Year-end Contest

Mind on Money readers are hereby invited to enter our first-ever year-end contest.

The point of the contest is to come up with a reason for a market anomaly to exist when it shouldn’t.  This is the kind of situation that costs people like me lots of money when trading.  It’s a prime example the old saw that the market can stay irrational longer than you can stay solvent.

I’m talking about the 30-year swap rate.

For the past year, 30-year swaps have been trading at a lower yield than 30-year Treasuries.  When I was new to the bond business and reading the bond market primer “Inside the Yield Book,” it was a given that higher yield is associated with higher risk.  Risk takes many forms, but the truth of this assumption about yield spreads was never questioned.

So what, pray tell, is the risk of 30-year Treasury Bonds the market believes is higher than the risk of 30-year interest rate swaps?

It can’t be currency risk, because both are quoted in US Dollars for all payments, both principal and interest. One thing I’m sure the Treasury will always have is access to Dollars to pay its debts.  I’m not saying anything about what those Dollars will be worth when you get them, but I’m comfortable assuming the Treasury will definitely pay me $1 million US Dollars for my million-dollar Treasury Bond when it comes due.

It can’t be option risk, because neither one is callable or extendable.  Is the market saying the Treasury will decide to force investors to take a ten-year extension when the Long Bond comes due?  Is it saying that the Treasury will force borrowers to take their principal early?  Is the market simultaneously saying that private market counterparties like banks wouldn’t do the same if some strange circumstances brought about these actions?

It can’t be credit risk, at least the way I look at it.  If Treasury Bonds were denominated in Yen or Euros, I might be able to swallow the idea that some bank counterparty could more easily pay the interest and principal than the Treasury.  But these aren’t in another currency.  The Treasury has the printing press, so Dollar debt simply can’t be safer from a credit perspective when it’s issued by some bank or brokerage firm.

That leaves me with only liquidity risk among the Big Four risks in bond world.

Perhaps the market is remembering the October Surprise of 2001, when the Treasury announced they would stop issuing the 30-year maturity.  Of course that announcement came when there was a budget surplus, and the operating cash for the Federal budget did not include as much as $400 billion a year “borrowed off budget,” a fact I suggested at the time meant that Enron was controlling both our energy policy and our Treasury policy.

Still, the staggering size of the swap market for interest rate derivatives gives some credence to this theory.  Put simply, the Treasury doesn’t borrow enough to satisfy the demand for non-callable long term investments, so the swap market has to step in.  The fact of lower rates on swaps says that even that market can’t satisfy the demand.

While not a more liquid market on a daily basis (though nobody knows for sure), we can say that there are far more 30-year swaps than there are 30-year Treasury Bonds.  The liquidity risk idea gets a resounding “maybe.”

Still, if a company or a strong bank or insurance company can get 30-year funding cheaper than the US Government, why aren’t they doing it?  Seems like they’d do it until the yields went up (at least until they couldn’t turn around and buy Treasuries to lock in a 30-year profit).

What about the credit issue?  Given the fact that nobody is willing to take a counterparty risk for 30 years these days in the private market, I am sure virtually all of the 30-year swap market includes collateral posting agreements to maintain credit standards.  So all you have to depend on with your 30-year swaps is that your counterparty will actually post collateral when they have to (the mechanism by which AIG counterparties got so many tens of billions of bailout dollars).

Since I’m not convinced, I’m proposing a contest.

‘Splain it to us dummies:  Why do 30-year Dollar Swaps yield less than 30-year Dollar Treasury Bonds?

Please attach your theory in the form of a comment, or, if you’d like, as a Blowback private e-mail.  Extra points for humor and originality.

First prize, for best rational explanation, is a guest column at MindOnMoney (if you want it).  The best humor/off the wall explanation will also be awarded a guest column, again at the sole discretion of the entrant.


3 Responses to Year-end Contest

  1. Gary says:

    In the short term, there are still loads of structured deals that rely on swaps for functioning (including the early 2000s favorite of hedge funds, swapping 30yr munis into synthetic money markets). The natural swap payors (mortgage hedgers) are basically out of business.

    Long term, there is an “option play” you aren’t considering: while the US Treasury will presumably give you $1 million dollars for your bonds (however worthless those dollars turn out to be) — the same cannot be said for swap obgligations.

    No one does a 30yr swap with itsy bitsy bank — they are (were) done with the big money center banks that are now, directly or indirectly, nationalized.

    After the Chrysler and GM fiasco, no one has any confidence in how the politicians will handle claims if (when!) these banks are recognized as insolvent. Your “senior” claim can be arbitrarily nullified by executive decree, by an inept/corrupt Treasury Secretary, a state attorney general, etc. The swap market is supposed to be between AAA institutions — not between insolvent banks run at the whim of politicians.

    Hence, the liquidity in long dated swaps has really dried up. Its pretty much insolvent bank “A” trading with insolvent bank “B”, using swap spreads determined by the Treasury/Federal Reserve, which owns both banks anyway.

    A pension fund isn’t going to defease long term liabilities using a contract with an insolvent institution. In most states, doing so is a prima facie breach of fiduciary duty.

    Further, if you can defease said liabilities cheaper using actual US Treasuries, why would you take the risk?

    Long dated swaps are essentially illiquid. Yes, I know there is whatever daily volume — but if you subtract out unwinds of earlier trades and interbank trading to get “market priced” trades, the market is essentially dead.

    There are no “true” AAA institutions with which to trade swaps, hence no market

  2. Li says:

    I think it is primarily due to the liquidity risk you described, but there still could be a small credit component.

    The liquidity component is just a market inefficiency. The need of life insurers and pension funds to buy long term assets have not changed much. But the credit spread for even highest rated corporates have widened dramatically in the long end of the curve. This leaves the long bonds the easiest thing to buy. But long term investors can outperform the long bond by buying high quality short to medium term bonds and swap it out to 30 years, e.g., 5 year AAA credit card at L+100bps plus 30 year swap (pay floating receive fixed). Although the asset swap gives back 15 bps (the current negative swap spread level) it still produces T+85 return, PROVIDED the 100 bps spread stays the same. The risk of under performing buying the 30 year bond outright is when the investors roll the 5 year cash bond forward the spread won’t be as high as 100bps. Obviously the AAA cards or other high quality investments in the short to medium term horizon is not the same credit risk as the treasury, but the investors obviously have room in their portfolios to take those credit risks.

    The credit risk component is a bit unconventional. Note the 30 year swap is far different from the 30 year funding cost of AA banks (which is the typical ratings of the banks that make up the BBA LIBOR poll). Rather is the (PVed) average of a string of 3 months borrowing costs of a group of big banks with acceptable credit profile. (The weak banks do get kicked out of the BBA poll.) Therefore if the market is somewhat rational when it priced the 30 year treasury bond at a higher rate than the 30 year swap, maybe it is expecting at some point during the next 30 years the US treasuries 3M borrowing cost will be higher than that of the 3M borrowing cost for the big banks. Obvious it is pretty far fetched since we just had negative rates on the T-bills. But still the ability for the US to borrow in dollars is not unlimited. And it is also conceivable that the government may choose not to print money even if it can print it. So the investor might feel more comfortable in evaluating the credit risk embedded in a 5 year non-government bond every 5 years (and do the swap to lock in the long term risk) than investing for 30 years.

  3. barryzee says:

    OK, Howard here goes. Regardless of what the current CDS rate is on the 30yr UST it is on the entire stream of principal and interest from the bond. The UST credit is for all the flows.

    Now in a swap the functional credit granting and therefore risk is only on the difference between the flows. The credits of the two parties to the swap cancel out for each except for the difference between them.

    The net difference between the flows can vary, depending on the movement of interest rates, varying from a minute difference to a decent size difference, but always a small fraction of the total flows. The vast preponderance of the swap flows are riskless. What is left after the riskless portion is eliminated is so small regardless of the imputed CDS rate the swap will have less risk in absolute dollars than the whole 30yr UST.

    Barry Zee

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: