Seeking Alpha
Dividend growth investing, retirement
Profile| Send Message| ()  

Background

OK, we have got five more myths to discuss. This has become a popular exercise. The first three articles on this subject have garnered more than 1,100 comments altogether.

Before getting started on the latest five myths about dividend growth investing, let's review the first 15. In my first myth article, back in 2012, we considered these five falsehoods:

1. The only sensible goal in investing is total return.

2. Dividend growth investors don't care about total return.

3. Income from "selling a few shares" is the same as dividends.

4. Dividend growth investors think they have found a free lunch.

5. Each dividend growth stock must maintain its performance for 30 years.

A short while later, in the second article, we looked at five more:

6. You need more money to live off of income than withdrawals in retirement.

7. Dividend growth investors don't care about current yield, they only care about yield on cost.

8. Dividend growth stocks are all large-cap U.S. stocks concentrated in a few industries, therefore a dividend growth portfolio lacks diversification.

9. The S&P 500 is a good proxy for a dividend growth stock portfolio.

10. Dividend growth investors select stocks based only on yield; dividend growth investing = high-yield investing.

I honestly thought that the subject was exhausted after the first two articles. But by early last year, we had rounded up five more:

11. Dividend growth investing does not involve analyzing individual stocks.

12. You can replicate a good dividend growth strategy with ETFs.

13. Stocks with high yields have low dividend growth rates, and vice-versa.

14. Younger investors should take on more risk, because they have more time to make up losses.

15. Broad statistics are helpful in assessing individual stocks.

You can't make this stuff up. Every myth is something that I have seen in print, often from respected advisors, pundits and major firms. In the past year, five more myths have made their appearance. Here they are.

Myth #16. Dividends do not explain returns.

This is a mantra of Modern Portfolio Theory (MPT). To understand this myth, it helps to have a brief background in MPT.

MPT "explains" returns by examining "factors." As more and more research has been done, the following factors have received the imprimatur of academic approval, and they are now widely accepted as factors that explain returns.

  • Beta, which means the market itself. This factor explains most returns according to MPT. In the early days (about 50 years ago), it was the only factor recognized. This was the source of directives that exist to this day, such as "setting your asset allocation is the single most important decision you can make as an investor. [F]or the average investor, the asset allocation mix they choose…accounts for 100% of their return level." The idea is that the market determines the returns of each asset class, so whatever instruments you buy do not matter much. Just look for low costs and proper coverage of the asset class.
  • Value. In MPT parlance, value is revealed by ratios like p/b (price to book ratio), p/e (price to earnings ratio) and p/d (price to dividend ratio, which is the inverse of what most of us call yield). As value investors understand, better-valued stocks have a greater likelihood of future price appreciation. Before this factor was accepted academically, the success of someone like Warren Buffett was an unexplainable anomaly. Or as David Fish has put it, his success was a "nomaly." In MPT, anything unexplained by MPT is an anomaly. It is either the result of luck or of factors yet to be discovered.
  • Size. In a nutshell, small-cap stocks outperform large-cap stocks over time. The three factors listed so far comprise the famous Fama-French three-factor model, which was the state of the art about 20 years ago.
  • Momentum. Price trends tend to persist. Adding this produces the Carhart four-factor model.
  • Profitability. This has been added just in the last couple of years. While most stock investors know instinctively that profitability probably has something to do with a company's performance, it has only recently gained academic approval as a factor that explains returns.

With these five factors, academics have well over 90% of returns explained. This being the case, they maintain that dividends do not add any additional explanatory power to returns. In fact, they treat dividends as just one of several valuation ratios (p/d), and not the best one at that. So in the end, it is said that p/d is a poor value screen. The other valuation ratios are better, because they have more explanatory power.

There are a few problems with this line of reasoning. First and most obvious, dividends don't explain returns, they are returns. As David Fish put it recently:

I prefer the Two-factor model that explains 100% of returns 100% of the time: Total Return = Price Change + Dividends

That fundamental equation seems to be ignored in MPT. MPT deals only in total returns, as the thought is that the only return investors should be interested in is total return (Myth #1). Let's just accept that point for a moment, that you should only be interested in total return. Even if that were true, why would you dismiss one of the two major components of total returns? Most studies show that in an average year, dividends account for 40% of total return (price changes account for the other 60%). That seems like a big factor to ignore.

Now let's move beyond thinking that total returns are the only returns worthy of consideration. Many investors, myself included, are more interested in income - its generation, reliability and growth - than in sheer wealth. That does not mean that we reject wealth (Myth #2), but rather that our main focus is on the income that our assets provide.

Now we are in a position to turn this myth on its head. Instead of "Dividends do not explain returns," the proposition should be, "MPT does not explain dividends." And in fact, that is true. MPT consigns dividends to irrelevancy. In fact, there have been several academic studies trying to answer the puzzling question of why some investors irrationally are interested in dividends. Psychological factors are widely believed to be the culprit.

In consigning dividends to irrelevancy, or labeling an interest in dividends as irrational, MPT consigns the central goal of a significant number of investors to the junk pile.

Myth #17. The reduction in share price prior to ex-dividend day is permanent.

The following chart shows Johnson & Johnson (JNJ) over the past five years. During that time, they have paid 20 quarterly dividends and increased their payout rate five times.

Notice the reductions in JNJ's price four times per year when they pay the dividend and especially once per year when they raise it. Can't see that? Neither can I.

Modern Portfolio Theory says that the price of JNJ is permanently reduced by each dividend it pays. Over the course of the five years depicted, they have paid a total of $11.28 per share. According to MPT, that means that if they had not paid those dividends, JNJ's price would be $101.89 rather than $90.61. The reason that its actual price is lower is because each dividend permanently reduced JNJ's price by exactly the same amount.

JNJ's current price-to-earnings ratio (P/E), according to Morningstar, is 20.2. If JNJ's price were $101.89, its P/E ratio would be 22.7. Let's say that we went back 10 years instead of five. Its price would be even higher and so would be its P/E ratio at that price. Let's go back to the beginning of time and add all the dividends JNJ has ever paid to its current price. Its price would be astronomical and so would its P/E ratio.

MPT would say that a dividend payer will always have a lower price than it would have if the dividend had not been paid. The amount of the difference equals the total of all dividends paid out, and the gap will keep growing as more dividends are paid out over time.

Does that make sense? No, of course not. The market determines price, not an equation, so a better explanation of price comes from thinking about what motivates investors. How does the market decide what a share of JNJ is worth?

A far better explanation than the dividend-reduction theorem is that investors value a company based on its earnings and prospects for earnings growth. That is what Chuck Carnevale demonstrated in his article, Dividends Provide a Return Bonus. He said there:

[T]he market will value a given company's earnings based on their past and future prospects for growth,…regardless of whether a dividend is paid or not. …If you examine two companies with equivalent rates of earnings growth, where one pays a dividend and the other does not, the dividend payer will provide their shareholders a higher total return. In other words, we're suggesting that both stocks will provide equivalent capital appreciation when measured over a time period when ... both stocks were being priced at fair value in the beginning and in the end. Consequently, the stock that pays a dividend to its shareholders is providing them a return bonus or kicker.

Of course, an MPT-er might say that is believing that dividends are a free lunch (Myth #4).

An MPT-er might also say that JNJ's earnings would be much higher if it had retained all its earnings for all of those years, so its P/E ratio would still be 20.2 at a much higher price.

But an academic paper actually refutes this. In "Surprise! Higher Dividends = Higher Earnings Growth," the authors stated this:

The historical evidence strongly suggests that expected future earnings growth is fastest when current payout ratios are high and slowest when payout ratios are low ... Our evidence thus contradicts the views of many who believe that substantial reinvestment of retained earnings will fuel faster future earnings growth. Rather, it is consistent with anecdotal tales about managers signaling their earnings expectations through dividends or engaging, at times, in inefficient empire building.

In other words, dividends are not a free lunch. Rather, they often come from companies that are extremely good at allocating capital.

In addition to the findings of that paper, ask yourself whether it makes common sense that a company as established as JNJ could just increase its earnings proportionately if it simply retained all of its earnings rather than paying some out as dividends. If that were the case, it seems that is what JNJ would do. Why don't they?

It is far more likely that JNJ has found a good balance between retaining some earnings to fund the growth projects that make the most sense, and releasing another portion back to shareholders as dividends. JNJ has increased its dividend for 51 consecutive years, and one gets the feeling that they know how much to retain to fund growth and how much to pay out as dividends.

I will let you decide whether the MPT argument makes sense. I find it much more believable that the market prices stocks according to earnings and business prospects, and that price reductions at dividend time have no permanence.

Myth #18. Receiving a dividend is exactly like selling a share.

This is one of those myths where you can sort of see what is meant, but because it is so beguiling, it becomes dangerous, because it is flat-out wrong.

This myth says that a dividend is a disinvestment in the company in the same amount as the dividend received. In other words, a shareholder who receives a dividend is in exactly the same place as a shareholder who sells part of his stake in the company.

The thinking is this: When a company sends you a dividend, the company is worth less, namely its cash has shrunk by the dividend it sent out. That (the thinking goes) is reflected in its value. The math is simple: A company worth $100 per share sends out a $1 dividend. Now the company is worth $99 plus you have $1 in cash. Your total holding is still $100. It's as if you sold 1/100th of a share. Your overall position has not changed at all. You have the cash from the dividend, but you have an equivalent reduction in the value of the shares that you own.

If the price reduction were permanent (Myth #17), this would have some validity. But the whole concept ignores the impact of the market on the price of a stock. As we saw in #17, it is next to impossible to identify any reduction in price that corresponds to dividends being paid. To assert that the impact on price is permanent is just that, an assertion. It is theoretical. Academics think that because it is what should happen (according to their theories), it is what does happen.

What actually does happen was explained by Chuck Carnevale in the article quoted earlier. Market participants price stocks over time according to earnings and earnings expectations. It does not matter that some of those earnings are distributed as dividends.

That's why in FASTgraphs, Chuck stacks dividends on top of earnings. Dividends are a separate source of return, in addition to price changes, just as shown in the fundamental equation earlier.

The aqua area represents dividends paid. They are in addition to the price. (The aqua area could just as well be stacked atop the black actual-price line rather than the orange earnings-justified price line. That would show total return exactly at any point in time.) Investors receive the dividends in addition to, and independently of, the price they receive if they sell.

The latter fact is what makes the myth dangerously misleading. If you believe the myth, you may then also believe that selling shares as a source of "income" is the same as receiving dividends.

But they are not the same. If you sell shares, you have fewer shares. Is that not obvious? Since dividends are declared per share, the fewer shares you own, the fewer dividend dollars you will receive.

You can still sell the stock at any time to convert the price into cash. The number of shares you own times the price of each share equals the cash you will get from the sale. That is obvious too. So the fewer shares you can sell (because you no longer own them), the less they are worth to you.

Therefore, it seems clear that the number of shares you own is important. That number determines how much you will receive in dividends (if the stock pays dividends), and it also determines how much you will get when you sell your stake.

If you have been selling off shares a few at a time to create cash flow, your future ability to keep doing that will keep declining. At some point, you may run out of shares to sell.

Myth #19. Dividend stocks are all overvalued or in a bubble.

In compiling this year's Top 40, I found a fair number of well valued stocks. If you only think of stocks in groups or sectors, you will never discover individual bargains. Whether an index, country, sector or industry is "expensive" is not relevant to whether individual stocks are good buys or not.

Part of the way that I value individual stocks is through Morningstar's star system. Their system compares the stock's current price to their calculation of its fair value. Their calculations are among the most well-conceived that I have encountered, and I have a great deal of respect for the way they value stocks. They convert their valuations into a simple star system, where five stars means greatly undervalued and one star means greatly overvalued.

It is true that some popular dividend growth stocks are overvalued right now. Two stars means overvalued. Examples of popular dividend growth stocks that are two-star stocks right now are Altria (MO) and Genuine Parts (GPC).

But four-star stocks are undervalued. Examples of four-star stocks at the moment are Baxter International (BAX) and Philip Morris (PM).

The point is, the blended valuation of large groups of stocks, or sectors, or indexes, is important only if you are planning to invest in those groups, sectors or indexes. If you are planning to invest in one stock at a time, it is that stock's individual valuation that is important.

Myth #20. You can only spend total return.

This is sort of like Myth #1. But when it is stated in the way it was here, it becomes so jaw-droppingly inaccurate that it deserves its own recognition.

I feel a little foolish even explaining this, so please forgive me. I will keep it short.

Dividends are delivered to you (or your account) as cash. You can do whatever you like with the cash, including spend it.

That's the end of the explanation. But this might be a good place to remind that in order to spend the capital-gain portion of total return, you must first unlock it. The only way to do that is to sell it. Once you've converted an asset to cash, you can do anything you like with that cash too, including spend it. But there is no way to convert it except by selling it.

When you receive a dividend, you get to spend it and keep the shares too. What those are worth will be determined by the market, not by Myth #17. In any event, you can obviously spend dividends.

Source: Dividend Growth Investing: Myths 16-20