. 23
( 47 .)


deficit and debt figures change. If annual deficits and cumulative debt
Bonds 149

swell beyond current projections, they could cause interest rates to rise

Three Ways to Lose Money in Ultra-Safe
U.S. Government Bonds

The federal government deficit, as currently projected, does not seem to
spell doom for bonds. Those who argue that the deficit will cause prob-
lems have been making their bearish case for decades. In fact, Professor
Tobin™s quote regarding the crowding out effect was made in 1986. Eigh-
teen years later, Alan Greenspan said, “One issue that concerns most ana-
lysts, especially in the context of a widening structural federal deficit, is
inadequate national saving.”5 Those who fear that the large U.S. federal
deficits will eat up all the available savings might eventually be proven
right, but there is no evidence of a problem yet.
Even though U.S. government bonds are among the safest investments
in the world and there is no imminent risk from deficits, there are three
ways investors can lose money: (1) You never get your money back, (2)
you get worthless money back, or (3) you get much less money back than
alternative investments.

Bond Risk #1: Government Default
The most extreme risk is that the U.S. government defaults on its bonds.
Many corporations and other countries have defaulted on their bonds. It
is almost impossible to imagine the U.S. federal government defaulting.
In the most extreme circumstances, annual U.S. government deficits
could reach trillions of dollars. Even in such circumstances, the govern-
ment has unlimited ability to create dollars so there is essentially no risk
of default.
A U.S. government default is almost impossible. Those who worry
about such a default ought to be investing heavily in items such as guns
and food. I am not saying that it is impossible for the U.S. government to
150 Applying Science and Art to Bonds, Stocks, and Real Estate

default on its bonds, just that this is extremely unlikely, and if it happens
we will have much more to worry about than our investments.

Bond Risk #2: Inflation
The second risk to bondholders is that they will be repaid but that the dol-
lars they are repaid with might be able to buy very little. On this subject,
the science fiction legend Robert Heinlein (author of Starship Troopers,
Stranger in a Strange Land, and many other classics) wrote, “$100
placed at 7 percent interest compounded quarterly for 200 years will
increase to more than $100,000,000”by which time it will be worth
We already covered the risk of repayment with worthless dollars in the
inflation discussion. As of right now, there is no sign of dangerous levels
of inflation in the United States. Nevertheless, the risk exists and is one
of the most serious risks for bondholders.
In most circumstances the interest rate on government bonds exceeds
the inflation rate. While this has almost always been the case in the
United States, the amount of cushion”the difference between interest
rates and inflation”has varied dramatically. Figure 7.5 shows the inter-
est rate on the 10-year U.S. Treasury bond minus the rate of inflation.
The extra return on bonds above inflation has been decreasing for the
last 20 years. The 1983 bond investor received 7% above inflation while
the 2003 investor received only 1% above inflation.
In comparison to current inflation, bond buyers today are getting the
worst deal they have had in decades.

Bond Risk #3: Opportunity Cost
In 1989, I was the chief financial officer of Progenics Pharmaceuticals, a
start-up biotech company (now publicly traded with the stock symbol
PGNX). For no good reason related to my job, I wrote an analysis of the
RJR-Nabisco leveraged buyout. The Wall Street Journal published a
short version of my analysis as an editorial.6
Bonds 151

Real Interest rate (10-yr - CPI)

* part of year

FIGURE 7.5 Interest Rates Adjusted for Inflation Are Extremely Low
Source: Federal Reserve, Bureau of Labor Statistics

One of the consequences of this article was an invitation to give
my first lecture at Harvard. I was honored by the request so I flew to
Cambridge to present my analysis. How did my first Harvard lecture go?
The short answer is that it went amazingly poorly. Early in the lecture, I
asserted that RJR-Nabisco bondholders had lost $1 billion because of the
leveraged buyout. My calculation simply added up the loss on all RJR-
Nabisco bonds as quoted on bond markets on the day the deal was
announced. (These bonds traded actively so it was easy to get an accurate
measure of how much the price dropped because of the buyout
A student objected by saying that the bondholders had lost nothing.
She argued that the RJR-Nabisco bondholders were still going to get all
their money back. Accordingly, she said that the current price of the
bonds was irrelevant. I tried to argue against this view, but it was shared
by most of the students. After about 20 minutes of incoherent verbal flail-
ing, the professor had to intervene and say, “Please, let™s just assume that
152 Applying Science and Art to Bonds, Stocks, and Real Estate

Terry is right and move on.” This intervention allowed the lecture to con-
tinue, but obviously I had lost all credibility.
Let™s view this issue in the context of a $1,000 dollar investment into
a 10-year U.S. Treasury bond. Assume that the bond is bought with an
interest rate of 4%. The purchaser gives the government $1,000. In return
the government promises to pay $40 a year for 10 years, plus return the
original $1,000 at the end of the tenth year.
Now let™s consider what would happen if, soon after the purchase,
interest rates on 10-year treasuries jumped from 4% to 6%. What would
happen to our investor? The investor owns a bond that still promises to
pay $40 a year for 10 years and to return the $1,000 upon maturity. From
this perspective, the bondholder doesn™t appear to have lost any money
(this is the student™s argument from my lecture). On the other hand, the
rise in interest rates means that the market price of the bond would have
dropped by $150.
So how much money would our bondholder lose? Is it $0 or $150 or
something else?
The economist™s answer is that the bondholder would lose the full
$150 even if the government makes all the payments as promised. Where
does the loss come from? The loss is caused by the change in the “oppor-
tunity cost.” By investing at 4%, the bondholder has lost the opportunity
to earn 6% on the $1,000. And these are not simply losses on paper, these
are real dollars that the investor could have in his pocket but never will.
In my first Harvard lecture, and many since then, I have learned that
opportunity cost is a difficult concept to grasp. Even highly trained peo-
ple who understand the idea tend to overlook opportunity costs.
While the opportunity cost in financial terms is often misunderstood,
in other areas of life it is clear. A famous”and almost certainly fake”
wedding toast goes as follows: “Sometimes at rare moments in human
history, two people meet who are meant to be together forever. When
such romantic lightening strikes, I hope that the bride and groom have the
strength to say, ˜I am sorry, I™m already married.™ ”
Bonds 153

For those who are unwilling to divorce, the opportunity cost of marriage
is the forgone opportunities with other potential mates. A similar theme
is revealed in stories of a mythical culture where women were allowed to
have up to three husbands, but where divorce was banned. It was said that
women in this culture almost never had a third husband, and when they
did, he tended to be extremely handsome. By some calculations, the third
husband has the highest opportunity cost in this marriage system because
he rules out all future possibilities.
So the bondholder who locks in a 4% interest rate for 10 years loses
when the world changes to provide opportunities for 6% investments.

How Low Can Interest Rates Go?

U.S. bonds have had a 20-year run. When will interest rates begin to rise
thus ending the bull market in bonds?
Predictions, particularly of the future, are tough. (Economists have a
nearly unblemished record for predicting the past.) The most famous
wrong prediction of an economist is probably Yale Professor Irving
Fisher™s quote that, “Stocks have reached what looks like a permanently
high plateau.” Professor Fisher made this sanguine statement in October
1929 just before the stock market collapsed by 90% and the Great
Depression began.
Professor Fisher was, by some accounts, the world™s most famous
economist and he made his remarks at precisely the wrong time. This is
amazing, but not too different from the record of many economists.
One of my neighbors is a meteorologist (and don™t call her a weather
lady) for one of the Boston TV networks. She™s a bit of a celebrity in the
area. I see her from time to time in our building™s elevator, and my run-
ning joke with her has two themes (neither of them are at all funny). First,
I blame her for bad weather and thank her for the occasional nice day.
Second, I tease her for forecasts that often miss the mark. Actually, I used
154 Applying Science and Art to Bonds, Stocks, and Real Estate

to tease her; once she found out that I am an economist, the teasing had
to end.
Although their failures are usually less spectacular than those of Pro-
fessor Fisher, many seers in many fields have missed the mark by simi-
larly wide amounts. Some decades ago I read an article arguing against
immigration into the United States. The article said, look we™re all immi-
grants here in the United States (even Native Americans arrived only
15,000 years ago), but enough is enough. The country is finite and filling
up. It™s just common sense that the United States can no longer accept
huddled masses upon its teeming shore.
The punch line was that this anti-immigration piece had been written
hundreds of years ago when the country was empty by modern stan-
dards. By reprinting the article the newspaper was arguing in favor of
There are two ways to be wrong in making predictions. Irving Fisher
was wrong to predict no end to the trend that prevailed in the roaring
stock market of the 1920s. In my experience, it is even harder to predict
turning points in powerful trends. This was the mistake of the original
author of the anti-immigration piece who thought America was overpop-
ulated hundreds of years ago.
When will the bull market in bonds end? In spite of the hazards of pre-
dicting market turns, I have a clear answer: The bull market in bonds will
end soon, maybe not today, maybe not tomorrow, but soon and for the
rest of your life. Actually that™s Humphrey Bogart™s line as Rick near
the end of Casablanca. While prospects are not quite so bleak for bonds,
the bull market has largely run its course.
How low can interest rates go? The historical answer is that interest
rates can move substantially lower. In the Great Depression, the interest
rate on some U.S. Treasury debt fell to 2 basis points (that™s 0.02%!). At
this rate, a year™s interest on $100 is two cents! At around the same time,
the interest rate on 3 to 5 year Treasury notes was under 1%.7 Similarly,
Japan has had interest rates below 1% on some of its government debt in
recent years.
Bonds 155

Even the current low interest rate on U.S. treasuries isn™t the lowest
possible. Rates can go significantly lower. They cannot, however, go
below zero. That may seem obvious, but the proof is actually a bit subtle.
Would you give the U.S. government $100 in order to receive $99 back
in a year? The obvious answer is no. A far better alternative would be to
put the $100 in a safety deposit box and thus still have $100 in a year.
The cost of storage is the key to understanding the lower bound on
interest rates. In our society we can store money safely at very low cost.
Thus, we can always get at least $100 back in the future for $100 stored
away today.
Consider the very different storage world of a squirrel. Squirrels bury
food in good times and hope to retrieve some of it in bad times. On their
acorn investments, squirrels always accept negative interest rates. When
a squirrel saves 100 acorns by burying them, it always receive fewer than


. 23
( 47 .)