Improved Quantitative Strategy For Selecting The Best Dividend Growth Stocks

by: Zach Tripp

I would like to expand on a previous article I wrote in which I proposed using a weighted decision matrix to rank a list of dividend growth stocks. I used David Fish’s CCC list. After eliminating companies classified as being in the banking industry, the ADRs, any companies noted as splitting into two or being bought, REITs, and MLPs (since they take special analysis), I was left with 317 companies.

It is beneficial to eliminate stocks from the list of 317 companies that may not meet your entry criteria. This gives more weight to companies you would actually consider starting or increasing a position in.

Since the Vanguard Extended Duration Treasury Index ETF (NYSEARCA:EDV) currently yields 3.5%, I would eliminate any stock with a yield under 3.5%. Granted, the EDV carries a risk of the yields going down over time as opposed to up like a dividend growth stock but a low-yielding dividend stock could take a couple years of compound dividend reinvestment to get a yield up to 3.5%. This reduces the list from 317 stocks down to 92.

Inflation is the silent killer of income-producing portfolios. I prefer companies that have increased their dividends at a rate of 2x the rate of inflation (since there is fluctuation on the dividend rate increase). The rate of inflation over the past 5 years has averaged 2.28%; therefore, I look for a 5-year dividend growth rate of at least 4.6%. This reduces the number of companies down to 69.

Sept-Sept CPI Increase/Decrease

Source: Bureau of Labor Statistics

Finally, I require a safe dividend yield. I will eliminate any stock with a payout over 80%. Typically I look for 60% to 70% earnings per share (EPS) to dividend per share (DPS) payout, but since the payout can fluctuate year to year, I will use 80% to prevent good companies with bad single years from being eliminated. This reduces the list down to 50 stocks.

I will use the same weights as I did in my previous article and perform the same analysis, except this time with the remaining 50 stocks.

  • Weight 5: Payout Ratio
  • Weight 4: 5-Yr DGR
  • Weight 3: Yield %
  • Weight 2: NY EPS Growth %
  • Weight 1: 10-Yr DGR

Here are the results for the top 12 companies. Click to enlarge: Top 12 Dividend Companies

How can an individual investor implement this system? The real value in dividend growth stock investing is the compounding of reinvested dividends. Sharebuilder brokerage is a good choice for this. Sharebuilder allows an investor to buy in dollar amounts as opposed to whole shares. Also, dividends can get reinvested into additional shares free of charge. With regard to cost, for $144/year ($12/month or $1 per trade if you buy 12 stocks a month) you can make 12 transactions a month (as long as they are on Tuesdays). An individual investor can set up the automatic investment for his or her 12 stocks, putting more money toward the higher-scoring stocks and less money into the lower-scoring stocks. The investor can recalculate the 12 top stocks once a quarter and readjust the automatic investment, putting the fresh monthly money into the current top 12, keeping any stock that fell out of the top 12 in the portfolio with the dividends still being reinvested.

This would be an example of a monthly investment plan using the criteria I have selected:

Possible Investment Plan

One can imagine as stocks fall in and out of the top 12, the actual holdings of the portfolio can grow, so caution has to be taken to ensure the stocks still in the portfolio (but not getting new cash, just reinvesting dividends) do not fall subject to dividend cuts.

In conclusion, the above is not a buy list but a proof of a method for generating a buy list based on data as opposed to emotions. Each investor has to determine his or her desired minimum entry criteria and the individual weights of the evaluation criteria.

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

Disclaimer: Author is not responsible for typographical errors or errors in data, whether errors are from the source or author's calculations.