The graph below shows that rates on T-bills and notes are at record lows across
the yield curve. The graph covers the entire period that data is available for from the Federal Reserve of St. Louis. It shows
the history for 1-month, 1-year, 5-year and 10-year notes.
Investors are entirely
concerned right now with the return OF their money, not the return ON their money. If you plan on holding a T-note to maturity,
you can be assured that you will get your money back. You have absolutely no assurance of what the buying power of that money
will be, however.
If you have to sell before maturity, you can lose on your T-note
bets. Back in the 1970's, T-notes were routinely referred to as "certificates of confiscation." It is probably safe to assume
that those holding the 1-month, and perhaps even the 1-year bills, do actually plan to hold them to maturity. However, for
many holding longer term T-notes and bonds, this is not the case.
Right now, the
U.S. is going through its first real bout of deflation since the 1930's. Thus, on the surface the low rates for the short
end of the curve seem plausible. However, at rates this low, there is no real difference between holding a T-bill and simply
stuffing a wad of dead presidents into a safe deposit box. Physically it might be tough to simply put all the billions that
Paulson & Co. have showered on the banks into the vault, but economically the banks are doing the same thing by hoarding
cash in the form of short-term T-bills.
Holders of longer term T-notes and bonds
are playing a far more dangerous game. It strikes me as highly unlikely that the U.S. will experience deflation for 10 years.
While historically it has happened, most notably in the late 19th century, we were under a gold standard then, not under paper
To buy a 10-year T-note today, one has to assume that over the next 10 years
inflation will average less than 2.18% -- and average less than 2.66% for the next 30 years in the case of the 30-year bond.
That does not even consider the effect of taxes.
Meanwhile, last week the Fed made
clear that it will do everything in its power to prevent deflation. The amount of money the Fed can create is infinite, so
one has to believe that in the end it will be successful in putting an end to deflation. It might take a few quarters to do
so, but in the end the Fed will be successful.
Given that monetary policy always
works with lags, and the data that the Fed is working from is far from complete and accurate, the danger that the Fed will
overshoot is huge. There is at least a 1/3 probability that by 2011 inflation will surpass its late 1970's peaks. Just take
a good look at what sort of yields T-notes had back then.
It is true that right now,
owning T-notes is just about the only thing that is working. However, do you really want to be holding a 30-year bond (27
year bond by 2011) with a yield at cost of 2.66% when comparable bonds are yielding over 15%.
insurance companies are traditionally some of the biggest holders of long-term treasuries. I would avoid names like Hartford (HIG), Met Life (MET) and Conseco (CNO).