The New Normal

I sometimes hear people at rating agencies talking about when credit conditions “get back to normal”. Do any of us really know what “normal” means, though?

The issue here is that investor incentives are simply not what they used to be. For example, in the good old days, a bond holder wanted to get repaid. Nowadays, the bond holder can go out and enter into a credit default swap (hereafter, a “CDS”), and offset (or maybe more than offset) any losses on his bonds just in case the bond issuer goes bankrupt. Needless to say, this state of affairs completely changes the bond holder’s incentives. And changed investor incentives will sooner or later change the very structure of whatever market the investors happen to be in. The important thing about that is that when the structure of the capital markets change in a fundamental way, nothing “gets back to normal” ever again. “Normal” is now something completely new.

Let me explain what I see as the new “normal”, using the following hypothetical. Suppose I have about $1,000,000 that I have to invest. One thing I could do is that I can buy a “risk free” 10-year U.S. Treasury that will yield 2%. I have another choice of investment, which is that I can buy a 10-year senior secured General Electric (GE) bond that yields 10%. GE is riskier than the United States government, and hence the extra yield on the GE bond. So far, sounds pretty “normal”. Let’s keep going.

Let’s suppose that I can buy a CDS insuring $1,000,000 worth of notional principal on GE debt, for $10,000 – call it a 1% spread. Setting aside my counterparty risk on the CDS, if I buy $990,000 worth of GE bonds and $10,000 worth of CDS, I have just turned my GE bonds into “risk-free” bonds because if GE defaults, I make $1,000,000 back on the CDS. Now here is where it gets interesting. See, Treasuries pay me 2%, but after I back out the cost of my CDS, the GE stuff is paying around 9%. Get it? Same risk on the GE bonds and CDS investment as I would have on an investment in U.S. Treasuries, but I pocket an extra 7% owning GE bonds insured with a CDS. Nice arbitrage opportunity.

When most of us think of “normal”, we think that when the market gives you an opportunity to make money risk-free, that opportunity will get arbitraged out of the market quick. Real quick. Investors will just buy up GE bonds, driving down the yield, and will buy up CDS, driving up the spread, until the risk-free US Treasury yield is going to be almost exactly the same yield as a GE bond + a CDS.

But that is not what is happening. There is a dichotomous force at play, which is that in today's “normal” market, when the price of a CDS goes up, that “normally” signals to investors that the creditworthiness of the bond issuer is going down. And when the creditworthiness of a bond issuer goes down, lots of investors sell the issuer’s bonds, and that, in turn, means that the yield on those bonds goes up.

See, what we have here is that we have this arbitrage opportunity (the one I describe where I make 7% extra on GE bonds hedged with CDS compared with what I’d make on US Treasuries), but instead of this state of affairs driving up the demand for GE bonds, it’s actually driving the demand for those bonds DOWN, because people see that increasing CDS spread and they freak out and say “oh golly, GE is going bankrupt, time to dump GE bonds!” And in so doing, the yield on those bonds goes way up (let’s say, hypothetically to 15%), and now the risk arbitrage opportunity is even larger.

Well, okay. Maybe yes, maybe no. Maybe, as the yield on GE bonds in this example pops to 15%, the spread on the CDS goes way up also – assume, hypothetically, way up to 8%. So, a risk arbitrage guy maybe isn’t making an even larger spread, but for those with a somewhat longer term strategy, this is a case of spending a little bit extra to get a lot extra. These chaps actually WANT to pay more – a lot more – to buy CDS insurance policies if by doing so, those increasing spreads will make some pension funds dump their GE bonds like a pack of lemmings diving off a cliff.

We've all seen how that works. Welcome to the new “normal.”

Who can say how it will play out – I’m not here to offer predictions. My only point is that the incentives and opportunities today are different from yesterday, and so are the rules. To make heads or tails out of why GE just lost its top credit rating, or why Berkshire Hathaway (BRK.A) just lost its own, an investor needs to consider that it may have nothing much to do with these companies’ balance sheets or the economy. And don’t let some analyst at a rating agency (stellar track record that these guys have when it comes to pricing risk) tell you any different.

As I said earlier, when incentives change, rules change, and when the rules change, the structure changes. We're at the stage where the structure is just starting to shift, and it's tough to really see any place for rating agencies.

Disclosures: The author owns shares of General Electric and Berkshire Hathaway.

Comments

[morph366:]

I don't quite follow the alleged arbitrage opportunity you have cited in your illustration.
Buying a CDS would protect you from default on the GE Bonds (perhaps even on a Treasury issue if you so desired - although that raises other issues). But what it won't protect you against is a decline in the price of the GE bond i.e. when you want to sell it in the secondary market prior to maturity. The reason why the yields on bonds of non-sovereign issuers like GE may keep going up is because traders in the secondary market may still want to see lower prices on these instruments. This has to do with perceptions of credit quality, market risk etc.
You would need more than a CDS to hedge that risk - so where does the arbitrage opportunity arise?

Mar 15 08:06 AM

[User 366136:]

There is another credit plat at hand isnt there: if you bought the CDS from some other counterparty, dont you take their credit risk on the ability to pay if GE fails? Isnt that the whole problem with AIG, they sold a ton of CDS and now cant pay in the events of default?

Mar 15 08:53 AM

[manya05:]

Good article. Thanks! What is not clear to me in your example is this: once they are issued, the supply of GE bonds is finite, but the supply of CDS for the GE bonds is potentially infinite, right? as long as somebody is willing to underwrite the risk and take my money for it...so, doesn't it sound to you like a good opportunity to manipulate a market? Again, an honest question out of ignorance. Thanks again.

Mar 15 09:03 AM

[morph366:]

Re User 366136
AIG can't pay but the US taxpayer can.

Mar 15 09:08 AM

[dawase@gmai...:]

Possible outcomes of the hypothetical:

1. No bond default
2. Bond issuer defaults, CDS issuer pays par
3. Bond issuer defaults, CDS issuer defaults

You would effectively need a CDS-squared against the counterparty that issued the CDS in the first place. Otherwise you're exposed to the possible outcome path of 3. Bond issuer defaults and CDS issuer defaults.

I leave it as an exercise for the reader to determine whether P(2) > P(3)

Mar 15 09:16 AM

[Alpha Econ:]

Makes sense to me. The only argument against this arbitrage is the risk of a downside to capital gains if you are forced to resell prior to maturity. But the same risk applies to equities. So even assuming risk premium on, say GE equity over bonds goes to zero (an extremely unlikely assumptions that disfavours bonds), you have equivalent capital losses/gains in two asset classes, while positive yield differential on bonds compared to shares. Unless, that is, there is some return to a strongly positive dividend yield curve in any time in the near future...

Mar 15 09:33 AM

[a. palmer jr.:]

If things get back to "normal" it should go back to how it was in the 50's and early 60's (that's as far back as I remember) when interest rates on savings was 5.25 percent for several years on end. CDs were insured and at 7 percent or so. You couldn't get loans just by signing your name; they wanted either collateral, a big down payment or both, plus you had to show the ability to repay the loan. They actually were concerned about getting their money back!! I didn't play the stocks back then, didn't need to. Of course gasoline was .26 a gallon also. My house cost $7500 and wasn't a fixer-upper.
My version of "normal" won't come about ever, barring a nuclear war or something, then we'll have a new kind of non-normalcy to worry about.

Mar 15 09:55 AM

[{5U}:]

I've been around the CDS market. It's the worst thing ever.

It dwarfs the actual cash market and its that leverage that ratcheted credit spreads so tight and also allowed the financial system to get torn apart.

The mentality of CDS traders were to get as big as possible. It was a game of balance sheet warfare where trades could only show a nickel of risk on the balance sheet for each $1 of real risk they took.

Fortunately, this will all change. The government induced housing derby was like the introduction of cocaine and the CDS market is like the invention of the crack.

Mar 15 10:39 AM

[@TexasER:]

Sometimes “experts” are so close to the details they can’t see the forest for the trees. And it seems they are always drinking the same kool-aid.

These swaps didn’t used to exist at all, and markets functioned just fine without them. It used to be a bank’s job to assess risk, loan money, assume the risk and then profit if they do it all well.

But here’s the CDS world. Banks assess risk, loan money, outsource that pesky risk part, and then of course profit. Of course why bother doing a good job on that risk assessment part, it's not OUR risk.

Swaps are the pyramid-shaped, aluminum foil hats that protect you from risk. And monsters.

Mar 15 11:35 AM

[367851:]

Maybe that means the taxpayer won't be able to pay either, or maybe the spread will be inflated away.


On Mar 15 09:08 AM morph366 wrote:

> Re User 366136
> AIG can't pay but the US taxpayer can.

Mar 15 11:36 AM

[manya05:]

a. palmer jr., what you describe as "normal" is not too far from how it is in Europe. They pay you interest when you save (not much, but they pay). The bank provides a service when you move money around, so they charge you a monthly fee for checking accounts (a hefty fee). Credit cards are rare, and if you get one, credit limits are puny. Most people carry a debit card, and if the bank extends credit on it...it is just until the end of the month and for a very low limit. You want to buy a house? 20% down minimum, and if you are over 50, 30% down and forget a 30 year loan, it is 20 year, 25 if you smile. This is not rocket science...the US got in the mess it got into because of the sheer incompetence of its banksters in collusion with their bought-out politicians. Let's be realistic, it is a corrupt system.

Mar 15 12:07 PM

[LJR:]

It seems that normalcy is a state of confusion, and it gets more so every day.

Mar 15 01:39 PM

[John Lounsbury:]

dawase - - -

A comment on your thought about CDS squared. It is not likely to be a successful strategy. CDS's can be like dominos: when one falls it starts a cascade.

That is why Lehman was such a shock when it failed. Lehman was a small CDS player (I believe about $3-4B, but with its failure trillions of interrelated CDS arrangements came into question. The shock waves were carastrophic.

In fact CDS squared (or cubed or more) may be part of the tangled web that no one can figure out.

Mar 15 01:47 PM

[vaultvix:]

Not to be cynical...but I get the point. I have seen it made before, and there is validity to it.

Markets work. If the government (which is the biggest player) wants to intervene, it can. In 1998, the Hong Kong Monetary Authority decided that it wanted to "discourage" speculators in the stock market (it had spent US$1 billion defending its currency--successfully... but shorting other financial assets could have the same effect--it intervened and bought stocks directly. People who were short stock lost money, and, several years later (governments are the ultimate long term investors), the HKMA had made US$1bn on the trade.

The US government could step in and write CDSs on any companies it felt were under attack from this assymetric situation where the supply of bonds is fixed but the supply of CDS is potentially infinite. It just has to write the CDS all day. Underbid the current writers. It might lose on some--BRK might default, but how likely is that? Citi might default, but how likely is that? In the latter case, the government would basically be writing credit default insurance against itself. But from the buyer's perspective, what better counterparty than to be buying from the government? It can just raise revenues to cover its liability.

Here is where the cynical part comes in. Who in the government would do this? The only person who would propose it is someone who knows how these things work. But those people have all been told, "No, please do not apply for jobs with the Treasury because we don't want people from Wall Street."

As you like. Just keep drip feeding.

Mar 15 04:18 PM

[shorty735261:]

Your whole article is based on a false premise, and is wrong. Credit default swaps on GE are 8x more expensive than your example shows. I.e., they are priced correctly and there is no opportunity for arbitrage vs. Treasuries.

www.reuters.com/articl...

Mar 15 06:46 PM

[Alex Trias:]

I'm assuming a hold to maturity scenario, although I should have made that assumption more explicit in the article. And the arbitrage I refer to in the hypothetical is the hypothetical investor gets a lower yeild on an investment in a "risk free" US Treasury than he gets on an investment in a GE bond + a CDS. My point is that since both investments are equal in terms of default risk, both should have the same yeild to maturity, and if they don't, the investor can gets what I call a risk-free arbitrage by investing in the GE bond + CDS.

What I wanted to write about more specifically, which may go to your point if I understand where you're going when you ask "where's the arbitrage" is this. If I see two investments with identical default risk, but one has a high yeild, and the other has a low yeild, what I can do from an arbitrage standpoint is to buy the cheap risk (the one with the higher yeild) and simultaneously sell the expensive risk (the one with the low yeild. Using the hypothetical investments in the example I gave, I suppose you could give a simplified example where the investor sells a bunch of US Treasuries short, and then takes the sales proceeds and invests in GE bonds and a CDS insuring GE debt.

As you know, there are some real practical constraints with that specific trade, which is why I left out any discussion of it.

Does that answer your question? If not, please let me know and I'll see if I can't flush this out a bit more.

Thanks for taking the time to read and comment on the article.

- Alex

On Mar 15 08:06 AM morph366 wrote:

> I don't quite follow the alleged arbitrage opportunity you have cited
> in your illustration.
> Buying a CDS would protect you from default on the GE Bonds (perhaps
> even on a Treasury issue if you so desired - although that raises
> other issues). But what it won't protect you against is a decline
> in the price of the GE bond i.e. when you want to sell it in the
> secondary market prior to maturity. The reason why the yields on
> bonds of non-sovereign issuers like GE may keep going up is because
> traders in the secondary market may still want to see lower prices
> on these instruments. This has to do with perceptions of credit quality,
> market risk etc.
> You would need more than a CDS to hedge that risk - so where does
> the arbitrage opportunity arise?

Mar 17 02:31 PM

[Alex Trias:]

Thanks for the link to the Reuters article, which very much supports the exact point I am trying to make - the pricing of CDS can be appropriately explained not in terms of a company's ability to avoid defaulting on it's debt, but rather, as a function of arbitrage opportunities that would otherwise result if the CDS were priced lower. In case it was not clear, I used a hypothetical scenario as the basis for this article for that reason.

Thanks for taking the time to comment and for pointing me towards the reuters' article.


On Mar 15 06:46 PM shorty735261 wrote:

> Your whole article is based on a false premise, and is wrong. Credit
> default swaps on GE are 8x more expensive than your example shows.
> I.e., they are priced correctly and there is no opportunity for arbitrage
> vs. Treasuries.
>
> www.reuters.com/articl...

Mar 17 02:38 PM