Fischer Black
on
MARKETS
Vol. 1, No.1
February 17, 1976
IS THIS THE TIME TO SWITCH TO BONDS?
From the start of 1926 to the end of 1974, stocks grew (with dividends
reinvested) at an annual rate of 8.5%.* Other starting and ending points
would give quite different rates of growth, because this period includes
some major upward and downward moves in the stock market. Over the same
period, a continuous investment in long term government bonds would have
returned 3.2% per year, while a continuous investment in short term Treasury
Bills would have returned 2.2% per year. Thus stocks returned 5.3 percentage
points more than long term governments, and 6.3 percentage points more than
Treasury Bills.
Currently, long term government bonds are yielding up to about 8% per year,
while Treasury Bill rates are less than 6%. If we compare a yield of 8%
with a return on stocks of 8.5%, we might well think that bonds are a better
investment, since they are much less volatile than stocks. Similarly, if
we compare bond yields with bill yields, we might think that the longer term
bonds are clearly better than shorter term bonds such as Treasury Bills. But
before calling a broker, let's think about these comparisons more carefully.
The Future for Stocks
To some extent, the future for stocks will undoubtedly be like the past.
But there's no reason to believe that the numbers will be identical, no matter
how far back we look in the past, or how far ahead we look in the future.
More generally, there is no reason to believe that the average rate at which
investors expect stocks to grow (including reinvestment of dividends) stays
fixed.
Fischer Black on MARKETS Page 2
We can't tell directly how rapidly investors expect stocks to grow because
we can't see into their minds. But there are some factors that we would
expect to be important in guiding investors' expectations. And no matter
now many factors we think of, there will be others that we didn't think of
causing expected returns on stocks to change over time.
One factor that investors doubtless consider is the rate of inflation. An
increase in the rate of inflation may be associated with a fall in the market.
But after the market falls, if there is no further increase in the rate of
inflation, investors will likely expect a higher growth rate for stocks. One
reason for this is that inflation may mean higher dollar figures in the corporation's
income statement. Another reason is that the market can grow
faster if it starts from a lower point, even if the dollar figures in its
future income statements are unchanged.
Another factor affecting expected returns on stocks is the level of the
market generally. After stocks rise, investors as a group may be more willing
to risk their money, and they may invest in stocks even though they expect
slower growth for stocks than they expected before the rise. After stocks
fall, investors as a group may be less willing to risk their money, and it
may take a higher expected return on stocks to coax them into the market.
While I think this is the most likely direction of the relation between past
returns and future returns on the market, it is also conceivable that the
reverse is true: that a rise in the market leads investors to expect a
further rapid rise, while a fall in the market leads investors to expect
only a modest rise in the future. For the moment, the important thing is
that the rate at which stocks are expected to grow is related to what the
market has been doing.
A change in the volatility of the market or in how investors feel about the
volatility of the market may also be associated with a change in the expected
return on the market. When the market is more volatile, either because companies
are using more leverage or because uncertainties about fundamental
Fischer Black on MARKETS Page 3
factors affecting future earnings are higher, it may take more to bring investors
into the market than it takes at other times. If investors did not
expect more of stocks at such times, there would be an imbalance between the
amount of stock investors want to hold and the amount of stock outstanding.
It's likely, too, that changes in the tax laws affecting investors in
stocks and changes in the amounts that corporations payout in dividends will
have an impact on the expected returns on stocks. It turns out to be difficult
to predict the direction of these effects. Surprising as it might seem,
higher taxes may increase the level of the market and reduce future before-tax
expected returns, and higher dividends may decrease the level of the market
and increase future expected returns. For now, let's just note that there
will be an effect of one sort or the other.
The Volatility of the Market
The volatility of the market can be observed more easily than the market's
expected rate of growth. We know that the typical move in the price of a
stock (as a fraction of the price) was higher in the depression of the 30's
than it was either in the 20's or at any time since the depression. We can
see changes in the volatility of a stock that seem to occur gradually over
time, and we can see changes that depend on specific events such as a recapitalization
that changes the company's debt-equity ratio.
We also observe a rather strong relation between changes in the level of the
market and changes in the volatility of the market. When the mar et goes up,
its volatility (expressed in percentage terms) goes down, and when the market
goes down, its volatility In fact, the effect seems so strong
that an increase in the level 0 the market may be associated with at least
a temporary fall in the dollar volatility of the market. A stock that moves
$0.50 on a typical day when it sells for $20 may move only $0.375 on a
typical day when it sells for $40.
Fischer Black on MARKETS Page 4
There are several possible reasons for a relation between stock prices and
volatility. Changing stock prices may cause changing volatilities, or
changing volatilities may cause changing stock prices, or there may be underlying
factors causing changes in both stock prices and volatilities, or all
of the above may be true. We don't need to know why the effect is there to
know that volatilities change, sometimes dramatically.
Interest Rates
Interest rates are even easier to observe than the volatility of the market.
No one doubts that interest rates change over time. (In fact, it is a
general rule about markets that nothing can safely be taken as constant
over time.) We can even use the yields on long term bonds to forecast
changes in the yields on short term bills. That's why an investor may buy
bills yielding 6% when bonds are yielding 8%, even if he doesn't mind the
fluctuations in bond values. If the average yield on bills over the life
of the bonds is expected to be about 8%, then many investors may be indifferent
between holding bills and holding bonds.
Interest rates seem to be affected by all the things that can cause changes
in the expected return on the market. An increase in the rate of inflation
seems to be associated with an increase in short term interest rates. S0
does an increase in the volatility of the market. And in recent years, a
decline in interest rates has been the occasion for many a striking rally in
the stock market (though the cause may well be more from stock market changes
to interest rate changes than the other way around).
From the start of 1926 to the end of 1974, the rate of inflation (as measured
by the consumer price index) averaged 2.2% per year. * The consumer price
index has many defects as a measure of the level of prices or the rate of
inflation, so we should not make too much of the fact that the average
return on Treasury Bills for this period was also 2.2% per year. But if it
makes sense to talk about a real interest rate at all, it seems clear that
the average real interest rate has been a fairly small number.
Fischer Black on MARKETS Page 5
There is some evidence that changes in the rate of inflation that investors
expect are reflected approximately point for point in changes in Treasury
Bill rates. If an investor is in a 50% tax bracket, this means that a
one point increase in the rate of inflation is associated with a half-point
decrease in the "real" after-tax interest rate. Half of the one point increase
in the nominal interest rate is taken away from him in taxes. In
other words, for an investor in a 50% tax bracket, an increase in interest
rates associated with higher inflation may mean a lower effective interest
rate for him. When the nominal rate goes up, his real after-tax rate may go
down. If he's a lender, he may be sad when rates go up; while if he's a
borrower, he may be happy when rates go up.
What Should We Do?
Interest rates are now higher than they were from 1926 through 1974, partly
because the rate of inflation is higher. But higher inflation probably means
higher nominal returns for stocks in the future, too.
The last decade has been a bad time for stocks, especially relative to
bills. From the start of 1965 to the end of 1974, stocks returned an average of
1.2% per year, while Treasury Bills returned an average of 5.4% per year.
But it seems plausible that a relative decline in the market means higher
relative expected returns on the market in the future.
On the other hand, the volatility of the market has been lower in recent
years than it was for much of the period from 1926 to World War II. A lower
level of volatility would suggest lower relative expected returns on the market
in the future. Taking the recent performance of stocks and the recent volatility
of the market together, I can see no reason to forecast either a
higher relative return or a lower relative return than we have experienced
in the past. Thus the expected return on stock~ may be about 6.3 percentage
points higher than interest rates on Treasury Bills. With Treasury Bills
Fischer Black on MARKETS Page 6
at current levels, this implies an expected 12% per year for stocks in
the short run. Since bond yields imply that investors think short term
rates will rise in the future, the long run expected return on stocks may
be even higher than 12% per year.
Thus today's invpstor may have about the same wealth, and may face about
the same volatility and difference between stock returns and bond returns
as the investor of 10 years ago. Such an investor may decide to keep in stocks
the money he currently has in stocks.
What's more, shifting money between stocks and bonds because of changes in
an investor's beliefs about their relative future returns can have high
transaction costs and can introduce a form of poor diversification over
time that is much like poor diversification across stocks.
In sum, this is probably not the time to switch to bonds. Nor is it likely
to be the time to switch to stocks. Some investors should do one, and
some investors should do the other, but the typical investor should very
likely stay put. And I expect to continue to believe this no matter what may
happen to the stock market or to interest rates in the future.
* The figures in this letter are taken from "Stocks, Bonds, Bills, and
Inflation: Year-By-Year Historical Returns (1926-1974)," by Roger G.
Ibbotson and Rex A. Sinquefield.
issue of the Journal of Business.
It appears in the January, 1976
If you have any questions or comments, please call me at 617-253-6691, or
write me at 50 Memorial Drive, Cambridge, MA 02139.