Thursday, September 17, 2015

Survivorship bias in Dividend Investing - Part 2

This is the second part on suvivorship bias in dividend investing. In part 1, I laid the grounds that investors who put their money to work in dividend growth stocks are not suffering by survivorship bias ( despite the efforts of greedy money managers to portray ordinary investors in a negative light)

I wonder if the same type of logical analysis on survivorship bias is performed by investors who are encouraged to invest in US Equity markets, when they are told how an investment in the S&P 500 or Dow Jones Industrials average would have performed over the past 10, 20, or 50 years. If you consider the event of purchasing shares of Johnson & Johnson (JNJ) due to its history as an example of survivorship bias, then you should not be using historical data on S&P 500 or Dow Jones Industrials average over the past 50 years either in order to prove your point about equity investing. Somehow, this point is lost on so many investors. When discussing long-term returns on equities over the past two centuries, you often hear about the US or UK stock performance. However, you never hear about the performance of a Chinese stock investor or a Russian stock investor from the middle of the 19th century till now. My great-grandfather was born and lived in Eastern Europe more than 100 years ago, saved almost his entire salary working in the coal mines and invested his savings in agricultural land. Unfortunately for him, the communists nationalized his land when they came to power. If he were in the US, he would have died a rich man after decades of compounding. Too bad he weren't.


That being said, finding companies that have managed to grow dividends for a long time is not an example of survivorship bias. Rather it is a first step in a long process to vet companies that might be suitable for investment. I have looked at the original dividend aristocrats and dividend champions lists from the early 1990s ( the first time these were available), and know for a fact that these companies in aggregate are good investments for long-term holders. Not all of them were good investments, but the aggregate performance has been satisfactory. If you buy and hold a diversified portfolio of dividend champions for the next 20 – 30 years, sell after dividend cuts and replace with new dividend champions, you are very likely to earn more dividend income, and a pretty decent amount of unrealized capital gains. Of the dividend champion companies from 1991, many are still around, while a larger portion was either acquired, split, reorganized etc. A very small minority of investments actually failed outright, thus resulting in dividend cuts and substantial impairments of capital. For investors in Anheuser Busch or Wrigley, both of which were acquired at premium prices, acquisitions were not that bad of a deal.

(Fun fact - there were 83 dividend champions in 1991, according to the old stock manuals from that time. Of them, only 2 failed outright, resulting in 100% loss for shareholders- KMart and Winn-Dixie. Fourteen of the 83 remained as Dividend Champions as of late 2014.  The rest ended up being acquisition targets, merged with other companies, spun-off divisions, froze dividends from time to time, or cut dividends only to start growing them again).

So basically what I am trying to say is that statistically speaking, if you buy portfolios of dividend champions/achievers, you are very likely to do well in most situations. Statistically speaking, according to studies by NDR, over the past 40 years dividend growths stocks have produced much higher total returns than non dividend paying ones and those that merely paid a dividend. Now you will have people that do particularly poorly or exceptionally well in any aspect of life. Dividend growth investing is no different. However I do believe that dividend growth stocks do provide investors with an edge, as I think that once you set the dividend growth in motion, it can stay in motion for many years. Even if the dividend is frozen, it can be raised again, as Kellogg (K) and Hershey (HSY) have shown to their long-term investors.

(Even index investors end up with different results than those of the index they are investing in. According to Morningstar, investors in the Vanguard S&P 500 Index fund (VFINX) have earned 1.37%/year over the past 15 years. The index fund however earned 4.24%/year over the past 15 years.)

If you have an edge, the longer you keep investing, the more likely you are of earning a profit (through dividends and share price appreciation). If you do not have an edge, you could get lucky once or twice. But, if you keep playing, you will lose money in the long-run. If you do not believe me, go to your local casino, and play any of the games every day for one year or buy a lottery ticket every month. That is an example of a game with bad probability of success.

This is because rising dividends are a symptom of companies that have managed to grow earnings over a certain period of time. When managements expect further increases in earnings, they get optimistic and raise dividends as a result. Most companies that manage to grow earnings over time, and pay a growing dividend, tend to have some sort of a competitive advantage that allows them to grow in the first place. It is quite possible that some companies manage to increase dividends because they were lucky, and were in the right place at the right time. However, luck eventually runs out at the end of one or two economic cycles, which is why I require at least 10 years of dividend growth, before researching a company. What you should be focusing your attention is not on outliers, but situations where the odds are high that a large portion of companies will keep delivering the results you expect them to, for extended periods of time.

Only a certain type of company with a certain type of strong business model can afford to grow dividends for over 25 years in a row. That's not enough, and is just the first step in analyzing each prospective investment. In my case, I have discarded companies from consideration, even when they had managed to grow dividends for over 50 years in a row.

You also increase odds in your favor by analyzing companies one by one, in order to determine if there are factors for future earnings growth. You also increase your odds of success by requiring a certain entry valuation, and also by diversifying risk by allocating cash slowly into at least 30 – 40 companies.

In addition, many dividend growth investors are not emotionally attached to the companies they own. Rather, they buy for the rising dividend if it is attractively priced and if they believe that future earnings growth can support future dividend growth.If the dividend is cut, they are out of the door and tend to reinvest their sale proceeds into another quality dividend growth company selling at a reasonable price. I have added to my position in PepsiCo (PEP) because of its ability to grow dividends for extended periods of time due to its wide moat and strong competitive advantaged. However, while I own PepsiCo which has raised dividends for 43 years in a row, I would not hesitate to pull the trigger if it cuts distributions for me. A dividend cut would most likely be a symptom of a bigger problem at the company, and show that things might have changed. In the case of competitor Coca-Cola, I have not added to my position for quite some time, due to stagnation in earnings per share. I am patiently holding on the sidelines however, and allocating Coca-Cola dividends selectively into my best and attractively priced ideas at the moment.

For those reasons, I do not regard dividend growth investing as an example of survivorship bias. This is because income investors have a positive expectancy that a portfolio of dividend growth companies will keep growing dividends, they also monitor company fundamentals, and do not merely buy stocks without understanding future growth potential as well as current valuation.

Full Disclosure: Long JNJ, KO, PEP, GE, WFC

Relevant Articles:

Where are the original Dividend Aristocrats now?
Historical changes of the S&P Dividend Aristocrats Index
Should Dividend Investors own Non-Dividend Paying Stocks?
How to be a successful dividend investor
Warren Buffett’s Dividend Stock Strategy

Popular Posts