In one of the most surprising research developments over the last four decades, Robert Arnott and Clifford Asness published an article in the January/February 2003 issue of the Financial Analysts Journal entitled “Surprise! Higher Dividends = Higher Earnings Growth.”
The article carries more than normal credibility in that Mr. Arnott is a prior editor of the prestigious journal, and Mr. Asness is on the editorial board. In the 2003 publication the authors stated, “The historical evidence strongly suggests that expected earnings growth is fastest when current payout ratios (of dividends) are high and slowest when payout ratios are low.”
This would appear to be a form of academic betrayal, as virtually every textbook and article on dividends suggests that low dividend-paying stocks provide the greatest growth potential. The typical academic literature is even backed up by the “sustainable growth model” measure of valuing stock prices, which suggests that future growth is largely supported by the percentage of retained earnings that is reinvested in the corporation (and not paid out as dividends).
The academic rebels, however, back up their high dividend, high earnings evidence with the argument that companies that pay high dividends are generally confident in their ability to provide strong earnings growth in the future. Were this not the case, the authors suggest, the firms would hoard each dollar of current earnings in a “so-called” contingency fund to protect against future developments.
Another explanation by the authors is that high earnings retention and a low dividend payout may signal an attempt at empire-building by current management. In order to develop a larger enterprise and the higher executive compensation that often goes with that, management may engage in developmental projects that do not represent the best interest of stockholders. While this is not always the case, it happens often enough to capture investors’ attention. One need only look at the evidence on corporate divestitures and spin-offs of unwanted divisions to realize that many prior empire-building mistakes by management have to be corrected at a later date.
The controversial 2003 research by Arnott and Asness was further affirmed by Ping Zhou and William Ruland in the May/June 2006 issue of the Financial Analysts Journal. These authors went from the aggregate market data used in the initial study to a company-by-company analysis and reached the same conclusion. In the introduction to their study, the authors state: “Our tests also show that high-dividend-payout companies tend to experience strong, not weak, future earnings growth.”
While the evidence from these two comprehensive research studies questions the traditionally assumed relationship between dividend payout and growth, it is not intended to override the evidence on companies in the early stages of growth. No one would suggest that Oracle or Microsoft would have been well advised to pay dividends to stockholders as they were just beginning their road to success. The thrust of the two cited studies applies more to stocks that are in their mid-growth phase and that are listed in the S&P 500 index, the Dow Jones industrial average, or similar stock market indexes.
This is where the great debate lies. There are enough research studies on this topic to fill up a university library. Suffice it to say that the research study results are mixed (going all the way back to the 1960s), with as many studies supporting the dividend as value-enhancing as those that take the opposite viewpoint.
In this article, I present additional data that is more up-to-date and of potential interest to individual investors. The emphasis is on dividend yield—the annual dividend divided by the stock price.
One way to assess the relationship between dividend yield and stock market performance is to look at stocks in the Dow Jones industrial average. The timeframe used here is the beginning of 2004 through June 30, 2008. The 30 stocks and their performances over that 4½-year time period are shown in Table 1.
In examining Table 1, you can see that the higher-yielding stocks do not necessarily appear to be sacrificing capital gains, nor do the lower-yielding stocks appear to have an exclusive claim to high capital gains.
|Bank of America||4.25||–43.9|
|Johnson & Johnson||2.20||26.7|
|Proctor & Gamble||1.86||23.6|
To further examine the issue of dividend yield and capital appreciation (or loss), the 30 stocks were divided into those with a dividend yield of 2% or greater (higher-yielding stocks), and those with a dividend yield of less than 2% (lower-yielding stocks). As indicated in Table 2, the higher-yielding stocks had an average gain over the 4½-year time period of 32.0% percent (with a midpoint return of 19.7%); the lower-yielding stocks had an average loss of -1.4% (and a midpoint return of 2.2%). Over this time period, the S&P 500 index advanced 15.2% and the Dow Jones industrial average was up 8.6%.
|Higher-Yielding Stocks (2% or Greater)|
|Bank of America||4.25||–43.9|
|Johnson & Johnson||2.20||26.7|
|Lower-Yielding Stocks (Less Than 2%)|
|Proctor & Gamble||1.86||23.6|
Based on this data, it is apparent that higher-yielding stocks did not provide less capital appreciation. Also, keep in mind that the higher-yielding stocks provided more dividend income to go with capital appreciation.
The higher-yielding stocks paid an average total dividend over the 4½-year period of $5.72, while the lower-yielding stocks provided average total dividends of $3.43. These numbers further enhance the case for higher-yielding stocks over lower-yielding stocks based on total return.
The importance of dividends to total return was particularly evident in the case of JPMorgan Chase. As can be seen in Table 2, the firm had a 14.2% loss over the 4½-year time period. Although not shown in the table, its stock value went from $40 to $34.31. However, the cash dividends paid out over the time period were $7.14, and on a total return basis, there was a net gain of $1.45 (+$7.14 in cash dividends minus $5.69 in stock value decline). Based on an initial price of $40, there is a total return of 3.62% ($1.45 ÷ $40). While the return is modest, it is still a positive return for a company that traded in negative territory based on price change alone.
Firms such as AIG, American Express, Home Depot, Intel, and Pfizer, which had low dividend yields, did not have a sufficient dividend payout to cover their losses in market value.
While reasonable dividend yields may provide benefits and signal a corporation’s confidence in its ability to sustain its earnings in the future, an excessive yield may have the opposite effect. That is, a very high yield may signal that the dividends may not be sustainable and are likely to be cut in the future. A company tends to become suspect when its yield goes to 6% or higher.
This guideline tends to apply to most areas except REITs, where high yields may not be a warning signal.
Having stressed the potential importance of dividends, I should point out that the 2000s have been a declining decade for dividend payouts. In 2007 and the first half of 2008, the dividend payout ratio on the S&P 500 index hit an all time low of 30%. Historically, companies have paid out 40% to 60% of earnings in dividends. Also, the dividend yield on the S&P 500 index shrank to 1.8%, versus the historical norm of 3.5%.
The lower dividends and payout ratios are not a problem when stock prices are booming. For example, in 2003, the S&P 500 index was up 26.38% and the dividend yield was 2%, for a total return of 28.36%. For the most part, investors are not concerned about the dividend yield when they are receiving substantial capital gains but, of course, this is not always the case
Table 3 shows three contrasting decades in which dividends played a different role in total returns. In the 1940s toward the top of the table, the yield was high—it averaged 5.87% and capital appreciation averaged 4.10% for an annual total return of 9.97%. In this decade, dividends provided investors with over half (59%) of their total return (shown at the bottom of table).
|Percentage of Returns Attributable to Dividends|
|Source: Ibbotson SBBI, 2008, Classic Yearbook, Market Returns for Stocks, Bonds, Bills, and Inflation, Morningstar, Chicago -2008.|
Similarly, dividends accounted for 38% of total return in the 1960s, and a mere 14% in the 1990s. It is not surprising that investors entered the decade of the 2000s with less than normal interest in dividends.
However, between 2000 and 2007, the S&P 500 was flat, going from 1,469.04 on January 1, 2000, to 1,468.36 on December 31, 2007. And then came the bear market of 2008. One observation that can be made about this decade so far is that any positive element of total return had to come from dividends. The only other decade so dependent on dividends for total return was the 1930s.
As the dividend yield and payout ratio have been cut in half over the last few decades, an obvious question is: Where have these funds gone?
A desirable answer, from an investor’s view, would be toward reinvestment in plant and equipment, new products, research and development, and so on.
However, the main recycling has been toward share repurchases. That is, funds that were once used to enhance dividends are now being deployed toward buying back shares. The evidence can be found in Table 4, which shows the total of stock repurchases versus dividend payouts for companies in the S&P 500 in recent years. Note, in particular, the value for Net Repurchases—it exceeds Dividends Paid in all three years of the table.
|Gross Repurchases ($ bil)||351||494||430|
|Equity Issuance ($ bil)||110||123||125|
|Net Repurchases ($ bil)||241||371||305|
|Dividends Paid ($ bil)||202||224||252|
|Dividend Yield (%)||2.1||1.8||1.8|
|*The year 2008, which is an aberration, is not shown.|
|Source: Morgan Stanley Online, 2007.|
In looking at all sides of the argument about share repurchases, one could say that companies that were repurchasing their own shares during the bull market of the 1990s looked smart as the value of their shares continued to go up, and foolish a decade later in the bear market of the 2000s as their shares declined in value. One could hardly credit the management of General Motors for repurchasing shares in the middle of the current decade at $35 to $40 per share, only to see the stock fall to $1.50 and $2.50 a few years later.
While dividends have always been important to the investor, they are especially significant in the current market environment. Because of the many false market recoveries we have seen in 2009, it is important to buy (or hold) stocks that have downside protection in case of a false signal that the market has bottomed out.
Stocks such as General Electric, Dow Chemical, Pfizer, and others are providing historically high yields and associated downside protection. This safety cannot be found in high-flying, low-yield growth stocks. The high-dividend-paying stocks also have good upside potential, as many are trading at 50% or more below their highs of a year or two ago.
High-dividend-yielding stocks also are appealing in the low-yield environment for money market funds, CDs, etc. The investor is able to pick up four or five additional percentage points while also participating in the opportunity for capital gain.
One final point to consider with dividend-paying stocks is the ability to reinvest the dividend without paying a commission. Many corporations offer dividend reinvestment plans in which the dividend is automatically reinvested in the firm’s stock at no cost to the investor. This can be particularly advantageous when small amounts are involved. Some corporations even offer direct sales purchase plans, which allow investors who are not yet stockholders to buy stock directly from the company so they can begin to take advantage of the dividend reinvestment plan.
There are a number of puzzles surrounding the payment of cash dividends by corporations.
The first has to do with recent research that indicates that high dividend payments lead to strong future earnings. This theory is at variance with traditional thought, and at least forces the growth-oriented investor to consider investing in stocks that pay moderate dividends.
There are an equal number of research studies that argue that dividends increase or diminish stock value. I address this issue by looking at contemporary data regarding dividend yield and stock performance of the firms in the Dow Jones industrial average between 2004 and mid-year 2008. Actually, the firms with dividend yields over 2% provide better capital appreciation performance than those with yields under 2%.
However, one caveat is that stocks that pay abnormally high dividends (6% or higher) may be giving off signals of future problems and that the dividend is not sustainable. This warning does not apply to REITs or other normally high-yield sectors.
Another puzzle surrounding dividends is the aggressive growth of stock repurchase plans in preference to cash dividends. There is little evidence that the repurchase alternative provides superior returns. Perhaps, corporations and investors need to rethink this strategy.