Seeking Alpha
Long-term horizon, dividend investing, long only
Profile| Send Message| ()  

Fairly often investors who are chiefly concerned with a growing stream of income quibble over which is more desirable: a higher initial dividend yield or a higher dividend growth rate. Often the deliberation comes down to one's time horizon. For example, if you have a shorter investment time horizon then it is expected that you might prefer a higher yield. On the other hand, if you have decades on your side it's presumed that the advanced rate of payout growth will eventually win out. It seems logical enough. In fact, I have previously commented on both "quantifying the yield vs. growth tradeoff" and "finding the sweet spot." Yet I think I can add another lens from which to view the debate.

Let's start with a quiz. Would you rather have a 3% current yield with a 45% payout ratio or a 2% starting yield with a 30% payout ratio? Now unless you have a one-year time horizon, which is focused solely on income, then it seems like you would want more information. For example, perhaps you would like to know the relative value of each security. Although certainly not perfect, one way to judge this might be the price-to-earnings ratio. What's interesting to me is that I've already told you that both securities have the same P/E. Here's the math:

Price can be restated as the dividend divided by the dividend yield (D / Y). For example, a $2 dividend with a 2% yield indicates a price of $100. Earnings can be restated as the dividend divided by the payout ratio (D / Payout). For example, a $2 dividend with a 30% payout ratio indicates earnings per share of $6.67. Taken together, the commonly cited P/E ratio can be restated as (D / Y) / (D / Payout). Using some middle school algebra, this equates to the payout ratio divided by the current dividend yield (Payout / Y). In other words, using a price of $100 and an EPS number of $6.67 equates to a P/E of 15. Yet you can also get to that number by dividing the payout ratio by the current yield (30% / 2% = 15). So in our example both the 3% yield with a 45% payout ratio and the 2% yield with a 30% payout ratio have a P/E of 15 (45%/3% = 30%/2% = 15). The same holds true for a 1% current yield with a 15% payout ratio or a 4% yield with a 60% payout ratio.

So why does this matter? Let's use a simplifying example to better illustrate my point. We'll stick with the same 2% yield and 3% yield along with their corresponding payout and thus same P/E ratios. Imagine that both of these companies are expected to have the same payout ratio in the future, EPS is expected to grow at the same rate and people are willing to pay the same multiple. To quantify the example we'll say both companies will pay out 50% of profits, earnings will grow at 10% a year, the market will price each at 15 times earnings and our time horizon is 5 years. Here's what that would look like for the 2% yielding security:

Year

1

2

3

4

5

Price

$100

$110.00

$121.00

$133.10

$146.41

Dividend

$2

$2.50

$3.12

$3.90

$4.88

Payout

30%

34%

39%

44%

50%

EPS

$6.67

$7.33

$8.07

$8.87

$9.76

Notice that the price was arbitrarily set to $100, but the 2% yield and 30% payout ratio are in place. Earnings per share begin at $6.67 and grow at 10% a year. At the end of year 5, this security is paying out half of its profits in the form of a dividend and thus the dividend has grown at nearly 25% per annum. Now let's take a look at the 3% yielding security:

Year

1

2

3

4

5

Price

$100

$110.00

$121.00

$133.10

$146.41

Dividend

$3

$3.39

$3.83

$4.32

$4.88

Payout

45%

46%

47%

49%

50%

EPS

$6.67

$7.33

$8.07

$8.87

$9.76

Observe the commonalities: both securities start with a price of $100 and earnings per share of $6.67. Both companies end with a price of $146.41, a 50% payout ratio, $4.88 in annual dividends and $9.76 in EPS. The only difference that resulted was that the 3% security only grew its payout by about 13% a year. So it is true that the lower-yielding security provides a much greater dividend growth rate. Yet look what happens to the actual dividend payouts. Due to the linear math, the payout of the higher initial yield is always greater until the final year. In fact this holds true no matter what time horizon you choose. Now assuredly a variety of assumptions were made. Namely the fact that it assumes both securities would act precisely the same in the future. I acknowledge the limitation of a thought experiment. Yet given the concept that we cannot predict the future, I don't think it's all that unrealistic to assume that similar securities might have similar future prospects. In this way, provided the option of two securities with the same P/E and growth prospects, one would always prefer the higher initial yield. The ending price and EPS is the same, so capital appreciation is equivalent. And while the ending dividend payout is the same, the interim payouts will always be higher with the higher initial yield.

But we're not done quite yet. If we were to increase the payout ratio with the 2% yield, the higher initial yield would continue to win out. Yet what happens if we decrease the payout ratio? For example, what if a security yielding 2% had a 29% payout ratio (and thus a 14.5 P/E)? On a nominal basis, given that the two securities perform equally and market participants price the securities the same, now the 2% yield wins out. On a present value basis the required P/E discount, or required payout ratio discount, would have to be even lower, say 25%. But here's the takeaway: when comparing like prospect companies there exists a "breakeven" payout ratio whereby the lower yield must trade at a discount to the higher yield in order to make up for the lower initial payouts. Luckily it's pretty easy to determine what this guideline is. For example, on an implied P/E basis of 15 this is what the nominal "breakeven" payout ratios would look like:

Yield

Payout

0.50%

7.5%

1%

15.0%

1.50%

22.5%

2.00%

30.0%

2.5%

37.5%

3.00%

45.0%

3.50%

52.5%

4%

60.0%

You simply multiply the yield by what you believe to be a reasonable P/E ratio. So for example if you looked at a security with a 3.5% yield and a 52.5% payout against a 1.5% current yield and a 22.5% payout ratio then the higher yield would always win out if you assumed similar businesses results over your contemplated time horizon. Or said differently, in order to choose a security yielding 1.5% over a 3.5% yield with the same prospects, you would need the payout ratio to be under 22.5% (and thus the P/E to be lower than the 3.5% P/E). Perhaps I'm getting wrapped up in the math, but to me that's useful information. If I'm deciding between two like companies then I know there exists a certain payout ratio (or equivalent P/E) that I need the lower initial yield to be under in order to select it. And of course this ideology has practical application as well. For example using David Fish's Champion, Contender & Challenger spreadsheet as a source let's say we're presented with this dilemma, our investable choices include:

Energen (EGN) with a 1.1% yield and a 17% payout ratio.

Medtronic (MDT) with a 2% yield and a 31% payout ratio.

Wal-Mart (WMT) with a 2.5% yield and a 37% payout ratio.

Chesapeake Utilities (CPK) with a 2.9% yield and a 45% payout ratio.

Maiden Holdings (MHLD) with a 3.4% yield and 51% payout ratio.

SCANA (SCG) with 4% yield and a 61% payout ratio.

Now given the previously cited math, each of these combinations roughly translates to a P/E of 15. So if you think the prospects of each company are the same, you would favor the highest yield among the bunch. Said differently, you would require a lower payout ratio (thus a lower P/E) in order to invest in a lower initial yield. Of course the comparison becomes much more difficult if you believe the business of a lower yield will grow at a faster rate. For that matter, the comparison becomes even more complicated when you throw in companies like Aflac (AFL) with a 2.5% yield and a 22% payout ratio. Compared to say Wal-Mart and its 2.5% yield with a 37% payout ratio it must follow that the market expects the business of WMT to outpace AFL. Otherwise you may have found an interesting proposition.

Finally, I would like to point out that a consistently low current dividend yield coupled with a high dividend growth rate leads to a great deal of capital appreciation. It seems people dismiss low yields too often. For example, I think IBM is a great illustration of this:

(click to enlarge)

At the end of 2002 one could have purchased shares of IBM at a current yield of 0.8% - certainly nothing to text home about. Today, you can purchase shares of IBM with a current yield of 1.95% - surely I haven't roused your current income excitement yet. But what's not readily apparent is that during this time the dividend grew from $0.63 to $3.80 or by a yearly average of about 18%. As the current yield only jumped by about 8% a year, this translates to the 10%+ return that you see. The same holds true for other lower yields. Just because the yield is low, doesn't mean that the underlying investment might not be worthwhile.

In sum I think it can be all too easy to make simplifying assumptions about the dividend growth prospects of a company from the initial yield alone. For example if you're deciding between a 4% yield and 2% yield, one might be automatically inclined to favor the 2% yield if they're chiefly concerned with a higher payout growth rate. Likewise, those transfixed on current income might habitually choose the 4% yield. Yet, depending on your timeframe, either one might eventually provide the greater stream of income. If you believe the prospects of the companies are similar, I've presented a maximum payout ratio (and thus maximum comparative P/E) for the lower yield investment. Perhaps this is obvious, but if you're not considering the role of the payout ratio in the dividend growth or yield debate I believe you're selling the analysis short.

Source: Dividend Yield Or Growth: The Role Of Payout Ratios