, that I pulled out of the air--not based on the actual relationship between ratings and default probabilities.)

If two B-rated bonds are independent of one another, then the probability that both of them default is .01 times .01, or .0001, which is the same as the probability of a default on the AAA-rated bond. And if you have three independent B-rated bonds, the default probability gets even lower. Where this is leading is that with enough independence, a diversified portfolio of low-rated bonds can be created that has lower risk and a higher return than a AAA-rated bond.

The regulatory anomaly is that a bank or pension fund cannot take advantage of this. The regulators will see B-rated bonds in the institution's portfolio and penalize the firm for taking risk. The regulators overlook the diversification.

Into the breach steps a AAA-rated insurance company, selling credit default swaps on B-rated bonds. The bank can buy a B-rated bond, protect it with a credit default swap from the insurance company, and have the regulator treat the bond as at least AA. The insurance company performs the diversification function, selling swaps on a boatload of different bonds. The insurance company gets to keep most of the extra return that is created by having the banks take on low-rated bonds.

It turns out that at least two things can go wrong here. First, the bonds may turn out to be more highly correlated than was thought. It could be that if some of them go bad, then a lot of them go bad. Second, even if defaults stay under control, an increase in the probability of default can force the insurance company to put up more margin (collateral), and this can put the insurance company in dire straits.

The regulatory issue is this: why are banks not allowed to take the risk of (a) holding a diversified portfolio of low-rated bonds, when they are allowed to take pretty much the same risk by (b) purchasing credit default swaps? If they do (a), then if the bonds turn out to be highly correlated, then the portfolio will not be so well diversified and the strategy will fail. However, if they do (b), then if the bonds turn out to be highly correlated, then the seller of CDS is unlikely to be able to perform as promised.

Regulators have to take a point of view on the issue of how highly correlated are bond defaults. If they take the view that they are not highly correlated, then banks should be allowed to take on diversified portfolios of low-rated bonds. If they take the view that bonds pretty highly correlated, then banks should not be allowed to claim that the purchase of credit defaults swaps provides insulation from risk.

The inconsistency in regulation is what accounts for the rise in the credit default swap market.

I think if you are going to regulate, you have to be really careful to trace through all of the consequences of regulation. If you see a financial innovation taking off like crazy, there is a good chance that it is being used to exploit a regulatory anomaly.

Effective regulation is really easy after something blows up. In real time, it strikes me as a darn hard problem.

Comments and Sharing

COMMENTS (25 to date)
Mike Hammock writes:

I'm guessing these regulations of bond holdings are not new. If that is the case, then why did credit default swaps arise now, instead of earlier? If these regulations are new, why were they enacted?

Nathanael Snow writes:

Patenting of financial instruments is relatively young. (early 90's?)
Patenting a financial instrument may provide a buffer from competition, which could reduce opportunities for flaws in that instrument to be rooted out.
Chances that these elements contributed to problems?

sohaib writes:

"I think if you are going to regulate, you have to be really careful to trace through all of the consequences of regulation. If you see a financial innovation taking off like crazy, there is a good chance that it is being used to exploit a regulatory anomaly."

This is such a simple but excellent and important paragraph. It should be in every economics textbook.

blink writes:

I second sohaib; I definitely learned something from this post. My only question is: If CDSs can be explained so easily, why isn't this common knowledge?

Bill writes:

From April:

CDS seems like another example of regulatory shopping.

Call it a 'swap' rather than 'insurance' , and miraculously, you get different regulation.


tjames writes:

"If two B-rated bonds are independent of one another, then the probability that both of them default is .01 times .01, or .0001, which is the same as the probability of a default on the AAA-rated bond."

Just a technical point for this example: There is also the possibility that just one of the 2 B-rated bonds will fail while the other will not, which is 0.0199. The probability of total failure (i.e. both B-rated bonds) may be the same as the AAA-rated bonds, but there is overall an increased risk of failure of some part of the B-rated bond portfolio using the numbers from your example.

El Presidente writes:

To paraphrase Adam Smith: when you buy a second lottery ticket, you double your chances of losing.

CDSs are bets. Just like in Las Vegas, the natural seller of the opportunity to bet is the "house"; somebody who can in reality, or in the perception of the buyer, cover the payout and maintain liquidity while they wait for the exogenous event to occur. This is freelance banking. If I can sell you a bet, I can create credit. If I can create credit, I can use it as leverage for purchasing real assets. If I can purchase real assets, I can launder the bet in the interim and escape some or all liability for actual future losses. If I can fix the game (i.e. slot machines, table limits, and point-shaving; AKA market manipulation), I don't need to launder the bet, I just need to hire somebody who can break kneecaps.

Fractional reserve banking is, in essence, the same in that one takes deposits and makes bets with the hope that the bets will cover the demands against deposits. This is why we have to regulate it, in order to achieve good faith in credit markets and stability in input and output markets. We've seen this before when local banks had the privilege of printing their own currencies, thereby augmenting the money supply and cornering markets without regard for the stability of real output or employment. The circumstance preceding Shay's Rebellion is a poignant and familiar historical example for Americans. That sort of thing led to the inclusion of exclusive monetary authority in the Constitution (Article I, Section 8). The rebellion itself was contemporaneous to the creation of the Constitution. Now we have competition for partial control of the money supply.

We have, in the end, rogue bankers who have created money, taken a profit, and left the rest of us to clean up the mess. It is not that CDSs are fatally flawed. It is instead that they are designed to make losers out of everyone who accepts them (and a few who issue them recklessly). Just like Poker and Roulette are not investments, CDSs are games. If we have an urge to play games, "game on". If we are looking for investments, we need to look elsewhere.

Elvin writes:

One reason for credit swaps is that the corporate bond market is too thin. The big fixed income managers (i.e. PIMCo) use it to adjust their corporate bond portfolios all the time.
Most of the products of these managers are unlevered or levered slightly with some overlay strategies, so the chances of a true blow up are remote. However, compared to the standard fixed income portfolio, they can and do underperform significantly at times.

Grant writes:

Thanks Arnold, that makes a lot of sense.

I'm wondering if the scenario played out something like this:
Financial institutions want to purchase lower-rated bonds and diversify their risk. Ideally they'd be able to buy bonds with different rates of defaults. The risk levels here would be relatively transparent, because the bond holder and the bond issuer would be the only financial intermediaries involved.

Regulators prevent this from occurring, so the financial institutions have to purchase lower-rated bonds that have been insured and turned into higher-rated bonds. This adds an additional intermediary to the whole process: the insurer. This reduces transparency, and the ability of the bond holder (and bond rater) to estimate the default correlation of their bonds (as they can't know how diversified the bond's insurer is).

This makes me wonder: Why didn't the holders of insured bonds tread more carefully due to the uncertainty of their insurer defaulting? I can understand that laxer lending standards and low interest rates caused an increase in supply of mortgage bonds with a high default rate (presumably making their price more attractive), but I don't understand why investors would not try to estimate the correlated risk of their insurer.

Of course, mortgage bonds can be highly diversified over many small housing markets. Only a widespread downturn in home prices could cause defaults across all those markets and put the insurers and bond holders in trouble. Naturally that is exactly what happened, but again I find it hard to believe that bond holders did not at least try to estimate the counterparty risk from their insurers.

Why did the buyers of insured bonds systematically underestimate the risk of their insurer defaulting? Was it really as simple as not believing a nation-wide downturn in home prices could occur?

Arnold Kling writes:

Mike H,
The Basel international standards for capital regulations are in fact relatively new. The agreement was adopted in 1988 and was gradually implemented in various countries.

You are quite correct, which was why I made the vague statement that with enough diversification and independence you get a better risk-return profile. The example itself, with only two bonds, does not show a better risk-return profile.

The argument that CDS should be regulated as insurance is interesting. The analytical reasoning has some validity, and in addition there is the fact that a big seller was AIG, the insurance company. However, there are many products that can be deemed insurance, including mortgage securitization.

Boonton writes:

I'm not sure I understand what the regulation has to do with the problem in the big scheme of things. Suppose the regulator let the bank own two or three B bonds. The bank will have no need to buy insurance on those bonds from a CDS supplier but the bank is acting as insurance.

Again if the same things go wrong we still have the same problem. If the bonds are correlated with each other (say one os a bond for Ford, another for a auto parts supplier to Ford and the third a large dealer of be a little silly but to drive the point home)...the bank is in trouble.

One could argue that at least with the insurance there is at least one line of defense for the bank. Before the bank gets hit both the bond AND the insurance company have to default. Presumably the insurance company, to preserve its A rating must keep some reserves.

At the end of the day you have the same problem. The banks think they are safe either because they have 'insurance' that makes their holdings A rated or because they think their financial model makes the multiple B bonds in their books A rated.

Grant writes:


I think its easier for bond holders to judge the risk correlation of bonds than it is for them to judge the risk correlation of bond insurers' holdings (see my comment above). As Arnold often points out, more financial intermediaries typically means less transparency.

Boonton writes:


Yes but from what I'm understanding here the banks are the bond holders, they are just trying to buy insurance on their holdings so they can be treated as A level risk instead of B. So the bank must still judge the risk of a default on the bonds and then the risk of a default of the insurance company. From the banks POV the insurance is at least another line of defense against a massive default.

All in all, even if the insurance company defaults and only pays $0.10 on the dollar of their obligations that should still be a measure of risk reduction for a bank that is properly evaluating the bonds it decides to buy. (Also assuming, of course, that the premium on this insurance is not too expensive)

It seems to me as if the the problem is not so much the regulatory structure but the theory. The three B level bonds DO NOT net out to become just as good as one A level bond and betting money that it does is going to be just as foolish as betting against a casino's slot machine.

Boonton writes:

Let me just point out that while King's post makes a lot of sense, it seems to be missing something important, why didn't it work?

Let's say that buying three B level bonds is just like having one A level bond. In a free unregulated market the B level bonds should only yield a little bit more than A level bonds. Why? Because there is always some systemic risk thatyou cannot diversify away.

Because of regulation, though, certain big buyers cannot buy B level bonds as if they were A levels. This should depress the price of B level bonds driving up their yield. So a third party not subject to the regulation should be able to buy the B level bonds, earn the higher returns and itself be rated an A level risk. The third party can secure funding by selling the banks the higher returns they are not allowed to enjoy directly (and, of course, taking a cut for themselves).

Theory wise the only problem here seems to be that this set up is a bit inefficient since it puts a middleman between the banks and some of the B level bonds. So what? The yields of B level bonds should have fallen until the point where it deviated only by two slight measures from the A for the risk that could not be diversified away and another for the fees charged by the third parties (the loss due to the poor regulatory structure).

Yes the third party has some issues over transparancy but again so what? Rating agencies can demand to see all holdings or refuse to rate them. Banks can pay higher fees to third parties who allow their books to be openly inspected. Competition should, in theory, address that problem as well unless you think that huge amounts of bank money was being run by idiots who were suckered by con men (something theory again says should not happen in a competitive market).

Grant writes:


Yes, but its hard for a bond-holder to judge the risk of their insurance defaulting. There is no way to know how insurance default might be correlated without knowing the insurer's holdings.

floccina writes:

Grant wrote:

Why did the buyers of insured bonds systematically underestimate the risk of their insurer defaulting? Was it really as simple as not believing a nation-wide downturn in home prices could occur?

I think that this is great question. I think that the answer is that people like me knew that home prices would not keep rising but that we thought that we could make make money without working and so we did not adequately monitor our investments. The people in the banks and financial institutions saw that we were not looking and so they took what they could get (protected by the corporate structure). So it is my/investors fault, it was our money that was at stake. BTW it is a curiosity to me that so many people use loans from strangers (mortgage companies, banks. etc.) with the great inefficiencies that creates. I borrowed the money to buy my home from my father. He got a high rate of return and I got no closing costs.

flocci na writes:

Addendum IMO no amount of regulation will allow people to invest without much effort, without great risk.

Boonton writes:


I think you missed the point of my confusion:

You asserted the bond holders are good at being able to "judge the risk correlation of bonds".

1. Well then what's the problem? I may be silly in assuming my auto insurance or homeowners insurance company is solvent but that doesn't mean I want my house to burn down or I want my car to be totalled. I'm still going to try to avoid those things if I can.

2. This assertion would imply that the bond holders knew their stuff was going to burn down but thought that they could essentially rip off their insurance companies. But wait a minute, how could bank managers believe hundreds of billions of dollars worth of bonds were going to go bad BUT the insurance companies would remain solvent? If I believed that a plague was going to kill 40% of humanity in the next three years I certainly wouldn't be wasting any money buying life insurance on myself or anyone else. I certainly don't need to see any insurance companies private accounts to know that and I'm hardly smart enough for anyone to hand me tens of billions of dollars to play with.

At the end of the day the problem is this argument doesn't make sense. It's not the credit default swaps that were the problem nor the regulators who wouldn't let the banks treat three B bonds on their books like it was one big A bond. It was the belief that the three B bonds were just like one big A bond that was the problem. If that belief was not held, the 'insurance companies' would have charged much higher premiums for the 'insurance' & banks would have found the fantastic high returns of those B bonds to be much less. If that belief was not held, the insurance companies would have been charging higher premiums and would have been providing protection against a smaller amount of bonds. Banks, likewise, would have been putting fewer bonds into their books which would have meant mortgage mills would have had to have waited longer to turn over the loans they wrote to be packaged by Wall Street.

Since the problem was a false belief, the question isn't so easy as finding some regulatory policy to scapegoat. The question is why would smart people, whose income supposedly depends on making the right calls in an intensely competitive market subscribe to such a false belief?

Grant writes:


I'd meant that bond holders didn't know the insurer's positions, and thus couldn't judge the effect of a general turn-down in a market (such as housing). That is, how would a bond-holder know how many mortgage securities AIG had insured? What if they'd insured more corporate paper? If they hadn't insured many mortgages, they wouldn't have a problem with a general mortgage meltdown, and so the bond purchaser might find the insurance worth purchasing. How is the bond-holder to know?

I think its safe to assume that bond purchasers have a comparative advantage in judging risk correlation of their bonds. Throwing bond insurance into the mix would seem to add an additional lair of uncertainty.

Boonton writes:

Let's say the 'uncertainity level' without insurance is 100. The worse case is you buy insurance and the company is unable to pay you anything. Beyond that, even if the insurance company is only able to pay you $0.10 on the dollar you are still less uncertain than when you had no insurance.

If your assumption is true the worse that insurance could do is cause bond holders to waste a relatively small amount of money on premiums. The bond holders assumed their bonds were good and were just buying insurance to make regulators happy & to show their investors an 'A' rating rather than a 'B' one. If you rented a time machine and told these people their bonds were going to default in the next month or two their reaction wouldn't be 'no problem we got insurance'....their reaction would be 'ohhh sh*** we are screwewd!'

Grant writes:

If you rented a time machine and told these people their bonds were going to default in the next month or two their reaction wouldn't be 'no problem we got insurance'....their reaction would be 'ohhh sh*** we are screwewd!'

Only if they thought their insurance would default as well. Without being able to know the details of their insurer's positions, how could they know this?

For example, say an investor judges the possibility of a bond defaulting at 50%, and the possibility of its insurance defaulting also at 50%. He thinks he'll get paid 75% of the time, and he believes purchasing the bond is a wise decision if it has a 70% or better default rate. So he purchases the bond.

Of course, a general downturn in home prices wiped out his insurer. This is worse than simply wasting the insurance premium because it means he miss-judged the risk on the bond, and thus took on too much risk. The solvency of his entire portfolio is threatened. The problem in this scenario is that the investor has no way of knowing that his insurer isn't capitalized for widespread defaulting of mortgage bonds.

Of course, I'm not saying this happened. It could be as simple as both insurers and investors betting that home prices wouldn't drop (this seems a bit daft to me though), when they all knew the insurers wouldn't stay solvent in that case.

Dan Weber writes:
I think if you are going to regulate, you have to be really careful to trace through all of the consequences of regulation. If you see a financial innovation taking off like crazy, there is a good chance that it is being used to exploit a regulatory anomaly.
This is good, but how does it apply to your own recommendations; i.e. that FM and FM should only buy mortgages where the owner has put 20% down (and occupies and pays a fixed rate, but I'm ignoring those points for now)?

Wouldn't we see people making "secondary" loans to people so that they can put their 20% down? This financial innovation would spread because people will get massively better credit if they can meet this arbitrary (in their eyes) requirement.

I had to check a box on my mortgages that my downpayment was all my own, but that's a pretty minor check.

Mr. Econotarian writes:

I think this whole thread reflects the fact that overshooting the value of real estate by $5 trillion is bad, and at the end of the day, it doesn't matter what you do.

We (regulators or the private market) won't see the next bubble coming either.

All we can hope is to keep the economy flexible enough to roll with the typical punches. Avoid regulations that boost up one industry and keeps down another, because you never know where your exit route might be.

Boonton writes:

I had to check a box on my mortgages that my downpayment was all my own, but that's a pretty minor check.

I suspect that check was minor because your down payment was plausible. If you had the profile of someone whose bills and consumption eats up 99% of their weekly income you might have to do a bit more to prove that the $35,000 that suddenly appeared in your checking account two weeks ago wasn't really a loan from a friend or relative.

Maybe ten years ago when my gf brought her first house she had to present 401K statements to prove she hadn't tapped those balances.

Mr. Econotarian,

I think all in all you are correct. This was a bubble but I think Kling overshoots a bit here by trying to blame regulation on it. The fact is that the CDO's were 'logical' in the old regime but they weren't connected to reality.

RMullins writes:

Arnold, I would like to know if the Senate legislation passed in 2000 in an Agricultural Bill S.2697 Commodity Futures Moderization Act of 2000 has anything to do with the problems we are having in the Financial Market and other problems with the credit crunch?
The reason I am asking is because 60 minutes did one of their segments on the Credit Default Swaps and was equating it to problems with Bucket Shops back in 1913 and sighted Commodity Futures Moderization Act of 2000 which reversed the lessons learned from 1913.
If this is the case, has Congress done anything to reverse out the Commodity Futures Modernization Act of 2000? I would think that would be the first step at attacking the possible root cause. I am attempting to understand how these bets are being made with little collateral and the the payouts can't possibly be covered. Seems like futures betting gone wild.

Comments for this entry have been closed
Return to top