Wednesday, December 9, 2009

When to break your rules

As a dividend growth investor, my strategy is picking the right stocks that provide a decent balance between dividend yield and distribution growth. Thus I have maintained a rigid requirement for a 3% initial yield before investing in a dividend growth company’s securities.
Most dividend investors look for yield when purchasing income securities. Most dividend growth investors purchase securities so that they could enjoy a rising stream of dividend payments over time. Thus, maintaining a proper balance could be a challenge that could make or break your portfolio.

I realize that using a strict yield criteria I could miss out on potential dividend growth stories such as Wal-Mart (WMT) for example. Wal-Mart has never yielded 3% since it went public in the 1970s. The 29.1% annual dividend growth since 1975 has been truly spectacular however. This means that Wal-Mart’s dividend has doubled every 2.5 years for 34 consecutive years. Wal-Mart has delivered a 23.40% dividend growth since 1985 and a 20.20% dividend growth since 1995. Check my analysis of Wal-Mart.

My rationale behind selecting a minimum yield is to provide me with an adequate margin of safety should the stock stop raising dividends and should the stock price fall or remain flat for a large period of time. In the case of Wal-Mart, the stock has been trading in a range over the past decade. Back in 1999 the stock fluctuated between $70.25 and $38.68 and closed at $69.12. The stock wasn’t yielding much back then – about 0.30%. Even if the dividend were doubling every 2.5 years, it would take a retiree almost 13 years in order to reach a yield on cost of 10%. At the current dividend rate, the stock is actually yielding 1.60% on cost, assuming that you purchased it on the last day of 1999. The actual dividend growth over the past decade comes down to 20.80% per annum, which translates into the dividend payment actually doubling every three and a half years.

Now the down side to my having a strict initial yield requirement for entry is that I would miss out on some huge gains, which could lead to early financial freedom. If one had purchased Wal-Mart stock at the end of 1984, the tenth year in a row in which it increased its dividends, their entry price would have been $1.18 (adjusted for five stock splits) and their initial yield would have been only 0.55% at the time. Fast-forward 25 years and the yield on cost comes out to almost 100%.

Many dividend growth investors tend to project past dividend growth rates into infinity, which seems unsustainable to me. If a company with $1 billion in profits enjoyed a 15% annual growth forever, it would double its net income almost every five years. In reality, as the companies grow larger they would find less opportunities that could sustainably earn them higher incremental returns on investment. For example, with a company like McDonald’s (MCD) people could only eat so much burgers and fries. After a company hits a plateau, EPS growth could largely be sustained by increasing efficiencies, raising prices, repurchasing shares or buying other competitors.

This analysis is not meant to be used as a weapon against Wal-Mart or McDonald’s, which are fine companies. It just goes to show that once shouldn’t solely rely on past data in their investment decisions. Furthermore, projecting past data into the future, without adding a what if analysis of your common sense could prove costly in the long run. In addition, purchasing stocks solely for the dividend growth is as dangerous as chasing high yielding stocks blindly.

As far as my strategy is concerned, I am considering lowering the entry yield criteria to 2% for stocks, which appear to have a sustainable above average dividend growth ahead of them. My target allocation for such stocks would be half of what I would normally allocate to such dividend growth champions such as Johnson & Johnson (JNJ) or Procter & Gamble (PG) however.

Full Disclosure: Long JNJ, MCD, PG and WMT

Relevant Articles:


  1. I was wondering how you might feel about bending your ten year rule of consecutive dividend increases by using the same pairing method that you are using with Wal-Mart and AT&T. In that arrangement the pairing is based on balancing out WMT's low, fast growing yield with T's higher, slow growing yield.

    It seems the same type of pairing could be based on years of consecutive dividend increases instead of yield. Stocks that are close to or have the potential to become dividend achievers such as COP, KFT, RTN, YUM, GIS, INTC, MSFT could be paired with the longest paying dividend aristocrats PG, JNJ, KO, EMR, MMM. Or is the ten year rule one you would not consider breaking? This probably would not be a great strategy for those closer to retirement. However, for those with a longer time horizon would it be worth trying to find these stocks a little early?

    Also I'd be curious to dig a little deeper into your thoughts on Wal-Mart circa 1984. If such a stock existed today would you consider it speculative to invest in such a low priced, low yielding stock. Or was Wal-Mart exhibiting some characteristic(s) that would have been identifiable to dividend investors in 1984? If so would it be worth putting a small amount of money in a few stocks that exhibit similar characteristics to Wal-Mart in 1984, or is there too much risk trying to project such outsized dividend growth like Wal-Mart and McDonald’s so early in the process.

    Kudos on the Original Dividend Aristocrats and the Dividend Grouping posts. Very insightful. I have also really enjoyed your posts that focus on value investing (what I struggle with most in the market these days) so I hope you'll keep writing more of these. Thanks.

  2. You're smart to broaden your search for dividend growth.
    What I'd give to own 100 shares of MCD from the first time my father took us to McDonald's for 15 cent burgers!

    To the previous poster's point, I'd encourage you to look for smaller cap growers that are already paying dividends. It's the only way you'll identify the next Wal-Mart or McDonald's, and I think you'll find the joy of discovery as satisfying as continually plumbing the depths of the dividend aristocrats.

    (And if you'll accept some advice from an old-timer, don't tell yourself that you're relaxing your criteria or you'll revert back to your current criteria with your first mis-step. Invent a new strategic approach like dividend asset allocation or something similarly positive. Massage it and take pride that you're pioneering a new concept.)

    Keep up the fine work.

  3. Consider analyzing a potential dividend growth investment in terms of its potential "Yield on Cost" (YOC) at various future dates.

    Case 1: Stock A is currently yielding 4%, but has grown its dividend by 7% a year for the past ten or more years. If you believe this dividend growth rate will continue for the next ten years, the dividend will double in ten years. Ten years from now, in other words, your YOC will be 8%.

    Example: You buy $10K of stock A and it yields 4%, or $400, in year 0. Ten years from now, the dividend has doubled to $800, which is an 8% YOC.

    Case 2: Stock B is currently yielding 2%, but has grown its dividend by 14% a year for the past ten or more years. If you believe this dividend growth rate will continue for the next ten years, the dividend will double in five years and then double again at the ten-year point. Five years from now, your YOC will be 4% and ten years from now, your YOC will be 8%.

    Example: You buy $10K of stock B and it yields 2%, or $200, in year 0. Five years from now, the dividend will have doubled to $400, which is a 4% YOC. Ten years from now, the dividend will have doubled again to $800, which is an 8% YOC.

    Ten years from now, both stocks will be yielding the same annual dividend payment ($800), which is the same yield (8%) on the original cost of $10K.

    Upside: If you believe these dividend growth rates will continue beyond ten years, stock B will continue to yield a higher YOC because its dividend growth rate is higher than stock A's dividend growth rate.

    Downside: The higher the dividend growth rate, the greater the chance it might slow down in future years because higher dividend growth rates are harder to sustain. But if a business is exceptional (i.e., great MOAT, well managed, great opportunities to expand), it might be able to sustain a high dividend growth rate for a long time.

  4., which advocates a strategy similar to yours, has a post up which compartmentalizes companies neatly into three groupings of yield and growth. It's worth a read.

  5. Dj,

    I think that requiring at least 10 years worth of dividend increases would be a good screen to weed out companies which have simply been "lucky" in raising distributions. Many companies started paying out dividends in 2003 and raising them because of the favorable dividend tax. If this tax expires, these companies might decide to stop raising dividends.
    Also some companies which have flat earnings but low dividend payouts could raise dividends for many years until dividend payout ratio is unsustainable.
    You also want to avoid getting whipsawed into cyclical stories which are about to burst.

  6. Dividends value is good site on investing. DV and I are one of the founders of the dividend and value investing network.

    The reason why I wrote that I might consider lowering yield criteria is exactly because WMT in the 1980's yielded less than 1%. The dividend growth was high however. In hindsight buying WMT in 1980s would have been a great investment. Of course you also have to consider all the companies that had similar characteristics to Wal Mart in 1980's which are either bankrupt today or have stopped paying distributions. Kmart and Winn-Dixie are some examples of former dividend growth stars which have crashed and burned after losing their dividend streak status...

  7. You're right, of course. It's easy, in hindsight, to highlight stocks like WMT and MCD, but they are truly the exceptions that prove the rule. Far more companies would have been failed investments.

    Finding "future exceptions" like a WMT will probably require additional filters--a blending of growth and value criteria, along with a strong, judgmental conviction about the story underlying the stock.

    It's an intresting quest. I'm sure, somewhere, others have gone back and tried to model WMT and MCD and applied their findings to stock selection. Guru investing might be a place to start.

    Good luck and keeping breaking your rules!

  8. Thanks for informing on your smart tips on when to break one's rules.


Questions or comments? You can reach out to me at my website address name at gmail dot com.

Popular Posts