A friend joked that we’re overly-fixated on dividends. The thing is, many studies show that dividends can add significantly to an investor’s returns.
Consider a study by Krishna Ramaswamy and Robert Litzenberger. They found a significant correlation between dividend yields and subsequent returns. James O’Shaughnessy found a similar result. From 1957 through 1994, the 50 highest-dividend-yielding big stocks had a return that was 1.7 percentage points higher than the market. Professor Jeremy Siegel writes: “The historical analysis of the S&P 500 Index supports the case for using dividend yields to achieve higher stock returns.”
Higher payout ratios plus higher earnings
Robert Arnott and Clifford Asness examined 130 years of dividend payout ratios from 1871 to 2001. (A payout ratio is the percentage of earnings paid out in dividends.) What Arnott and Asness found was that the higher the payout ratio, the higher the earnings growth over the following 10 years. The lower the payout ratio, the lower the earnings growth. In fact, the bottom quartile (or the 25 per cent of companies with the lowest payout ratios) saw their earnings fall at an average yearly compound rate of 0.7 per cent.
While Arnott’s and Asness’ study covered 130 years, they focused on the period from 1950 to 2001. They believe that this period contains the most relevant information for today’s investors. This data is based on Standard & Poor’s 500-stock index. But since Canadian stocks move in the same direction as US stocks, it matters to Canadians too.
Arnott and Asness made 485 different rankings based on rolling ten-year periods. That is, at the beginning of each month, they again divided the 500 companies into four groups based upon their payout ratios and then looked at the subsequent 10-year earnings growth of each group.
Even in the worst 10-year period of earnings growth of the highest payout ratio group, it beat the average earnings growth of the lowest payout ratio group. Similarly, the average 10-year earnings growth of the top payout ratio group beat the best 10-year earnings growth period of the bottom group.
Why do high payouts lead to earnings growth?
Arnott and Asness offer some potential reasons to explain this relationship between dividend payout ratios and subsequent earnings growth. One is that a high payout ratio shows that management expects the company’s future earnings to rise. A low payout ratio, by contrast, suggests that management thinks the company’s earnings are likely to flat-line or decline.
A second possible explanation is that “in any year, there are only a few truly compelling internal reinvestment opportunities in most companies. The first dollar of retained earnings will presumably be invested carefully and wisely for future growth. Each subsequent dollar will, of necessity, pursue a successively less-compelling internal project or investment, until the return on those projects falls off a cliff. In essence, when most of the earnings are retained instead of paid out, resources are available to fund marginally less-attractive projects.”
A third potential explanation is ‘empire building’ or what we call the ‘Big Shot’ factor. Some managers want to impress everyone with just how important they are. So they use the shareholders’ money to hire more people than needed. After all, the more employees at their disposal, the more of a ‘Big Shot’ they are. They enjoy ruling over fawning minions.
The study also conducts seven “robustness tests” to make sure that its findings are valid. It examines, for example, the possibility that a short-term drop in a company’s earnings may distort the data.
Robustness tests supported the findings
When a company’s profits fall temporarily, it’s unlikely to cut its dividend. Since dividends become a larger percentage of the reduced profits, the payout ratio rises sharply. This can even bump a company into the group with the highest payout ratios. Then, as profits recover from a low base, it makes the company look like it has faster growth. So Arnott and Asness used statistical techniques to eliminate this factor. They found that their conclusion remained valid: higher payout ratios lead to faster earnings growth over the following 10 years.
A second robustness test considered stock buybacks. Some firms see this as better than a dividend increase. (We disagree.) It can raise a firm’s earnings per share without triggering taxes the way dividends do. After controlling for this factor, the study says “there is considerable stability in the basic message: when payout ratios are low, future earnings growth tends to be slow, and high payout ratios go hand-in-hand with rapid subsequent earnings growth”.
This is an edited version of an article that was originally published for subscribers in the January 18, 2019, issue of The Investment Reporter. You can profit from the award-winning advice subscribers receive regularly in The Investment Reporter.
The Investment Reporter, MPL Communications Inc.
133 Richmond St. W., Toronto, On, M5H 3M8, 1-800-804-8846