Last night, I was reading the forum on bogleheads.org that discusses the advantages of index investing over individual stock picking (as you might guess, the website is named after Jack Bogle, the founder of the low-cost index fund provider Vanguard). Even though I currently own no bonds (besides the occasional savings bond I received as a gift growing up) and do not currently own any index funds, I enjoy reading the intelligent arguments from people that reach the exact opposite conclusions that I do. It's one of the best ways to uncover potential holes in my own thought process.
Anyway, when I was reading one of the forums discussing the advantages of an index strategy over a portfolio of dividend stocks, I came across one comment that I believe articulates a prevalent yet misguided comprehension of whether a portfolio stuffed with cash-generating common stocks works:
"But dividend-based strategies crop up every so often for purely non-rational reasons (not unlike the Permanent Portfolio but for different reasons)…
Dividend stocks, like every other security "characteristic" there is, will do better than average from time to time in an almost completely random and unpredictable fashion. So after a few years of this, performance-chasing investors will hop on dividend-based strategies believing their outperformance will continue."
First of all, I agree with the general notion that, in the short term, the difference in stock performance between a company that pays a dividend and one that pays no dividend is completely unpredictable. I have no idea whether Johnson & Johnson (JNJ) will outperform Google (GOOG) in 2014, or whether Coca-Cola (KO) will outperform Amazon (AMZN) in 2015. Since stock prices only represent what other people are willing to pay for a given share at a given point in time, I have no ability to make predictions about what other people will think about a given stock in a short period of time.
But over long periods of time, things get more interesting. Over 10-20 year periods, the price of a stock quote is much more a reflection of the underlying business performance than it is a reflection of the whims of market participants. Whether you will outperform, underperform, or perform in line with the S&P 500 hinges upon the price you pay for a particular company, and its operational results thereafter until the end of your holding period (provided stock prices are loosely rational at the time you decide to sell).
I hope none of you reading this article agree with the argument that "dividend stocks will do better than average from time to time in an almost completely random and unpredictable fashion."
First of all, the classification of "dividend stocks" does not really mean anything by itself. There is no such thing as this "abstract dividend strategy" in existence, but rather, there is such a thing as unique, highly individualized portfolios that consist of cash-generating assets hopefully purchased at attractive prices. This distinction may seem trivial, but it is not. If I buy an apartment complex for $1 million that only generates $25,000 in annual rents, it is not proof that a real estate strategy does not work. Rather, it is proof that paying too much for a particular real estate asset does not work. Likewise, if I paid $500,000 for an apartment complex that was generating $50,000 in annual rental income with a reasonable likelihood of adding tenants, it is not proof that a real estate strategy is superior. Rather, it is a piece of evidence that demonstrates the success of buying a lucrative cash-generating asset at an attractive price.
I mention this for one reason: there is no "arbitrariness" or "randomness" to whether a dividend-paying security will do well over the long term. It is all about determining each stock on an individualized basis in relation to the future profits it will generate (in addition to some other not insignificant side factors such as debt loads on the balance sheet and change in earnings quality).
I'll give a couple of examples. Right now, Hershey (HSY) is approaching a valuation of 30x earnings. That is the company's highest valuation since the dotcom boom. Value Line predicts a long-term P/E ratio of 19. You don't need me to tell you that Hershey is an excellent, stable cash cow. Twizzlers, Ice Breakers gum, Reese's, Kit Kats, Milk Duds and Hershey's Kisses will likely be with us for some time to come. But that is not the point. As Professor Benjamin Graham pointed out, the price you pay determines your future return. Sure, the company could grow earnings at above-average rates over the next 10 years. But if the company delivers the kind of growth over the next 10 years that it gave shareholders over the past 10, it should not surprise anyone if Hershey underperforms the market.
Likewise, when you look at something like Royal Dutch Shell (RDS.B), you will see that the company is trading at about 7-7.5x earnings, a decent amount below its historical range of 9-10x earnings. Plus it pays a $3.44 annual dividend (currently offering a 5.34% yield at present market prices) that grows as long as oil prices do not spike downward (when oil prices go down 20-25% or more, the company has a track record of freezing or cutting the dividend by a commensurate amount).
If oil prices rise or stay at the current levels for the next five years, I would expect Royal Dutch Shell to outperform the other S&P 500 components (fun fact: if you are familiar with the research of Dr. Jeremy Siegel, you know that Altria (MO) has been the best-performing American stock in the 20th century because it offered investors high growing dividends and a depressed stock price. This same phenomenon explains what has allowed Royal Dutch Shell to compound at 10% from the summer of 2004 to the end of 2012. It's all too easy for a lot of investors to ignore a stagnant stock price and a 5% dividend that gets habitually reinvested at 7-10x earnings).
And to give a final example, General Electric (GE) is my largest stock position after Johnson & Johnson. From what I can tell, the company is neither cheap nor expensive. Historical P/E ratios are of limited use with this company because it has gone from crazy overvaluation during the dotcom days to the low of $6 during the financial crisis, but a long-term P/E ratio of 15-17x earnings seems reasonable for a large, industrial conglomerate. That means that my expectations going forward are equal to the operational results of the company. If the company grows 7% annually, that will be my annual return over the long term. If it grows by 11% annually, that will be my total return over the long term.
The point is this: "Dividend stocks" are not some abstract classification that can be proven to either work or not work. It all hinges upon specific companies, and the specific price that you pay to acquire ownership. Likewise, these companies are not lottery tickets that move in an arbitrary and random direction over the long term. It does not take a genius to figure that paying $19 per share for Coca-Cola stock in 2009 would lead to very nice returns come 2019. Likewise, Brown Forman (BF.B) has been raising its dividend every year since before I was even alive, but it also trades at 24-25x earnings. Paying that high valuation today has consequences. If the total returns of the company lag the overall earnings growth of the firm, it should not shock anyone. With a lot of these companies, the classification as a "dividend stock" has little to do with whether or not the strategy proves lucrative. It almost always comes down to the price paid for each security, relative to future cash flows.