Please Note: Blog posts are not selected, edited or screened by Seeking Alpha editors.

Dividends: Who Pays? Who Doesn’t? Who Should?

THIS IS A REPRINT OF AN ARTICLE THAT INITIALLY APPEARED ON FORBES.COM

There are two great risks an equity income investor takes. One is the obvious scenario, the one all should and usually fear; the prospect of a dividend cut or omission. We know how to manage this. The other is subtle, the risk of having one's judgment clouded by nonsensical rhetoric. To manage this one, we need, every now and then, to refresh our basic understanding of what dividends are and why they are, or are not paid.

What are Dividends?

Contrary to what the New York Times suggested on 2/13/16, dividends are not a "status symbol," and certainly not a "battered status symbol." They are a portion of corporate profits, more specifically, the portion paid directly to shareholders as an alternative to retaining them for reinvestment in the company's business.

Profits sometimes go up and sometimes go down so in theory, dividends should do likewise. But as a practical matter, companies try as best they can and as often as they can to keep them on a stable or rising trend. They can usually do this thanks to the flexibility they maintain by paying out less, much less in many cases, than the entire amount of profits they earn (thus saving some "retained earnings" for distribution on the proverbial rainy day). That this doesn't always succeed (sometimes, companies wind up having to cut or omit payouts) does not in any way diminish the reality of this investment cash stream. So contrary to New York Times-style hand wringing, this ever-present risk instead calls for mature and rational analysis of factors that associated with increased probability of adverse outcomes, much the way lenders are supposed to analyze credit (cough, cough, wink, wink).

Others, following the lead of folks like Eugene Fama and Kenneth French, see dividends as a statistical series, a set of numbers. That is an unfortunate aspect of the "physics envy" problem that tends to beset not only the social sciences but finance too. This has lulled some into going so far as to suggest dividends actually damage corporate wealth (if you can figure out the logic, good luck) and that their validity should turn on the efficacy (alpha) of a naïve higher-is-better strategy. That's false. Investors are not in the business of validating simplistic numerical factor sorts. For income seekers, higher yield is better only when we limit consideration to a pre-qualified universe consisting of stocks for which analysis leads one to believe the risk of cut or omission is modest.

The quant ideas are well and good - if you're writing a grad-school paper. In the real world, however, they distract us from the less-than-gloriously-mathematical reality that dividends are a share of corporate wealth, again, the share withdrawn from (as opposed to reinvested back into) the business.

Good Dividends versus Bad Dividends

Whether dividends are good or bad is determined, on an individual-company basis, on whether the payout or reinvest decision was wise. If we assume the penultimate measure of what makes for corporate (fundamental) good is return on equity (ROE) then we have our answer. (By the way, we should assume this: How can a company that uses $100 of capital to produce $20 of profit not be better than a firm that gets only $8 from the same war-chest!) If a dividend enhances overall shareholder wealth (the combination of what they have on their own and what they have as a result of stock ownership in the company in question), then the dividend is good; and vice versa.

Ideally, a company will, when considering a dividend policy, estimate what the company's ROE would be if earnings are retained (i.e. reinvested into the business) and compare that to the ROE that could be produced by alternative investments their shareholders could make with the proceeds of dividends they receive.

It's hard to say how many companies actually go through the nuts and bolts of such exercises (aside from what summer interns might do in order to show off the knowledge they picked up last semester in school). But there are things we can examine to give us clues as to the extent to which such outcomes are materializing one way or another. We can get at this by comparing ROEs of companies that pay out a lot of earnings, versus the ROEs of companies that are more inclined to retain and reinvest profits. I did this using a set of Portfolio123 screens.

Knowing how often and how aggressively net income, a number that is critical to both the ROE and dividend payout ratio computations, can be distorted by unusual developments (writeoffs, gains or losses on asset sales, etc., etc., etc.), I massaged the data: Rather than using standard values for ROE or Payout, I recomputed both.

  • II defined Payout as Cash Dividends paid divided by Cash from Operations (the bottom-line figure from the Cash portion of the Statement of Cash Flows)
  • I defined ROE as adjusted Pretax Income (adjusted to exclude Special Items) divided by Equity; I use a pretax figure because unfortunately, the accounting regulators allow companies to refrain from disclosing the tax impact of these items
  • I'm sorting based on 5-year change in payout rations; stingy companies are those who are gradually reducing the percentage of profits they pay out and generous firms are those paying increasingly larger portions of earnings.

Table 1 shows the results for a universe that approximates the constituent list of the broad Russell 3000 Index.

Table 1: Russell 3000 type companies

 

Change in Payout Ratio: Median

Highest 20%

Lowest 20%

Payout Ratio last 12 months

56.1%

Nil

     

ROE last 12 months

11.6%

6.2%

ROE 5 years ago

12.4%

8.0%

We see that there some deterioration among both groups in terms of ROE change over the course of five years; this likely reflects broadly applicable economic and capital-markets factors. There's a bigger difference in terms of which companies are doing what. Generous dividend pays are substantially more profitable in terms of (pretax operating) ROE than are their stingier brethren.

We can now get into a good chicken-and-egg type of controversy. Do generous dividend payers wind up with higher ROEs because they are better per se? Or are the ROEs higher because of the mathematical inevitable outcome of a smaller equity base (i.e. the earnings that are paid out would, if retained, remain in the Equity balance-sheet account)? We'll see.

Table 2 shows the same information but for a Russell-2000 type, small-cap, universe. Size-based distinctions really matter, and I'm not talking about a statistical Fama-French "small-cap effect." Smaller firms tend to have more room to grow (NASDAQ:GOOD), more strategic and operational flexibility (also good), harder times covering fixed costs (not good) and less internal diversification and smaller cushions that shield them during tough times (less good).

Table 2: Russell 2000 type companies

 

Change in Payout Ratio: Median

Highest 20%

Lowest 20%

Payout Ratio last 12 months

24.3%

7.9%

     

ROE last 12 months

12.4%

13.3%

ROE 5 years ago

12.8%

12.8%

So much for the mathematical inevitability of higher payout ratios leading to higher ROE! The impact of differing payout practices among this group is not nonexistent. But it is modest.

What little variation we see, however, is intriguing. ROE improved slightly for the stingy group, the earnings reinvestment prone firms. But it slid a bit, just a wee bit, for the generous group. I see two potential takeaways from Table 2:

  • Within the small-cap arena, where size-based growth prospects are likely to be more significant, we probably have to evaluate the reinvest-versus-pay decision on a company-specific rather than macro basis.
  • That reinvestment-heavy companies are able to generate any improvement at all over time in their ROEs (given that they are allowing their Equity based to get bigger) suggests there is reason to believe a goodly number of such companies are making effective capital allocation decisions (i.e., choosing to reinvest because ROE prospects really justify it).

Table 3 focuses on a larger-cap group, a Russell-1000 type universe:

Table 3: Russell 1000 type companies

 

Change in Payout Ratio: Median

Highest 20%

Lowest 20%

Payout Ratio last 12 months

24.5%

13.7%

     

ROE last 12 months

14.8%

19.7%

ROE 5 years ago

15.9%

19.6%

Here, there is no doubt that the ROE edge goes to the firms that prefer to reinvest more of their profits. This is a bold leap even further away from the simplistic notion that companies are pumping up ROE us by pushing capital out of the business.

What we're actually seeing here is something we should want to see. Companies capable of generating higher ROEs more inclined to retain and reinvest earnings. We also see that firms less able to reinvest at a high ROE are more likely to refrain from unproductive reinvestment and instead, favor distribution of more profit to shareholders.

Kudos to the group; on balance, the companies are getting it right.

Table 4 takes an interesting side-trip into the S&P 500. We might think of this as a super-large-cap group but that's not necessarily so. The S&P and Russell indexes are independently constructed, so we cannot see the former as being the larger half of the latter. While size (and other objective trading related characteristics predominate S&P's eligibility criteria, the firm does retain some qualitative wiggle room that facilitates its goal of having an index of "leading companies in leading industries." (There's a reason why the S&P 500 is nicknamed the "Blue Chip" index.)

Table 4: S&P 500 Constituents

 

Change in Payout Ratio: Median

Highest 20%

Lowest 20%

Payout Ratio last 12 months

26.5%

20.8%

     

ROE last 12 months

16.4%

18.6%

ROE 5 years ago

20.2%

22.9%

That's interesting. Maybe it raises the questions of whether shareholders should feel blue over blue-chip stature. We see here that the difference between the median payouts in the generous and stingy groups is smaller than what we saw previously, and that companies in both parts of the blue-chip portion of the large cap group might have cause to rethink their reinvest-versus-payout policies. In both cases, we're seeing ROE trend downward from ballpark similar levels. This seems to provide modest comfort to critics who accuse executives of leading companies of empire building.

Dividends are Profits, Not Numbers

It seems that anyone who tries to make any argument at all to the effect that dividends are inherently desirable or undesirable is barking up a wrong tree. Dividends are what we think they are, a portion of corporate profits. And they can be good or bad depending on whether the payout or retention decisions are effective in preserving and building wealth. Maybe this is why the most successful investors we know of tend talk about fundamentals, rather than factor analysis, regression, correlation, T-tests, etc.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.