Before expanding on my previous article, I wanted to quickly check what articles have been written on this topic and cite the works of other authors (where appropriate). As they say, once bitten, twice shy. As I started researching more on the topic of dividend stocks, I've come across a numerous Seeking Alpha authors. There is definitely a lot of demand (seen by a number of comments) and supply (seen by a number of articles) for articles discussing dividend investing.
To ensure that you don't end up picking expensive stocks, David suggests considering "value" metrics. Similarly, to minimize the possibility of buying a falling knife that draws up dividend yield, David suggests "quality" metrics. What could be better than trying to expose yourself to numerous factors simultaneously; i.e. yield, value, and quality. Of course, David goes beyond merely focusing on factors. For more detail, please refer to his article "Dividend Newbies: 5 Tips For Avoiding Common Mistakes".
Christopher, on an other hand, showed that income factor might have different market sensitivity in the US market (beta lower than 1) vs. international markets (beta slightly higher than 1). This finding reinforced my view that focusing solely on a single factor in choosing stocks and disregarding resulting portfolio characteristics might expose an investor to "unplanned" exposures.
Ploutos explored how combining two indices - the Dividend Aristocrats and Equal Weighting - in equal proportions has beat the S&P 500 in 14 out of last 16 years. Ploutos provides the list of ETFs that could be used to replicate this strategy.
Of course, the list of works above is not exhaustive, but I thought would relate to few ideas I want to discuss in further detail in this article:
- Combining factors (e.g. value, size, quality, low volatility, market) with dividend yield
- Topics related to practical implementation
Dividend investing comes in different shapes and sizes.
Few possible approaches involve focusing on high:
- Approach #1: dividend yield only,
- Approach #2: dividend and dividend growth rate,
- Approach #3: dividend yield and buyback yield,
- Approach #4: dividend yield, buyback yield, and dividend growth, and
- Approach #5: cash flow yield.
You might be asking why cash flow yield comes into picture anyways. We will cover it in detail below.
We have briefly discussed "Dogs of Dow" in the previous article, which involves targeting high dividend yield stocks. One might argue that focus on dividend yield only exposes investor to potential issues.
Dividend yield might be high due to negative price momentum; it is not fun to catch a falling knife! High dividend yield might not be sustainable. Perhaps, one should be cognizant of payout ratios.
In spite of the arguments above, Jeremy Siegel's findings indicate that historically high-yielding stocks generated sizable alpha (>3% over the long term) at a lower beta (~0.82). I guess as long as you use at least 10 or so stocks to get exposure to "high yield," idiosyncratic risks would be somewhat diversified.
As discussed in the previous article, one might argue that focusing only on dividend yield is missing important dividend growth component. Proponents of dividend growth model would have an entire academic model to support the assertion that dividend growth is an important input to consider. As we have seen in very limited backtesting presented in the previous article, the performance could be improved by taking into account dividend growth.
Approach #3 and #4
Those approaches were briefly mentioned in the previous article as well. One would need to conduct a thorough analysis of the impact of buybacks. Theoretically, buyback yield increases shareholder returns by increasing EPS as long as P/E ratio is not negatively impacted. There is a possibility that large buyback program can result in such an increase in financial leverage of the company that cost of debt can jump up putting pressure on investor confidence, and potentially bringing down P/E in larger proportion than improvement in EPS.
Additionally, I did not find it useful to research the usefulness of incorporating dividend buybacks. The reason: buybacks, typically, are not a reliable source of shareholder return. Companies tend to engage in buybacks during later stages of the cycle and buybacks are the first ones to get cut before companies consider cutting dividends. Given that buyback yield could be pro-cyclical in nature (this statement needs to be validated), adding buyback yield to dividend yield (and dividend growth) might be detrimental to performance.
Cash flow yield?! Why would one consider cash flow yield when discussing dividend investing?
- For dividend stocks, cash flow yield could be used to quickly assess the ability to continue paying similar to payout ratio
- For non-dividend stocks, cash flow yield is indication of value-generation capacity of the company
So you want me (dividend and/or income investor) to consider non-dividend paying stocks?
Of course, as an income investor, you should focus on total return! Discounted cash flow model would argue that cash flow yield is one of the key inputs in value creation and, hence, expected a total return.
You can still get your "income" with non-dividend stocks. This would involve selling your stocks that likely (and hopefully) generated long-term capital gains. If you are investing in your taxable account, you would be subject to similar tax rate whether you get it in the form of a dividend or long-term capital gain. Of course, one might argue against "eating your principal" and/or make a point about using principal during market downturns.
I hear that using principal is mentally and emotionally disturbing. However, let's clarify that one should think of his principal as his tax base/initial acquisition price. Capital gains are not your principal. If anything, focusing on non-dividend stocks gives you the flexibility to choose when you want the access to your money instead of being forced to receive your dividend. This might be more attractive from tax planning perspective.
But, we have a historical analysis showing that high-yielding dividend stocks outperform. I am not sure that high cash-flow stock would outperform. Do you have a proof?
This brings the entire discussion back to one of the key assertions I've made in the previous article. Generally, dividend stocks exhibit "value" factor (e.g. most of the dividend ETFs show value tilt). What is high cash-flow yield? Well, it is an indication of "value". Whether you get "value" through P/D (price to dividend ratio; inverse of dividend yield) or P/CF (price to cash flow), you are still generally exposed to "value" factor. You might want to consult works of James O'Shaughnessy; he showed that filtering stocks using cash flow ratio resulted in a superior historical return. However, his work goes well beyond the cash flow ratio only and would greatly benefit the interested reader.
Combining factors with dividend yield
It is not hard to find high-yielding stocks that also showed lower historical beta and trading at lower P/E. For instance, a sizable number of large-cap dividend aristocrats show low historical beta ("low volatility"). Some of the names would be consumer staples (which represent ~25% of dividend aristocrats and the most overweight sector vs. S&P 500) that can provide you with additional "wide moat" exposure as well.
The beauty of such multi-factor approach is that typically factors exposures in such cases would be reasonably stable. On the negative side, you might end up with low-return and even lower risk stocks; i.e. you would probably generate alpha, however, the overall absolute return could be low. In this case, one can always argue that you cannot "eat Sharpe ratio". Unless, of course, you are ready to leverage up the portfolio.
This brings me to my next point. One might consider investing in "low volatility" and "wide moat" staples as a safe bet for volatile periods and recessions. Historically, staples would outperform any other sector during recessions. However, they don't bounce back as some of the more cyclical sectors. This could be attractive to investors who rely on their dividends for their income requirements: they go through a smaller roller-coaster while having the stronger conviction that those companies are likely to withstand the turbulence.
For an investor with longer time horizon, "low volatility", "wide moat", and potentially "quality" staples combined with the reasonable amount of leverage could be of interest as well. We will discuss how to "eat your Sharpe ratio" in the next section.
Even though I've mentioned before that high-yielding stocks tend to exhibit "value" exposure, this might not be the case for a large number of stocks. For instance, as of this writing, only four of ~50 dividend aristocrats have P/E below 15. Therefore, before concluding this section, I wanted to highlight that I agree strongly with David Van Knapp that one should not overpay for any stock, including dividend stocks. As an example, I've reviewed how Core 10 portfolio (consisting of highest yielding dividend aristocrats) might be further augmented by using simple P/E and P/B metrics (similar to the one advocated by Graham).
How to "eat your Sharpe ratio"?
As discussed above, some groups of investors might benefit from levering up the safer portfolio of "low volatility", "wide moat", and potentially "quality" staples to meet their risk appetite. This would allow an investor to maintain expected alpha while not coming shortly on absolute return (a leverage is not suitable for a large cross-section of investors; please consult your judgment and/or your financial advisor!). There are following approaches to leverage the portfolio using:
- Margin loans
- Levered ETFs/ETNs
Futures and margin loans might be too much of a hassle for many investors and introduce liquidity risk (margin calls in the worst possible time). To pursue leverage through those instruments, one should be comfortable with extra work involved and have resources that will ensure that he/she will not be squeezed out of their positions.
This leaves me with deep in the money LEAPs. The downside is somewhat larger transactions costs (bid-ask spread), which can quickly erode your returns. Compared to LEAPs, leveraged ETFs/ETNs might be easier to put in place and manage. However, one needs to understand the implication of performance decay of daily leveraged products (same applies to monthly leveraged products to a lesser degree).
Personally, I use LEAPs and levered ETNs/ETFs. I plan to use margin account and futures as my portfolio increases in size.
How to combine factors?
You might, of course, consider using stock screeners to come up with the list of stocks showing exposure to numerous factors including the dividend. However, one should keep in mind that factor investing might end up exposing you to some "unplanned" risks. One of the examples is sector concentration risk: e.g. chasing dividend yield might tilt your portfolio heavily toward utilities and telecom. "Quality" might end up pushing you in the direction of larger caps unless you control for "size" factor.
Therefore, one would be well-advised to keep total portfolio characteristics in mind as well. For example, when it comes to sector allocations - you might want to ensure that sector allocation stays in line with your benchmark. In others words, if you want to tilt your portfolio to high dividend stocks, you might want to pick few stocks from each sector.
Some ETFs, such as SDOG, would do it for you. SDOG takes each sector and chooses five highest-yielding stocks from each; each position is equally-weighted. As a result, you end up having equally-weighted sector exposure to high-yielding stocks. As you might have heard, historically, equal-weighting sectors in S&P 500 outperformed market-cap S&P 500. This is achieved through not being over-exposed to the booming sector (e.g. Tech during early 2000s and Financials in 2008).
If you like the idea of equally-weighting sectors (and style) and prefer to have minimum involvement in the process of daily management of the portfolio, you might find Personal Capital as a useful tool. Personal Capital provided very interesting whitepaper on its website that you might be able to replicate yourself, as long as you have the willingness and large enough portfolio.
Don't blindly buy highest-yielding stocks. You might do better by considering additional inputs (e.g. dividend growth) and factors (e.g. value, quality, size), and don't forget to check overall portfolio exposures so that you are not unintentionally over-exposed to specific sectors.
ETF discussed: SDOG
Disclosure: I/we have no positions in any stocks mentioned, but may initiate a long position in SDOG over the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.