The investment value of a stock is the present worth of all future dividends to be paid upon it . . . discounted at the pure [riskless] interest rate demanded by the investor.
Before his final doctorate oral exam, Williams did something seemingly foolish: he sold the rights to his dissertation before he had been granted a degree.
John Burr Williams was no ordinary student. He was a successful Wall Street investor. He had gone back to Harvard to learn what caused the 1929 stock market crash and subsequent Great Depression.
Since he was attending more for the knowledge than the degree, Williams wanted to share his findings as soon as possible. Harvard University Press, who had purchased John Burr Williams’ dissertation, published it as a book called “The Theory of Investment Value.”
Through that book, John Burr Williams’ ideas would rock the foundations of the investing world for decades to come.
The Harvard faculty was irate that Williams had published his dissertation early—without their approval. Not only had he subverted the traditional process, but his ideas went against most academic theory. Williams’ ideas were such a radical departure from existing investment thoughts that he didn’t think Harvard would ever award him a doctorate. His troubles with the dons of the school of economics were centered in two areas:
1. Williams claimed that the correct way to determine a company’s intrinsic value was to calculate the present value of its future dividend payments. The universal belief at the time was that earnings were the real driver of stock prices; dividends were an afterthought.
2. He voiced skepticism of John Maynard Keynes’ theories and the socialist programs of President Franklin Roosevelt. Williams disagreed that government policies led to prosperity; he believed government mismanagement was partly responsible for the 1930s depression.
It wasn’t until his book gained accolades from the financial community that John Burr Williams was finally granted his doctorate in 1940.
Academics weren’t the only group to scoff at Williams’ ideas. His revolutionary ideas about valuing stocks with dividends offended the very group they most sought to benefit: Wall Street itself.
John Burr Williams said that thinking was wrong. But he didn’t stop there; he went on to blame Wall Street, at least partly, for the stock market crash:
The wide changes in stock prices during the last eight years, when prices fell by 80% to 90% from their 1929 peaks only to recover much of their decline later, are a serious indictment of past practices [of Wall Street]. Had there been any general agreement among analysts themselves concerning the proper criteria of value, such enormous fluctuations should not have occurred, because the long-run prospects for dividends have not, in fact, changed as much as prices have. Prices have been based too much on current earning power, too little on long-run dividend-paying power.
Williams’ ideas were rejected, even ridiculed, by Wall Street elites. Over time, however, they steadily gained a following. One of his students was Arnold Bernard, founder of perhaps the most famous of all independent investment research firms: The Value-Line Investment Survey. Bernard echoed Williams’ concerns in his 1959 book:
Williams postulates that the value of a stock is the sum of all its future dividends discounted by the present interest rates. . . . Because there is no generally accepted standard of value, the market prices of stocks fluctuate far more widely than their true values. The wide fluctuations have in the past imposed a heavy burden on the general economy and undermined the faith of many people in the free market economy. The need, therefore, exists for rational and disciplined standards of value that cannot lead to the wildness of 1929 or 1949 or the present.
Bernhard and Williams pointed out the need for generally accepted criteria to value stocks. The effect of not having such criteria resulted in excessive stock market volatility, which damaged investor confidence in the stock market, but also in the free-market system that encouraged global economies to thrive.
Unfortunately, there has been little progress in a standard method of valuing stocks over the decades. That was clear in the 1987 stock market crash, the 1995-2002 Technology Bubble, and housing bubble of 2008-09. Time will tell whether our current 2022 bear market joins the list of volatility-induced panics and subsequent economic crises.
To both John Burr Williams and Arnold Bernard, the valuation method that most promoted market and economic stability was dividends:
In our own experience, during periods of inflation as well as at other times, in this country and abroad, it has been found that dividend-paying ability is the final determinant of the price of a common stock. Whenever, over a period of years, the dividend, or the ability to pay dividends, went up, so too did the price of the stock. When the dividend-paying ability went down, so did the price of the stock, inflation or no inflation.
In applauding Williams’s theory, however, Bernhard inserted a subtle twist to Williams’s basic premise by adding the words, “the dividend, or the ability to pay dividends.” By adding just these few words, he re-entered the world of earnings and left behind the “dividends only” world that Williams described as so important in determining long-term intrinsic value.
Warren Buffett, the most famous investor of modern times, credits Williams’ rationale for intrinsic value. But, like Bernhard, he also makes a subtle twist:
The value of any stock, bond, or business today is determined by the cash inflows and outflows—discounted at an appropriate interest rate—that can be expected to occur during the remaining life of the asset.
Neither Buffett nor Bernhard nor many of the thousands of others who have quoted John Burr Williams over the years is saying the same thing. Williams was speaking of dividends alone, not earnings, cash flow, or a combination of the two.
Williams went so far to keep dividends at the center of his methodology that he included a chapter in his dissertation titled, “A Chapter for Skeptics.” There he explained that he was aware that without earnings and cash flow there would be no dividends, but he steadfastly asserted that earnings and cash flows mattered only if you owned the whole company. In response to arguments for earnings or cash flow instead of dividends, he offered the following:
Earnings are a means to an end, and the means should not be mistaken for the ends. In short, a stock is worth only what you can get out of it.
He then added the following poem:
Even so, the old farmer said to his son:
A cow for her milk, A hen for her eggs, And a stock, by heck, For its dividends.
An orchard for fruit Bees for their honey, And stocks, besides, For their dividends.
The old man knew where milk and honey came from, but he made no such mistake as to tell his son to buy a cow for her cud or bees for their buzz.
We began using a dividend-oriented investment style in the early 1990s as a higher-yielding alternative for bonds. Our bond investors had experienced a precipitous fall in interest income during the prior years and needed to generate income to pay their bills in retirement.
We relied mostly on high-yielding, slow-growing companies because we did not have the capability to calculate the present values of lower-yielding, faster-growing companies. When we discovered Williams’ ideas, it unlocked a whole new world for us.
We constructed our first multi-stage, dividend discount model of the type that Williams described. Even with the more sophisticated models, we could not produce reliable intrinsic values for companies that paid little or no dividends.
Since almost all technology companies and many other high-growth companies paid no dividends, we joined so many before us to perform what we now see as a regrettable twist of Williams’s principles: We changed our model from a dividends-only model to what we called a Modified Cash Flow model, which we described in our January 1997 quarterly letter.
First, we were delighted that our new valuation tool could calculate the intrinsic value of almost any company. But, as with many one-size-fits-all models, weaknesses soon appeared.
As the market began its tremendous ascent in the late 1990s, we realized that even among companies that were paying a generous dividend, few were paying at the level of our standardized projections, which were based on historical trends. Dividends had fallen out of favor among many corporations, who were opting for share buyback plans. We had to make continual adjustments to re-calibrate our model to the widening gap between earnings growth and dividend growth.
Many companies were also taking massive write-offs. These write-offs were reported as one-time events, but they lowered the past years’ earnings. Since our model was based upon the adjusted earnings, which excluded the write-offs, our model assumed companies could pay larger dividends than reality.
At the beginning of 2000, we realized that we had come face to face with the essence of the old farmer’s poem.
Through a subtle and innocent twist, we were now investing in the “bees for their buzz,” instead of the “bees for their honey.” Williams believed that earnings are a best guess, but always subject to change. Dividends, on the other hand, are real money consciously and publicly distributed to shareholders. Once paid, they are not subject to restatement.
Dividends can change, of course, but investors expect these cash payments in perpetuity, so boards of directors reduce dividends only as a last resort. And, since they are so reluctant to reduce dividends, they are reluctant to increase them unless they have confidence those dividends reflect sustainable earnings power.
We decided to use the Modified Cash Flow model only for companies that did not pay a dividend. For those that did, we would use a dividend-only model. We also decided to cut back on stocks that paid little or no dividends. This shift back to the actual dividends in our models enabled us to determine intrinsic values with much more confidence. This helped us more accurately value companies that were paying dividends.
At first, we were concerned that this dividends-only approach was too narrowly focused. It excluded the non-dividend-paying, high-growth stocks that were doing so well at the time. The narrowing of our focus turned out to be a blessing. During the bear market of 2000-2003, dividend-paying stocks performed much better than non-dividend paying stocks.
Unfortunately, Williams’ ideas about market volatility have not been remedied. If anything, they have only accelerated. Today’s technology allows stocks to be traded in seconds; many of these trades are now entered by computers rather than humans.
What Williams observed about the 1929-1937 Great Depression and market still applies today:
Such enormous fluctuations should not have occurred, because the long-run prospects for dividends have not, in fact, changed as much as prices have. Prices have been based too much on current earning power, too little on long-run dividend-paying power.
His words have proven true over the last century. Dividends tend to be much more stable than earnings. Since 1928, the S&P 500 earnings have been nearly 3x more volatile than dividends[1]. This is particularly true in times of crisis. In 2008, for example, earnings fell by 77.5% while dividends increased by 2.3%.
Most investors pay attention to quarterly earnings, so the volatility of stock prices tends to follow earnings. We think focusing too much on earnings is a mistake. A portion of earnings must be reinvested into the business to generate future growth. Therefore, a portion of earnings is not available to be paid out to shareholders and should not be considered as the basis for value.
We are not suggesting non-dividend paying companies have no value. Companies that responsibly earn and reinvest their earnings at high returns on that capital are able to compound faster than those distributing a portion of that to shareholders. However, the basis for the value of any business—publicly traded or privately held—is the cash they will return to shareholders in the future discounted back to the present.
Consider a company like Netflix, which shows positive earnings per share. If we base the value of Netflix on its earnings, we inevitably overvalue the business. Why? Because the entirety of Netflix’s earnings must be reinvested back into creating new content to maintain its subscriber base and, ultimately, drive earnings.
The case for investing in Netflix, therefore, must be with the eye toward a future when only a portion of its earnings must be reinvested in new content with the remainder available for returning to shareholders.
If that never materializes, then the only investors who will have made money on Netflix are those who successfully sold its stock at a profit. The rest, inevitably, will have paid something for a business that ultimately went to $0. Without a dividend, how else could investors—in aggregate—have made any money on it?
Let’s assume, for a moment, that Williams was correct in saying that the value of a stock is its future dividends discounted back to today. If that is true, then a stock’s price over the next ten years should grow at a similar rate to its dividend. That would have two major implications for you and investors, as a whole:
First, it means that stocks should produce total returns equal to the long-term dividend growth rate plus the starting dividend yield. If long-term dividend growth is 5% and starting dividend yields are 2.7%, investors would earn a total return of 7.7%. Investors who buy a 10-year US Treasury bond today will earn the same starting income rate, but they will not have the growth; 2.7% annual return will be the fixed return for the life of the bond.
Second, you now have a way to make money in the stock market without relying on price growth. If you invest in companies that pay dividends in the future, stocks no longer become flashing numbers on a screen; they are no different than real estate properties that generate rent, farmland that generate crops, or bees that generate honey.
These reliable income streams come regardless of what happens to the price of the building, the land, or the hive.
Don’t make the same mistake that the farmer made. Cows are not for cud and bees aren’t for their buzz. Buy cows for their milk, bees for their honey, and stocks for their dividends.
Original article written by Greg Donaldson & Mike Hull, April 15, 2004.
Updated by Nathan Winklepleck, July 25, 2023.
[1] DCM calculation. Data source: Multpl.com
This report was prepared by Donaldson Capital Management, LLC, a federally registered investment adviser under the Investment Advisers Act of 1940. Registration as an investment adviser does not imply a certain level of skill or training. The oral and written communications of an adviser provide you with information about which you determine to hire or retain an adviser. Information in these materials are from sources Donaldson Capital Management, LLC deems reliable, however we do not attest to their accuracy.
An index is a portfolio of specific securities, the performance of which is often used as a benchmark in judging the relative performance to certain asset classes. Indexes are unmanaged portfolios and investors cannot invest directly in an index. An index does not charge management fees or brokerage expenses, and no such fees or expenses were deducted from the performance shown. Past performance is not a guarantee of future results. The mention of specific securities and sectors illustrates the application of our investment approach only and is not to be considered a recommendation by Donaldson Capital Management, LLC.
S&P 500: Standard & Poor’s (S&P) 500 Index. The S&P 500 Index is an unmanaged, capitalizationweighted index designed to measure the performance of the broad U.S. economy through changes in the aggregate market value of 500 stocks representing all major industries.