Wednesday, November 26, 2008

Proof we are not in Kansas any more

Brad Delong thinks that the failure of Citigroup is about “risk premium”.  I think it is about trust – and I guess in some sense these are the same thing.  However it is about trust then no plausible amount of capital solves the problem – whereas in the Brad DeLong model there is a reasonably estimable and plausible quantum of capital that is sufficient.

Anyway today was a deal that proved to me just how strange these markets are. 
 
Goldman Sachs got to issue some FDIC bonds.  These bonds are guaranteed by the FDIC and are hence full faith and credit of the US government.  They were issued at over 200bps of spread.  The spread traded to about 200bps.

That spread makes no sense whatsoever unless you really believe that there is a chance that the US government is going to selectively default.

Once upon a time Long Term Capital Management traded “on-the-run” versus “off-the-run” Treasury bonds, noting that a 29 year bond (which was off the run) had a different yield to the current 30 year bond (which was the reference bond and hence more liquid).  If you went long the illiquid bond and short the liquid bond you might make 16-20bps.  Lever that 100 times and you get quite a nice return on equity.  However if the spread between them (call it a liquidity premium) blows out you can get into some trouble.  

Now we have an FDIC bond with a spread of 200bps or more.  No need to lever 100 times – ten times will make you a super-profitable hedge fund.  And you are leveraging 10 times to term matched US Government obligations which should be less risky than the slightly term mismatched positions taken by LTCM.  

Of course it is a little harder these days to borrow the government securities needed to pull the LTCM stunt.  (There has been a problem with fails in the Treasury market.)  But the margin leaves me gob-smacked.  

If anyone has an explanation other than trust (a belief the US government might default on full faith and credit obligations) or liquidity let me know.  If Brad Delong is right – and it is about capital then this should be able to be levered 5 times as there should be little capital risk.  

But I don’t think it is about capital – it is about something else – something altogether darker and harder to solve.

In the Wizard of Oz all you needed to do was click your red shoes together.  

I have been trying all day!





John Hempton

15 comments:

Mark Cheshire said...

I have been considering buying regular Goldman Sachs bonds, yielding around 9%. The rationale being that the US Government would not dare let GS fail. However if the fully FDIC guaranteed variety provide 2% spreads, perhaps there is a risk too dark to contemplate in the former.

Cheers, Mark

Anonymous said...

These FDIC insured bonds do not qualify for Govy or Agency repo programs. They can only be financed through a corporate repo. Current haircuts on corporates (even highly rated ones) are huge, up to 25% in investment grade land. So even a 10 leverage is going to be difficult with this product. Buying Agencies even without their government guarantee can be far more efficient, as they can be levered far more.

MW said...

"...you are leveraging 10 times to term matched US Government obligations"

See John Jansen's blog: GS issued a 3.5y bond, which you're comparing to a 3y bond! http://acrossthecurve.com/?p=2179

Bakken BEx said...

I don't think people are buying these with yields of 3.5% expecting default. And if the claims paying ability of the US government were in question, why would treasuries be in such demand? It seems the spread would imply a perception that somehow the government could default but then would not get around to paying off the FDIC TLGP bonds until a bit after treasuries. There's just a pulling back of capital from all markets right now, and these bonds, like many many other things currently, just do not make sense. Probably no inherent logic behind their current pricing that implies anything sinister in my opinion.

Unless....
http://www.reuters.com/article/marketsNews/idUSN2446382720081124

Anonymous said...

you know what the to dark to contemplate is...the US falling apart at the seams from unending corruption, and ineptitude of unbelievable proportions.

job loss, revenue loss, massive tax losses, a new socialist administration which will likely spend with little reserve...

choose a prefix...hyper or de
2% would speak to expectations of deflation. I for one think its going to be a very bumpy ride..buckle up

Tim said...

Even after the fed announced they were coming into the CP market, 90 day GE CP was coming in at 390 basis points, a huge yield even though the Fed was effectively saying "we'll buy this up come maturity if no one else does." Granted, not the same as being officially backed by "full faith and credit" but still a big handle.

John Haskell said...

John-
it's called mark to market risk.

Beyond that google "FIRREA" for a trip down memory lane, particularly as it pertains to government changing the rules as they go along (we haven't seen any of that in this cycle though have we?)

David Merkel said...

John, I used to do Credit Tenant Leases for an insurance company, often to agencies of the US Government -- they are illiquid, but secured by the property leased to the government, and I would get an extra 2% over Treasuries. Oh, rental payments were not subject to appropriation -- Congress did not have to approve payment.

Also due to illiquidity, I would get some full faith and credit bonds -- Title XI (shipping), and OPIC (among others)at spreads of 1-2% over Treasuries.

But with the GS and MS bonds, there is a lack of trust in the US Government. What if they walk away from their guarantees? I know, unlikely, and the courts would have something to say, but the US Government has been behaving erratically of late.

Were I back running a life insurer's investment portfolio, I would be scooping some up now as a substitute for Agencies or Treasuries. But if the creditor CUSIP (first six digits) indicates a GS or MS creditor CUSIP, many insurers may be full on the names now, and might not want to add to names where they would have to put up disproportionate capital, or, kick out straight GS or MS debt and take losses.

But I'm with you, these explanations don't go far enough to explain the high yields.

Anonymous said...

Here's my explanation: Aside from a bunch of financiers who run around arbitraging small spreads and getting their fingers burnt on a regular basis, most people take into account things like:

The name on the bond
The length of time this particular type of bond has been trading
etc.

Now that deleveraging has stripped the market of arbitrageurs, we are back to the world of thirty years ago where real money investors are looking to buy and hold. These guys move more slowly and if they miss a good deal, well at least they didn't lose money by rushing.

In short, I think we are in Kansas -- and prices will be set on Kansas time. I bet you'll see some convergence in six months and maybe even in six weeks.

Blank Xavier said...

In my entirely lightweight opinion, I think it's fear of Government. Government has been so variable and is so unpredictable, you can have no faith in the future, because you have no idea what they're going to pull on you next.

Tim said...

Is the reason for the spread that hedgers can no longer hedge? The price is based on the coupon the buying public will accept? No more massive leverage to wring out a few basis points of mispricing? If you try it the margin call wipes you out the next time the market goes stupid.

williamdb said...

Here's my take:

Once upon a time, up until about 20 years ago, investments were judged by their returns without leverage.

Over the last 20 years an excessive amount of investable capital led to diminishing returns. Investment rules changed and leverage became completely embedded in ROI calculations.
Your analysis ('I can leverage by 10, or even by 5 and make decent money') betrays this.

I believe the investment community is starting to recognise the fact that leverage ads to investment risk, and that acceptable investments whatever they may be, are those that offer 'decent money' without leverage!

It follows that any investment that requires leverage to make an investor decent money must be a bad one!! Now go figure that!

I suspect this must be absolutely shocking to a lot of people, particularly those who started their careers in the 90s or later.

In my opinion this shift is taking place right now. And I suspect no matter how much money Ben pumps into the system to keep asset prices inflated, the 'old' investment rules are likely to prevail again.

Thanks for your blog BTW, it's a gem.

John Hempton said...

I appreciate the comments here - and apologise for taking so long to moderate them - but I had a good sleep in Australia.

The comparison between 3 and 3.5 year treasuries was a mistake - but the issuance was still at more than 200bps of spread which remains strange.

Some people (other blogs, this one) have noted that there are plenty of Full Faith and Credit (FFC) obligations that trade at spreads though 200bps is indecently large.

The general view however appears to be trust in the US Government.

Of course the US Government can solve this by buying the FDIC insured bonds themselves and selling Treasuries. There is 200bps of arbitrage there - and if that is a free lunch I do not see why the taxpayers shouldn't have it.

Of course the Treasuries they issue will go on the national debt - whereas the FDIC obligations are obligations of Goldies and do not. The government seems indecently concerned about government accounting and not about the economics. If the government were concerned about the economics then the Treasury, and hence the taxpayer, would be enjoying the free lunch.


John H

cgaros said...

I'll agree with those who blame the spread on uncertainty about future government actions. Let's say Obama & Co. comes in next year and says that some of the recent financial decisions should have been subject to Congressional approval/court review/more oversight, or simply starts revoking dumb decisions made in haste by the previous administration.

The current administration has said these are "full faith and credit" obligations, but there are lots of others in the government who could say they didn't follow the correct process for taking on a government obligation. It's become apparent that the rules are whatever those in power say they are and change daily.

Treasuries have enough tradition and settled law behind them that they aren't going to default outside of a total meltdown that destroys all investments, so the yields stay down. The DGP is a very new FDIC program coming at a time when the FDIC is already stretched thin by bank failures and the increased deposit guarantee. The FDIC has no funding mechanism other than charging fees to banks (good luck) and begging Treasury for help. What happens if Treasury refuses to bail out the FDIC next year? I don't know, and I assume the spread is due to the fact that no one else does either.

Anonymous said...

Could the wide-to-agencies spread on the new GS FDIC-backed bond just be simple market window dressing? The dealers (and govt) want blowout demand so they can claim the program was a success, and continue to sell tons of this as cheap money for investment banks. So they have the first GS and MS bonds come cheap to agencies so they can get the deals upsized (which they did: GS went from $2.5 B to $5 B).

General disclaimer

The content contained in this blog represents the opinions of Mr. Hempton. Mr. Hempton may hold either long or short positions in securities of various companies discussed in the blog based upon Mr. Hempton's recommendations. The commentary in this blog in no way constitutes a solicitation of business or investment advice. In fact, it should not be relied upon in making investment decisions, ever. It is intended solely for the entertainment of the reader, and the author.  In particular this blog is not directed for investment purposes at US Persons.