Hacker News Reader

That's an interesting analysis. Isn't that same as burning down one's house to get insurance payout? In this case, burn down a 3rd party's house or entice the 3rd party to burn down their own house.

Yes, that's exactly why you can't insure a 3rd party's house or life.

I guess insurance is more regulated. But maybe you can get a CDS on someone's house burning down?

Mortgage default swaps (and other asset backed default swaps) are exactly this.

All life insurance is in a sense third-party?

Employer's can take out life insurance on an employee (provided consent is given). Wondering if there are exceptions to this rule?

I think it's about not creating adverse interests. Generally, a company is much more interested in a living employee than they are the payout from one dying.

The credit event auction means that you can't just get someone to write a CDS, create a default, and collect money. If everything is trading at par then the auction will be at 100 and you'll get nothing.

The trick here is that they have bonds that are trading way below par. Honestly this seems like a story about someone falling asleep at the wheel.

Below link has a primer on auctions as well as results:

http://www.creditfixings.com/CreditEventAuctions/fixings.jsp

Is the implication that Goldman should have bought up those bonds until the price was higher, to protect their position? Do other sellers of these securities do the same? In that case, this would seem to be yet another example of Goldman seeking profit outside the game of finance. It worked for them in 2008...

It means that the level where they trade the CDS should account for this. TBAs are another good example where this stuff is important.

As in passthrough TBAs? How so?

It's literally the same effect. The exact security isn't specified so you're going to get the "worst" one that is eligible. If you misjudge how bad the "worst" security is then you'll get screwed.

Similarly, these CDS don't specify the exact security so you need to think about the worst case. The major difference is that the auction only happens if there is a credit event.

Maybe here's an easy non-financial way to think about it. Suppose we sign a contract that says that I need to delivery 1 ton of at least 90% gold to you. When you price the contract, you probably shouldn't assume that I'll deliver 99% gold.

Ah, yes. Lots of stories about contras calling up and thanking you for letting good stuff out the door on A-day.

Mistakes like this happened -all- the time at both the banks I worked at.

This sounds like a fairly obvious scenario that the lawyers who define the CDS would’ve considered. Is that true?

I was thinking the same thing, based on some discussion we had about a big contract we entered into that can terms where if we as the supplier got into financial difficulty, our customer could release our code from escrow and run it themselves. The concern was that the customer might deliberately try to push us into financial difficulty (e.g. by not paying us on time) as a way to in effect permanently avoid paying for our service. So our legal team added wording to the effect of "except if you caused it".

Nope! Blackstone has successfully done it before: https://www.bloomberg.com/view/articles/2013-12-05/blackston...

Thanks for the link. That article makes this intriguing point:

“But you shouldn't overstate that value. The reason that Blackstone made money on its credit-default swaps is not just that they were triggered by this clever maneuver. Just triggering CDS is not a big deal, because CDS pay out based on the difference between the face value of a bond and its post-default trading value.”

Is the same thing going on in this new deal?

All of the outstanding bonds except two of their smaller issues are being quoted above/close to par. CDS has not been triggered yet so can't really say but around those prices BX would probably lose money.

I've always been so puzzled by these "synthetic" financial products. If someone wants to "insure" a corporate bond, why don't they just swap it with a bank for a cash flow stream guaranteed by the bank, minus some premium?

... which guarantee would then be hedged by the bank with a CDS.

Who would take the other side of that trade ... ?

That's basically the same thing as selling the corporate bond and buying a government bond of similar maturity and coupon frequency in the same currency.

Though I don't get why you're puzzled on synthetic derivatives, since what you've proposed is a synthetic instrument itself. It's an exchange of upfront payment for an annuity stream.

And it's basically the opposite sign of the off-market swaps people in this HN community have already criticised banks for providing as a means of off-balance-sheet funding to some clients. (Eg, Greece and Italy entering into off-market swaps with banks, which is what your proposed contract is, except receiving upfront cash in exchange for the liability of a stream of future payments).

Separately, the bank could wrap your annuity stream as into structured note to make it seem more like a cash bond. But then you're back into something even more bespoke and financially complex than a standard contract CDS.

> I've always been so puzzled by these "synthetic" financial products.

Well what you are describing with your swap scenario is the very definition of a synthetic product, so your question as posed doesn't really make any sense:)

The real answer probably has a few reasons

1) There are more benefits to owing a bond than just the coupon and principle payments, why give those up by handing ownership of the bond to the bank?

With a CDS you still own the bond and the insurance in the case of a default, and remember a default doesn't mean the bonds go to zero.

What if the bond doesn't default but goes way up in value, you've handed that value over to the bank rather than getting it yourself.

What if you want to sell the bond before the expiry date.

All of these are answerable on their own but you can see how this actually rapidly complicates things.

2) and this is probably alot more important, What if I don't have the bond itself but still want default insurance on the bond? I might be a large counterpart to the company, htey could be my largest client or supplier.

Also, sometimes there can also be tax advantages to not selling something (realizing gains) and instead buying insurance on it (possibly deductible expense).

Uniformity is a major advantage of CDSs. Bonds are idiosyncratic- every one will has its own maturity date, coupon payments, and lots of minor details in the bond covenants. That makes bonds hard to trade. CDS are far more uniform- they trade with standard terms and maturities. Bonds are primarily a way to provide companies with funding, CDS are primarily an instrument for financial companies to trade.

Could a company "insure" a bond by writing a swap with a bank? Sure, a bank would be happy to write a bespoke swap with you, but they're going to charge more than a CDS would cost, and you're not going to be able to trade the swap to someone else.

CDS actually make it easier for companies to raise money. If you can hedge your weird, illiquid bond with a nice liquid CDS you're more likely to buy the bond in the first place.

FWIW I'm not a specialist in fixed income, so don't take this as gospel, but I think the general point is right.

Not sure I necessarily agree with the point about making it easier for companies to raise money. I agree with it at face value but not based on what usually happens in the market. Chances are, if there's liquid CDS for an issuer, they're not going to be issuing weird, illiquid bonds. On the other side, if all an issuer has are weird, illiquid bonds outstanding and without knowing anything about the issuer, I would say it is negative to have a liquid CDS market as that's probably a negative creditworthy signal.

I'm sure some academic has done a study about how CDS has shaved a few bps off borrowing costs for large IG issuers though.

Agree with everything else in there.

Thanks for the info!

To answer several instances of the same question at once:

Regardless of whether you think "pay me $COUPON every 6 months until date D when you pay me $PRINCPIAL" might be considered "synthetic" surely you must all admit it's far less synthetic than "Pay me $X if and when CORPORATION defaults"!

I still don't understand your objection to GSO's opportunistic trading. Triggers and credit events were provisions agreed on by both parties - it's a bespoke CDS after all - the only unintended consequence would be for future CPs to read the annexes to what they are trading, can't say it will be bad.

RE: 1, single name bespoke CDS are rarely traded nowadays especially given that more and more names are cleared now.

If this is allowed to happen then who in their right mind would ever again underwrite a CDS for a companies debt if some other company can so easily force a default event.

Would this be terrible?

Yes, incredibly.

CDS's are a form of insurance. It can be hard for someone not in the industry to see the importance of these as it doesn't affect yoru day to day life.

Just imagine your day to day life where you can't get any isurance at all.

No car insurance, no house insurance, no medical insurance.

That's the extreme case, in all likely hood the more likely scenario would be all your insurance tripling in price as no provider can hedge their risk vai reinsurance, etc.

To me issuance is very important and one of the pillars of modern society. If insurance going away, or atleast being inaccessible to 99% of the population that doesn't seem to be "terrible" to you then I'm not sure what explanation I can give you to help you understand :)

If CDS market disappeared or dried up - the broader insurance market will survive and exist just as it had prior to 1994 when the first CDS arrived.

Credit default swaps were invented in 1994. Liquid bond markets existed for centuries without them (1602 for the Amsterdam Bourse and the late middle ages for the Italian city-states).

CDS's are a form of insurance -- that's true. But in the case of a regular property/casualty insurance, you can't typically insure something you have no financial interest in. For CDS's, that's the base case.

It is claimed that CDS's originally were created to allow investors to insulate themselves against the economic shock of defaults. That's barely true. CDS's are, and have always been, instruments of leverage (gearing). They allow credit traders to earn a higher return by taking on the credit risk with a smaller outlay of capital.

I certainly don't want them to go away, but let's not kid ourselves that 1) they are fundamental to the credit market, or 2) they're not subject to potential abuse.

Alright I'll bite.

I fully agree with this.

> 2) they're not subject to potential abuse.

But to be fair, no one claimed otherwise;)

I'd appreciate it if you can change my mind but you really haven't' laid out a good argument yet.

My thesis is that CDS's allow the credit markets to be larger than they would be without them, not that credit markets can't function without them. Obviously credit markets have existed before the ability to easily hedge out risk

I mean I think you'd have to agree that hedging and risk minimization is a good thing and something that allows the market to be alot larger and more liquid due to the ability to hedge out risk with them.

If swaps go away tomorrow, what replaces them? How do people hedge out credit risk without any form of swaps?

My argument is that they are not now, and never were, primarily a hedging instrument. While they can be used for hedging, like options for instance, they are primarily used for leverage. That is, it is much cheaper go long or short a particular name with CDS than it is just trading bonds. (Especially short which used to be nearly impossible.)

It is argued that this actually allows for better price discovery more efficiency in the credit markets. But that has be balanced with the fact that the CDS's have become untethered from the actual bonds that they purport to insure, and the tail has begun to wag the dog (as evidenced in the original article.)