I'm not a fan of back testing data. I don't like the possibility of bias, the connotation of positive outcomes or even the implied usefulness of a test's results. Instead, I focus most of my time searching for the best businesses that I can find and then I try to determine whether or not those companies will remain wonderful in the future. Yet it's important to realize that there is at least a foundation upon which history might serve as a guide. Or as Warren Buffett would indicate: "In the business world, the rearview mirror is always clearer than the windshield." Sometimes to see where you're going you have to know where you've been. We know that, say Colgate-Palmolive (CL) provided a high return on shareholders' money over the past decade or that Wal-Mart (WMT) has grown earnings like clockwork each year. What's uncertain is whether or not these types of powerful businesses will remain fundamental.
With regard to my personal strategy, I find an increasing dividend to be a strong - although certainly not perfect - piece of a company's history that might shed some light into the future prospects of the firm. If Procter & Gamble has increased its dividend payout for 57 straight years then it seems especially likely that it would do so in the 58th year as well. Or at the very least, it seems plausible that management has a strong propensity to continue that streak of payment increases.
With the above being said, I'm about to present a reasonably compelling back test. Believe me; I understand the limitations of such information. So take the following with a healthy dose of skepticism if you must, but I would at least hope that you find the results interesting.
In my view the absolute best resource for finding companies that consistently increase their dividend payout is the Champion, Contender and Challenger spreadsheet (which can be found here) compiled by fellow Seeking Alpha contributor David Fish. As of latest update, there were 105 companies that have not only paid but also increased their dividend for at least 25 consecutive years. Familiar names like Coca-Cola (KO), Procter & Gamble (PG), PepsiCo (PEP), McDonald's (MCD) and Johnson & Johnson (JNJ) come to mind.
As I review the "CCC" spreadsheet monthly and regularly dedicate my investing funds towards a dividend growth strategy, I thought it might be interesting to see how these types of companies had performed in the past. Without any strategy whatsoever, I decided that I would compile the performance results of the 105 "Champion" companies over say 15 years and see what the end result happened to be.
Now compiling each set of data points - start and end prices, along with a decade and a half worth of dividends - for 105 companies doesn't exactly sound like the most exciting Sunday afternoon. Luckily, I have the helping hand of F.A.S.T. Graphs which makes relatively quick work of the process. For example, I have included the 15 year performance chart for Colgate-Palmolive to demonstrate my process:
Using this table as a guide, I simply clicked through the 105 tickers and inputted the return data generated by the F.A.S.T. graphs table into excel. Of the 105 Champion companies, 3 had somewhat odd histories that Mr. Fish was able to correct for in his spreadsheet but I passed over - thus leaving 102 companies with 15 years of performance results. Or I suppose more specifically - 14 years, 9 months and a business week - as the table presents return results up to the current date.
Now the benchmark can be widely debated, but for arguments sake we'll say that the S&P 500 is a decent proxy. As we can see in the CL performance chart, a $10,000 investment in the S&P 500 on 12/31/1998 would have turned into $15,587.37 - or roughly a 3.1% annualized return.
I won't bore you with the complete Dividend Champion past return performance table, but I will get to the punch line. If you aggregate the 102 companies, on average, they would have turned a $10,000 investment into $35,195.52 - or roughly a 9% annualized rate. In addition, $18,977.74 of that gain would have been from capital appreciation, while the remaining $6,217.78 would have resulted from a growing dividend. Said differently, the average dividend Champion would have nearly tripled your underlying principal while giving you over 60% of your initial investment back in just 15 years.
The obvious and overwhelming objection to these results would be survivorship bias. If a company is doing poorly and cuts its dividend, it's no longer on the list and thus not included in my return results. The argument is a strong one and should not be discounted. Yet, I would still advocate that these returns from the dividend companies are quite substantial in relation to the overall index. For example, without going back and including the failed payout raisers, let's simply assume that you picked a random collection of average dividend growth companies and 20% of your picks went bankrupt. Highly unlikely I would argue, but I believe it will get the point across.
In keeping the same dividend Champion averages, on a $10,000 investment your total return would now be about $28,000 with almost $5,000 of that still coming from dividends. In other words, one out of five of the companies in this scenario could go bankrupt and you would still oust the market over that time by about 4% a year. Perhaps this is the DGI in me, but it seems that at least in the past there was something to be said for owning a collection of companies that increased their payout each year.
Even though I didn't necessarily "stage" my back test experiment, it's also interesting to note that this data comes after starting with high P/E ratios at the end of 1998 and going through 2 recessions. A lot of people get caught up in the day-to-day fluctuations of the market. Yet if we're investing in for the long-term - regardless of the underlying strategy - an investor will likely do just fine.
Interestingly, there were 24 companies that put together total returns over 10% annually, with 8 of those companies - Raven Industries (RAVN), Eaton Vance (EV), Mine Safety Appliances (MSA), RLI Corp (RLI), UGI Corp (UGI), Nucor Corp (NUE), Federal Reality (FRT) and HCP Inc (HCP) - that actually returned more in dividend income than one's initial investment.
Furthermore, of the 102 Dividend Champions 94 of them provided returns that beat the S&P 500 over the last 15 years. Or course that also means that there were 8 companies that did not beat the index - Telephone & Data Systems (TDS), Nacco Industries (NC), AT&T (T), Diebold Inc (DBD), Cintas Corp (CTAS), 1st Source Corp (SRCE), Coca-Cola and Tootsie Roll (TR). Of those 8, just two - TDS and NC - would have lost investor money. Said differently, less than 1 in 50 of the present Dividend Champions made you nominally less wealthy over the last decade and a half.
Perhaps the most interesting company on the "did not beat the S&P 500 list" is Coca-Cola, long regarded as one of Buffett's preferred picks and a personal favorite of mine. To understand the reasoning behind this underperformance, I'd like to take a look at Coca-Cola's history through the lens of F.A.S.T. Graphs:
Here we see that the operating history of KO has been very strong, increasing by about 7.5% a year for the last 15 years. In addition, there has only been one down year for earnings during this millennium and shareholders have been consistently rewarded through increasing dividends for half a century now. The business of Coca-Cola is just fine. Thus the underlying reasoning behind the poor performance results of KO must be a consequence of market pricing. And that's exactly what we see: despite the current premium P/E ratio around 18 today, the market was pricing Coca-Cola at roughly 48 times earnings moving into 1999. Frankly even a positive return when moving from a P/E of 48 to today's 18 is impressive. It's hard to make investment headway when you start out with a 2% earnings yield.
Now perhaps you're still skeptical about the underlying tenets of holding a collection of companies that reliably increase their payouts each year. Conceivably, you don't place much faith in the backward looking results I've presented. To be blunt, I would not only encourage this but also welcome different opinions and strategies. Most approaches with a reasonable basis - whether they include dividends, a pure growth plan, using options to supplement income, or something else - will likely do just fine over the long-term. There are many ways to find success. For that matter, there are many ways to define success.
Yet the point of providing this obviously arguable backward looking data was not to influence you or necessarily sway you to the "DGI side." Rather the point was that some common misconceptions about focusing on a growing stream of income, at least in the past, held up quite well to the inherent criticism. For example, investor's who aren't focused on income might indicate that the dividend growth investor is obsessed with a growing payout and thus loses out on total return. Well, as demonstrated by this article, there were at least 90 companies that increased their dividend each year and simultaneously trounced the market's total return. It's not that we don't care about total return; it's just that the dividend growth investor knows that a growing stream of income eventually necessitates capital appreciation.
Or some might say that the results are all good and well, but it's not that easy to find these types of companies. But again there were at least 90 companies who had increased their dividend for at least 15 straight years. It's not like they are trying to hide from you. Check out David Fish's lists and you'll get a good idea of the types of companies who at the very least have the propensity to continue to increase their payouts.
Finally, I believe the Coca-Cola example is especially paramount. If you can buy a company at 48 times earnings and almost mirror (2.3% total return versus 3.1% total return) the index 15 years later I think you might be on to something. Not in that you're trying to mirror the index, but instead that paying a reasonable price for a wonderful company will often lead to business results or better. In turn, if companies like the Dividend Champions are outperforming the S&P, a good deal of companies must underperform the index. Our job, if we elect to select individual securities, is to simply find the best companies at reasonable prices.