Dividend Growth Investing: How I Derive The Top 40 (Part 2)

by: David Van Knapp


Finishing the screening process for Company Excellence.

An introduction to grading companies.

The results.

Dividend Growth Investing: How I Derive the Top 40 (Part 2)

In Part 1, we reduced the number of stock candidates from more than 11,000 to 589 by using the CCC spreadsheet. Then we eliminated about 300 of those by using screens covering dividend increase streaks, overdue increases, structural changes, yields, dividend growth rates, and over-saturation in single industries. As we pick up today: Companies remaining: 159.

Finishing the Screens

Screens are binary, pass-fail tests, although exceptions can be made (as I illustrated with a couple of judgment calls in Part 1). Screens are useful when you want to cut through massive swaths of stocks based on a single factor that you wish to "enforce" with a pass-fail approach.

Screening for quality

We did not get to quality last time. My initial quality screens utilize credit ratings from S&P and Morningstar, which do not appear on the CCC. So let's do those next and cut the candidates down further.

S&P credit ratings are available for free from S&P. You can also find them in other sources, including FASTgraphs, which I use. They appear in the white portion of the right-hand information panel.

Morningstar credit ratings are found on the Quote page for each stock under "Morningstar's Take." Morningstar does not have as wide a coverage universe as S&P, so often you will find an S&P credit rating without one from Morningstar.

You will recall that BBB- is considered the lowest "investment grade" rating. It is the ones below that that we will eliminate.

I don't know any way to do the credit ratings except to look them up one at a time. So I create two new columns in the spreadsheet to record them.

Now I will go look them all up. Excuse me for a while.


I'm back from Starbucks. It took about 2 hours to look up all the credit ratings and insert them in the spreadsheet. The time spent on Morningstar seems wasted. There are only 1-2 instances where they had a rating in the absence of one from S&P. Most of the time S&P had a rating and Morningstar did not, or neither one had ratings (46 times). Most ratings were identical or within one level of each other.

So I sorted the spreadsheet on the Scredit column. Here is what I found.

The ones marked in yellow (except for OHI and QCOM) are hereby eliminated. I am puzzled by the lack of a rating for QCOM. Number deleted: 59. Companies remaining: 100.

There is one more screen that I use for quality. But it is time-consuming, so I hold it to the very end, after I have used scoring grades to winnow the stocks down further.

The final screen is each company's Story. The test is simple: If I cannot understand how a company makes money, I will not buy the stock.

The Story is a paragraph about the company's business model. What does the company do? How does it make money? Why is it likely to continue to succeed?

I look for companies that are dominant in their fields. I like legal monopolies and duopolies; lack of hindrance by (or even assistance from) regulation; great brands; timeless and everyday "necessity" products and services; and sustainable competitive advantages.

I want to avoid companies that are subject to technological disruption, shifting tastes, fads, and product obsolescence.

I believe in writing out each company's Story in a few sentences. I try to explain each company's merits in simple language, being crystal clear about what the company does, how it makes money, and why I think its business is sustainable. This is not a marketing piece, it is analytical. Include risks and concerns.

Grading vs. Screening

Grading is a different way of evaluating stocks, and I switch over to it after the screens have knocked out enough companies to make the remaining work enjoyable and rewarding.

To grade stocks, I have created grading scales for particular factors. After playing with these scales for years, I have gained confidence that they do a good job of identifying companies that I want.

Here is a simple example. One grading factor is how many years the company has been increasing its dividend. We already used that as a screen (requiring 5 years minimum), now we will use it as a ranking device.

The scale for this is simple: One point for each 5 years of increase streak, with a cap of 8 points. So just divide the increase streak by 5, drop the decimal, and you have the point value.

The CCC shows the increase streaks. Using my new skill of hiding columns, I have hidden the columns that we won't use and ordered the Years column from high to low. Here is what the top of the spreadsheet looks like now. The last column is one I created to record the scores.

I will henceforth do the same thing for each factor.

Making the Grades

The following table shows all 13 of the scoring factors that I used last year and the maximum points available from each factor. I have highlighted in green the 8 factors whose data is available from the CCC.

Here is an example of the scoring scale that I use for yield. Similar scales exist for each factor.

2.0% to 2.3%: +1

2.4% to 2.6%: +2

2.7% to 2.9%: +3

3.0% to 3.4%: +4

3.5% to 3.9%: +5

4.0% to 4.4%: +6

4.5% to 4.9%: +7

5.0% to 5.4%: +8

5.5% to 5.9%: +9

6.0% to 6.4%: +10

6.5% to 6.9%: +11

7.0% to 7.9%: +12

8.0% to 8.9%: +13

>9.0%: +14

To save space, I will not show all the scoring scales here. I will say that I do try to make them intelligently proportional to each other. For example, a stock can earn a lot more points from yield (maximum 14) than beta (maximum 5). I strongly encourage anyone to add or drop factors to suit themselves, and also to adjust the point scales to your liking.

The scoring system is deliberately open-ended. That means that grading is not on a curve that must total 100 points or any other set number. That allows maximum freedom in adjusting scale values, adding or dropping factors, and so on.

Finally, there is no "passing" grade. The idea is to rank the stocks against each other, not against an abstract scale. Even the best-scoring stocks do not score a high percentage of the total points available. I have never seen a company that I love everything about. There is no perfect investment.

To give you a sense of how it works in practice, I have graded the remaining stocks on the factors colored in green above. I broke up the screen shots for clarity.

Total points are shown in the right-hand column. You can see the wide separation that begins to emerge between high-grading and low-grading stocks. Just on the factors scored so far, the highest-scoring company receives 35 points, while the lowest gets 14, almost a 3-to-1 difference already.

Miscellaneous Notes

1. Once all 13 factors are scored, the order of rankings might change considerably.

2. In constructing the Top 40 (or a portfolio), you might not simply take the highest-scoring stocks. For example, you might make sure that you have representatives from most sectors or industries. I usually do that.

3. You might want to change the point scales. For example, many of the highest-scoring stocks (so far) earned 0 points on beta. Maybe beta is really important to you. So instead of using a 5-point-max scale as I did, you might want to use 10 or 15. Or you could simply eliminate all stocks with betas greater than (say) 1.1. That alone would eliminate 41 companies from the 100 remaining. But be careful: The reason we switched to grading was so that a single factor could no longer single-handedly decide a company's fate.

That completes the company scoring system. I call this rating Company Excellence. But as we all know, not all great companies make great investments. A huge step still remains: Valuation. That will be the subject of the third and final article in this series.

Disclosure: The author is long PG, KO, JNJ, KMB, PEP, MCD, AFL, T, HCP, CVX, BBL, MSFT, PM, OHI, HAS, LNT, O.

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.