Dividend Growth Rate

[Roadmap2Retire]

This might sound like a dumb question, but how do you calculate the dividend growth rate?

Ive noticed different websites use different numbers for the same company (when writing about 5-yr DGR or 10-yr DGR) and wondering if everyone has their own take of calculating the number. Looking for input on how the folks here calculate their DGR numbers.

[rapidacid]

I use what stockrover uses, which is the compounded annual growth rate

CAGR = ( Ending / Beginning ) ^ ( 1 / years ) -1

Use WAG:

End of 2013 they had a div / share of $1.18. 5 years prior, at end of 2008 they had a div / share of $0.41

CAGR = ( 1.18 / 0.41 ) ^ ( 1 / 5 ) - 1
CAGR = 2.878 ^ 0.2 -1
CAGR = 1.235 - 1
CAGR = .235
CAGR = 23.5%

Edit: Wolfram Alpha Calc

[Dividend Watcher]

R2R, I use the same method as rapidacid when I want to calculate it myself. You will see different numbers depending on what periods they use to cakculate it. For example, I believe Mr. Fish uses the dividend paid through Dec 31st of the applicable years for the CCC list so it many not be the same if you're calculating June to June.

[Roadmap2Retire]

I noticed that Mr. Fish also uses the same CAGR formula.

I think you are right - the problem seems to be which period folks use. Thanks for the input.

[Dividend Watcher]

R2R, let's add some nuances here since I was too tired when I got home to fully explain it.

One instance that may skew the numbers is if the first dividend payment in 2011 was actually accelerated into 2010 (I think those were the years) to ameliorate some changes in the tax laws affecting dividends. Mr. Fish does compensate for this but some sources may not and it will look like a dividend cut and skew the CAGR if you begin or end your calculations in those years. A real pain in the keister because now you have to go look at the company's IR website to see if they detail the dividend payments.

Another would be companies that pay a special dividend. Do you include it in the CAGR formula or not? I prefer it not be considered because special dividends are usually used to either 1.) reward shareholders for an exceptionally good year and not part of the board's stated dividend policy or 2.) reflect a one-time event such as a sale of a major portion of the business or a product line. I can't speak to any one source as to whether they include those amounts or not.

Also, you want to watch for companies that make a large increase in the dividend for one year to realign their payout with a board's new strategy or in a reaction to activist investors. For example, PEP increased their dividend 15% early in 2014 in what some say is a response to activist' calls to split the company. They were following a 5-9% increase schedule for a long time and I'm expecting it to drop back to the former level going forward. MMM, MDP, CBRL and HP come to mind also. In these cases, I either like to look at what was happening prior to the big jump or after (if there's enough data to make a valid trend) and expect that to be the trend in the future barring any other information.

To illustrate, the latest CCC List shows a 1/3/5/10 DGR for Helmerich & Payne (HP) of 364.3%, 80.8%, 46.9% & 23.3% respectively. Certainly you can't expect these types of growth rates to be normal. Right next to those numbers, the yearly dividend payment shows their dividend went from $0.28 in 2012 to $1.30 in 2013. Now if I go back to the 12/30/2011 CCC List, I see HP's dividend growth rates as 18.2%, 11.0%, 8.5% & 5.7%. Their latest was 10%, if I recall correctly. That's still acceptable to me, especially for their business model and industry, when you consider you're starting with a >3% yield if you bought some at these prices.

So, if you're relying on the dividend growth rate (or chowder number) to help winnow down your list of candidates, you'll want to go back year by year for a few years to see if there were any "abnormalities" in their payouts. I use the S&P tear sheet and/or longrundata.com to check these.

[Roadmap2Retire]

Thanks for the details breakdown DivWatcher. Good point on going back to older versions of the sheets and seeing what happened. Reading what happened just before and after the raise is definitely a great idea.

Thanks for some thought provoking writeup.

regards

[cannew]

I prefer to track the annual increase each year. Just determine the % change year over year. If the % is constant or growing I'm happy.

[Kerim]

(11-20-2014, 12:36 PM)cannew Wrote: I prefer to track the annual increase each year. Just determine the % change year over year. If the % is constant or growing I'm happy.

I also use calendar years (though I might tinker to adjust for a tax-accelerated dividend). I keep track of quarterly dividend payments in my own spreadsheet, generate a total for each year, and just do a little math to come up with the annual percentage increase and whatever multi-year average(s) I am interested in average.

(And then, in evaluating a stock, I give a lot more weight to recent years. for example, I almost completely ignore the 10 year dividend growth rate.)

[DividendGarden]

(11-25-2014, 11:02 AM)Kerim Wrote:

(11-20-2014, 12:36 PM)cannew Wrote: I prefer to track the annual increase each year. Just determine the % change year over year. If the % is constant or growing I'm happy.

(And then, in evaluating a stock, I give a lot more weight to recent years. for example, I almost completely ignore the 10 year dividend growth rate.)

Same here, Kerim. I usually use the 3-year average. Some companies stagger larger & small increases, so a 3YR normalizes for the ups & downs, while also only considering the recent history. The 10YR DGR is really just a statistic.