I found Tom's observation about higher rates due to economic improvement and their likely effects on HY bond yields to be very interesting. I think it was him that made the point recently that high yield bonds don't correlate to treasuries. It was those comments that got me interested in buying HYG and JNK for yield plays..............................................Xot
The most common topic of the e-mails I get from RealMoney readers
is how to best short the bond market. This is somewhat ironic, as for
most of the last three years, I've been predicting rates would fall. But
now, I am finally going to tell you how to best short Treasury bonds.
Hint: It isn't the ProShares UIS 20+ Treasury Trust (TBT - commentary - Trade Now).
Also this week: what to take away from the jobs number, corporate bonds
on fire as portfolio managers give in to the interest-rate reality, MBS
recovering and weak-handed muni shorts getting exposed.
Dallas Fed President Richard Fisher asked in an Oct. 7 speech: "The
vexing question is: Why isn't this liquidity being utilized to hire new
workers and reduce unemployment?" Indeed! Although today's employment
number produced some jobs, it remains far below the level needed to
bring the unemployment rate down. Fisher goes on to question why the Fed
should keep pumping in liquidity if it doesn't seem to be producing
I'm supportive of additional quantitative easing, and I will not rehash the argument in this space (see here for details). However, readers should bear in mind that the goal of QE2 is not to create additional jobs per se.
It is well established that employment lags economic growth. Therefore,
any policy decision based on employment statistics would necessarily
lag reality and, thus, be ineffective and potentially dangerous.
Looking at the employment picture as reported Friday morning, one
should look past the headlines and examine whether the statistics
indicate that the capacity of existing laborers is well utilized. In
other words, employers are only likely to hire more if their existing
labor force does not have capacity to produce more as it is. Average
hours worked is one indicator of this. So is productivity. We've seen
average hours improve from a low of 33.7 to 34.2 hours a week. Still,
this figure has been basically flat for four months now, which isn't a
good sign. Pre-recession average hours were running at 34.5-34.8.
Productivity did decline last quarter, which tends to be an indicator
that existing employees cannot produce more on their own. However, that
broke a streak of five straight quarters of huge gains, and I'm not
terribly confident that the Q2 decline wasn't just an anomaly.
This is why we are likely to see a mild
acceleration of job gains in 2011 to a monthly average around 150,000.
Unfortunately, this will remain much too low to bring unemployment down
materially and will certainly have no bearing on whether the Fed moves
forward with additional policy measures.
Treasury bonds: Eat my shorts
Long-time readers of my work probably think me a consummate
bond-market bull. Not true. It is just that for the last three years,
there has been overwhelmingly bearish sentiment in bonds. Particularly
in the post-crisis period, I've been arguing not that bond market
valuations are attractive, but more that the risks were, at worst,
evenly balanced. All the supply and demand indicators told me that rates
should be extremely low (not enough bonds) and, therefore, the only
bearish argument for bonds was that rates were already low enough. To
me, that was always a poor argument to justify selling bonds and holding
cash, or, even worse, taking an outright short position.
Despite that, I am constantly getting e-mail asking if now is the
time to short bonds. To this I say, yes. In a sense, I've shorted bonds
in my professional portfolios, adding aggressively to the positions in
But hold on there, my bond-bearish friends. Don't go out and buy
TBT just yet. Let's consider the problems with an outright bond short.
First, when you are short bonds, you have to pay the carry. Therefore,
every day you are wrong about rates rising, it costs you not only in
price, but directly in terms of interest income. And once rates do rise,
you don't get back the interest you've paid out, so your upside is
always going to be less than your downside in any bond short. And timing
In addition, the "not enough bonds" argument will probably take a
long time to reverse. Fear has driven investors into holding more
bonds. That fear has hardly dissipated, despite being more than two
years out from the depths of the financial crisis. On the supply side,
not only are consumers and corporations borrowing less, but even
Treasury issuance is slowing down. Neither supply nor demand seems
poised to turn in a bond-bearish way anytime soon. Perhaps worse, you
have the Fed and its limitless printing press standing in the shadows,
just waiting to buy more Treasuries. Trying to stand in the way of that
locomotive seems awfully dangerous to me.
So, how to realize an effective bond short without
taking on the problems of a bond short? Consider that interest rates
can rise for two reasons. First, real interest rates can rise. This
occurs when investment opportunities improve, which, in turn, is a
function of economic growth. Second, inflation can accelerate. Now,
consider that you can make money by taking long positions in the right
bonds given either scenario. If real rates are rising and the economy is
improving, then it is likely that high-yield bond spreads will tighten.
In fact, I'd argue that they will tighten by at least as much as rate
rise. And, oh, by the way, you will earn around 8% in yield while you
If inflation increases, then TIPS will perform well. Granted, you
aren't earning much while you wait -- currently around a 0.5% nominal
yield on 10-year TIPS. But at least that is positive carry, as opposed
to a Treasury short with negative carry.
Perhaps most importantly, long positions in high-yield bonds and
TIPS are not negatively exposed should the Fed announce a very large
asset-purchase program. In fact, both have had a very nice run since the
beginning of September as it became more and more apparent that QE2 was
Now, I know neither of these trades will satisfy the
dollar-devaluation or America-equals-Greece crowd. Those people should
just hang on to their Treasury shorts and assume all the money they lose
is a sacrifice to the goddess of austerity. I'm going to go with the
high-probability trades, because, you know, I actually want to make
money more than I want to prove a point.
Corporate bonds: Capitulation
Corporate bonds had a very strong week, led by bank and finance
bonds, which were anywhere from 10-20bp tighter. We obviously had a big
up day in stocks on Tuesday, but even Wednesday and Thursday, while
stocks were flailing around, near unchanged, corporate bonds were
What you are seeing is a capitulation by
real-money investors. We've seen this exact trade many times over the
past two years. Real money has been fighting falling rates that whole
time, at times refusing to buy anything under the assumption that rates
are about to rise. Then when they don't, it becomes a scramble to get
cash to work. The imminent implementation of QE is just today's
particular excuse. But the underlying reason for corporate-bond strength
is the huge flows into mutual funds. This past week, $2.3 billion
flowed into taxable bond funds, according to AMG. As long as this trend
keeps up, there will be a strong bid for corporate bonds.
MBS: Conditional surrender
When I say "conditional surrender", I mean not quite a
capitulation, but related. MBS had a pretty strong week, with Fannie Mae
4% 30-year MBS up three-quarters of a point versus just half a point
for the five-year Treasury. Some of the factors benefiting corporates
are also benefiting MBS: As flows move toward bonds, it is a rising tide
that lifts all boats. But the trouble with MBS is that you are very
exposed to a Fed-created rally in rates. Thirty-year MBS with a 4%
coupon typically have underlying borrowing rates of 4.5-4.75%. As of
Oct. 7, Freddie Mac reports the average 30-year borrowing rates to be
4.25%. So, the 30-year 4% MBS is fairly close to being in the money. As
such, further rallies in rates won't benefit the price of the MBS, as
the market will assume that the borrowers will refinance. I wrote two
weeks ago that I thought MBS were oversold, but fundamentally, this
isn't a sector I like. We're max underweight.
Municipals: False signals
Municipals had a pretty poor week. Ten-year AAA muni rates fell
3bp for the week, according to Municipal Market Advisors, compared with a
decline of 10bp for Treasuries. Despite this, it was a big week for CDS
spread tightening. The Muni CDX (a basket of 50 muni CDS contracts)
dropped 16bp, the biggest one-week drop in over a year. As muni CDS have
widened post-crisis, many press reports have tried to imply some
fundamental implication to the muni CDS market. The municipal
credit-default swap market is a strange place. Most of the players are
either investment banks hedging interest-rate swap exposure they have
with municipal customers, or hedge funds speculating. The problem with
this market is that there is no natural short side to the CDS contracts
(that is, long side to the credit). Those that understand CDS don't
understand munis, and those that understand munis don't understand CDS.
Anyway, what results are some weak-handed longs in
the CDS contracts (i.e., short the credit). The speculative traders in
muni CDS don't have any notion that an actual default will occur, merely
that some greater fool will come along and buy protection at an even
greater price. So, when sentiment improves even a little, muni CDS come
crashing tighter as the weak hands are exposed.
The CDS will tend to move with general risk aversion and, therefore,
we might see some correlation between, say, Illinois or California cash
bond spreads and the CDS. And it is interesting to look at the relative
spreads of various individual CDS contracts, such as whether Illinois
is wider than California, or vice versa. However, be very careful
comparing muni CDS to corporate CDS and/or to historic averages. It
isn't very meaningful.