Credit Default Swaps Under Fire

The WSJ reports that:

International momentum is building for stricter oversight of derivatives trading, as a top U.S. regulator recommended new limits on credit-default swaps and European leaders pushed for a ban on speculative bets against government debt following recent financial turmoil in Greece. ...

German Chancellor Angela Merkel said Tuesday that her government is backing an initiative to curb the credit-default swaps market, together with France, Greece and Luxembourg, and she suggested Europe would forge ahead on its own even if the U.S. didn't go along.

José Manuel Barroso, president of the European Commission, the European Union's executive arm, said the commission would examine closely the possibility of banning outright "purely speculative" trading of the swaps. ...

As financial problems mounted for Greece and other euro-zone countries in recent months, prices of swaps insuring against debt default by those nations soared, drawing attention to the troubles and raising questions about whether speculation was worsening them.‬‪

This is just absurd.

Let's review what credit default swaps are and how they work:

Credit default swaps (CDS) are a form of insurance. Let's say you borrow money from me. I'm worried that you might default. So I hedge that risk by purchasing a CDS. If you end up unable to pay me back, the seller of the CDS will cover my losses. (The insurance analogy admittedly is not exact, but it suffices for present purposes.)

As the Journal explained, banning the use of CDSs as a hedging device would have adverse consequences, just as banning insurance would:

Any attempt to restrict CDS trades could result in unintended consequences such as more risk for the financial system and higher borrowing costs for a range of nations and companies, some analysts and investors warn.

Restricting credit-default swap trading could push up borrowing costs for various nations if investors feel they have fewer ways to protect themselves if the bonds' prices decline.

The trouble is that CDSs can also be used to speculate. Suppose I think that Greece is going to end up defaulting on its bonds. Accordingly, I don't own any Greek bonds. In order to gamble on the possibility of default, however, I buy a "naked" CDS:

It is legal for someone with no underlying exposure to the credit instrument to purchase CDS protecting against the default of that instrument. This is known as a “naked” position in the CDS. Such positions are taken by those who are making a bet either on a default, in which case the CDS would pay off, or increased perceptions of a default in which case the CDS would appreciate and could be sold to someone else seeking protection against default.

It is this use of CDSs that the regulators are most upset about.

To revert to the insurance analogy, the purchaser of a life insurance policy must have an "insurable interest" in the life of the person being insured, We don't want Tony Soprano taking out an insurance policy on Big Pussy right before he whacks him. Yet, if Tony can do so, his incentive to whack Pussy becomes even stronger.

The problem with CDSs is similar. It is not the CDS itself that contributes to Greece's problem. A bet on a football game doesn't change the outcome of the game, as a poster at Seeking Alpha explains:

If I buy CDS on a company without owning any of the underlying debt, I cannot effect the health of the company. Note that I don't have an insurable interest, but it doesn't matter - the "perverse incentive" is a pipe dream, because I can't act on it.

It doesn't matter how much CDS I buy - I could own a gazillion dollars worth of insurance on the debt of a given company - but that still doesn't give me any say, any seat at the table in a restructuring negotiation scenario. There is a similar analogy with short selling stocks, and it's the reason why people who blame short sellers for the demise of companies are generally nutjobs: short sellers cannot and do not effect the health of a company.

Similarly, CDS levels do not effect the health of a company - they REFLECT the health of a company, or at least the market's interpretation of that health. Some companies will see the value of their CDS widen when people fear for the company's financial health. The CDS is a reflection of the fear, and not the cause of the company's problems. To suggest that panic from widening CDS levels causes companies to collapse is like saying that avoiding marking assets to market makes them worth whatever we want them to be worth - limiting CDS trading would not alter the underlying health of the company, it would only mask it.

So, since one cannot effect the health of the company by owning CDS - since one cannot murder the company (or burn it down) via a CDS position, the "burning down your neighbor's house" analogy goes up in smoke as a straw man fallacy.

Instead, the problem is that the holder of a naked CDS has an incentive to do other things that increase the probability of default. Fraud, manipulation, and the like come to mind.

Banning naked CDSs thus suffers from three problems:

1. It's at best a misdirected prophylactic, If the problem is fraud and manipulation, ban fraud and manipulation. Don't ban gambling, ban point shaving.

2. There is no evidence that holders of naked CDSs have done anything to contribute to the problems faced by Greece and other sovereign borrowers. Indeed, there's no evidence that people are even using them to speculate in Greek bonds. As the Journal explained:

[A] study released Monday by Germany's financial regulator, BaFin, found no evidence that credit-default swaps have been used to speculate against Greek national debt. The study showed the net volume of outstanding credit-default contracts on Greek national debt has remained unchanged since January at about $9 billion. This compares to total Greek government debt of about $400 billion. "The market data do not show massive speculation in CDSs," the regulator concluded.

Instead, as Megan McArdle observes, Greece's problems are entirely of its own making:

I expect that Greek debt will be carrying a substantial risk premium for quite some time, reflecting the fact that the debt is, well, riskier than the debt of bigger and richer nations. The Greek economy remains quite dependent on tourism and agriculture, both of which are subject to rather sudden shocks. Its institutions are often quite weak, and corruption and tax evasion remain serious problems.

3. A ban on naked CDSs would be unworkable, as a poster at Seeking Alpha explained:

In testimony before the House Financial Services Committee on Thursday, March 26, Treasury Secretary Geithner made the following statement regarding naked CDS positions:

I know there are strong opinions on this issue, so I say this with some trepidation. My own sense is that banning naked (CDS) volumes is not necessary and wouldn’t help fundamentally in this case. It’s too hard to hard to distinguish what’s a legitimate hedge that has some economic value from what people might just feel is a speculative bet on some future outcome. If we could find a way to separate those two types of transactions from each other, we would have done that a long time ago across a whole range of financial innovations, but it is terribly hard to do. But we will listen carefully to any ideas in this area and understand why people feel so strongly about this.

Our view is that the absolutely essential thing is that there is more capital held against these positions so we never again face the situation where those types of judgments could imperil the system.

The position Secretary Geithner is taking ... [is] that it would be next to impossible to differentiate between naked CDS positions and positions intended to hedge underlying exposures.

London's City AM similarly reports that:

The crackdown on speculators also got short shrift from bond market experts who said it would be impossible in practice to unravel the complex interrelated web of CDS positions.

Unless policymakers make it absolutely clear what does and does not count, uncertainty will cause extreme volatility, and they run the risk of ending up with worse problems than they started with,” said Simon Thorp, head of fixed income at Liontrust Asset Management.

Update: Todd Henderson blogs that:

As I explain in this paper, credit derivatives are merely a financial tool that can be used by those exposed to credit risk, say a default by the Greek government or General Electric, to share that risk with others. This lowers the costs of borrowing and helps spread risk. In addition, third parties with no exposure to the particular credit risk can bet on whether the Greeks will default. These secondary-market transactions are the same as an individual buying stock in General Electric betting it will rise. Importantly, these bets provide a liquid market for credit risk, which lowers the cost of hedging for those with primary exposure, and provides the market with better information about whether Greece or General Electric is a good credit risk. Those who might lend to the country or company, those conducting other business with it, and those who might face the risk of default in other ways, can use this information to better plan their activities. For instance, those disbelieving a country or company’s claim of financial soundness, say because of funny accounting (think: Enron or, dare I say, America) can use credit derivatives to short debt, something that was impossible before credit derivatives were invented. This makes debt prices more accurate and holds borrowers, be they sovereigns or corporations, better to account.

Of course, there is the possibility for abuse. Another New York Times article from a few weeks ago highlights the possibility for abuse in this market. (Note the parallel between the conflict of interest across departments at Goldman Sachs and those in the investment analyst scandals from a few years ago.) But abuse is possible in all markets, and everyone should be in favor of vigorous enforcement against those who try to manipulate markets or trade on undisclosed conflicts of interest. The existence of the potential for abuse, however, is no more an indictment of credit derivatives generally than it is of the stock market or any other useful tool of society than can sometimes be abused.

It is the ultimate irony that politicians are blaming their problems on a tool that helps reveal their tricks and mistakes. This is akin to a burglar blaming an alarm system for being caught. Sure, the burglar might have been better off without it, but the homeowner and everyone else is glad it was installed.

Comments

[C.E. Petit]

I think the problem here is that legally, a failure to ban is also treated as a positive endorsement. In other words, the securities bar (not alone, but in particular) uses purely Aristotelian logic, and defies the concept of the "excluded middle." Too often -- and insider trading theories in all forms are an excellent exemplar -- one sees the argument that since a particular trading practice or instrument is not affirmatively prohibited, it can therefore never be used as evidence or part of a scheme to defraud.

Now that I think about it, there's an area of law where this is actually much worse than in securities fraud: Copyright infringement claims based upon prior, obscure publications and similarity analysis.

[save_the_rustbelt]

Right, how can rich guys loot the economy is there is no CDS market?

Oh,...

[Lindasy]

I believe Seeking Alpha means 'affect,' not 'effect.'

[Warren Bonesteel]

You know, Professor, the problem with smart people is that they're often the last to face reality...because they're so smart that they can convince themselves that reality just ain't so.

see:
MISH'S Global Economic Trend Analysis
Calculated Risk
Zero Hedge
Market Ticker

[Steve White]

It's at best a misdirected prophylactic, If the problem is fraud and manipulation, ban fraud and manipulation. Don't ban gambling, ban point shaving.

Well sure, except that it's a easier to ban gambling than it is to root out point shaving. The more gambling we allow, the more likely it is that sharp operators will look to evade whatever surveillance and penalties we put on point shaving so as to make a lot of money. After all, the criminal mind weighs reward and risk differently than you and I do.

With a naked CDS, there is plenty of incentive to manipulate the market, many ways of doing so (some of which are quite subtle), and (given the usual state of political affairs) little attention paid to rooting out manipulation until a crisis occurs, when it is too late.

The harm to society is also much greater: if an NBA referee participates in a scheme to help gamblers shave points in a series of games, the losers are the game, the fans and the other honest gamblers. Society itself doesn't take a drastic hit even if the Lakers lose. In contrast, a naked CDS manipulation has the potential to torpedo a national economy.

That Greece has clearly created most of its own problems doesn't change the fact that a naked CDS manipulation could bring down the Greek economy -- indeed, the incentive there is higher than trying a similar act of fraud against the US economy, since the Greeks are in more trouble, the fraud would be easier to pull off (fewer dollars up front invested by the manipulators) and the fraud would be more likely to succeed.

Your position is that of a purist, mine is that of a cynical realist: I understand just how big the incentives are to rig a naked CDS, and I don't like the consequences of a successful one. I would ban them and require anyone buying a CDS to have a clear interest in the financial instrument being insured.

[David Alexander]

I don't understand why the insurance analogy is not apt here. Murder is illegal anyway, yet we ban those without an insurable interest from buying life insurance policies to avoid the moral risk of murder; arson is illegal, yet no insurance company will sell a $300K policy on a $100K house, to avoid the risk of fire from "friction" between the value of the house and the value of the policy. So if there is a recognizable risk of fraud and manipulation by holders of naked CDSs, why not avoid it in the same manner? What is the distinction between the moral risks of CDS fraud and insurance fraud that makes it acceptable to address one but not the other?

[Paul Snively]

I wouldn't be surprised if http://www.freedom-to-tinker.com/blog/appel/intractability-financial-derivatives also turned out to be relevant to CDSes as well as CDOs, but I haven't actually attempted a direct comparison, so caveat emptor.

[Jeff]

You can always buy a long term OTC put option against a companies fixed income issues to "bet" on a failure. In some ways a CDS is just a special type of put option. Usually triggered by a default as opposed to a specific strike price but the same kind of bet.
At this point in time I am not actually aware of any CDS ever having been paid off, i.e. a company defaults and someone has to pay off the CDS's.
As I understand it the AIG problem was a margin call that they could pay.

[Stunned and Amazed]

Explain the taxpayer bailout of AIG please.

[Arthur Pendennis]

I thought the greater problem in the past was that some of those writing the insurance did not have reserves to back them up. Wasn't that the problem AIG had?

Though I am also not sure that, where a country as small as Greece is concerned, the risk that someone having purchased a large amount of insurance might try to manipulate their debt has been sufficiently addressed. In other words, I don't feel entirely comforted by the argument in point 1 where current laws against point-shaving might not be working that well. Perhaps if adequate disclosure was made on who was wagering what on whom, I might feel better about detecting or deterring the point shaving.

[matt]

Having had this discussion with SEC Chairman Cox in 2008, the fractionization and hiving of various revenues and income streams from the underlying securities led to a market in which the interrelationships of those derivitives became opaque. The consequence was a meltdown of the financial system with trillions of dollars lost by investors.

When pricing risk into the CDS's there were inadequate or nonexistent benchmarks against which to measure that risk. In the end, no one knew who owned what, and there was a cascading effect as a single event would affect multiple financial instruments.

The problem still exists and has not been resolved.New CDS's are now being offered to the market. The issues are transparency, leverage and the complex interrelationships of derivatives.

[AD]

The Economic Meltdown...

Don't blame me, and don't blame thee, blame that CDS behind the tree.

[CHM]

The ole Saturday Night Live crew had a scit about razor blades years ago after Gilette came out with a -2- razor blade. At that time their joke was we will do -5- razor blade. The point being enough is enough, overkill. Insuring the insurance and insuring the insurance is OVER THE TOP. You make a decision on your INSURANCE by the rate of interest charged up front. To have people placing/taking bets on debt they don't have an interest in is just gambling and serves no economic value.

There is a reason we do not allow people to purchase LIFE insurance on individuals whom they have no vested interest.. i.e. family members,valued employee, business partner....

[Wes]

'If I buy CDS on a company without owning any of the underlying debt, I cannot effect the health of the company.'

Gawd... pet peeve. Affect and effect are 4 different words. When effect is used as a verb it is essentially equivalent to saying 'bring about'. I'm sure the writer meant 'affect' as in 'alter/impact', but wanted to impress.

[Doug Collins]

From what I have heard, there is another problem with Credit Default Swaps that is more or less unrelated to debt- They can be used as a sort of uncontrolled, print-your-own money.

Charles Kindelberger in Manias, Panics, and Crashes: A History of Financial Crises, said (my summary), most of the bubbles through history were caused when a new form of money/value was invented or discovered. Initially it is uncontrolled by either law or any other feedback mechanism. I write you an IOU for a million dollars and you write me one.

Voila! We are both millionaires. The IOU might be in the form of tulip bulbs, stock margin borrowing or naked CDSs. As long as they don't get called, we are all good. If the crash comes quickly, then the punters and their creditors are hurt, but the damage is fairly limited. Everyone must believe that the lender of last resort will not act, then the lender must time it to act at the last moment. There can be no "Too Big to Fail" entities. If they are protected and bailed out too soon, all those IOUs start to be integrated into the world money supply and massive deflation can be the result when the IOUs are finally called and found wanting. All that money vaporizes.

Unfortunately, the recent bailouts may have made a future deflation more probable, rather than counteracting a present one. And, of oourse, the bailouts themselves are another kind of IOU.

After the crash, the most recent version of money is made illegal or tightly regulated and things are fine for about another generation. Then some financial wizard thinks up a new kind of money and the cycle repeats. This time it appears to be CDS's.

[Sean]

Call me crazy, but wouldn't outlawing CDS's basically put more of the risk (and the associated costs of said risk) back on the issuer and buyers of bonds?

In which case, Greece won't be able to borrow any more money at "X", but will have to price the risk in at "2x", for example, since no one else is around to assume the risk of their default.

In the end, Greece still couldn't borrow money at the rate they wanted, CDS's or no CDS's. But without CDS's, other folks looking to borrow money would have the risk factored into THEIR issuances.

Instead of speading the risk of default out wide, it's limited now to the buyer and seller of the bonds.

When are voters going to learn that lawyers and lawmakers have no clue about financial markets.

[furytrader]

I think there are a couple of things to clarify here.

1) The claim that "CDS levels do not effect the health of a company - they REFLECT the health of a company, or at least the market's interpretation of that health" is not always true. For example, in the case of CDS written against debt issued by a bank or other financial services firm, whose "good name" and trustworthiness is essential to its continued operation, if people start to drive up CDS prices for the sole purpose of making them go higher still (i.e., to generate momentum), it can be taken as a signal to other market participants that something is wrong, the firm's trust gets destroyed, credit lines start to get pulled and no one will trade with that financial firm. It can no longer do business. It goes bankrupt.

This contrasts with selling short the stock of a company that holds no debt but owns a lot of assets. There comes a price level where the value of the company's stock is less than the liquidation value of the enterprise - giving other market participants a reason to buy the stock below that price, almost as a risk-free trade. In this case, the short selling of a stock cannot drive this company's stock to zero.

The difference in these two cases is that the first company's very existence is based on the trust the market has in its continued operation while the second company's value and existence is based on tangible assets.

2) A CDS is nothing more than a put option on a bond. Can someone describe what kind of hedger would SELL (or write) CDS? I'm having trouble imaginging who that would be ...

I don't see why put options on bonds aren't traded the same way put options on stocks are: through a central exchange with daily market to market, position reporting and, as a result, greater liquidity. Seems like a no brainer, except that it would demolish the spreads enjoyed by the current market makers in CDS.

[Don Meaker]

Rich guys can't loot the economy WITH a CDS market. They can only loot the guys selling the CDS, and the CDS seller must be a willing partner in the looting. When too many people buy CDS at a given price, that should be a signal that the CDS seller may have calculated the odds of default incorrectly. When that happens, the CDS seller should refigure the odds (as bookies do) or limit his exposure (no more CDS on that available). If the CDS seller is right, he wins, if he is wrong he loses.

[Don Meaker]

Banning CDS means banning the transation that only takes place with a CDS. Result: Less money in circulation, deflationary prices, less growth, less opportunity for new starts.

Old money wins. New money doesn't get a chance to get made.

[Doc Merlin]

CDS and similar information markets are an easy way to make a market measure of a country's short to medium term fiscal soundness. This terrifies unsound countries. They would rather people not know.

[ThomasD]

I agree with Doug Collins. While on the one hand CDSs are described in terms of insurance, but when you look at how they actually function in markets they are really a means of facilitating transactions that might not otherwise occur without them. Effectively they represent increased liquidity - ie. they are money.

They are only considered insurance right up until the point when they would be called upon to cover systemic loss. Then the fiction is revealed, as it was with the mortgage industry bust.

If they ever really were insurance - adequate reserves determined by sound actuarial principles - then losses would have been covered (or at least substantially reduced) and we wouldn't be having this discussion.

[Lots of Reasons]

And what if no CDS is available for a bond I own, but there is another CDS available on a bond whose default risk is highly correlated to my bond? Am I one of those terrible "speculators" now?

BTW, as far as financial firms go, worrying about the "good name" of financial firms being tarnished by naked CDS holders, would you ban analysts from issuing negative research reports on those companies (whether they own CDSs or not)? If I bet you that Goldman Sachs is about to implode, does that make Goldman more likely to implode? Does just saying it here in this comment make it more likely?

[ian527]

I have no particular problem with CDS's. They are an insurance policy and as such the company that issues them has to be adequately capitalized so that if they had to pay out on all of them (worst case) that were issued, they could. This was clearly not the case with AIG.

[Wannabeer]

Let's assume the insurance contract model is the appropriate one. In that case, the entity buying or selling insurance should have an "insurable interest" in the underlying reference obligation. In other words, there is a true hedging of exposure. It's in cases where there was no such interest (ie, where there was total speculation), that things got out of hand. Contrary to your analogy, it's not like "buying a stock because you think the price might go up." In that case, you actually have a real interest in the asset. For the same reason, it's not even like naked shorting of a stock, where at least you have a derived interest in how it does. Moreover, because shorting happens in a regulated, transparent market, I suppose we could make the argument that the shorts are a way of price discovery and actually help the market by correctly pricing the security.

Those who just "clipped the coupons" -- selling protection without any hedging, were the real risk. (Hello AIG). We couldn't identify it because the market was opaque, and even if we could have id'd it, no one had the power to do anything about it. You can solve the clipping coupon problem by requiring adequate reserve to meet the obligations on which protection is sold.

In other words, as is true so often -- Wall Street objects WAY too much when it comes to regulation. Face it, we have just come through a profoundly unregulated time. The result was profound, and fell mostly on people who couldn't afford it and didn't create the problem. So suck it up.