# How can the CDS market possibly be \$46 trillion?

Most every time I read an article about the financial crisis, the article notes that the credit default swaps (CDS) market is \$46 trillion. This number is emphasized to show how potentially catastrophic the events on Wall Street can be.

\$46 trillion is a huge honking number, and it occurs to me that it is, quite possibly, an impossible number. World GDP for 2007 was approximately \$45 trillion. I know that GDP does not measure the entirety of wealth, but still, how can one market be worth the same as global GDP? Obviously, I’m missing something. Could someone please explain?

Sua

What they’re talking about is the underlying asset value of the swaps, but not the actual exposure to either party.

For example, suppose there’s a basket of \$1,000,000 worth of bonds that pay 6% per year. (Or \$60,000 per year.) I agree to pay you \$60,000 per year, and you agree to pay me some floating interest rate (say it’s pegged to the prime rate) per year, based on \$1,000,000 of principal. So if the rate goes to 6.5%, you pay me \$65,000, and I get \$5,000 in profit. If it goes to 5.5%, you pay me \$55,000 and I lose \$5,000.

Here’s the kicker: The value of the underlying assets is just used for the interest rate calculation. So this is called a million-dollar swap contract, even though the actual exposure is just a few grand a year. The \$45 trillion figure is the total of all these base principal amounts; it’s a largely meaningless number.

For swaps that have a default clause, where one party has to turn the defaulted bond over to the other in exchange for the principal amount, the exposure is higher. Especially if the bond buyer bought the bonds at a large discount (as many FNMA investors did.) Still, it doesn’t mean that \$45 trillion of money is going to evaporate.
This is all based on my limited understanding of this market; I may have gotten some of the details wrong.

friedo, you’ve described an interest rate swap. Credit default swaps are a little different.

A credit default swap is best thought of an insurance policy. You pay a premium, and get a payout if a particular company defaults. It has legitimate insurance purposes if you actually own the debt of that particular company and want to protect yourself against default.

Unlike insurance, however, it can also be used to speculate. You can’t do that with ordinary insurance–I can’t buy insurance that will pay me if you suffer a financial loss. (Unless, obviously, you’re my spouse or parent, in which case I have an insurable interest in your financial health.) I can, however, buy a CDS which will pay me if you default, even if I don’t own your debt.

Apparently a lot of hedge funds like to do this to hedge themselves against (or to bet on) interest rate spreads–that is, the difference between risk-free and risky borrowing. I claim no expertise as to exactly how or why one would do this. I’ve never felt the urge to do it myself, but then I don’t have \$64 trillion.

The \$64 trillion represents the total value of the debt “insured”. Keep in mind that the same debt can be “insured” many times, because again, you can “insure” it even if you don’t own it. Nobody is going to have to pay \$64 trillion unless everybody in the world defaults on every dollar, yen, and euro of debt that they owe. But there will be more defaults than usual, and even a small fraction of \$64 trillion is a big honking lot of money.

Oops.

The total amount of “assets” included in all the fancy new unregulated credit instruments is certainly hundreds of trillions of dollars, and may exceed a quadrillion dollar worldwide. The number is not real for a variety of reasons, but try this simple analogy to help explain how you can multiply a few real things into a mythical total.

A baseball team sells 50,000 season tickets. They cover 82 home games. Therefore the official attendance for the season is 4,100,000. That’s true even if some tickets holders don’t show up to every game, or give their tickets to other people, or even if nobody shows up at all.

The total of all assets, continually being made, swapped, bought, sold, and subdivided, is far more than the amount of individual money that is put into play at any instant. This is money creation rather than wealth creation, which is what the GDP reflects. Investors long ago figured out that the two were different and could be played with in different fashion, but that understanding hasn’t filtered down to the general public.

This vastly oversimplifies the game being played, but console yourself that the heads of these firms don’t know the real value of their assets either.

But if the 64 trillion was all “insured” or hedged on junk mortages or assets rather than quality ones then it would be entirely possible for their to be 64 trillion of outstanding debts.

But assuming the US goverment does do a full bailout - does that mean there will be some very rich winners as many people must have hedged that the units were going to fail?

I don’t believe, however, that that’s the case.

That would probably be a true statement, yes. If you’re on the winning side of a lot of these default bets, you’re feeling good. If you find that you’ve driven the opposite party into bankruptcy, you’re not feeling good. If you find that the government will bail out the opposite party, you’re feeling good again.

I’m a credit trader weighing in here…

That huge \$70 trillion number massively overstates the real condition. Basically, most of the swaps just cancel each other out. Example: You and I do a \$10 million swap – you “buy” and I “sell”. A while later, I do the exact oppisate trade with someone else where I “buy” and the other guy sells. Net result: I’m neutral – any money you owe me gets forwarded to the third guy and vice versa.
But when the newspaper calls all the banks to find out the size of the swaps market, they’ll call you and hear you have 1 \$10 million swap. They’ll call me and hear I have 2 \$10 million swaps for a total of \$20 million, and they’ll call the third guy and hear he has 1 \$10 million swap. Net total: \$40 million. Even though in reality, it all nets down to a single \$10 million swap.

The amazing thing about credit default swaps is that the money basically just goes in a circle. A handful of people trade the same swaps back and forth between them. The net risk is actually much, much lower than the \$70 trillion figure you’ve heard.

Why do they do this?

Cause the price fluctuates all the time, so by buying and selling the same thing back-n-forth, you hope to lock in a profit.

A credit default swap is an insurance contract. You pay me \$50k/year, and if some company (say GE) goes bankrupt, I pay you 10 million. Right now, you and I estimate GE’s odds of bankruptcy as tiny but non-zero.
Tommorow, GE announces lower profits. They’re still very unlikely to go bankrupt, but the odds are slightly higher than yesterday. So now, the “insurance company” will demand a higher premium since the odds of them paying out are higher. You and I are still locked in to our “\$50k a year for \$10 million of insurance” deal, but either of can still buy/sell a different insurance contract to somebody else.

So here’s what you do – you’re currently locked in to paying me \$50k and you’ll get \$10 million from me if GE goes bankrupt. But the fair cost of insurance has gone up now, so other people are now willing to pay \$60k for the same thing. So you do a credit swap with them where they pay you \$60k and you pay them \$10 million on bankruptcy.
Net result: if GE goes bankrupt, I owe you \$10 million and you owe the other guy \$10 million, so you just pass along the money and it doesn’t cost you anything. Or, if GE doesn’t go bankrupt, then you owe me \$50k a year, and the other guy owes you \$60k a year. You pocket the difference. Score!!! Guaranteed no-risk profit!

As for me, I’m still getting your \$50k a year, but I’ve “lost” money in the sense that my odds of paying out are now higher. I’m afraid: “oh crap, I really don’t have \$10 million, I’d better protect myself just in case GE bankruptcy happens.” So I cut my losses by buying insurance from somebody else, just in case it becomes necessary for me to pay that \$10 million to you.

Final result: you and I both have 2 swaps on, that just cancel each other out. We’ve both essentially transferred our original swap to somebody else. Now, imagine you and I do that ten times a day. You can see how the numbers start to make it sound much worse than it is.

Let me just add a little more to doubled’s excellent explanation of a very complex financial instrument.

You can enter into these swaps without actually owning the stock or the \$10 million or whatever is underlying the contract. This means you can enter into more and more swaps with nothing backing it up but a nominal amount of money and your promise to pay if necessary. Then when the company goes bankrupty suddenly you find you have 100000 people all demanding their \$10 million dollar at the same time. This is what (almost) happened to AIG and is why the government had to bail them out. If they had defaulted on these swaps it would have had worldwide reprecussions. They got greedy and keep selling these ticking time bombs with no one seeming to care about risk.

I think counterparty risk is a big looming problem which the government will have to address and soon.