In my last post I switched from screening for current ratio to using a filter for quick ratio. Today, I would like to address another change in screener filters - dividends.
When I first started using the set of screener filters that met my criteria for profitable, well-run companies, I noted that focusing on returns, profit margin and debt management left us with a lot of companies - in the neighborhood of 950. In applying a specific targeted dividend yield, we could cut that number dramatically, but I don't really think targeting a specific div yield is a good idea.
As it is, dividend yield is false data. Companies do not explicitly say "we will maintain a dividend yield of X%." Dividends are always expressed as an actual amount of money to be paid per share. Dividend yield comes from dividing the paid dividend by the present share price. This is because dividends are paid on the basis of share ownership, not share price; share price, after all, is variable, and is subject to wide fluctuations over any given period of time - something not always within the power of the company to control.
Keeping this in mind, as it turns out, one can achieve a dividend yield of, say, 4% on a dividend of $0.80 for a share at the price of $20.00. Let that share price increase to $40.00, and that same $0.80 dividend represents a yield of 2%. If the share suddenly drops to $10.00, the yield jumps to 8%. However, each share, whether bought at $20, $40 or $10, only receives $0.80. What matters to the individual investor is the yield at the time they bought their shares.
I bought 100 shares of PFE @ around $21. The div is $0.88, or - at the price I paid - about a 4.2% yield. PFE recently traded above $25, where the div was represented at a 3.5% yield. I still get a 4.2% yield, however, because I bought @ $21. Someone buying @ $25 will only get a 3.5% yield. But we both get the same $.88 per share dividend payment.
With this in mind, it is not really worthwhile to look for a specific yield on a screener (unless you are looking to maximize yield-per-share price). What does count, however, is whether the company in question can actually pay the dividend, or at least whether the company can pay the dividend from its earnings, rather than gradually depleting its assets to compensate for shortages.
This is why I have started focusing on dividend coverage - the extent to which EPS covers dividend payments, coverage being equal to the EPS divided by the dividend. If (EPS / Div) > 1, then the company is earning more than enough to pay its dividends. However, if (EPS / Div) < 1, then the company is faced with a choice: cut the dividend, or accrete funds from some other source (i.e., assets).
Major problem: "dividend coverage" is not a ratio typically used in the U.S.; it is more customary in England. In the United States, the similar concept is "payout ratio," which is determined by dividing the dividend payment by the EPS - the inverse of the dividend coverage. What makes this something of a problem (at least, for me) is that while the higher the dividend cover the better, the reverse is true for payout ratio - the lower the ratio the better. Ideally, you want to see a payout ratio of 100(%) or lower. Anything over 100 means the company's earnings do not cover the dividend.
I do think it is relevant to identifying an effective investment (having whatever yield) to be able to determine if the company being investigated is able to cover its dividends. Since the U.S. insists on doing things differently than the more sensible British, I propose to add "payout ratio" to my screener criteria, screening to make sure that dividends are covered.
This leaves us with the following criteria:
ROE > 1 (ROE, ROA & ROI could also be set at "> 0")
ROA > 1
ROI > 1
Net Profit Margin > 0
Quick Ratio > 1
Performance YTD > 0
Performance Q >0
Payout Ratio < 101
Disclosure: I am long PFE.