Dividend Investing Part 4: Do Dividend Stocks Outperform?

Investors usually choose a dividend strategy because they believe that dividend stocks will outperform. Investing for dividend income may have some psychological advantages that help real-life investor performance. Dividends may also be tax-efficient for personal investors in low tax brackets. Those investing through a corporation might even be able to have more capital invested, resulting in higher dividends. Whether that is tax-efficient or not depends on multiple factors. Further, the irrelevance of dividends means that we are essentially exchanging dividends for capital gains as part of the total return of a stock or fund. Capital gains benefit from tax deferral until realized, and are only partially taxed, whether personally or through a corporation.

However, if we are selecting dividend stocks, do they really have the same total return, or do they outperform the broader markets? If dividend-paying stocks have a higher total return, that advantage may still be overshadowed by the earlier or higher taxes incurred on dividend income. People love stories, and the narrative that dividend-paying companies are better investments is a powerful attractant for dividend investors. Some narratives can be very expensive stories for investors. Is the dividend narrative one of them?

The Dividend Narrative

The dividend narrative goes along the lines of only strong companies with good cash flow pay dividends. Therefore, they are better investments and dividend stocks outperform the broader market. There is some evidence to support this notion. For example, in a whitepaper by Hartland Advisors, theoretical portfolios built from stocks within each quintile of dividend-payers showed that those paying higher dividends had greater returns based on 90 years of historical data. With the notable exception of those paying the highest dividends, they all had a higher total return than dividend non-payers. This finding also held up when they looked at 20-year rolling periods.

---

High Yields & Payout Ratios

One of the reasons why the quintile portfolio of the highest dividend payers did not have the highest historical return is that a very high dividend yield may not be sustainable. Dividend yield is the dividend paid divided by the stock price. If the future cash flow of a company looks to be worsening, the price will drop because the market is forward-looking and accounting for that. This causes the yield to increase, and an unusually high yield may be a warning sign.

Another purported warning sign is if the dividend payout ratio is high. The dividend payout ratio is the ratio of the dividend relative to the company’s cash flow. A higher payout ratio indicates that the company is allocating a greater portion of its free cash flow to support the dividend. That leaves less wiggle room for the company in the event of a cash flow crunch. What is considered a high payout ratio varies by industry, depending on how reliable cash flow is. So, rather than a specific number, comparison to peers or a dramatic change is what may be useful.

---

Dividend Growers, Stagnators, & Cutters

Predicting whether a stock will be a dividend grower, stagnator, or cutter is important. As shown in the data below of stock performance based on their dividend policy in a given year, changes in the dividends paid are strongly related to the total return. Dividend growers are rewarded while dividend cutters are brutally punished relative to the broader market.

Markets price future risk efficiently using all available information. So, unless you know something about the future that the market does not, you cannot predict whether the dividend of a company is going to be a grower, stagnator, or cutter moving forward in time. Regardless, dividend investors try to use historical dividend policy, payout ratios, and yield in an attempt to see the warning signs that the market may be sending. Even if they do not know the story behind it yet. By the time it is fully known, it is too late.

Dividends Are a Surrogate Marker

Companies recognize that changes to their dividend policy will have an impact on stock price and the company’s ability to access capital. Particularly, that dividend-cutting will be punished. So, they are often reluctant to initiate or grow dividends unless they are certain of cash flow. You can’t easily put that genie back in the bottle. They are also reluctant to cut them when there are problems, hoping that they are transient. This makes dividends more of a surrogate marker for outperformance and more loosely tied to a company’s health.

Astute and nerdy readers may have recognized that the data from the preceding chart, showing quintile portfolios of dividend payers outperforming, used Kenneth French’s data. Kenneth French is famous for the Fama-French five-factor model that explains most of the historical stock returns. Why then are dividends not one of his five factors that explain portfolio returns? The reason is that dividends are only a surrogate for other factors that better predict outperformance than dividends do. Dividend stocks tend to tilt towards value pricing (HML), conservative investment management (CMA), and profitability (RMW). When those factors are accounted for in multivariate analysis, dividend yield is not a statistically significant predictor of return.

Accounting for those factors, dividends are no longer a significant predictor of stock outperformance. That means that there are stocks that do not pay dividends that have those positive features. Conversely, there are also stocks paying dividends that do not have them. Therefore, targeting value, conservative management, and profitability directly would be a more effective way of capturing premiums compared to targeting dividends. Here is an example to illustrate this from Larry Swedroe.

Practical Challenges

Dividend stocks with the right characteristics have historically outperformed the broader market. While directly targeting factors would be the preferred strategy, in keeping with the above research, it has practical challenges. Once a factor is widely known, it likely becomes less profitable to target. The efficient market arbitrages it away as investors price it in. However, the factor may persist if there are other reasons why it is difficult to capture. Screening stocks for these factors is a labor-intensive process. That either means more work for investors or paying a fund manager to do it.

The factor studies used a long-short methodology. They go long stocks that have the desirable factor and short the ones at the other end of the spectrum. In real life, shorting has extra trading costs. There will also be trading costs and potentially taxes triggered as stocks move in and out of the portfolio as their characteristics change. The research portfolios are typically reconstituted monthly, and depending on the factors, may change a little or a lot.

Put together, how an investor or fund operationalizes its factor strategy may eat up the benefit. Dividends may seem a simple alternative. However, higher costs for choosing a portfolio of dividend stocks likely to maintain or grow their dividends also eat into performance. That could come from transaction costs and behavioral errors for individual stocks, or from the higher MERs associated with dividend-focused ETFs. Further, the excess return to capture may also be less since dividends are only a surrogate marker. We’ll examine these dividend investing strategy implementation issues further in the next post.