To all - text of earlier WSJ piece (3/30/98) by same two bullish guys.
March 30, 1998
Commentary
Are Stocks
Overvalued?
Not a Chance
By JAMES K. GLASSMAN and KEVIN A. HASSETT
The Dow Jones Industrial Average has returned more than 200% over the past
five years, and the past three have set an all-time record. So it's hardly
surprising that many observers worry the stock market is overvalued. One of
the most popular measures of valuation, the ratio of a stock's price to its
earnings per share (P/E) is close to an all-time high. The P/E of the average
stock on the Dow is 22.5, meaning that it costs $22.50 to buy $1 in profits--or,
conversely, that an investor's return (earnings divided by price) is just 4.4%,
vs. 5.9% for long-term Treasury bonds.
Yet Warren Buffett, chairman of Berkshire Hathaway Corp. and the most
successful large-scale investor of our time, told shareholders in a March 14
letter that "there is no reason to think of stocks as generally overvalued" as long
as interest rates remain low and businesses continue to operate as profitably as
they have in recent years. Investors were buoyed by this statement, even
though Mr. Buffett provided no analysis to back up his assertion.
Widespread Misunderstanding
Mr. Buffett is right--and we have the numbers and the theory to back him up.
Worries about overvaluation, we believe, are based on a serious and widespread
misunderstanding of the returns and risks associated with equities. We are not
so foolish as to predict the short-term course of stocks, but we are not
reluctant to state that, based on modest assumptions about interest rates and
profit levels, current P/E levels give us no great concern--nor would levels as
much as twice as high.
The fact is that if you hold stocks instead of bonds the amount of money
flowing into your pockets will be higher over time. Why? Both bonds and
stocks provide their owners with a flow of cash over time. For bonds, the
arithmetic is simple: If you buy a $10,000 bond paying 6% interest today, you'll
receive $600 every year. For equities, the math is more complicated: Assume
that a stock currently yields 2%, or $2 for each share priced at $100. Say you
own 100 shares; total dividend payments are $200--much lower than for
bonds.
But wait. There is a big difference. Profits grow over time. If that dividend
should increase with profits, say at a rate of 5% annually, then, by the 30th
year, your annual dividend payment will be over $800, or one-third more than
the bond is yielding. The price of the stock almost certainly will have risen as
well.
By this simple exercise, we can see that stocks--even with their profits
growing at a moderate 5%--will return far more than bonds over long periods.
Over the past 70 years, stocks have annually returned 4.8 percentage points
more than long-term U.S. Treasury bonds and 6.8 points more than Treasury
bills, according to Ibbotson Associates Inc., a Chicago research firm.
But isn't that extra reward--what economists call the "equity premium"--merely
the bonus paid by the market to investors who accept higher risk, since returns
for stocks are so much more uncertain than for bonds? To this question, we
respond: What extra risk?
In his book "Stocks for the Long Run," Jeremy J. Siegel of the University of
Pennsylvania concludes: "It is widely known that stock returns, on average,
exceed bonds in the long run. But it is little known that in the long run, the risks
in stocks are less than those found in bonds or even bills!" Mr. Siegel looked at
every 20-year holding period from 1802 to 1992 and found that the worst real
return for stocks was an annual average of 1.2% and the best was an annual
average of 12.6%. For long-term bonds, the range was minus 3.1% to plus
8.8%; for T-bills, minus 3.0% to plus 8.3%.
Based on these findings, it would seem that there should be no need for an
equity risk premium at all--and that the correct valuation for the stock market
would be one that equalizes the present value of cash flow between stocks and
bonds in the long run. Think of the market as offering you two assets, one that
will pay you $1,000 over the next 30 years in a steady stream and another that,
just as surely, will pay you the $1,000, but the cash flow will vary from year to
year. Assuming you're investing for the long term, you will value them about
the same.
What valuation level of the stock market would equalize the income flow from
stocks and bonds? To keep the calculations simple, we will pretend that you
can buy a perpetuity (an annuity that lasts forever) that pays 5.9%--about the
current long-term T-bond rate. Assume that the flow of payments you receive
when you buy a stock is also a perpetuity. Also assume that after-tax earnings
are a reasonable estimate of the cash flow from a stock, that future inflation
will be 2.8% annually, and that real earnings will grow at the same rate as the
rest of the economy, 2.1% a year. (Currently inflation is far lower and growth
far higher, but we're using the long-run assumptions of the Congressional
Budget Office.)
Using a simple and accepted formula, we find that the P/E that would equalize
the present value of the cash flow from stocks and bonds is about 100. By this
measure, the stock market is undervalued by a factor of about four. At current
prices, the flow of cash associated with holding stocks for many years is four
times as high as that for bonds.
Does that mean the Dow ought to be at 35000 instead of 8800? Not quite.
There are three important qualifications. First, the present-value calculations
will change a great deal if the real growth-rate assumptions change only a little.
If growth falls even a little short, the cash flow from stocks and bonds looks
more similar.
Second, we assumed that there was no risk premium at all in the long run, as
Mr. Siegel's research shows. But that may overstate the case. For the premium
to vanish, investors must hold their shares in a diversified portfolio for at least
20 years. In practice, investors often waver. A sensible question, therefore,
would be: How would our calculations change if we introduced a risk
premium, which would increase the rate at which you discount expected future
cash flows? As it turns out, a fairly modest risk premium of about 3% would
imply that the market is currently valued correctly. Again, however, that
premium is higher than history indicates it should be.
Finally, earnings might not be the best measure of the cash flow associated
with holding a stock. Suppose, for example, firms are retaining lots of money
today because they anticipate having to make big capital expenditures in the
future in order to maintain growth. It may be that dividends are a better
measure, and dividends are much lower than earnings. Dividends are probably
an underestimate of the cash flow, since many big profitable companies, such
as Microsoft, pay none at all. But making the same growth and inflation
assumptions as before, a payout of merely 1% would equalize the present value
of cash flow from holding stocks with that currently received by bondholders.
The Dow now yields 1.6%. Adding an estimate about the current level of share
repurchases (the equivalent of dividends) the total cash flow from firms to
current shareholders is probably about 2%. That means that even by the
dividend measure, the market is undervalued by about 50%.
Taking the Long View
Allow us now to suggest a hypothesis about the huge returns posted by the
stock market over the past few years: As mutual funds have advertised the
reduction of risk acquired by taking the long view, the risk premium required
by shareholders has gradually drifted down. Since Siegel's results suggest that
the correct risk premium might be zero, this drift downward--and the
corresponding trend toward higher stock prices--may not be over.
In order for the risk premium to go all the way to zero, the market would need
to rise by between 100% (the dividend measure) and 340% (the earnings
measure). We wouldn't bet the ranch on such an enormous and immediate
increase. After all, subtle variations in parameters we cannot possibly predict,
such as the growth rate or inflation rate, lead to big changes in conclusions.
The cash flow between bonds and stocks would also be equalized, for
example, if the stock market stayed where it is and the interest rate on long
bonds climbed to about 9.5%. However, in the current environment, we are
very comfortable both in holding stocks and in saying that pundits who claim
the market is overvalued are foolish.
Mr. Glassman is a resident fellow and Mr. Hassett a resident scholar at the
American Enterprise Institute.
Copyright © 1998 Dow Jones & Company, Inc. All Rights Reserved. |
