Dividend growth portfolios are a common subject on Seeking Alpha. As common as they are though, there doesn't seem to be an accepted definition or objective for dividend growth portfolios. Neither does there seem to be a good way to compare one portfolio relative to another. For a standard portfolio, if I may call it that, where the sole objective for the collection of stocks is primarily an increase in the net asset value, a common method to showcase the performance is to use a 10-year bar graph of the value change year by year. Every prospectus I receive for every fund I receive includes such a chart. In addition to the chart, it also usually shows the annual total return for the 10-year period. The annual returns and the average 10-year return are then usually compared to an appropriate index, such as the S&P500 or Russell 1000, or any index that is considered most relevant. As such, I consider this the accepted methodology to evaluate a fund's performance and to compare one fund to another.
For a dividend growth portfolio, since the main objective is an increasing income each year, it makes sense that the appropriate chart to show the portfolio's performance would substitute income growth for growth in value. As the mindset of a dividend growth investor should be to concentrate on the growth in income rather than the portfolio value, such a chart would seem appropriate. Also, as most dividend growth investors tend to use a selection of U.S. large caps, the appropriate benchmark should be the S&P 500.
Let me try to define the objective of a dividend growth portfolio, DGP. A DGP's objective should be to have
- the highest dividend yield possible, consistent with having
- the highest dividend income growth rate, while having
- the lowest possible year to year volatility of the dividend growth rate or the highest possible minimum expected growth rate.
All three factors -- high yield, high dividend income growth rate and a low standard deviation of the growth rate over a 10-year period (or a period of 9 growth years) are critically important to the dividend growth investor. A portfolio with a very high dividend growth rate that did not have a high yield and had large variations year to year in the dividend income would presumably not make a very good dividend growth portfolio. All the stocks in the S&P 500 are a good example and I will start with this as it will be the benchmark.
A bar graph showing the annual percentage change in the sum of all the dividends for all the 500 stocks in the S&P 500 portfolio was shown in a recent article (and can be seen as part of the final two bar graphs in the current article). The average and mean dividend growth rates for the S&P 500 are actually pretty high over this period, 7.5% and 10.9%, respectively. These are growth rates that would likely be acceptable to most dividend growth investors. But the S&P 500 doesn't make a good dividend growth portfolio for 2 reasons. The yield is a low 1.9% and the standard deviation of the growth rate over this period is high, 9.8%, meaning that minimum dividend growth rates can be negative.
Using a Bell curve and standard deviation approach to evaluate a dividend growth portfolio gets interesting. In my opinion, perhaps the most important feature of a dividend growth portfolio is growth year in and year out. A negative growth rate in any year is to be avoided. With that reasoning, I think a good criteria for evaluating a dividend growth portfolio is to determine its statistically minimum growth rate expected. A good way to define a minimum dividend growth rate is to set it as the mean minus 3 standard deviations (also known as 3 sigma). The chance of having a year in which the dividend growth rate would be below this minimum would by definition be only 0.3%, a satisfactorily low probability. For the S&P 500 it corresponds to 10.9% minus 3 times 9.8% or -18.5%. That explains why the S&P500 would make such a bad dividend growth portfolio; dividend growth rates are so volatile that a significant negative growth rate is possible. It should be pointed out that the negative number for the S&P 500 is largely spurred by the negative dividend growth in the year 2009, and the low growth rates in the neighboring years, 2008 and 2010.
Now that we have looked at the S&P 500 with its dividend growth portfolio characteristics of 1.9% dividend yield, 10.9% median annual growth, a standard deviation of 9.8% and a minimum dividend growth rate expected of -18.5%, let's look at two different dividend growth portfolios, mine, which I will call the BD Portfolio and what I am calling the DH Portfolio, the core or foundational dividend growth stocks selected by Dividend House. See this article for Dividend House's description of the core holdings of her DGP.
My personal dividend growth portfolio has the following stocks. The table below shows the dividends awarded each year for each of the stocks in the BD portfolio from 2005 to 2014. In the table, dividends are summed up for each year and the growth rate of those dividend numbers are used to define the annual dividend growth rates of the portfolio. The stocks are not weighted in any other way. By doing it this way, it is as if each stock in the portfolio has the same number of shares.
|1||Johnson & Johnson||JNJ||1.28||1.46||1.62||1.80||1.93||2.11||2.25||2.40||2.59||2.76||3.00|
|2||Proctor & Gamble||PG||1.09||1.21||1.36||1.55||1.72||1.89||2.06||2.21||2.37||2.53||3.36|
|7||Illinois Tool Works||ITW||0.61||0.75||0.98||1.18||1.24||1.30||1.40||1.48||1.60||1.81||2.02|
|11||Avalon Bay Comm||AVB||2.84||3.12||3.40||3.57||3.57||3.57||3.57||3.88||4.28||4.64||3.03|
|12||T. Rowe Price||TROW||0.49||0.59||0.75||0.96||1.00||1.08||1.24||1.36||1.52||1.76||2.56|
|18||Omega Health Investors||OHI||0.85||0.96||1.08||1.19||1.20||1.37||1.55||1.69||1.86||2.02||6.00|
|20||Extra Space Storage||EXR||0.91||0.91||0.93||1.00||0.38||0.40||0.56||0.85||1.45||1.81||1.81||2.76|
|minimum dividend increase||1.2%|
The performance of the BD Portfolio to the S&P 500 benchmark, from a dividend growth perspective is shown in the following bar chart. The data labels are only shown for the BD Portfolio for clarity's sake. (Please look at my previous article, to see the data labels for the S&P500).
Reviewing this graph, it should be clear that the BD Portfolio has behaved much more like a dividend growth portfolio than the benchmark. Dividends have grown each year. The minimum growth was in 2009 and was 3.3%. The best growth year was 2007 with a growth of 10.7%. As can be seen in the table above, the BD Portfolios' current dividend yield is 3.26% and the standard deviation of the growth rate is 2.3% much smaller than the number for the S&P500 and that means, based on the 3 sigma, the minimum growth rate expected would be 1.2% in any given year. Dividend growth should be reliable in this portfolio. But is it the best possible dividend growth portfolio?
Let's look at the same charts for the DH Portfolio.
|3||Johnson & Johnson||JNJ||1.28||1.46||1.62||1.9||1.93||2.11||2.25||2.4||2.59||2.76||3.00|
|Proctor & Gamble||PG||1.09||1.21||1.36||1.55||1.72||1.89||2.06||2.21||2.37||2.53||3.36|
|25||Owens & Minor||OMI||0.35||0.40||0.45||0.53||0.61||0.65||0.80||0.88||0.96||1.00||3.02|
|minimum dividend increase||3.2%|
Comparing the DH Portfolio to the BD one, it has a higher yield, a lower mean growth rate and a higher minimum dividend growth rate. A table summarizing the critical characteristics for easy comparison is shown below.
A rating system to compare the portfolios would be based on a weighting of the three critical parameters. An individual investor might have his own weightings. For me, I would use the following.
Dividend Yield Weightings
0 points if below 2%
1 point if between 2% and 3%
1 additional point for each 0.5% increase
Dividend Growth Rate Weightings
0 points if below 3%
1 point for each additional %
Minimum Dividend Growth Rate Weightings
-10 points if negative
2 point for each % increase above 0
Using this weighting system for the BD, DH and S&P 500 portfolios, leads to the following ratings shown in the table.
The S&P 500 rating is negative, declaring that it is not an appropriate dividend growth portfolio. Also, the DH portfolio is rated higher than the BD portfolio mainly due to the importance of high dividend yield and minimum dividend growth rate. For me this is an important conclusion and suggests that I should seek changes to my portfolio that would both increase yield and increase the minimum dividend growth rate that I could expect.
The use of the BD and DH portfolios are intended to illustrate the method. Other portfolios could have been considered. Further, the actual weightings on the stocks held in the portfolios could have been included rather than the assumption of equal shares.
The main purpose of this article is to argue for the need for an accepted way of evaluating dividend growth portfolios and to suggest one approach. Personally, I would like to see included in all ETFs and Mutual Funds that purport to be dividend growth investments, like NOBL or SDY, a bar graph in their prospectus like I have shown here to describe past fund performance in addition to total return.
In the comments section, I would encourage readers to comment on the desire to have a rating system for dividend growth portfolios and alternate approaches for doing so.
Disclosure: The author is long CL, ED,NOBL, SDY,T, WEC,AVB, CLX, ITW,O,PG,TROW,BDX, EXR,JNJ, OHI, RSG,SRE, VTR,BF.B,, NEE, VZ. The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.