In mid-August, I created three model Dividend Growth stock portfolios based on a U.K. research study’s findings regarding subgroups of DG stocks that historically outperformed indices and the overall DG stock performance on a total return basis. Since that time, I also developed some working theories on a potential indicator (“-20% rule”) of future dividend cuts and/or price declines.
The table below highlights the performance of these three portfolios, as well as a variation on the Income-Growth portfolio that utilized the 20% rule to actively remove and replace stocks that exhibited that pattern.
All of the model portfolios started with $300,000 on August 16th, 2011, and have outperformed the S&P (SPY) and S&P Dividend ETF (SDY) over the last quarter. In addition, with the exception of the small cap portfolio, they did so with lower standard deviation of returns than the S&P as well.
My intent is to rebalance the model portfolios once per year, as the U.K. study did. This article addresses the rebalancing of the Small Cap portfolio. The DA+ portfolio will be rebalanced in January after the new Dividend Aristocrats list is published, and the Income Growth portfolio will be rebalanced in early February. Staggering the rebalancing dates will make the process more manageable for me, and allow me to provide a monthly update with a rotating focus (i.e. a quarterly report on one model), while noting the results of all three. The screening process for these models relies on David Fish’s CCC list, so again, thank you to him for putting it together each month.
Changes to the Screening Process
Since the original DCC-SmallCap portfolio was developed, I have received helpful input from SA contributors and also performed additional research on dividend cuts that will be applied to this model going forward. The main changes are highlighted below:
- MLPs and REITs are no longer excluded. The original U.K. study excluded these securities, but since many SA income investors use them and they demonstrate long histories of consistent dividend growth, there is no reason to exclude them here. They may get removed due to one of the process filters though.
- The portfolio will now have 40 stocks instead of 30. Given the intrinsic volatility of smaller cap stocks, increased diversification should help to moderate volatility. Having less money invested in each stock insulates the portfolio from a steep decline in a particular issue. More holdings also increase the opportunity to benefit from a buyout; the portfolio missed out on owning HGIC due to the 30-stock limit.
- The universe now starts with all CCC list stocks with 8 or more years of annual increases. Previously, the cutoff was 10 years, based on the U.K. research's methodology, but since I expanded the final portfolio, I wanted to have a larger sample group.
- Dividend growth rates are now part of the screening process, not just yield, payout, and market cap.
- The -20% gap rule, which I discussed in my articles on why dividend investors should not ignore price (part 1 and part 2), will be applied to all of the models. In short, if a DG stock underperforms the S&P by a margin of 20 percentage points for 4 consecutive weeks (using weekly data), there was a high probability of a future dividend cut and/or a significant price decline. These stocks will be removed from the portfolio and replaced with the next best option from that sector. If the stock proceeds to increase dividends, it will be considered again at the next rebalance; for example, Meredith Corp (MDP).
- Micro-caps (Market Cap < $250MM) were removed. One concern in the original small cap DG portfolio was that I was unfamiliar with most of the stocks produced by the filtering process, as many were very small firms with limited information or brand. While this could be a source of alpha, it can also be a source of volatility, especially without doing more in-depth research. Removing these stocks should reduce volatility and produce a list with more familiar stocks, adding to investor comfort and the ability to find information. The average market cap of the CCC stocks is still much higher than the $5.8B average for this portfolio, so relatively speaking, it is a smaller cap DG portfolio.
- Stocks with average daily volume less than 50,000 shares were excluded. Some stocks in the original portfolio traded very few shares daily, making it impractical for investors to follow this strategy due to a lack of liquidity. This also results in larger spreads, which are less efficient. Only a few stocks were impacted by this decision.
The Screening Process
This section outlines my screening process. I have tried to make it as objective as possible; readers should be able to replicate the results. The number in parentheses indicates the number of stocks remaining after each filter step. For reference, the initial universe of 338 stocks had an average yield of 3.1%, and 1-, 3-, and 5-year DGRs of 9.0%, 10.2%, and 12.4%.
- I began with the latest CCC list, combining the champions, contenders, and challengers with 8 or more years of dividend increases (338 stocks in total).
- I sorted the list by Market Cap, and removed those with a market cap less than $250MM. (308)
- Next, I deleted the top 25% by market cap, since this portfolio focuses on smaller cap stocks, which the U.K. research found to outperform larger caps. (231)
- I sorted this list by current yield, and proceeded to remove stocks with a yield less than 2%. (153) The U.K. research found that higher yielding stocks tend to outperform, and also, the purpose of this portfolio is to provide investors with a decent yield as well as growth. Individual readers could choose a higher cut-off, but ultimately, I reviewed survivors starting with higher yielders anyway.
- From this list, I remove 9 stocks because they were overdue for a dividend increase or because they were in the process of being acquired. (144)
- Lastly, stocks with a 5-year DGR under 4.0% were removed, as I am seeking stocks with higher growth rates. (113).
Screened Group Characteristics
The 113 stocks that survived the screening process had an average yield of 3.75% (close to the 4% that many DG investors look for), and 1-, 3-, and 5-year DGRs of 9.0%, 10.2%, and 13.3%. The screen increased the average yield, while maintaining the historical growth rate averages. I sorted the list by industry type and then added a column for Sector, as I use a formula to determine how many stocks to purchase in each sector. After classifying each stock, I re-sort the list by sector and by yield. The table below shows the group characteristics. Click to enlarge:
I calculated the percentages for each sector based on market cap and by representation (count). Note that banks make up 25% of the list, but only 15% by market cap.
Since small firms tend to be more volatile, I chose to favor larger caps by using a 2-to-1 average weighting for each sector. I multiplied the average weightings by a factor until I reached a total of 40 stocks; this column also shows how many stocks from each sector to select. Relative to the S&P weightings, the portfolio is lighter on energy, health care, and technology, and heavier on financials and especially utilities.
Time will tell if having 17.5% in utilities hinders total return. Dividend-focused investors should be comforted by the lower volatility though. Individual investors can make their own decisions on how to weight the sectors, I was trying to follow the characteristics of the screened universe (count and market cap).
To decide which stocks to select from the screened list, I added 5 columns (Yield Rank, 5-yr DGR, Payout, 2012 EPS Growth, and 1-yr DGR) and assigned point values to help me compare the stocks. More points are better!
- Yield: Stocks with a yield < 3.25% received a 1, >= 3.25% received a 2.
- 5-yr DGR: Since a primary consideration is DG, stocks with a 5-year DGR < 6% received 1 point, between 6-12% received 2 points. If the rate was over 12%, I did not consider that sustainable, so I used the same point system based on the 1-year rate. I wish there was a metric for projecting future growth, but for now, looking at historical data will have to suffice.
- Payout: 2 pts for a payout <=50%, 1 pt for <= 80%, and 0 if above 80%. This was a secondary determinant. For stocks with a zero, I examined the last four quarterly cash flow statements and if operating cash flow covered dividends paid in at least 3 of 4 quarters, the stock was not eliminated.
- 2012 EPS Growth: Negative growth rate (-1), less than 5% (0), between 5-10% (1), over 10% (2). This was also a secondary metric; as long as the payout ratio was low, slow or even slightly negative EPS growth would not likely hinder dividend growth, but it could slow total return.
- 1-yr DGR: Less than 4% (0), between 4-8% (1), over 8% (2 pts). This was used more as a tiebreaker and for reference.
With this information, I proceeded to select stocks from each sector, starting with those that had the highest yield (2 pts) and highest DGR (2 pts). If the payout ranking was a zero, I verified the cash flow, and removed the stock if it did not pass that test.
Before final acceptance, there were two additional checks to see if the average trading volume was under 50,000/day (liquidity requirement) or if the stock’s performance was 20% below the S&P’s for four consecutive weeks (“-20% gap rule”). If either criteria was met, the stock was removed from consideration. This process was repeated (high yield with moderate DGR came next) until the quota for that sector was reached. I favored stocks with higher yield first, then DGR. The final list of 40 stocks is presented in the picture below. Click to enlarge:
The final portfolio has an average yield of 3.90%, and its 1-, 3-, and 5-yr DGRs are comparable to the initial universe at 7.2%, 9.7%, and 12.3% respectively. Each of these DGR values is higher than the original DCC-Small Cap portfolio, though the original portfolio had a higher yield of 4.1% at inception. (Note: If I had stuck to 30 stocks, the new portfolio yield would have been 4.2%.) The average 1-yr and 5-yr EPS estimates are 7.8% and 9.8%, and the beta of the portfolio is rather low at 0.80.
I have updated my virtual portfolio with the new holdings as of the closing prices on December 2, 2011. The total account value was $323,576.56, so approximately $8084 was invested in each of the 40 stocks after commissions (~$537). I will continue to track this portfolio and report on its performance relative to the S&P (SPY), S&P Dividend ETF (SDY), and the other DG model portfolios.
The next full rebalance will occur in one year, unless there are major market movements that necessitate the need for action. If any stocks cut their dividend, get bought out, or exhibit the -20% gap rule, they will be removed and replaced with another stock from the same sector based on the screening process.
I welcome feedback on this screening and selection process, as I continue to refine it based on comments, observations, and new learning. I hope SA members find the process and the recommended list useful for identifying potential candidates for their portfolios. I view this work as action research, an attempt to test out the recommendations from prior research and the comments that followed from it.
I use this model as a benchmark for comparison, and always recommend doing additional research on individual firms before purchasing their stock, though if a stock can survive all of the criteria used, I’d like to think it is above average to begin with! These results are based on historical data though, so knowing more about future prospects would be beneficial. Personally, I tend to prefer mid-to-larger cap DG stocks, but there are some names on this list that I own or have become interested in.
Disclosures: I own shares of NHI and O, and have limit orders in place for CBU and NUE.