Swedroe: Rethinking Dividend Strategies

Is getting cash flow from dividend-focused equity strategies as attractive as it seems?

During bear markets, the dividends thrown off by companies provide the cash flow required, while a total-return approach requires one to sell shares to provide the cash flow—a clear advantage of dividend-focused strategies that those who favor them are quick to point out. This blog addresses that issue specifically.

We’ll begin our discussion by pointing out that any strategy that focuses its screen on dividends is likely to result in a portfolio that pays higher dividends than a strategy that doesn’t. This is true of any strategy based on a single metric. For example, until recently, the small value fund of Dimensional Fund Advisors (DFSVX) used a single screen of price-to-book (P/B) ratio (it recently added a profitability screen).

On the other hand, Bridgeway’s small value fund (BOSVX) uses four different screens, of which P/B was just one. (Full disclosure: My firm Buckingham recommends Dimensional and Bridgeway in constructing client portfolios.)

While Bridgeway’s funds typically had lower prices relative to earnings, sales and cash flow, the P/B ratios were typically almost identical, or they even slightly favored DFSVX. Thus, from the perspective of the other value metrics, or from the perspective of a combined measure, BOSVX was “more value-y” than DFSVX, and would have higher expected returns. However, if you only looked at the P/B metric, you wouldn’t find that difference. In fact, as I mentioned, at times it favored DFSVX.

Similarly, a dividend strategy such as an investment in the SPDR S&P Dividend ETF (SDY | A-66)—it seeks to closely match the returns and risk of the so-called High Yield Dividend Aristocrats, which Morningstar classifies as a large value fund—is going to have a higher dividend yield than a similar large value fund such as DFA’s DFLVX. SDY has a dividend yield of about 2.7 percent, while DFLVX’s dividend yield is only about 1.7 percent. However, in terms of the other value metrics, DFLVX is much more “valuey,” with much lower P/E, P/B and P/CF ratios.

What you also observe is that there really isn’t that great a difference in yields between the two value strategies—about 1 percentage point. In other words, not screening for dividends doesn’t mean you only buy the stocks of nondividend payers, it means you don’t avoid them as a dividend strategy does. You will also buy the stocks of dividend payers if they meet your other buying criteria.

In this case, if you needed to generate 2.7 percent cash flow from your portfolio, SDY would provide the full 2.7 percent. If you owned only DFLVX you would have to sell just 1 percent of the value of your shares.

However, as I have pointed out, the prices of the stocks of SDY are now lower by the 2.7 percent dividend SDY paid out, while the stocks of DFLVX are lower by just 1.7 percent. Thus, while you will have fewer shares of DFLVX, because the dividend was smaller, its price will be relatively higher, offsetting the reduction in share ownership. Your expected returns will be the same whether DFLVX paid out 2.7 percent in dividends or it paid out its 1.7 percent dividend and you sold 1 percent of value of your shares.

The other problem typically cited is that it’s one thing to sell shares during bull markets, but when you sell in bear markets, you’re selling at the worst time. There are two problems with this line of thinking.

The first is that dividends are the equivalent of a “disinvestment” in a company in the same way as is a self-made dividend (via the sale of shares). The second is an even more important one: Those who rely on a total-return approach don’t need to hold all their investments in equities.

They typically also own bonds, just as do some investors using dividend strategies. They also adhere to an asset allocation plan that requires them to rebalance on a regular basis. During the accumulation phase, rebalancing is typically done by buying the assets that had underperformed, of if need be, to sell some of the outperformers. In other words, you wouldn’t sell stocks during a bear market; in fact, you’d be a buyer.

Sensible Drawdown Approaches

Now let’s look at the withdrawal stage.

In a bear market for stocks, you’ll be meeting both your cash flow needs and the need to rebalance by selling the outperforming safer bonds, not your stocks. Thus, in bear markets, unlike those that rely on dividends for the cash flow who are in fact disinvesting from their equities, a total-return approach would avoid such sales.

We’ll also take this opportunity to provide two examples of why dividends reduce the value of a company.

Company A is deciding whether to pay a dividend. Instead, it chooses to use the cash it would have paid out to buy back its stock. In either case, the company’s future earnings prospects are exactly the same. However, under the buyback scenario, it has fewer shares outstanding. Thus, its earnings per share are now higher. Thus, since all else is equal, the company’s stock will trade at a higher price.

Alternatively, the company could choose to pay back some of its debt. In that scenario, its future earnings are also higher, because interest expenses are now lower. In addition, it would have a lower leverage ratio than if the company had paid a dividend. That makes it a less risky company.

Again, since all else is equal, both factors should result in a higher stock price. In fact, the authors of the 2012 study “Enhancing the Investment Performance of Yield-Based Strategies” found that by expanding the definition of “dividend yield” to include three alternative measures of dividend yield, the explanatory power of the dividend yield could be improved.

The three alternatives were:

• PAY1: Dividends plus share repurchases.
• PAY2: Dividends plus net share repurchases.

• SHYD: An acronym of sorts for shareholder yield that includes net-debt paydown as part of the yield calculation. Net-debt paydown yield is measured as the year-over-year difference in the debt load of a firm, scaled by total market capitalization.

The authors found both that there’s no evidence that high-dividend strategies systematically outperform. They also found that SHYD was the top performer in virtually all subperiods, providing a much better explanation of returns. This supports the evidence from two other recent studies.

The study “On the Importance of Measuring Payout Yield: Implications for Empirical Asset Pricing,” found that the dividend yield does a poor job predicting future returns in a sample that runs from 1972 through 2003. In addition, the 2003 study “Predicting the Equity Premium with Dividend Ratios” found that that dividend yield has little predictive ability out of sample.

For those interested in learning more about the differences between a cash-flow and a total-return approach (using homemade dividends when needed to supplement the interest and dividends generated by the investment portfolio), Vanguard’s research team has a paper on this subject.

Their recommendation, like my own, is to use a total-return approach, with the investor first deciding on an asset allocation that is based on his or her unique goals and objectives, and their ability, willingness and need to take risk. “This decision should be the investor’s highest priority.”

Investors should then stick to their plan, rebalancing along the way as needed.

Larry Swedroe is director of Research for the BAM Alliance, a community of more than 130 independent registered investment advisors throughout the country.