The On-the-Run Premium on Treasury Securities

Felix Salmon argues, contra Paul Krugman:

The TED Spread and the Flight to Liquidity - Finance Blog - Felix Salmon - Market Movers - Portfolio.com: Well, yes, up to a point. But "liquidity issues" includes a hell of a lot more than just credit risk, otherwise there wouldn't be any spread between on-the-run and off-the-run Treasuries. (Which spread, as I recall, was in large part responsible for the implosion of LTCM)...

Here is what he is talking about:

The on-the-run U.S. Treasury bond is the bond issued at the most recent auction. The off-the-run bonds are the bonds issued at all the other auctions that are still outstanding. The on-the-run bond has a higher price--a lower yield. The yield-to-maturity of the on-the-run bond averages some ten basis points--some one-tenth of a percentage point--lower. If you are thinking of buying-and-holding the ten-year on-the-run Treasury to maturity, you can buy-and-hold the off-the-run and then rollover your money for the few months until the on-the-run matures, and wind up on average some eight-tenths of a percent richer: the on-the-run bond costs an average of eight-tenths of a percent more than it "should."

The magnitude of this on-the-run premium is made even more awkward for those of us who worship at the altar of the efficient market hypothesis by the short duration of the premium. Every three months the on-the-run Treasury is replaced by a new issue at a new auction. This suggests a trading strategy: (1) sell the on-the-run issue short, (2) buy the first off-the-run issue with the proceeds, (3) hold for three months, (4) liquidate, and (5) repeat. This strategy would appear to promise an average pre-expenses return of three percent of your gross position a year with (a) very little initial capital outlay (your short sale pays for the long leg), (b) very little risk (fluctuations in the on-the-run premium are the only source of risk), (c) low administrative costs, and (d) low transaction costs (the bid-ask spread on the ten-year Treasury is typically 1/32 of a percent of value, less than 1/20 of the on-the-run premium.

So why is this money left lying on the sidewalk? What risks and costs shut down this trading strategy that are not obvious to me?

Whenever I talk to people on the Street or to academics (see, for example, Krishnamurthy (2002) http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6VBX-473MBMS-C&_user=4420&_rdoc=1&_fmt=&_orig=search&_sort=d&view=c&_acct=C000059607&_version=1&_urlVersion=0&_userid=4420&md5=7a7a89cd60f0a2920b1c38c67d5ce1dd) I seem to understand their explanations of the on-the-run premium while they are making them, but afterwards I am once again dazed and confused. Explanations in terms of "special repo" costs seem inadequate--that they are not causes but rather consequences of the off-the=run premium.

At the very least the U.S. Treasury should be making a pretty penny off this premium, for it can construct out of its own bond inventory synthetic securities with exactly the same payment characteristics down to the day and the penny as the on-the-run security: it can go short the on-the-run with no risk at all whatsoever.

---

Note: In journalism speak, "up to a point" means "no." In Evelyn Waugh's novel Scoop, there were only two things you could say to Lord Copper, proprietor of the daily Beast: "definitely, Lord Copper," or "up to a point, Lord Copper."

Comments

[Bruce Webb]

Question from the unwashed singletude.

Does the premium account for transaction costs? I know maybe less than nothing about either the auction (on the run) or the bond market (off the run) but it would seem that there would seem to be more economic friction in the latter than the former, i.e. commissions and fees. Unless they actually get included in the 'bid-ask' spread you cite.

Also while your five step process could be automated to some degree, it still requires more attention and so more staff than a policy of buy and hold. In the context of multi-billion dollar investments (say via the Chinese national bank) an extra account manager here or there wouldn't rise to the level of a rounding error but in an operation where the investments were in the millions and perhaps split across multiple clients the pure office costs might rise to the level of significance.

I don't know the answers but I did come out of a governmental organization where the cost of the person in the seat could well be dwarfed by the cost of the seat (software licenses, space costs, energy costs). Which is to say the answer to your question may not lay with the guys in finance but instead with the office manager.

[n]

dunno - first time i've looked at it. maybe 3% does not get people excited - there are less risky ways to get 3%. You would need collateral to cover for the time when the return is negative (there would be opportunity cost for collateral).

What about short treasuries and long muni bonds?

[Colin Danby]

has to be something about the way the on-the-runs are taken up -- what are the institutional differences in the buyers? The spread is remarkably durable!

Can't be the arguments thus far. If there's any asset mkt with low transactions costs it should be this one. People arbitrage stuff a lot more complex than this & plenty of folks are already set up to do these trades. N, I think you're ignoring the "gross position" part.

[Matt]

Great, homework!

Here is my answer, I will be Hayek-like.

What path do I take to implement this strategy?

Hey, the bond trader can do this during work hours with his own money!
Almost, but he might lose his job.

OK, I hire the bond trader. Well, I have an entry fee, and probably a minimum transaction size/rate restriction. Almost.

And on down the ladder, until I discover that the bond market delivers, oh my god, almost the same 3% return as any other long term stable business in the economy

[3% on the *gross* position, not 3% on the invested capital. The expected return on capital depends on the degree to which you dare leverage yourself up, and is equal to 3% times your leverage ratio plus the 3% safe bond rate (which is where you keep your capital reserves).

The variance... is the variance...]

, and that return, by efficient market hypothesis should approach the aggregate productivity level.

The atomicist in me says that any atomic configuration of the economy will have some commonalities, like the periodic chart, and one of these is the pervasive use of money technology that this industry always sits at the same quantum location, about 3/4 of the way down the right, on X, and the wealth function will have moderate cyclic variation of something like 1/3 of the state spectral line. And so on.

[David Merkel]

Brad, if you are a bond manager for a major institution, you are often asked when doing a trade, "Do you want to do Treasuries against that?" The two acceptable answers are:

Yes, (I don't have the cash in hand for the trade yet) and
No, just spotting (I have the cash, let's agree on the price/yield of relevant Treasury Note that this bond is spread off of.)

"Doing Treasuries" means taking a short/long position in the on-the-run Treasury that the bond being bought/sold is spread off of. (The spread is the difference in yield between the bond and the Treasury.)

This puts a lot of pressure on the on-the-runs. It places them in hot demand to be controlled, so that they can be easily used for lubricating the purchase and sale of other bonds.

Personally, if I were the Treasury, I would try to take advantage of this, and issue more notes/bonds when they trade cheap/special. There are a number of bond firms that try to make money off this phenomenon, but it is not a sure thing -- it is not always easy to hang on to a short position in the bond market, thus making the arbitrage tough.

If you want to talk more, just e-mail me at the web address at my blog.

[Arb Master]

You should also refer to Andrei Shleifer's book Inefficient Markets (2000), esp. Chs. 2 and 4, and Shleifer and Vishny's article "The Limits of Arbitrage", Journal of Finance 52, 35-55 for a discussion of risk adjusted returns for arbitrage in markets with noise traders. Noise traders are understood to be more prevalent in new issue markets, which is one of the reasons the securities laws continue to make a distinction between seasoned and unseasoned issues.

[john jansen]

With all due respect your initial set of facts is incorrect.There is no 10 basis point on run/off run curve in the market currently and while it sertainly happens it is unusual.I am taking these from Bloomberg at 850AM on a Sunday morning in New York at think they are correct. i did miss the Friday afternoon bond close so I am trusting the good folks at Bloomberg.

At the close on Friday the off the run 2 year note yielded 2.124 percent and the on the run 2 year traded at a pickup in yield as it yielded 2.134 percent. In fact when the on the run issue traded as a WI bond it was as much as 3 basis points cheap.

The off the run 5 year closed at a yield of 2.901 and the on the run 5 year finished at a slight premium at 2.894 percent.

The off the run 10 year yield s 3.711 versus 3.708 for the on the run 10 year.

The off the run long bond yields 4.487 and the on the run bond yields 4.50 percent.

[gorobei]

John Jansen's numbers seem closer to reality than the graph.

[Ummm... The graph, it may have escaped your attention, shows the pattern of the on-the-run off-the-run spread for ten-year Treasuries since *1961,* not a snapshot of the current moment...]

But in any case, the following seem to hold true:

1. If two things are not identical, they may trade with a spread.
2. On and off the run are not identical, even if their payouts are. Due to the market, the on-the-run is slightly easier to trade, ans so commands a slight premium.
3. The spread is small, so most people have to lever aggressively to make the trade attractive.
4. The spread is non-constant, so trading it should have the risk of capital calls or margin requirements on a levered position.

So, to make the trade, you either need a lot of cash on hand, or a big line of credit from some bank. As such, the trade falls into the broad class of "seling liquidity." Certainly profitable, but you'd better have the cash when the market gets fearful.

[bill]

If I remember correctly, didn't LTCM try that exact strategy as part of their fund? And it blew up on them. The spread got wider and wider until their capital was gone. I think.

[fe]

If I recall correctly, the unprecedented expansion of the swap spread and the blowup of one large merger arbitrage trade were more likely the proximate causes for LTCM's problems. I don't doubt that the on/off the run spread increased during this time, likely as a result of LTCM, but I haven't heard anyone argue that it was causal. Somebody who has actual facts, please correct me if I'm wrong.

[gorobei]

I'll admit I only looked at the last ten years or so of the chart. The numbers seemed higher and more volatile that I recalled seeing during that period. Although, perhaps, this was due to sampling bias on my part: the only days when I had time to go off and look at bond arbitrarge strategies were days when the markets were particularly quiet.

Accepting the correctness of the graph, it's pretty clear what the risks are to a hedge fund. E.g. take that spike in 2003 from 15bp to 40bp. For a target 15% return, the fund has to lever around 80 times on the spread trade. A 25bp move is a 20% loss, and you're forced to unwind positions thus widening the spread even more. Either you successfully call a lot of capital from your investors, or your deeper-pocketed creditors just make you an offer you can't refuse.

[dave]

Repo issues cannot be dismissed lightly. Financing that strategy is far more complication than it looks. Further the treasury market has been hit by market collusion, large players and BD's taking huge positions in an issue and refusing to lend in a blatant attempt to manipulate. If the off the run market ever goes electronic part of the prem will disappear overnight.

[James Bond]

The premium is 100% due to repo specialness, and squeezes perpetrated by the primary dealers. Back in 2006 (or very early 2007, don't have a link to the news item), there was an investigation by the Treasury into the squeezes - and the premiums of the on-the-runs dropped like a rock.

If you have to finance a short position in the on-the-run, some times it can cost you 300 bps carry on the repo financing. Even though this cost would only be run for a short period, the high magnitude matches the premium spread over the life of the bond.

Other than shutting down the squeezes by the dealers (which they largely did), there's not much the Treasury can do. (The Fed does help as well by lending out its inventory.) Remember that when the Treasury issues the debt, they're shorting 100% of the on-the-run's outstanding size! There's limits to what a bond issuer can do to guide trading in the secondary market of its debt.

[john jansen]

If you have to finance a short position in the on-the-run, some times it can cost you 300 bps carry on the repo financing. Even though this cost would only be run for a short period, the high magnitude matches the premium spread over the life of the bond.

That is from the comment of one James Bond. I think the reason that at the current time there can be no huge on the run/off the run premium is that there cant be a huge repo special with the funds rate this low. If repo is 2percent then the special can only be 200 basis points. When repo was 5 percent or 6 percent the special could be 500 or 600 basis points.

What does that mean in the real world? If the dealer is long the off the run he finances it at the general collateral rate of whatever. Lets say 6.00 for the sake of argument. Lets assume the bond pays 7.00 percent. He earns 100 basis points of positive carry.

On the other side he needs to borrow the on the run which he is short.Assume it has a 7 percent coupon also. The person who owns the bonds says to the short who needs to borrow the bond,"I will lend this security to you but I want to pay you just one percent interest on the money".So the dealer is losing the 7 percent coupon accrual every day and earning just one percent on his money lent. That is negative carry of 600 basis points. so one leg of the trade loses 600 basis points and the other leg makes just 100 basis points. You cant do that for too long and stay in biz. A very hypothetical example and a little exaggerated to make the point but you can see the problem.

So the reason that on the run/off the run cant go too crazy now is that with repo at 2 percent the specialness can never be more than 200 basis ponts.

Seprately, the spread can be effeted by supply considerations. Hot runs will begin to suffer as on the run supply ramps up to pay for guns and butter( I am having a flashback.) But this month the Treasury will sell about $30billion 2year notes and $20billion 5 year notes. As recently as last September that was $18billion and $13 billion respectively.