We’ve been questions regarding all of the dividend data that we use to analyze different stocks when we try to find the top ones (see our Top US Dividend Stocks and Top Canadian Dividend Stocks), especially about the payout ratio. It seems simple enough but the reader was (rightfully) confused about seeing different numbers between two websites and was wondering which one was right and what could cause these differences. It seemed like answer that might be of interest to quite a few readers so we decided to publish it here.
What is a payout ratio?
It is the basis isn’t it? Said simply the payout ratio is simply the portion of a company’s earnings that it pays out as dividends or:
Payout ratio = dividends / earnings
Or
Payout ratio = dividend per share / earnings per share
How could anyone get It wrong?
Not that simple
There are many different aspects that make the calculation of this ratio much trickier than you could initially think. We decided to take a few and hopefully that will help you better understand the numbers and how to interpret differences between data sources for the payout ratio.
Leading or trailing dividends?
If you have a dividend portfolio, you hopefully hold a majority of stocks that are raising dividends in a consistent way. The question would become, what number would you use in the formula above is this company is raising dividends. For example, suppose we are at 201-Q4 here and 2011 are your best estimates.
Quarter Dividend Per Share
2010-Q1 $0.20
2010-Q2 $0.22
2010-Q3 $0.22
2010-Q4 $0.24
2011E-Q1 $0.24
2011E-Q2 $0.26
2011E-Q3 $0.26
2011E-Q4 $0.28
Would you use a dividend per share of:
$0.88 (realized dividends last year)
$0.96 (last dividend annualized)
$1.04 (actual dividend expected in 2011)
You could make a case for all 3 and I can tell you that there is no “right” answer. Some use each one of those numbers and as you can imagine, it can make big differences.
Leading or trailing earnings?
The same situation occurs for earnings but the effect can be even more dramatic. Why? Because earnings tend to fluctuate and one “fluke” quarter with a big loss or a big gain can make a world of difference. In our case, here are the earnings:
Quarter Earnings Per Share
2010-Q1 $0.68
2010-Q2 $0.75
2010-Q3 -$2.24
2010-Q4 $0.82
2011E-Q1 $0.82
2011E-Q2 $0.88
2011E-Q3 $1.33
2011E-Q4 $0.95
Would you use a earnings per share of:
$0 (realized last year)
$3.26 (last dividend annualized)
$3.98 (actual dividend expected in 2011)
This one is much more difficult to judge but I would personally say that you tend to take out “extraordinary events” and use a number that is as representative as possible of what can be expected in the future.
Special dividends?
Many companies occasionally pay out special dividends for various reasons. These are certainly nice to have but it’s difficult to “assume” they will be paid out in the future and counting them in the payout ratio or any other dividend statistic can be a difficult decision. For some companies, the “special dividends” are a big part of their payout every year, especially in funds that out most of their earnings out. They will pay a conservative dividend for the first parts of the year and then pay out excess earnings at year’s end.
Errors
Researching this type of financial data as well as manipulation to include or exclude special factors certainly opens the possibility for human error. On all of our analysis we do our best to double check everything but we are always more than happy to look into specific numbers, you can simply contact us.
Which one is right?
In the specific case that the reader was looking into, the other source was well known newspaper which seemed to be off. I’m not sure what cause the problem but it could simply be a different method, different estimates or even a different analysis. As always, we recommend that you do your own research!
Thanks for the explanation of how it works and the subtle issues behind the numbers. Another example of not taking numbers for granted. Without knowing how they were calculated, they are meaningless.
Nice article. I think it’s all about consistency. Whenever you are tracking EPS and dividends per share one should use consistent data when tracking month to month, year to year or even comparing different companies. If you’re going to use TTM numbers, then that’s fine, but I think it’s best to use TTM numbers for all your analysis, numbers and comparisons. Makes it more apples to apples. Good stuff.
Hi Mike,
I think the problem is you are using the values from the Globe and Mail website (correct me if I am wrong). These values are not accurate when compared to other financial sites, TD Waterhouse data, other stock screening services etc.
I’m quite uncertain how the Globe and Mail calculates their DPR, but many companies show an innacurate DPR when compared to other screeners. But for sake of simplicity for readers, the quick and easy DPR is calculated as:
Dividend Payout Ratio = (Annual Dividend / EPS) * 100
@Dividend Ninja,
I use the G&M data for several things but when I actually pull out massive data like the DPR, I use other softwares.
The question actually came from a reader using G&M gold data ;-).
I was thinking of buying Telus but the payout ratio is 94, so according to this article , this is a no go!