“The deepest sin against the human mind is to believe things without evidence.” Thomas H. Huxley
I hate to be one of those bores who keeps repeating the same thing repeating the same thing, but yet more research shows that dividend stocks generate simply superb long-run returns, not only beating the market but absolutely walloping non-dividend payers.
What’s so intriguing about this newest research (pdf) is it also calculates dividend stocks’ alpha – a fancy rocket-science statistic measuring risk-adjusted excess returns. While mathematically complex, alpha’s purpose is straightforward. It provides an objective measure of whether investors taking higher risks are being rewarded with higher returns.
In a stunning reversal of the conventional tenets of risk vs. reward, the research shows dividend stocks clobber the market, yet expose investors to substantially less risk.
Is that a sweet deal, or what?
The study, conducted by Miller-Howard, also digs into risk vs. returns for low, medium and higher-yielding stocks. A bit of bad news: high-yielders were the one classification of dividend stocks with a negative alpha. Good returns, but at too high a risk.
Let’s look at the study background, then the results. Of course, this is just one study (one more study, to be precise). But that doesn’t mean we can’t add it to our store of evidence and consider it in the context of our individual goals and judgments.
The research crunched numbers from 1993 through 2007, spanning the ramp-up to the tech-bubbly late 1990s (which was no fun for dividend investors), the subsequent crash and the five-year bull market rebound. Some might argue that ending the study before the 2008-2009 dividend meltdown unfairly skews results. But high-quality dividend stocks, represented by ETFs indexed to Dividend Aristocrats (SDY) and Dividend Achievers (PFM), actually outperformed the market during that downturn.
The number crunching produced total returns and risk statistics for:
- The S&P 500
- The largest 1,000 stocks in the Zacks database
- All dividend payers
- Dividend payers yielding less than 3%, 3% to 6%, and 6%+
- Non-dividend payers
The study’s total return results won’t surprise regular readers of my articles. We’ve been there, done that. Dividend stocks outperformed the S&P 500 (SPY), with annual total returns averaging over 12%, compared with less than 10% for the benchmark. Stocks in the Zacks 1000 database also lagged dividend payers, but stayed closer at over 11%.
And non-dividend payers? With less than 9% annual total returns they underperformed everything in sight, despite two bull markets during the time frame under study.
The risk statistics provided the study’s surprises.
Overall, dividend stocks’ market-beating returns, coupled with their low riskiness, created a champagne-popping positive alpha of nearly 3%. That means these stocks’ returns were nearly three percentage points per year higher than the fair return for the risk taken. Hedge fund managers with 3% alphas can buy enough Porsches to always have at least one that matches the ties they’re wearing.
Non-dividend payers, the sad sacks of investing, were exactly the opposite. In addition to lagging the benchmarks, they handed investors far more risk than their relatively low returns justified, producing a skid-row alpha of negative 5%. (The S&P maintains a benchmark alpha of zero; the Zacks 1000 shows a barely positive alpha.)
Among dividend stocks, the sweet spot was 3% to 6% yields. These stocks scored better long-run total returns (nearly 13% annually) than their lower-yielding or higher-yielding cousins, as well as a higher alpha (2.7%). Archetypes here might include reliable, low-volatility classics like Johnson & Johnson (JNJ), Procter & Gamble (PG) and other dividend favorites.
Stocks yielding below 3% did almost as well, scoring nearly 12% returns and a 2.5% alpha. This group includes some now well-known low-yielders that skyrocketed over the study period, such as T. Rowe Price (TROW) and Jack Henry & Associates (JKHY).
And those 6%+ yielders? Average annual total returns were 11%, about in line with the Zacks 1000, and better than both the S&P 500 and non-dividend payers. But a negative alpha of about 1% says these stocks, in general, are a bit riskier than their returns justify.
That doesn’t mean good high-yielders make bad investments, of course. What’s not to like about 11% returns? But it does mean you might need to kiss a lot of frogs before one magically turns into your prom date. Because the mix of winners and losers averaging to 11% includes plenty of losers, such as Standard Register (SR), for example, rather than just perennial winners like Altria (MO).
So what can investors do about all this? Beyond the obvious step of sticking with dividend stocks so you can fill your garage with Porsches, I think a couple of things.
First, favor quality dividend stocks with sweet-spot yields between 3% and 6%. And as I’ve written before, if the dividends are growing, you’ll get even bigger benefits. When you venture into lower-yielders, the risk-return tilts slightly, so pay extra attention to valuations, cash flow, debt, return on equity and other indicators of entry point and quality. And those 6%+ yielders? Of the three dividend stock categories, they showed the lowest returns and highest risk, so choose carefully.
And in all this, as always, homework and due diligence make all the difference. Even great dividend growers can become great disappointments. So there’s no substitute for digging into a stock’s fundamentals before digging into your wallet to buy.
Finally, spread the word: those who are truly seeking alpha should seek out dividend stocks. In another article, I spelled out some reasons that dividend stocks, especially those with growing dividends, trounce the market decade after decade.
But I won’t go into that again here. After all, I’d just be repeating myself repeating myself.
Notes from the Math Club: The statistic alpha is a coefficient denoting the intercept derived from modeling deviations in portfolio vs. market returns over a specified time period. A positive value for alpha means the slope of the best-fitting model intercepts high on the y-axis, indicating better investment returns at every point on the fitted slope. The slope itself models the passive relationship between portfolio (or individual stock) and market variation, and is known as “beta.” Bored masochists can read more here.