There is an abundance of articles in Seeking Alpha's archives outlining the benefits of a dividend growth system where you accumulate stocks with consecutive years of dividend increases. One popular advocate of this system, David Fish, also provides a list (the CCC list) of such companies putting them into 3 categories depending on the number of years in which the dividends has increased.
In a recent Seeking Alpha article, I examined the stocks discarded by dividend growth investors and found these to contain even more outperformance on average than the dividend growth universe. The rule for these discarded stocks is that they have at least one year where the dividend stayed the same, but no year in which the dividend decreased. Sometimes a company may choose to 'anchor down' in bad conditions and maintain the dividend, which would exclude it from a dividend growth list.
In this article I want to examine the impact of the length of dividend history. I will look at these 'discarded dividend growth' stocks with a minimum 3, 5, 7 and 9 year dividend history where at least two years had the same dividend paid out.
In addition to this, I will look at single factor sorting of the stocks for investors who may want to buy the stocks with the lowest PE ratio, or the highest yield and so forth. I will outline one of my favorite single factors that is not related to value or growth - but sentiment.
Last, I will show the effect of simply choosing the biggest and most liquid names when investing in this group of stocks.
How Many Years of Dividend History?
The setup for all of the backtesting and screening is complements of Portfolio123.
- I start the testing in 2001 despite P123 having the ability to start in 1999 because I have to reach back up to 9 years of dividend history before portfolio formation.
- I rebalance this portfolio every 3 months, it is equal-weight with dividends being re-invested and the minimum share price is one dollar.
The portfolio sorts below are not meant to be investable as they each hold hundreds of stocks. The goal is to first discover if there is any major advantage to the minimum number of years relating to our dividend requirements.
|Minimum Years of Acceptable Dividend History||CAGR||Max Drawdown||Sharpe Ratio|
|iShares R3K ETF||5.08||55.77||0.07||Benchmark|
While there is some additive out-performance with the less restrictive criteria, it is difficult to say over this short period of time whether or not it is significant. When you restrict the maximum amount of years of the dividend criteria to 9 years, however, this smaller universe of stocks do seem to out-perform. The next chart holds discarded dividend growth stocks with more than 3 years but less than 9 years of acceptable dividend history.
3-9 years of dividend history
(click to enlarge)
Single Factor Sorting
It is impractical for the individual investor to attempt an equal-weighted portfolio containing hundreds of stocks. Each investor will usually have his own preferred factors to assist in stock selection. Some prefer high yields while others favor low PE ratios. Some will use book value while others price to cash. This next list of charts will hold the highest ranked 50 stocks according to the single factors listed below. There is a minimum 5 years of acceptable dividend history (one year where dividend is maintained but no year where it decreases) and rebalancing is every 3 months.
Please note that instead of ranking each stock in our 'discarded dividend growth' universe against each other - they will first be sorted into sectors. Only 'discarded dividend growth' stocks in each individual sector will be compared against one another for the highest rank. For example, we will look for the highest dividend yield in the utility sector, then the highest in the materials sector and so on. This sector neutral ranking and allows for more 'apples to apples' comparison of stocks against their peers while providing more diversity in the portfolio.
High Dividend Yield in Sector
Highest Earnings Yield in Sector
Lowest Quarterly Price to Book in Sector
Lowest Price to Cashflow Quarterly in Sector
Lowest Short Interest (% of Outstanding Shares) in Sector
Lowest Short Interest Ratio (Days to Cover) in Sector
I find these single factor sorts interesting as, on average, lower valuations do seem to produce excess returns, at a price. It usually is accompanied with higher portfolio drawdowns in bear markets.
High dividend yields does not seem to produce the best returns overall.
And the set of metrics that I like to include in most of my portfolio creation processes is short interest. Choosing companies with the lowest short interest ratio seems to produce some of the highest CAGR numbers with lowest maximum drawdowns.
But what about choosing among companies with the highest liquidity and largest marketcaps? Will this provide safety?
Large and Liquid Discarded Dividend Growth
This screen will use the same setup as above but the single factors will be average dollar volume over the past 20 days and market cap. Notice what happens to the returns:
Now suppose you only wanted to purchase 5 stocks and you created a 2 factor model that considered both liquidity and marketcap as your only two sorting factors. This would be your portfolio before considering trading costs and taxes since 2001:
2 Factor: High Liquidity and Big Marketcap
The 5 current stocks on this list is as follows:
When you look at each of these stocks individually, they may look like fine stocks to hold. The list is:
- Schlumberger (NYSE:SLB)
- The Walt Disney Company (NYSE:DIS)
- MasterCard (NYSE:MA)
- The Boeing Company (NYSE:BA)
- Merck & Co (NYSE:MRK)
All, except Merck & Co., come with at least a buy rating or higher. Again, with the exception of MRK, the long-term growth rates are well into the double digits. When you look at the TTM sales growth, only MRK is negative with MA into the double digits and the rest in the mid to high single digits.
So what is the problem?
The most liquid and largest companies are the playgrounds of big money. Huge funds need to invest in elephants. The increase in competition will usually result in less alpha as more people are competing for the same profits. It may be that the more a stock is traded and the more visible it is, the more efficient it becomes - which means forget trying to find stocks that are undervalued.
But don't ignore all stocks just because they are highly traded or have big marketcaps. This final chart uses a 2 factor ranking system that looks for the highest market cap stocks in each sector and the lowest average dollar turnover stocks per sector. This is the same sector neutral process described earlier in the article. The 5 stocks with the best combined score are held in this portfolio.
Big Marketcap (sector neutral) and Low Liquidity (sector neutral)
- CAGR 15.36%
- Max DD 41.91%
- Sharpe 0.46
There are indications that stocks discarded by dividend growth investors may still be solid investments. While there is a small amount of improvement with shorter dividend histories, it is inconclusive as to the 'right amount of years'. The best returns seem to come from limiting, not the minimum amount of acceptable dividend years, but the maximum.
While sorting the companies based on traditional measures such as value and growth, you may also want to consider alternative factors such as sentiment using short interest.
Buying the most commonly traded stocks that are well-known has not be a historically sound method to outperform the market. But if you are constrained by market-cap, consider looking for the biggest names in each sector (GICS) and show preference for stocks not as highly traded as their peers.