Managing Risk: The Dividend Payoff
John Buckingham will speak at the 2015 AAII Investor Conference this fall; go to www.aaii.com/conference for more details.
Scandalous accounting methods, massive corporate failures and increased global economic and political instability over the past four years have produced one of the most volatile and generally bearish market environments many investors have ever experienced. The resulting greater risk built into the average investment has left many investors seeking methods to shore up returns while minimizing risk.
Historical evidence suggests, however, that an investment strategy focused squarely on dividend-paying stocks can offer a perfect blend of maximum returns at a below-market level of risk. That’s what we found based on a study of the returns of the largest 100 dividend-paying and 100 non-paying stocks by market capitalization since 1985. Similar results were found for the 100 largest mid-cap dividend and non-dividend paying stocks. Of course, it is important to note that the study period is short compared to the history of the market, and the findings would not necessarily be repeated in future periods.
It may be surprising for many investors to learn that in the past 15 years, the largest dividend-paying stocks posted superior returns relative to the market and to their non-dividend-paying peers while offering lower levels of risk. Lower risk and higher returns?
In the study, we used beta as a measure of risk. Beta measures the volatility of a security’s return over time relative to a market index. A security that has a beta greater than 1.0 means that, on average, when the overall market was moving up, the security moved up to a greater degree and when the market was down, the security was down to a greater degree. A security with a beta of less than 1.0 means the security saw a smaller decline than did a falling market, and a smaller increase than a bullish market.
In our study, we found that the beta on the returns of large-capitalization dividend-paying stock over the study timeframe was 0.88, compared to 1.57 for the non-dividend payers. Dividend stocks, therefore, were less risky investments over the study timeframe.
At the same time, however, the dividend payers achieved an average annualized total return (capital appreciation plus reinvested dividend income) of 13.4%. That exceeds the 11.7% obtained by their riskier non-paying peers. These results are featured in Figure 1.
It would be reasonable to assume the higher-beta, non-dividend payers would outperform the dividend payers. And, in fact, that was the case before the bursting of the bubble—the excessive growth in U.S. markets starting in late 1998 and ending in March 2000. You can see in Figure 1 that the largest non-dividend stocks began to significantly outperform the dividend payers in mid-1999. At that point, the higher risk of these stocks began to really pay off.
This is further illustrated in the left-hand panel of Figure 2: The 100 largest non-dividend payers returned an average of 27.7% per year in the pre-bubble period, the 10¼ years before April 2000, outpacing the dividend payers’ 20.7% per year. On the other hand, returns over this time period for the dividend payers still proved relatively less volatile, with a beta of 0.94 versus the non-payers’ 1.41.
Nonetheless, when the market turned south in 2000, lower volatility actually worked in favor of the dividend payers—as shown in the right-hand panel of Figure 2. Large-cap dividend payers as a group lost 1% annually in the four and one-third years since the end of March 2000 and the Wilshire 5000 lost 6.5% per year in that timeframe. But the non-dividend payers lost 18.2% annually. That’s just slightly less than the abysmal performance of the tech-heavy Nasdaq Composite index, which lost 18.5% per year since then. Once again, the dividend payers were less volatile in their performance, turning in a beta of 0.78, versus 1.81 for the non-dividend payers.
We extended our analysis to mid-cap stocks, for which we found a slightly different story in terms of overall returns—but one that still demonstrated a penchant for strong returns at sharply lower levels of volatility. These findings are detailed in Figure 3.
It would be unwise to assume dividend-paying stocks will continue to outperform the market and their more volatile peers while showing lower volatility, as the higher volatility experienced by the non-dividend payers should result in higher returns in the long run. In fact, current S&P 500 members that paid dividends in 2003 returned, on average, 34.1% last year, versus 62.7% for those that paid no dividend. But the former currently maintain an average beta of 0.8, versus 1.7 for the non-payers.
Therein lies the crux of the dividend investment story. It is their history of offering lower volatility and a proven stream of income—both of which can offer vital protection in a declining market—that arguably makes them so attractive.
Often, it is up to the market observer to draw tangential conclusions from actual observations. That is, it is easy to look at empirical data—existence of dividends and stock returns—to see that the two are related. It is far more difficult, though, to figure out why. Indeed, dividends are a fine example of this conundrum. Why do stocks in companies that pay dividends generally offer less volatile returns?
Most companies begin to pay dividends once they reach a level of business maturity where attractive investment opportunities are relatively less available, while cash flow generation is stable or growing more slowly than in the past. The payment of dividends thus conveys stability—or lower risk—within the enterprise.
In fact, a recent study (by University of Michigan Business School Professor Douglas Skinner) found that the reported earnings of dividend-paying firms are more persistent in future periods (meaning that changes in reported earnings are permanent)—an effect that increases with both the size of the firm and the magnitude of the payout.
Another recent study (by Duke University Business School Professor Campbell Harvey and four colleagues) shed light on the reasons why companies pay dividends, and offered further insight into the drivers of volatility in dividend stocks. The study involved a survey of 384 mostly U.S.-based finance executives and an additional 23 in-depth interviews with executives regarding their dividend-payout policies.
One of the key findings was that dividend-payout decisions are at least as important as decisions to invest in new projects. Further, about two-thirds of respondents said their firm would raise outside money before it would cut dividends. The question with the greatest affirmative response was whether managers actively try to avoid cutting dividends—94% of dividend payers either strongly or very strongly agreed that this was the case. As for initiating dividend payments, two core reasons rose to the top:
- First, companies will begin to pay dividends because institutional investors want them to;
- Second, companies will initiate dividend payments based on greater assuredness of a sustainable increase in earnings.
Regarding investors’ views of dividends, four-fifths of respondents said they believe dividends convey information to investors and about two-fifths said dividends make a stock less risky.
The case for dividend-paying stocks can also be attributed to another motivating factor. According to financial research firm Ibbotson Associates, dividends comprise a substantial portion of long-term gains. Dividend income for large-cap companies generated 41% of their average annualized 10.4% return from 1926 to 2003. That portion fell as company size decreased, but remained an important contributor. Without dividends, then, investors could lose out on two-fifths of their long-term gain potential.
Notably, the fact that the portion of long-term returns comprised of dividends is so sizable can be attributed both to the greater stability of dividend payers and to the fact that dividend payments have increased over time. U.S. dividend payments in the aggregate increased 183% in real terms (647% nominally) from 1978 to 2000. In fact, dividend payments have increased despite the overall decline of 49% in the number of firms paying dividends over the timeframe. That’s because nearly all of the companies that stopped paying dividends had originally offered very small payouts and dividend payments increased substantially among the largest firms, reflecting a very large increase in their real earnings.
Dividend stocks have historically offered lower volatility with potentially market-beating returns given the right selection strategy. Dividend income has also comprised a significant portion of long-term gains. An added bonus is that dividend gains may also receive more favorable tax treatment going forward.
All of these characteristics make a compelling case for a long-term focus on dividend-paying stocks for those investors seeking a lower-risk investment strategy.