Re: longbond follow up hummel/henwood

[wfhummel@xxxxxxxxxxxxxx (William F. Hummel)]

Bond traders at the screens of terminals in the bond trading
rooms of security dealers are focused on small changes in price
that might work for an hour, a day, but seldom much longer.  They
employ a large amount of capital and are mainly looking to make
quick, but not necessarily large, profits per round trip.  They
work primarily with breaking news, and gain their advantage by
being first to act on such news.  Breaking news could relate to
number of things -- an assassination, a natural disaster, a
president's illness, or a speech by Greenspan.

The hyperactivity of long bonds applies to long maturities where
the volatility is greatest.  Traders can't make much profit where
volatility is low, i.e. at the shorter maturities, including long
bonds nearing maturity.  There is a characteristic negative slope
to the yield curve near the long end of just a few basis points.
That is where most of the bond trading activity takes place and
where the spreads are narrowest.

My contention is that traders dealing in spot news and trading on
very short turnarounds have no lasting effect on the term
structure.  They are a noise on the signal.  Longer term traders,
i.e. those whose turnaround time is say at least three months,
begin to have some influence even though they are playing for
capital gains rather than income.  The reason is that they are
dealing in fundamentals rather than breaking news.  What this all
implies, I believe, is that the transactions in the bond market
that most influence the term structure are those made by parties
trying to maximize total return (mostly from coupon income) over
their own investment horizon.  That implies a sensitivity to
longer term inflationary pressures and results in the inflation
risk premium seen in the term structure.

Here are some interesting data relative to this point.

         Estimated Ownership of Treasury Securities
                                as of March 1997

                                             Billions     Percent
Total Privately Held          $3451.7         100.0

Commercial Banks              275.0              8.0
Individuals                              355.4            10.3
    Savings bonds                  186.5
    Other securities               168.9
Insurance companies         236.5               6.9
Money market funds             83.9                2.4
Corporations                        262.5                7.6
State and local govts          353.0              10.2
Foreign & international     1191.1              34.7
Other investors                    686.4              19.9

Source:  Financial Management Services of U.S. Treasury Dept

The category "Other investors" was not defined, but it would seem
to include pension funds, mutual funds (aside from MMMFs) and
bond dealers.  It is likely then that considerably less than 20%
of the privately held Treasury debt issues is involved in short
term trading.  This is consistent with my contention that the
average 30 day holding period of Treasury bonds must be highly
skewed by the enormous traffic in bonds due to bond traders.  The
30 day figure, if not properly interpreted, gives a highly
distorted picture of what is happening.

William F. Hummel