Dividend yield investing as described by investing legend Geraldine Weiss is well defined. Even better, it can work as a total return approach or as an income investing approach. But, well defined doesn't mean it won't take some effort to implement.
Fortunately, because it is an established approach, there are historical base rates for performance and expectations. Using the AAII strategy implementation, for example, suggests a base CAGR of 9.8% through 2016. The S&P 500 had a CAGR of 4.5% for the same period. If you are an AAII member, you can log in and see the performance chart as well. The chart shows the strategy and the S&P 500 equity curves together and provides a sense of the relative draw-downs.
NOTE: If you use stock screening tools that allow back-testing, you should be able to create an even better base rate; one that provides risk measures, trading stats, and average number of companies passing the screen. More on back-testing later.
So why bother to personalize a strategy with historical success? The most obvious reason is to improve alpha - the risk adjusted measure of out performance. For many of us, making changes to reduce expected draw-down and volatility is worth some effort. I am way past wanting lots of volatility and draw-down in my investing strategies.
Also, if you are looking at this for income investing, some different rules have been proposed by various authors.
But consider variations to the published strategy not just to improve performance and lower risk, but as a way to avoid being "one of many devotees". There is some evidence suggesting that an influx of too many look-alike investors can affect performance. The Dogs of the Dow is a good example. A great historical strategy, it was inundated with lots of investors - even mutual funds - and not long after took a hit. Even though the DJIA companies were very large, the size of the common universe was only 30 stocks. Not a big enough pond for everyone to fish.
Even if you go with the pre-built AAII strategy, there is nothing forcing you to use the screen on the last day of the month, you could simply pick a day of the month and run the screen on that day every month. Or not even every month. At the very least, you have a base rate.
There are 5 areas of Dividend Yield investing where variations are realistic.
In the case of dividend yield investing, it is unlikely as a retail investor you will be able to replicate the Weiss Blue Chip universe. As described in the strategy definition, you can start with a list of Dividend Champions and eliminate companies that do not meet the 12 year EPS growth criteria. If you don't have access to 12 years of data, use what is available. NOTE: If you are down-loading a list of stocks for a universe, you must determine what rules have already been applied. For example, dividend growth may already be applied. Don't assume Dividend Champions are the same as Weiss Blue Chips. Check.
Below is a back-test of the Dividend Champions from 1/2/1999 through 1/2/2017. PLEASE remember that this is very imperfect since the list of companies is static, not dynamic. These results would suffer from survivorship bias to some degree. But interesting.
Another possible change is excluding some companies. I normally exclude MLPs (Master Limited Partnerships) not because they are bad investments, but because the constant stream of revised K-1 tax forms they send is a pain when filing my taxes.
Trading liquidity (generally not a problem with dividend paying companies) and market cap minimums can be applied to the base universe. If you are looking at US markets, it is reasonable to exclude OTC stocks even if they pay dividends. Build the highest quality universe.
Tortoriello, in Quantitative Strategies for Achieving Alpha, determined that in a broad universe, dividend yield alone is not a superior factor. However, applied to the healthcare, consumer staples, and energy sectors it is excellent. So a universe constructed from specific sectors may be worth consideration.
Finally, consider a universe built from or filtered by proprietary factors if you have access to them. Morningstar's Economic Moat companies might be interesting. They don't all pay dividends but the analysts at Morningstar believe the Wide Moat companies will be around for the next 20 years. Even the Narrow Moat companies are expected to last for the next 10 years.
There are value, growth and financial strength factors in Dividend Yield Investing as defined by Weiss.
The Dividend Yield investing strategy compares the current dividend yield with the historical high and low average yield. Current yield closer to the high average yield is better; meaning the stock is undervalued. So, generally high yield is good.
It may be possible to build these averages (only Stock Investor Pro has it pre-built), but it will be a lot of work. We have not tried it in any of the screeners we have reviewed, but plan to simply use dividend yield instead. Higher is better.
Shareholder yield is the other obvious variation. Higher is better. If you use it, you need to remember to use a payout ratio that includes buybacks as well as dividends.
This is a dividend yield investing, so dividend growth makes sense. If you are looking at this strategy as income investing (i.e., you intend to use the dividends as income and not re-invest them) you should keep this as is. You will want your income to grow.
One possible criteria variation is to use earnings growth (historical or estimated) in place of dividend growth. The idea being that a dividend payer is likely to increase the dividend if earnings are growing.
Changing the dividend growth rule to an industry relative dividend growth comparison is a reasonable rule variation. The rule would only consider companies whose dividend growth met or exceeded the related industry dividend growth.
Of the 3 financial strength factors, the payout ratio seems like a must, and it must correspond to the yield being considered. For payout ratios, lower is better.
In What Works on Wall Street, O'Shaughnessy found evidence that a ratio between 10% and 70% worked; this could be a variation to the rule insisting that the payout ratio be less than 50%.
There are lots of choices in debt/equity type financial ratios. Do a little digging if you want to consider a variation.
An alternate to current ratio might be quick ratio. Keep in mind, though, a dividend yield strategy is not a high turnover strategy so long term stability and strength make sense.
You will need an approach for choosing the "best" stocks passing the buy screen since, at some point, you will end up needing fewer picks than the screen provides.
Ranking is the mechanism used to pick the "best" companies from all the companies that pass the screen. This is the one area where the Weiss Dividend Yield investing strategy was silent. Maybe not completely silent, but there was no quantitative approach to picking fewer companies than passed the screen.
A single factor ranking system using yield seems to test well. The chart below divides the Dividend Champions list into 5 buckets on a weekly basis and measures the performance of the bucket from 1/2/1999 through 1/2/2017. Again, PLEASE understand this is not a qualified test since the universe is statiic and also will suffer from survivorship bias. But it is interesting.
The dividend yield ranking system also seemed to work on dynamic universes where the value and financial strength ratios were used to reduce the size of the universe. Dividend yield seems like a good starting point. The ability to rank with regard to closeness to the historic average may also work well.
Some multi-factor combinations can be evaluated if you have back-testing capabilities in your stock screening tools. Just be careful not to over-optimize. A win ratio between 55% and 70% is generally acceptable. Anything higher is probably highly optimized and unlikely to continue performing. It is also important to determine if most of the back-tested gains come from just 1 or 2 stocks. If yes, eliminate those and see how the back-test performs.
Plan to hold positions up to 3 years but you should regularly check your holdings by running them against a "when to sell" screen.
For example, if you are sticking with the original factors, you could close immediately if any of the following occur:
To replace sold positions, run your "buy" screen against your strategy universe, and select as many of the top ranked stocks as necessary to replace sold positions.
Once again, you need the ability to quantitatively pick the top stocks - ranking.
When you look at a performance chart of the strategy, there may be times when it would have made sense to be out of the market, or to use the same timing system to hedge. But, hindsight is 20-20.
The whole idea of timing is generally condemned by large investment houses supplying mutual funds and generally practiced by hedge funds. Your call. You can check out the idea on our Market Timing pages.
If you are familiar with and intend to use a specific stock screening tool, it may dictate some of the variations you have available for the implementation.
Building a dividend yield screen to implement your dividend yield investing strategy is not difficult and is the next step in the investing process.
Once your implementation is in place, you should begin by buying the top stocks (perhaps all at one time or over a period of time) in equal dollar amounts until you have the desired number. The Weiss preferred number is 20 but you need to make that decision for yourself.
The following books (especially O'Shaughnessy and Tortoriello) are goldmines of information on the historical performance of metrics used in investing strategies. If you are going to alter the dividend yield investing strategy by replacing a strategy metric with a similar metric (or a set of metrics) with superior performance characteristics, these are indispensable.
Faber, Mebane. Shareholder Yield: A Better Approach to Dividend Investing (2013). Focus - using net share buybacks in addition to dividends paid to determine the yield. Shareholder yield can improve the performance of dividend and income investing strategies.
Gray, Wesley and Carlisle, Tobias. Quantitative Value, + Web Site: A Practitioner's Guide to Automating Intelligent Investment and Eliminating Behavioral Errors. (Wiley Finance, 2013). Focus - identifying the best performing metrics. Excellent section on quantitative fraud detection. Note that the studies are based on cap weighted universes rather than equal weighted.
O’Shaughnessy, James P. What Works on Wall Street, Fourth Edition: The Classic Guide to the Best-performing Investment Strategies of All Time (McGraw-Hill, 2011). Focus – quantitative analysis of financial metrics and stock investing strategies derived from that analysis. Includes multi-variable ranking approaches.
Tortoriello, Richard. Quantitative Strategies for Achieving Alpha: The Standard and Poor's Approach to Testing Your Investment Choices (McGraw-Hill Finance & Investing, 2009). Focus - quantitative analysis of financial metrics and groups of metrics. Metrics and resulting performance are applied to sectors as well as standard universe of stocks.
We value your suggestions, comments, and questions.
Our goal is to make this site as useful as possible.
Stock screening tools and consistent, sustainable investing processes.
Why build a website instead of a blog?
About this site