The day after Microsoft’s announcement, 202 million shares traded hands, as opinions on Microsoft’s action differed sharply. Some analysts concluded that Microsoft had finally recognized that it had few growth prospects and would not need to keep its huge horde of over $56 billion in cash. But bulls said that returning profits to shareholders is one of the primary functions of a corporation and that Microsoft’s action was good for investors.
Who is right? Those who bought because Microsoft was returning its profits to the shareholders, or those who sold because paying a dividend represented poor growth prospects? History has provided an unambiguous answer to this question: dividends have been the overwhelming source of stockholder returns throughout time, and firms that have higher dividend yields have given better returns to investors.
The Big Picture
Consider the following crucial fact about historical stock market returns:
From 1871 through 2003, 97 percent of the total after-inflation accumulation from stocks comes from reinvesting dividends. Only 3 percent comes from capital gains.
Take a look at Figure 9.1. The sum of $1,000 invested in stocks in 1871 would have accumulated to almost $8 million after inflation by the end of 2003.1 Without reinvesting dividends, the accumulation would be less than $250,000.
During these 122 years investors did collect about $90,000 in dividends. If these dividends are simply added to the price appreciation, the total would be about a third of a million dollars. But this total is still tiny compared to the wealth accumulated when dividends are reinvested. In terms of annual returns, the numbers are equally striking. Without reinvesting dividends, the average annual after-inflation return on stocks falls from 7 percent to 4.5 percent, a drop of over a third.
FIGURE 9.1: CAPITAL GAINS AND REINVESTED DIVIDENDS
Dividends are not just good for the whole market. They are also good for individual stocks.
I examined the record of firms in the S&P 500 Index from its beginning in 1957 to the present. On December 31 of each year I sorted the firms in the S&P 500 Index into five groups (or quintiles) ranked by dividend yield, and then calculated the return over the next calendar year. The first group comprises those firms in the lowest 20 percent of the dividend yields (many paid no dividends, especially in more recent years). The second group comprises the next 20 percent ranked by dividend yield, and so on up to the fifth group, which includes firms with the highest yield. I then calculated the return on each of these portfolios over the next year before I re-sorted the firms again, based on the same criteria.
Figure 9.2 shows how striking the results are. In strictly increasing order, the portfolios with higher dividend yields offered investors higher returns. If an investor started with $1,000 at the end of December of 1957, she would have accumulated $130,768 in an S&P 500 Index fund by the end of 2003, for an annual return of 11.19 percent. If instead she purchased only the highest 20 percent of dividend yielders in each year, she would have accumulated $462,750, more than three times the indexed accumulation. Although the risk of these high yielders was slightly higher than that of the S&P 500, its annual return of 14.27 percent was more than worth the extra risk. If instead, she bought the lowest-dividend-yielding stocks, she would have accumulated only about one-half the sum placed in an indexed fund—a return under 10 percent a year. Even worse for investors, the risk of the lowest-yielding group containing many no-dividend stocks was also the highest of all the portfolios.
Note that the technology bubble of the 1990s caused the return on the lowest-dividend payers, which at that time were almost exclusively firms that paid no dividend at all, to surge and almost equal that of the S&P 500 Index.
But history snapped back and punctured these high-flying tech stocks, sending their prices down to earth. The low-dividend stocks went back to the bottom rung of the ladder, as they had been before the bubble began.
FIGURE 9.2: CUMULATIVE RETURNS TO S&P 500 SORTED BY DIVIDEND YIELD (SOURCE: COMPUSTAT®)
Declining Dividend Yields
The importance of dividends in generating stock returns is not just historical happenstance. Dividends are the crucial link between corporate profits and stock values. Finance theory states that the price of any asset is the present value of all its future cash flows. For stocks, cash flows are identified as dividends, not earnings. Earnings are only a means to an end—and that end is to maximize the cash returns received by investors.
In spite of the overwhelming historical and theoretical support for dividends, the dividend yield on stocks has fallen to record low levels over the past two decades. Figure 9.3 shows the trend. From 1871 through 1980, the average dividend yield on stocks averaged 5 percent. This was 76 percent of the annual real return on stocks over that period.
But the 1980s started the steady downward drift of dividend yields, a drift that became a freefall in the 1990s. At the top of the technology bubble in 2000, the dividend yield on the S&P 500 fell to just over 1 percent and has only modestly recovered since then.
FIGURE 9.3: DIVIDEND YIELDS, 1871–2003
Why has this decline occurred? How could the most important source of stock returns through history suddenly become so unimportant?
There were three reasons. The first was the misguided belief that by paying dividends firms were forgoing valuable growth options that could be financed with earnings not paid out as dividends. The second was the U.S. tax system, which double-taxed the payment of dividends. And the third was the proliferation of management stock options, which shifted management’s attention from paying dividends to pumping up the price of the stock.
Growth Options and Dividends
Those who scoff at the idea of firms paying out their earnings as dividends argue that money used to pay dividends would deprive fast-growing firms of their principal source of capital.
Andy Kessler, a former hedge fund manager and co-founder of a money management firm that invests in technology companies, wrote in a Wall Street Journal article entitled “I Hate Dividends”:
Scan the list of companies worth over $10 billion that pay a high dividend and you don’t exactly come away with the future of America, more like old home week. Duke Energy (5.6%), Eastman Kodak (5%), Ford (4.1%), GM (5.4%), JP Morgan Chase (5.6%), SBC Communications (3.9%), and Verizon (3.9%). Dividends entice investors into debt-laden, slow- or no-growth companies, more likely to cut their dividend, burning investors worse than conflicted research analysts. Run away. They are wearing a scarlet dollar sign. You want yield? Buy a bond …
[D]ividends don’t create economic growth. Failing companies just bribe investors with dividends. Encourage companies with a future to invest in their operations, seeking high returns. If all that mattered were dividends, we … would still be investing in railroad stocks.2
As persuasive as Kessler’s arguments may sound, the hard evidence proves otherwise. I showed in Chapter 2 that on average the returns on older firms surpassed the returns on the newer firms. In Chapter 4 I showed that in nine of the ten industry sectors studied, the firms that were added to the venerable S&P 500 Index underperformed the original firms that were in the index when it was founded in 1957.
Even though Kessler pokes fun at those who invest in the railroads, over the last forty-five years the railroad industry outperformed the industrial sector and the S&P 500 Index. And technology stocks, which pay the lowest dividends, have scarcely been market beaters.
Unfortunately, the growth options that firms sank their money into often turned into money pits. I showed in Chapter 7 that the “capital pigs,” those firms that undertook high levels of capital expenditures, underperformed those that spent sparingly on such capital.
And what of those firms that carry huge hoards of cash waiting to buy other companies to boost their earnings? Not good for investors. A study by researchers from Ohio State University, the University of Pittsburgh, and Southern Methodist University entitled “Wealth Destruction on a Massive Scale” concludes that the shareholders of acqu
iring firms lost a total of $240 billion in the four years from 1998 through 2001: “The large losses are consistent with the existence of negative synergies from the acquisitions.”3 This research is supported by that of Jarrad Harford at the University of Washington, who published an article in the prestigious Journal of Finance concluding, “cash-rich firms destroy seven cents in value for every excess dollar of cash reserves held.”4
Having a large cash hoard in the corporate till is akin to a having pocket full of money—it encourages you to spend. Very often I am asked if a particular company is attractive because it is selling below the value of cash on its balance sheets. I reply, “Watch out!” The stock is only a buy if you can take control of the company and disgorge the cash to shareholders. That used to be a favorite technique of Warren Buffett’s mentor, Benjamin Graham, who went after companies that were not employing their cash in the interest of shareholders. If you cannot take control quickly, I would stay away. Management may squander much of that cash in the next few years.
Warren Buffett and Berkshire Hathaway
I can visualize the Warren Buffett fans shaking their heads at all this emphasis on dividends. As a great admirer of the man, I feel it is important to explain why his no-dividend, cash-hoarding policy works so well for his firm.
Buffett’s investment vehicle, Berkshire Hathaway, paid a single ten-cent dividend in 1967 and has never paid one since. Buffett, who has always opposed paying dividends on his stock, maintains that the directors of the firm voted the dividend in Berkshire’s early years while he took a bathroom break during a board meeting.5 Despite the lack of dividends, Berkshire Hathaway has been one of the best-performing stocks over the last four decades.
Taxes are a major reason why Berkshire doesn’t pay a dividend. In an interview with Ted Koppel on Nightline in 2003, Buffett said, “If we could pay dividends out tax-free, it would be good for [our shareholders].”6 Buffett rightly claims that even with the recent reform of the U.S. tax code, which I will discuss below, there is a tax incentive for firms to create capital gains, which are taxed only if and when investors sell their shares.
Dividends, by contrast, are taxed automatically upon receipt. As a result, if a firm can profitably reinvest its earnings, by either expanding its own operations, buying other firms, or repurchasing its own shares, then the tax that an investor would have to pay is deferred to the future.
But it is far from certain that most firms can profitably reinvest all their earnings. As noted above, all too often management spends the earnings in a way that reduces instead of increases the return to investors. Managements that are flush with cash often lose control over costs, spend money on perquisites, or engage in empire building. Economists call these wasteful expenditures “agency costs,” and they exist to some extent in all organizations where the owners are not the managers.
Admirably, Warren Buffett is one of those very rare managers who has the discipline, will, and incentive to avoid all these pitfalls. By aligning himself so closely with shareholders (virtually all of his own enormous wealth is invested in Berkshire), he acts directly in his and his shareholders’ best interest.
Buffett is extraordinarily forthright with the public, revealing to all exactly what is going right, what is going wrong, and how his earnings and capital are spent. He makes no attempt to sugarcoat bad news or pump the price of his stock. If Buffett thinks there are no attractive investment opportunities, he will refrain from investing even as substantial cash flows into his company. He regards holding cash as a valuable option that enables him to pursue special opportunities if and when they appear in the market.
Holding cash in lieu of paying dividends to shareholders makes sense with Buffett’s strategy. During periods when financing is hard to come by, some of the best deals can be made by those who have the cash to lend or to buy with. Buffett is particularly good at spotting those opportunities. In addition, if he thinks that his own shares are underpriced, he will buy them in the open market through share repurchases, as he considered doing in 2000 when Berkshire A shares were trading below $45,000.7
Furthermore, Buffett’s investment strategy focuses on buying stocks or businesses that generate healthy cash flows, the prerequisite of dividends. “Our acquisition preferences run toward businesses that generate cash, not those that consume it,” he stated in the company’s 1980 annual report. And in his 1991 report he reiterated that he searches for businesses with “demonstrated, consistent earnings power (future projections are of little interest to us, nor are ‘turnaround’ situations).” In other words, Buffett’s own investment objective of buying companies with consistent cash flows at reasonable prices mimics investors who reinvest their dividends.
If the management of other companies had the same close relationship with their shareholders that Buffett has, the importance of paying dividends would be significantly reduced. But often, the difference between the goals of the shareholder and those of management can be substantial. The commitment to pay earnings back to shareholders in the form of dividends reduces the potential for management to squander shareholders’ wealth.
Dividends and Corporate Governance
If management always acted completely in the interests of shareholders, dividends would not be important. But for the rest of the corporate world, dividends are a critical ingredient in generating trust between shareholders and management and confirming management’s statements about earnings. I realized how crucial this trust is when the Enron crisis rocked the confidence of the stock markets in the fall of 2001.
The story of Enron is fascinating. The firm rose from an obscure Houston-based natural gas pipeline operator to become the world’s largest energy trader. At its market peak in August 2000, Enron was one of the fifty largest firms in the United States, with a total market value of almost $70 billion, ahead of such giants as General Motors, Ford, and Chevron.
Enron was praised as an example of how old-line firms could adapt to the new market-based world of energy distribution. For six years in a row, Fortune ranked Enron near the top of its “Most Innovative Companies” and named it one of the five most admired firms in America. It was the prototype new-economy firm.
But this new-economy darling was also cooking its books, portraying steadily rising earnings while hiding its debts in off-balance-sheet entities. And investors were soon to learn that Enron was not an isolated example of earnings deception and manipulation. Other firms that had won the plaudits of investors, such as Tyco, WorldCom, Adelphia, and HealthSouth, also falsified their earnings. Even some of the earnings of the bluest of the blue chips, General Electric, were being called into question.
Despite the proliferation of earnings scandals, the vast majority of corporate management was not losing its moral bearings. I realized that the real problem was that the most basic source of stock values, dividends, had fallen victim to investors’ undue focus on short-run performance, the U.S. tax system, and excessive issuance of management stock options. These were the true culprits for the crisis in investor confidence.
The Dividend Deficit
I put my concerns in an article published in the Wall Street Journal in February 2002, entitled “The Dividend Deficit.” I pointed out that history provides us with important lessons about the sources of shareholder value.8 In the nineteenth century there was no Securities and Exchange Commission, Financial Accounting Standards Board, or any of the other numerous agencies that oversee and regulate our securities markets today. A firm released whatever information it liked whenever it wanted, and management didn’t fret that it would be taken to task for reporting a dubious number.
Given the total lack of standards back then, how did a firm signal that its earnings were real? The old-fashioned way, by paying dividends—an action that gave tangible evidence of the firm’s profitability and proof that the firm’s earnings were authentic. If there were no dividends, then the stock’s value depended on trusting management’s earnings reports. If there was no trust, there
was no reason to buy stocks.
In the article I fingered the U.S. tax system and the excessive issuance of stock options, particularly to top-level management, as the principal causes of the decline in dividends.
Taxes on Dividends
The role of our tax system in discouraging dividends is clear. At the time I wrote the article, dividends were fully taxable to the individual and not tax deductible to the corporation. In contrast, almost all foreign countries exempted some or all dividends from personal taxation, since they were already taxed as part of corporate profits.
I called on legislators to eliminate the double taxation on dividends, and I set forward my proposals for dividend tax relief in my article entitled “This Tax Cut Will Pay Dividends,” that appeared in the Wall Street Journal on August 13, 2002. Because of my previous advocacy of tax relief for dividends, I had been invited to President Bush’s Economic Summit held on that very same day to discuss my proposal with other economists inside and outside the administration.
I proposed allowing dividends paid to stockholders to be tax-deductible to corporations, just as all interest paid to bondholders is tax-deductible. This would put all corporations on the same footing currently enjoyed by real estate investment trusts (REITs), mutual funds, and Subchapter S corporations.
The Future for Investors Page 14
