To print article, click Print on your browser's File menu.

Go back

Posted 10/10/2002

Sponsored link

Buy “The Education of a Speculator” at

Price 38.060
Change -0.530
Research Wizard

Add to MSN Stock List

Message Board

Price 34.010
Change +0.530
Research Wizard

Add to MSN Stock List

Message Board

Price 0.770
Change -0.050
Research Wizard

Add to MSN Stock List

Message Board

Price 3.790
Change +0.080
Research Wizard

Add to MSN Stock List

Message Board

Price 24.050
Change +0.360
Research Wizard

Add to MSN Stock List

Message Board

Related Sites

"Stocks for the Long Run" at

"Triumph of the Optimists" at

Christopher Cox's bill to end double taxation of dividends

The Speculator

Recent articles:
• The earnings gimmicks roll on, 10/3/2002
• Count on a company's cold, hard cash flow, 9/26/2002
• Empty shelves signal a rising stock, 9/19/2002

sponsored by:
Click Here!

The Speculator
What dividends say about a stock
Surprise, surprise: Dividends are at a six-year high. But right now, they're just as important for the confidence they represent. An end to double taxation could make them a powerful long-term factor.
By Victor Niederhoffer and Laurel Kenner

Whatever you do, give us more Barber.
-- Beethoven to Rossini, composer of "The Barber of Seville."

"Give us more dividends!" Everyone wants them, now that stocks are in the dumps. A nice 7% yield on REITs, tobaccos or utilities looks awfully good compared with 2% in the money market, -30% in the S&P 500 ($INX) or -40% in the Nasdaq ($COMPX) year to date.

Money Plus.
Easy online tools
and free Bill Pay, too.

Dividends are such a hot topic that people are using them in all sorts of debates. PIMCO's Bill Gross, king of the bond traders, says dividend yields on stocks are so low that the Dow Jones Industrial Average ($INDU) should be at 5,000. A move is afoot to eliminate the double taxation of dividends. The academics are studying dividends to see if they give more reliable signals than the highly distorted earnings that companies have tended to report.

Amid this abundance of interest, one thing is missing: a clear understanding of the practical significance of dividends, and updated studies on their importance. One reader after another wrote us after last week's column on buybacks, urging us to give dividends a look.
    I have a suggestion for a new topic: dividends. I think that companies that pay dividends (or buy back their shares) are good investments for their shareholders, especially in these tough times. The specific topic I had in mind was the double-taxation problem with dividends. I think the government should reduce or eliminate taxes on dividends, which would encourage more companies to pay them for the benefit of the shareholders. -- Sam W. Goodner, CEO Business Strategy, Agere Systems, Munich.
Yes, indeed. And who are you going to call when you need a no-nonsense study that puts statistics on the table and takes account of the inherent variabilities and changing cycles? That's right -- the Speculator Duo. They're back, pencil and envelope at the ready.

Dividends are up
While no one was looking, and certainly no one is commenting, dividend yields have been creeping up. While we were writing this column on Monday, Oct. 7, the S&P 500's dividend yield reached the 2% mark for the first time since December 1996. That's up from 1.1% at the end of 1999 and for most of 2000. A chart of recent monthly and selected year-end figures appears below.

 Recent S&P 500 month-end dividend yields
Date Dividend yield (%)
10/7/2002 2.0
9/30/2002 1.9
8/30/2002 1.7
7/31/2002 1.7
6/28/2002 1.6
5/31/2002 1.5
4/30/2002 1.5
3/29/2002 1.4
2/28/2002 1.4
1/31/2002 1.4
Source: Bloomberg L.P., S&P Security Price Index Record, The Triumph of the Optimists (Princeton University Press, 2002)

 Selected S&P 500 year-end dividend yields
Year Yield (%) Year Yield (%)
2001 1.4 1990 3.7
2000 1.2 1980 4.5
1999 1.1 1970 3.4
1998 1.3 1960 3.3
1997 1.6 1950 7.2
1996 2.0 1940 6.3
1995 2.2 1935 3.5
1994 2.8 1900 4.3
1993 2.7
Source: Bloomberg L.P., S&P Security Price Index Record, The Triumph of the Optimists (Princeton University Press, 2002)

You might think, "Well, 2% is a six-year high, but so what? Dividend yields were as high as 7.2% in 1950, and 4.3% at the beginning of the century." And indeed, that's the reasoning of people like Bill Gross, who uses 100-year statistics on dividend yields and dividend growth to argue for Dow 5,000.

That's such a naïve argument that ordinarily one would be ashamed to make it. It leaves out the fact that companies are now spending as much to buy back shares as they pay out in dividends. It also leaves out that the long-term interest rate on the 10-year bond is 3.6%, the lowest level since the 1950s, meaning stocks don't have to offer much in the way of a risk premium to be attractive. Here's a table showing 10-year yields at year-end over the past four decades:

 Year-end yields for U.S. 10-year notes (1962-2001)
Year Yield (%) Year Yield (%) Year Yield (%)
2001 5.051 1987 8.859 1973 6.902
2000 5.112 1986 7.223 1972 6.412
1999 6.442 1985 8.986 1971 5.892
1998 4.648 1984 11.514 1970 6.502
1997 5.741 1983 11.801 1969 7.882
1996 6.418 1982 10.389 1968 6.162
1995 5.572 1981 13.982 1967 5.702
1994 7.822 1980 12.432 1966 4.642
1993 5.794 1979 10.332 1965 4.652
1992 6.686 1978 9.152 1964 4.212
1991 6.699 1977 7.782 1963 4.142
1990 8.067 1976 6.812 1962 3.852
1989 7.935 1975 7.762
1988 9.137 1974 7.402
Source: Bloomberg L.P.

But most important, the argument leaves out the tax disincentive for dividends. Dividends are taxed at the corporate level, and then they are taxed again some 40%, on average, at the individual level. Capital gains are taxed at only 20%. It's much more advantageous to all parties if corporations retain their earnings and use them to invest wisely in assets and then turn over the capital gains to investors. That's why the retention ratio -- earnings retained versus total dividends -- has increased from 30% in the 1950s to 70% today.

Dividends and returns
We next looked to see what happens to the S&P 500 when per-share dividends fall.
From 1938 through September 2002, S&P 500 dividends have declined in only six years. The table shows what happened to the market in the year of the decline and in the next year:

 Down years for dividends
Year of dividend decrease % decline in dividend S&P 500 return that year S&P 500 return next year
1938 -37 25 -5
1942 -16 12 19
1951 -6 15 12
1958 -3 38 8
1970 -1 0 11
1971 -2 11 16

In the six years that dividends actually declined, the average capital gain in the S&P was 17%. The next year, the gain was 10%. Including dividends would add 5 percentage points to those figures. The myth that stocks can't go up unless dividends do is clearly wrong.

Payouts and performance
The definitive word on this research is provided in that magnificent book, the one investment book Laurel would recommend to her mother and Vic would most want his six daughters to read: "Triumph of the Optimists," by Elroy Dimson, Paul Marsh and Mike Staunton.

"Triumph" examined the performance of the 30% of companies with the highest dividend yields at the beginning of each year and compared it to the performance of companies that were in the lowest 30% of dividend yield. From 1900 through 2000, results are as follows: Companies with the highest yield showed a total return of 12.2% a year. Companies with the lowest yield returned 10.4% a year. These differences might not seem like much. But over time, they come to gigantic differentials. Over the 101 years of the study, $1 invested in the highest-yielding companies would have come to $4,948, while $1 invested in the lowest-yielding companies would have increased to a mere $1,502.

Martin Lee, one of the readers who wrote us about our column on buybacks last week, observed that the dividend-paying stocks in his portfolio "seem to have a tendency to buy back their shares fairly regularly -- Philip Morris (MO, news, msgs) springs to mind."

The Dimson group's review of the literature bore this out. They found that most buybacks are made by dividend payers. They also found that share buybacks surged in the United States from the mid-1980s. And here’s a surprise: Taking into account both dividend payments and buybacks, total payout to investors as a percentage of earnings increased from 1972 to 1998.

Dividends as signals
Only 21% of all companies pay dividends at all these days, and it's no wonder. The double taxation of dividends makes it inefficient to do so. But dividends still serve a signaling function. If they're increased, presumably it means that the company is confident of future earnings. If they're decreased, presumably the company isn't overly sanguine -- but the effect on the stock price, of course, comes before the announcement. (Research shows that boards are more reluctant to cut dividends when earnings fall than they are to increase them when earnings rise.)

Are companies still actually increasing their dividends? And how do these companies perform subsequent to the announcement relative to companies that decrease, discontinue or omit dividends? The Spec Duo found just 14 S&P 500 companies that decreased dividends in 2002:

Bausch & Lomb (BOL, news, msgs)
Centerpoint Energy (CNP, news, msgs)
CMS Energy (CMS, news, msgs)
Dynegy (DYN, news, msgs)
Ford Motor (F, news, msgs)
Goodrich (GR, news, msgs)
Millipore (MIL, news, msgs)
Pall (PLL, news, msgs)
Qwest Communications (Q, news, msgs)
Transocean (RIG, news, msgs)
U.S. Steel (X, news, msgs)
Williams Cos. (WMB, news, msgs)
Xcel Energy (XEL, news, msgs)

Their performance, on average, was -20%, compared with -19% for the S&P 500. (The S&P figure is an average of the index's performance from each announcement date.)

On the other hand, 112 companies -- more than one-fifth of the S&P 500 -- announced increases in dividends in this difficult year. The performance of these companies subsequent to the announcement was -14%, significantly better than the average 22% decline for the S&P 500 over the relevant periods.

We are updating the Dimson group's study on the difference between high-dividend-yield companies versus the low-dividend-yield companies and will report the results in a future installment.

Double taxation and distorted incentives
The outcry over earnings fraud has highlighted the problem with the double tax on dividends. If companies had to come up with hard cash to pay higher dividends each quarter, they might have less latitude to use accounting tricks, and investors would gain more control over the assets that, as shareholders, they own. Institutions are already demanding cash on the barrel; research shows that the higher the percentage of institutional ownership of a stock, the higher the dividends relative to earnings.

Management, unfortunately, has natural incentives to avoid paying dividends. Dividends reduce stock prices -- and the value of executives' options. Indeed, the rise in executive pay over the past 50 years has closely tracked the decline in dividends. The IRS, by permitting deductions for interest but not dividends, only leads executives further into temptation. Investors and rating agencies dislike highly leveraged balance sheets, and some companies -- Enron (ENRNQ, news, msgs) comes to mind -- go to great lengths to hide their debt with dubious or fraudulent accounting. Wharton professor Jeremy J. Siegel, author of "Stocks for the Long Run," argues that any effort to fix the earnings credibility problem is unlikely to work without mitigating the double tax on dividends. A bill to eliminate the double taxation was introduced Sept. 4 by Rep. Chris Cox, a California Republican.

In the 19th century, before U.S. securities markets were regulated, a dividend increase was the best way for a company to signal good earnings. It looks like we've come back to that.

Final note
Bill Gross may be the king of bonds, but the emperor sees things differently. We asked Paul DeRosa, the smartest bond trader we know, to analyze Gross's recent indictment of equities (in his words, "Stocks stink.") We'll send DeRosa's brilliant skewering of Gross to any reader who e-mails us at with compliments, comments or critiques of our column.

At the time of publication, neither Victor Niederhoffer nor Laurel Kenner owned or controlled any of the securities mentioned in this article.

MSN Money's editorial goal is to provide a forum for personal finance and investment ideas. Our articles, columns, message board posts and other features should not be construed as investment advice, nor does their appearance imply an endorsement by Microsoft of any specific security or trading strategy. An investor's best course of action must be based on individual circumstances.