Credit Default Swaps - it's hopeless

[Meadmaker]

http://news.yahoo.com/s/nm/20080923/ts_nm/us_financial_news_6

U.S. Securities and Exchange Commission Chairman Christopher Cox urged Congress to plug a regulatory hole in the $58 trillion market for credit default swaps, insurance-like products that many say pose a systemic risk.

There ain’t a darned thing any amount of money is going to do to shore up this system. Here's what's happening: All those derivatives are just bets. They are a zero sum game. You think they are worth something, but if you make money, somebody lost it. It isn't like the stock market, where a rising tide lifts all boats and all that rot. There's 58 trillion dollars floating around in uncovered bets.

58 trillion dollars is an ungodly amount of money. The problem is that it's not invested in the economy. It's just being wagered. That in and of itself would be bad enough, but people are calling their holdings in these things assets.

It's obscene.

Or, maybe I'm wrong. Maybe I don't understand what a "58 trillion dollar market" actually means. Somebody, anyone, try to explain this to me, because unless I missed something, there's a huge amount of money floating around based on a whole lot of nothing.

So...can someone explain how to fix a 58 trillion dollar market?

 

[Francesca R]

Yes, it *is* a mixture of long and short bets on default events occurring, some of them covered by underlying positions the other way around, and some of them naked. And as these bets are settled, there is a wealth transfer from losers to winners. And the problem is that so many losers cannot pay, and the winners already allocated their winnings so they rather need the losers to pay. And there you have systemic risk, caused by faulty risk management (failure to make sure you have enough to cover the probability of owing money) and excess leverage (trading so large that there is no way you will have enough to cover potential losses).

How to fix that (IMO) is to constrain the leverage. The outcome of excess leverage is inefficient and a public bad: it imposes a societal cost (bailouts) that does not make the network of players in aggregate any better off than if they had pursued the "optimal" amount of leverage and been able to settle their losses and gains (because what happens with a bailout is that the losers are left with almost nothing--rather than negative worth--and the winners still did not get their winnings). And society can't "insure" against excessive leverage that poses this risk to society. Hence it is an externality that justifiably requires government intervention to counter.

Note that this does not amount to saying that the CDS market itself is flawed or should be abolished. It is not and it shouldn't. In the absence of excessive leverage and the attendant public bad of a meltdown, the presence of hedging and speculative markets is a net benefit for society IMO.

 

[Lothian]

How to fix that (IMO) is to constrain the leverage. The outcome of excess leverage is inefficient and a public bad: it imposes a societal cost (bailouts) that does not make the network of players in aggregate any better off than if they had pursued the "optimal" amount of leverage and been able to settle their losses and gains (because what happens with a bailout is that the losers are left with almost nothing--rather than negative worth--and the winners still did not get their winnings). And society can't "insure" against excessive leverage that poses this risk to society. Hence it is an externality that justifiably requires government intervention to counter.

That is just what I was thinking, the government needs to line these feckers up against the wall.

 

[Meadmaker]

And the problem is that so many losers cannot pay, and the winners already allocated their winnings so they rather need the losers to pay. .... In the absence of excessive leverage and the attendant public bad of a meltdown, the presence of hedging and speculative markets is a net benefit for society IMO.

I would generally agree with what you said, but I have a question. What concerned me, in reading the article, was that number. 58 trillion. I thought it must have been a typo, but I checked it out and no, that's really the number. Well, other articles said 62 trillion, but what's a trillion dollars between friends?

Am I missing something? When a 58 trillion dollar market goes bad, is there any possibility of fixing it? The losers can't pay that much, and the government can't step in and cover the losers' obligation, which means the winners aren't going to get paid.

And I'm not 100% certain what happens after that, but it seems very likely that it is bad.

The government can step in and fix things so that it doesn't happen again exactly like this, but this time around, unless I don't understand what it means to have a 58 trillion dollar market, the economy is screwed.

 

[Francesca R]

Am I missing something?

No, the market (in terms of the total notional exposure outstanding) is (approx.) that large, except I think it is a doube-count: it's the face value of contracts sold *plus* bought, and each contract has two sides. That means that the total amount of credit events that are being "bet on" (for plus against) is several times what the actual financial impact would be on the underlying markets and companies if every one of them collapsed to nothing, or defaulted with no recovery at all.

When a 58 trillion dollar market goes bad, is there any possibility of fixing it? The losers can't pay that much, and the government can't step in and cover the losers' obligation, which means the winners aren't going to get paid.

The market is very diverse. It is not all simply bets on companies defaulting on their debt (which they do not necessarily do even if they go bankrupt anyway). CDSs are bought and sold on governments, infrastructure, real estate, all kinds of securitised loans, inflation, and pretty much anything you want. They do not all go up and down in value all at once.

And I'm not 100% certain what happens after that, but it seems very likely that it is bad.

Aside from "58 trillion" (or, half that) not very likely to all become payable from losers to winners at once, the fact that CDS issuance on some company debt exceeds the total value of the debt is not, by itself, a problem. It just means that investors who bet that company X will not default may end up owing more--in aggregate--than the underlying default of company X represents. If both sides of a CDS trade can bear the possible loss, they are fine. But with excessive leverage, some of them can't. That's what makes the assets (the claims that the winners have) "toxic" and causes them distress. How much of that total now represents--or will represent--"can't pay" scenarios is the thing that few people know.

 

[Meadmaker]

It just seems to me that even if only a fraction of the total CDS pie is truly at risk, a fraction of 58 trillion dollars is still a whole boatload of money. And credit default swaps are only one part of the derivatives market. There are plenty of pies in the oven where the losers might lose so much that they can't pay the winners.

 

[Francesca R]

No, no. It'll be fine, don't worry. Just let us have the seven hundred yards of dollars and we'll not bother you again.

 

[IllegalArgument]

Until we screw up again and come begging.

 

[Francesca R]

With this package, we're good until the end of the month at least.