Let me start out by explaining why dividend growth is important:
- If the dividend doesn't increase, you may as well get your income streams from Fixed Income. At least with bonds, the coupon as a rule is fixed, and you get your initial investment back at par.
- Without an increase in the dividend, the purchasing power of your income will be eroded by inflation over time.
- If dividend increases occur as a result of an increase in the underlying EPS of the business, the amount of invested capital will naturally be worth more over time. Without it, reduction in capital value is virtually guaranteed.
Unfortunately, current conventional wisdom seems to be that dividend growth alone defines the merits of an income stock. Little focus is put on initial yield or current yield at purchase. In my opinion, both numbers need to be quality values; and in actual fact, growth in the dividend is not nearly as important as it first appears -- at least not compared to initial yield. Here's why:
In all probability, a higher current yield at purchase will define income streams in future years more than growth in the dividend. Unless you have unshakeable conviction that earnings/dividends in the low yielder will gain very strong momentum in the immediate future, you will probably be playing catch up to the higher yielding stocks for many years. Simply put, the math is against you from the start.
Maybe an example using two classic dividend stocks will help illustrate this:
Current Yield = 5.43%
5 Year Dividend Growth = 5.09%
Abbots Laboratories (ABT)
Current Yield = 3.24%
5 Year Dividend Growth = 10.23%
ABT's CY is 36% lower than T. On the other hand, ABT's dividend growth is double T's. For argument's sake, let's assume that the growth rates remain constant for each stock. With those growth rates, it would be natural to assume that ABT's payout would quickly catch up and surpass stodgy old T. In fact, despite a growth rate double that of T's, it would take 13 years for ABT to catch up. Further, assume an equal $10,000 investment in both stocks. At the end of the 13 years, T would have paid out $2,114, or 23% more than ABT over that period.
Acc Div Comp
Reinvest the $2,144 difference and the equation becomes even more heavily loaded in T's favor. Although these are real world examples, I understand I am assuming that a lot of things remain equal -- which in the real world they rarely do -- but you get the point. Strong dividend growth can disguise opportunity cost when matched against actual income flows starting from a higher yield.
The Market Compensates, Right?
It would be reasonable to assume that the market would reflect higher growth by assigning a higher value, i.e., a larger capital gain for the stock whose earnings and dividends were growing faster. Logically, the stock price should reflect the differential. Again, using ABT and T, we can see short term this is not so:
ABT - 52-Week Performance = 17.73%
T - 52-Week Performance = 15.83%
And over 10 years?
ABT increased the dividend fourfold the rate of T's for the 10 year period.
10yr Change Dividend (%)
Over the 10 year period with short-term noise and abnormalities ironed out, ABT's stock price performance does reflect its higher growth. Due to the special 2004 dividend of $2.84/share, ABT actually paid out more over the 10 years, yet an equal dollar investment in both stocks still does not result in significant outperformance by ABT.
Performance - Stock
Total div per share
* Based on equal 10K investment
This is not supposed to be an exhaustive study of the merits of one stock over another, but could it be that higher cash flow up front can be more valuable in real terms than growth potential in the dividend? A bird in the hand is often worth more than two in the bush.
Change In Attitudes
For reasons mentioned earlier, investors should not abandon dividend growth as a performance metric. Growth does matter. The danger is that the current zero interest climate has changed investor attitudes. Because there are few alternatives offering meaningful yield, investors appear to be increasingly willing to assume more risk to capture it. With 10- year Treasuries yielding less than equities for the first time since the mid 1950s, it seems to matter little where equity yields start at, as long as there is a history of growth in the dividend. But as The Economist pointed out, there is the danger that "low bond yields have become not a plus point for equities but a warning sign about future growth." Bond yields are telling us historic dividend growth is no indicator of future gains. So by accepting an initial yield that is too low, or by paying too much for the asset producing the income stream, you're already accepting too much risk up front vs. the promise of increases in the income stream. Even if the risk dissipates and growth materializes, the math of catch-up is still against you.
The Usual Suspects Have Become The New Bonds
10-year Treasuries currently yield 1.69%. 30-year Treasuries will pay you 2.83%. Businesses are taking advantage of current rates to refinance existing debt. Few are taking advantage of them to borrow to fund expansion, and are choosing instead to hoard their cash. This is particularly evident in the U.S. One of two things has to happen: Either expansion materializes and real demand for capital rises, pushing up rates, or QE continues as growth sputters and earnings fall. Either way, investors who have accepted low yields on their U.S. equity investments are not well hedged.
In the current environment, companies with long histories of dividend increases are being held up to be sure things. It's as if so-called income stocks have become the new bonds. Johnson & Johnson (JNJ) with a low beta and over twp decades of uninterrupted dividend increases, is considered safe. Even in 2008-09, its decline was minimal relative to the market. It is one of those "sure thing" stocks. It trades like a perpetual bond with a floating coupon of 3.5% (current yield). The dividend has annualized 9.11% growth over five years. It is also expensive, trading on a P/E of 21. If again we assume growth in the dividend remains constant at 9.11% annually, it would take 14 years to recoup an initial investment in JNJ at today's price level. This is not a commentary on JNJ as a business. But it is a commentary on JNJ as an income investment. Growth in its income -- or should I say potential growth in income, is not compensating investors adequately for risk of capital. The more income an asset pays up front, the more quickly capital risk is reduced, as the purchase cost is recuperated earlier.
There is much discussion these days as to how cheap the S&P 500 is compared to its historic mean. For income investors, it is not such a bargain. You might want to read my earlier article, "The S&P 500: No Bargain For Income Investors" for a more comprehensive discussion as to why. The main reason being average payout ratios at US companies, which have been steadily falling for 30 years. Management at U.S. companies, despite sitting on record amounts of cash, see little reason for returning much of it to stockholders. The more cash management retains, the more risk is shifted to stockholders. This risk comes from three main areas:
- Reinvestment of retained earnings is an income opportunity loss for stockholders.
- Capital risk, i.e., a lower stock price if management reinvestment of earnings fails to add value.
- Longer capital at-risk period, as lower payout lengthens the time needed to recapture initial investment cost.
Ironically, the risk of holding U.S. income stocks is not created by the market, it is created by management's refusal to pay out more of their earnings as dividends. A low payout ratio would be prudent if profitability at S&P 500 companies was not at an all-time high. Investors in U.S. equities do not appear perturbed by how little earnings are distributed to them in the form of dividends. Instead, they choose to focus on how much management increases the dividend YOY.
There is no better evidence of this than the emerging importance of the "Dividend Aristocrats." U.S. companies included into this Standard & Poor's benchmark "have followed a policy of consistently increasing dividends every year for at least 25 consecutive years."* What is telling is how S&P also goes on to state that inclusion is for companies that "have followed a managed dividend policy of consistently increasing dividends every year for a specified number of years." This is key. The dividend policy is "managed." It is not much of a feat to increase the dividend when payout ratios start from such a low base. Only large cap, blue chip S&P 500 companies are eligible for inclusion. In other words, well-established businesses operating in mature markets, with stable income flows. C.R. Bard (BCR) is a member. It has increased its dividend payout ratio for 25 years. It pays just 12.73% of its earnings as dividends. Its current yield -- 0.81%. But it's an "Aristocrat." I am not questioning whether an objective benchmark as offered by the Aristocrats has practical utility. But the current preoccupation of whether an income stock is or isn't included in this index ignores the much wider problem of earnings distribution.
Disney (DIS) and Caterpillar (CAT) pay out 18.94% and 13.96% of their earnings, respectively. Although not Aristocrats, I simply fail to understand why these large moat businesses can justify paying out so little of their earnings to stockholders. Equally, I fail to see why investors perceive "risk" with foreign stock ownership, yet will gladly take on board real risk by overpaying to stay at home. The markets of Europe, Asia, and increasingly South America, offer higher average yields and lower multiples. Only Japan is offering lower current yields than the U.S. The low payout passes more risk to the investor, leaving market valuations as the main potential for increasing wealth. Increasing the dividend will not meaningfully compensate for that.
* Standard & Poor's