But there are legitimate reasons to worry about the growth o the CDS markets, regulatory arbitrage first among them. And now here's another possible reason: they may be making it harder for firms to restructure short of bankruptcy:
This week, mall operator General Growth Partners (GGP) and newsprint maker AbitibiBowater both filed for bankruptcy, after failing to persuade bondholders to restructure voluntarily.
Now lawyers involved in these bankruptcy proceedings tell the Financial Times that the credit default swaps are the problem -- mainly, bondholders who have purchased CDS on this debt have little incentive to negotiate or play ball, since the CDS, if the counterparty honors the agreement, makes them whole.
. . .
If it is AIG [on the other end of the swap contracts], it means our bailout is pushing companies into bankruptcy that might otherwise be able to restructure.
Note that this has been alleged before, though previously with GM's ongoing failure to get its bondholders to exchange debt for equity. Now those involved in actual bankruptcies are citing it as a problem.
This is very troubling. We know from mutliple economic studies that systems that are too creditor-friendly have lower rates of entrepreneurship and innovation. We all have a vested interest in forcing creditors to the table short of liquidation (though to be fair, in this particular case, my sense is that the bankruptcy is expected to result in a reorganization, not a liquidation). Perhaps swap contracts should allow the issuers to get involved in these negotiations, the way insurance companies sit at the table during lawsuits.
Isn't this the same problem with home mortgages; or to flip it around, changing the bankruptcy laws on mortgages created the same kind of scenario, which is now difficult to back out of?
It seems the down side of credit risk-mitigation is no one to re-negotiate with when a borrower slips into risk of default.
I don't think it's quite the same. The issue here is that CDSs allow postmodern debtholders to have all the benefits of traditional creditors without any skin in the game, because they have offloaded all their economic risk. Think of it as bringing irony to the financial markets.
I think that's what I said, but you say it clearer. Thank you.
Also, "changing the bankruptcy laws on mortgages" happened in 1978 in the context of an overhaul of the system that made it overall more debtor-friendly; the mortage rule was altered precisely because the other changes would have made mortgages much more expensive if the rules on secured debt had been left unchanged. It's not a good candidate for the problems we're seeing now.
Wouldn't they have had counterparty risk with the CDS, if our government hadn't set a precedent for making CDS holders whole in the event that a counterparty was on the verge of insolvency (as in the case of AIG)? This seems to be of a piece with the precedent of making bondholders in insolvent institutions whole as well, at tax payer expense.
That counterparty risk can, in turn, be hedged, as it apparently was with AIG: Goldman Sachs's short position in AIG stock ensured it would earn at least enough from an AIG bankruptcy to cover a failure to settle the CDS's.
The AIG case will not represent a troubling precedent provided Treasury is given new authority to deal with insolvent life insurers comparable to those it has to deal with insolvent banks. This authority is provided in the legislation that Mr. Geithner has requested from Congress.
The government has not "set a precedent"
the reason the Fed, et al, ponies up money to AIG in the first place is they - AIG - would be in default if the did not POST COLLATERAL on CDS. If AIG, or a similarly situated derivatives counterparty, failed to post collateral they could be in default.
The government did exactly what they said they were going to do last fall. To prevent default, they would lend money AIG owed AS COLLATERAL to many major banks. Perhaps the government should have let AIG default. Despite the hysterical news stories to the contrary, the money Goldman Sachs and other major banks got AS COLLATERAL on the CDS is not necessarily Goldman's to keep. It certainly is not income or earnings for Goldman or other banks, again despite the hysterical news coverage.
If I have a contract that says you have to post $1,000 collateral with me if X, Y and Z happen, that $1,000 you post (if you pay what is owed) is now my liability. Just because I have that collateral does not mean I have earned the money. A credit event, default on particular bonds, probably has to occur before I have earned any of that money. That has not happened in the AIG case. Theoretically, if credit events do not occur on those contracts where AIG sold protection, all $80 billion that AIG posted with creditors could be returned to AIG when those CDS contracts expire. That is unlikely, but could happen. More likely is AIG will lose some of that money of CDS coverage it wrote on bad bets is due because of defaults on the underlying bonds.
As for the "government" making bondholders whole, that notion is rubbish. Are holders of Lehman, Indymac and WAMU bonds whole? The notion of making bondholders take a "haircut" is just another attempt at pocket-picking investors by the Joseph Stiglitz "let's nationalize all wealth" crowd. I hope Geithner and Bernanke resist such nonsense schemes.
pct - "CDSs allow postmodern debtholders to have all the benefits of traditional creditors without any skin in the game,"
well, no
There was a cohort of CDS that were simply priced too low. You really can't have all the benefits of traditional creditor - like getting a competitive yield and principal - without bearing the risk. The only way CDS would allow you to do this is if someone, an AIG for example, sold you credit protection too cheaply. That happened, as is pretty damned unlikely to happen again in the near future
there's noting "postmodern" about it 2+2 still equals 4
This sounds like snobbery to me. You identify a problem with CDSs that couldn't be more integral to their very existence, but you pillory "crazy ignorant hating on CDSs," which presumably means a reflexive vilification of the poor things. It seems to me that if a financial instrument creates incentives to disaster then it deserves some hating, even if it's not as economically literate as you might like.
"We know from mutliple economic studies that systems that are too creditor-friendly have lower rates of entrepreneurship and innovation. "
I agree but remember the original argument for CDS was that it allowed more credit to be available. Banks could lend more to company A knowing they could lay off the risk with a hedge fund, or similar non-bank entity. So I would argum the development of CDS, by bringing in more players, did expand credit capacity to lend thus stimulating more economic activity/entrepreneurship. Whether this capacity was used wisely is another discussion - eg expanding the supply of credit led to reductions in its cost (loans and terms were underpriced - see subprime lending, leveraged loan market). But directionally more credit was available.
I agree that directionally with CDS it is harder to negotiate an out-of-court restructuring which was easier to do when you had capital structures where you only needed to deal with a few people (though to get wonk-ish that only holds true for CDS that do not include Restructuring as a Credit Event - ie with Restructuring a distressed exchange outside bankruptcy is a Credit Event and can trigger protection claim).
However, even if we bankruptcy in these situations in and of itself as a negative (Another topic, but one could argue that the problem isn't that getting in bankruptcy as such is bad, it is the cost of dealing with a bankruptcy given the lawyers, cost, and incentives for participants to slow the proceedings) the negative has to be outweighed against the relative positive of increased credit availability.
I'd also add that CDS as such weren't evil, what was ill advised was that these were not treated like exchange traded instruments. If you trade pork bellies in the CME, you post margin if the position moves against you, CDS didn't require MTM collateral if you were dealing with say AIG. Had AIG been forced all along to post collateral I think its book would still have got in trouble but the book would probably have been significantly smaller and less threat to the system.
In summary we may get more MGM's, but we also may get more, say, Google's and we have to weigh the good with the bad.
I must admit that I don't really understand this complaint (or to be more precise, I don't believe it). I would be shocked if the writers of CDS weren't involved in the pre-bankruptcy negotiations, if only by participating behind the scenes with the bondholders themselves- afterall, the bonds revert to them in the case of a default, so they will definitely be at the table in a bankruptcy.
If a restructuring outside of bankruptcy makes the bonds more valuable than they would be in a bankruptcy, then both the bondholders and the CDS writers have an incentive to pursue this by splitting this additional value. What this story sounds like to me is equity holders/management complaining that they aren't getting anything.
Yes - Abitibi is a major supplier of newsprint, that borrowed money to make acquisitions while the newspaper industry went into free fall. GGP made debt financed acquisitions in retail malls at the peak of the market just before the bubble popped on consumer spending
'hoocodanode'
Wasn't the point of a CDS to dillute the concern (the risk) of any individual bankruptcy, by spreading it out like a skim of a drop of oil on a lake? Inversely, for the writers to go through the effort of negotiating on behalf of all those bond holders, each with only a small risk, seems as much work as trying to scoop the drop of oil from the surface of the entire pond. Not worth the bother. That's why restructuring outside of bankruptcy isn't happening. If there's whining on behalf of equity holders and management, perhaps that's a symptom of the problem Megan's trying to get at.
I struggle to understand these things; just when I think I've got it, I realize I don't. So any help you can give, I always appreciate.
My understanding is that once the bankruptcy (or other credit event) occurs the vast majority of CDS are settled in cash. Once there is an event there is an 'auction', conducted by ISDA (International Swaps Dealer Association) to determine the price. Basically people bid for the bonds, then the market clearing price is used as the price to settle the trades. So if you wrote $100, the recovery price is $25, you write a check for $75 and you are relieved of all obligations. Those that want the bonds get them for $25 then deal with the bankruptcy court. There is alot of data on the ISDA website (www.isda.org) on more details. this process was used for Lehman, WAMU, etc etc and seems to be understood by the parties to these contracts.
Zic, I think you are confusing a Credit Default Swap (CDS) with a Collateralized Debt Obligation (CDO). A CDS is essentially just an exchange between two parties, whereas a CDO bundles together debt from multiple parties and then resells pieces of that bundle to multiple other parties.
As a simplification, in a CDS, one party says "give me a percentage of any upside on the bonds you own, and I will make you whole if there is any downside on those bonds." This structure changes the economic bond-holder without changing the nominal bond-holder, thus peoples' complaint that the nominal bond-holder has no incentive to negotiate so long as the economic bond-holder is creditworthy.
As a simplification, in a CDO, the buyer says "I have money to invest, but I can't take the risk that any single loan will default, but I'm willing to buy 1/1,000th of a piece in each of 1,000 loans so that if any single loan defaults I'll only lose 1/1,000th of my investment."
I’d be surprised if the CDS contracts don’t have any provisions that cover the scenario where a creditor is trying to restructure, but perhaps the transaction costs are high enough that the economic creditor is better off just going into bankruptcy?
I think people complaining about CDS are right in one sense. Where the notion of an "insurable interest" was breached, then these things became weapons of financial mass destruction. I think they are probably okay as long as these transactions are made between parties with an insurable interest.
Outside an insurable interest, these become casion gambling.
The point about the need for insurable interest has been made, by George Soros, John Hussman, and others. But that's not the issue in this post. This post claims that credit default swaps can be bad when the CDS holder does have an insurable interest (i.e., he holds the bonds on which the CDS is written), because it makes the CDS holder/bondholder less willing to negotiate with the corporate borrower. Of course, it's possible that this sort of risk-averse bondholder wouldn't have bought the bonds in the first place if he hadn't been able to hedge his default risk with CDS.
There is also the legal rights of the insurer. Where an insurance company is on the hook, it usually has a clause in the contract that permits it to negotiate to mitigate its damages (which in this case would mean restructuring the loans while paying off the loan-holder, much the same way that the auto insurer makes all the repair decisions about your car if you're in an accident). Most CDSs don't operate that way. Since they can be traded, you might not even know who the insurer is.
That was Soros' point in the WSJ a while back, and for once I agree with him. AIG thought they were writing insurance, but other parties regarded them as very cheap Bear Warrants, with very limited downside and huge potential upside.
Isn't the problem here a pretty classic case of moral hazard? Having shifted the potential loss to someone else, the bondholder has less incentive to work to decrease the size of that loss. Perhaps a partial solution might be to either require the bondholder to retain some of the risk (like a deductable)? I guess in the case where my bond is likely to be redeemed at pennies on the dollar, but I have credit protection on it, I actually have an incentive to push the default forward rather than accepting any negotiation. (Is that right?) If so, it's like the case where I have $2M of fire insurance on a $1M building.
I'll admit I'm not clear on how the lack of insurable interest is relevant in this case, unless the folks buying credit protection without owning any bonds are somehow in a position to impede bankruptcy negotiations. The only relevance would come up if we required that when someone owned a bond and bought credit protection on it, the seller got to take their place in any credit negotiation.
Megan,
I assume you're referring to the self-righteous sneering of Matt Taibbi?
Remind me again why anyone would deal with Wall Street at all?
Derek
I believe I saw something in the WSJ last week about this and they had a name for it....very spiffy but sadly I can't remember it.
Basically, the problem is that CDS alter the ownership of legal rights and incentives. The bondholder has the right to force bankruptcy and has the incentive to avoid this unless it is the best option to recover his money. CDS, though, create an odd situation where the bondholder may actually have an incentive towards bankruptcy. Consider a hedge fund that uses CDS to make a huge bet that a company will default. It then buys a lot of the company's outstanding bonds. The normal incentive is for a bondholder to negotiate. But here the fund doesn't really want the company to make good on its debts. It makes more money by forcing bankruptcy and immediate default.
The other side of the CDS, of course, has an incentive for the company to avoid bankruptcy but they have no legal rights in the matter.
I would propose looking at what was done in the 9/11 victim compensation fund. Basically they figured out what the compensation would be for each victim and then subtracted off life insurance policies that the victim had. Why not do the same? If GM owes you $5B but you have $2.5B in 'insurance' your bankruptcy claim is only $2.5B. Hopefully your 'insurance company' will pay you the other $2.5B and if they go belly up, well you can be a creditor in their bankruptcy case. You would be required to disclose any CDS's you have against the money owed to you as a condition of showing up at the creditor's table in bankruptcy court.
Assuming the lawyers could make this all work, what are the economics of allowing people to buy more 'insurance' against default than they actually own in bonds or even stock? If I own 'insurance' that pays me $15B if GM defaults, should I be allowed to own a lesser amount of GM debt or even GM shares? I'd basically be in a position to push GM to adopt bad policies in order to make the bankruptcy happen. This would be different from a typical short seller, selling GM's shares short precludes you from being able to vote the shares. Perhaps it is the job of the other side of the CDS (the 'insurance seller') to demand that I disclose owning any assets that might give me the ability to force a default? If Goldman Sachs knows I own 60% of GM shares they might not want to write me a policy that pays $100B if GM goes bankrupt.
Boonton - the problem is that total ownership of CDS's can exceed the outstanding bonds without any one individual being in this situation. So the market prefers bankruptcy, even if no single individual does.
I suppose an argument could be made, similar to short sellers, that these arrangements force the company into what they already are. AbitibiBowater is a declining industry with high costs, facing competition from low cost producers elsewhere. In the heady days where money was cheap, deals were made that didn't make sense. The consequences have arrived, and CDS' are making it happen quicker than otherwise.
Was there some government money going to pulp and paper mills in the US?
Derek
The argument here seems to be analagous to saying, "if you have homeowner's insurance, you don't care if your house burns down."
The insurance is there to lessen the burden of the house burning down, but the difficulty in rebuilding/replacing your belongings keep the homewoner from lighting up the torch.
In other words, the transaction costs of using the insurance have always kept the claim process as an option of last resort.
CDS's are basically insurance against bankruptcy. If making a claim (i.e., accepting payment on the CDS) becomes just as easy as protecting the underlying asset (i.e., resructuring the company), then liquidity evaporates - bondholders no longer have an incentive to keep the company solvent. And since stockholders can be bondholders and bondholders can be CDS holders, confidence erodes and markets collapse.
President Obama (and others) talk alot about the need for new/additional regulation in the financial markets, but rarely say what those new regulations should be. The more I read/think about this, the more I think the regulations have to be around slowing down the near-frictionless markets we've built over the last couple of years (something akin to the "trading curbs" the NYSE has).
My understanding of this is that if the CDS will give you 90% of the insured value, but restructuring will give you only 85%, then forcing bankruptcy is the best option for you.
Megan,
This is a remedial finance education type question I've been puzzling over for awhile now. It seems like CDS provide a lot of the same function you could get from an options market in the bonds or whatever is being covered. That is, instead of buying credit protection on GM bonds, you buy put options on them for a year in the future. Assuming you could get those options on the market, how much more would CDS give you than you could do with the options contracts?
albatross,
It certainly has to be true that someone, somewhere along the insurance/reinsurance chain uses simple options and futures to hedge the risk involved in writing CDS themselves. Like you, I wouldn't mind an explanation of what benefit the CDS contract adds that couldn't be obtained before. I imagine there is some economy of scale involved that lowers transaction costs.
This seems to me more a case of poorly priced CDSes being the problem. Insurance companies were basically giving the things away before the bust.
If CDSes were regulated like other insurance, they'd have been expensive enough that a. not every bond would be CDS insured, and only strongly risk averse investors would have insurance, or b. companies likely to default would be unable to persuade people to buy their horrendously expensive insurance.
Actually, case b. looks like it could be a way to provide properly incentivized bond ratings, since the insurers have the right incentives to look at accurate default rates.
The argument here seems to be analagous to saying, "if you have homeowner's insurance, you don't care if your house burns down."
Not quite. A more correct way to say it is 'if you have a mortgage and homeowners insurance, the BANK does not care if your house burns down'. Which is true. And if the bank has a choice between taking a loss in a cram-down on your mortgage versus getting full value on your mortgage if the house burns down, then the bank will prefer that your house burns down.
Most CDS are not owned by people who have the debt. For these gamblers, who have no say in a bankruptcy, it is more like "If you have homeowners insurances on some guys house in Topeka, you dont care if the house burns down." In fact you prefer that the house burns down as it is the only way to get paid. CDS for people who have not issued the debt are bets on bankruptcy or to maintain the analogy, bets that a house will burn down
Credit Default Swaps generally will pay off both in a bankruptcy AND in a debt restructuring. So I don't see how it makes a difference either way.
So here's the problem, if you had homeowners insurance for 20 times the value of your house, you may *want* it to burn down. This would be a problem considering that you live in your house and you could *make* it burn down. So both the law and the industry make it hard for you to arrange things this way.
This problem is NOT like short selling. When you short a stock you sell its shares, you give up the right to 'run' the company. You make a bet the company will go down but you also give up your right to push the company to make bad calls.
Imagine, though, a person who amassed a huge CDS stockpile and also owns a large amount of stock or bonds in the company. If the company does well he looses (or doesn't gain what he could if the company fails). His incentive is for the company to do bad and by owning it or owning its debt he has some ability to force that to happen.
It's hard to own a controlling stock interest in a company and be hidden by it. I think the entities writing CDS's should be able to see it's not a good idea to let someone with a controlling interest in a company amass huge bets that it will default. I think in the case of bonds the solution is to break the legal right of bondholders to push the company into default.
To get something you have to give up something. If you want 'insurance' on your bonds then you should give up your right to sit at the bankruptcy table. The company with the bonds no longer owes you money but the one with the CDS does, push him into default.
Credit Default Swaps generally will pay off both in a bankruptcy AND in a debt restructuring.
I did not know that.
I have a question about the gamblers, the people who bought CDS on debt that they did not issue. Will they get paid off in restructuring or default? How does that work? Do they get paid off the difference between the settled amount of the debt and what the full debt was for even though they never owned the debt?
This conflict was discussed on the nakedcapitalism blog in Jan:
http://www.nakedcapitalism.com/2008/01/credit-default-swaps-increase-odds-of.html