Uncertainty in Stock Statistics

Yesterday Mike reposted an old blog post from my previous blog.  Commenter Adam correctly pointed out that it was quite muddled, to the point of almost being unreadable.  As a gesture of atonement, and to hopefully prove I’ve learned a thing over two over the last three years, I want to take another shot at what I was trying to get at in that post.  And I won’t use any parenthesis!  If you want to play along at home, read yesterday’s post before you read this one.

There are a massive number of statistics which are tracked for most publicly traded companies.  From price, to revenue, to dividends, to volume.  There are a number of ratios and other derived values which investors have also come to base investing decisions on:  from price-to-earnings, to dividend yield, to moving averages.

Different investing styles place greater, or lesser, weight on each of these values and ratios.  Moving average, for example, is a classic technical trader technique, while price-to-earning is a core component of fundamental analysis.  Each investor who follows one of these metrics has a great story why the indicator they believe in gives greater insight into the true value of a market and leads them to beat it.  Joel Greenblatt’s “The Little Book That Beats The Market” presents an investing strategy built on earnings yield and return on capital, Derek Foster’s first two books were built on the idea of investing based on dividend history, while Phil Town’s “Rule #1” details a slightly schizophrenic blend of technical *AND* fundamental techniques.  The famous Graham number, still in popular use by the likes of Tom Connolly and Warren Buffett, is calculated using earnings per share and book value per share.

All these strategies sound pretty impressive when you read about them for 150 pages or so.  Results are typically given which make the reader confident he’s figured out a back door to making big bucks in the equity market!  Sadly, things often don’t play out that way.  Any strategy can outperform in certain market conditions, but they also can take a beating and under-perform in other market conditions.

Beyond any specific problems with strategies, I’m suspicious about the data they’re based on.  Some metrics and the ratios based on them, such as dividend yield, are known.  Price is a record of the value a stock was assigned in a specific trade and is a known value.  Similarly, dividends are cash that actually show up in investor’s bank account.  These are known, real values.  In contrast, a number of metrics are self-reported by companies, who often have a vested interested in presenting a certain perspective on the company’s financial health to the public.  Beyond the self-reported values, other values are from stock analysts who make predictions on what they THINK the company will report in the future.  Clearly this is a spectrum of real, known values through to silly, fantasy numbers.  Some formulas combine silly numbers, magnifying the inherent margin of error.  The likelihood of  the numbers used in a strategy being correct or not clearly needs to be incorporated in the decision about whether or not to base investing decisions on them!

Between dividend yield and dividend growth, I think a case can be made that yield is the far more reliable of the two metrics.  Yield is based on the annual dividend paid divided by the stock price, while growth is the % increase over some period of time.  The yield isn’t certain moving forward, as there is no guarantee that the company will maintain the dividend.  The growth is definitely going to change.  Companies try to maintain dividend payments whenever possible, but they don’t try to maintain a precise dividend growth rate.  If the dividend growth of a company was a reliable predictor of future dividend payments, I’d agree that investing in stocks with high dividend growth is preferable to investing in stocks with high dividend yield, but I don’t believe that is the case.  Blindly investing in high dividend yield companies would be equally foolish.

As Homer taught Bart in the famous Simpson’s episode “Homer at the Bat“:  “Can’t win, don’t try”.  While this isn’t a great lesson for life, it is a good lesson with the stock market.  Instead of betting on an uncertain strategy based on uncertain statistics, investors can give up on beating the market and instead happily match it.  Passive investing lets other people do all the work of frantically appraising every gyration the markets undergo and simply reap the benefit of long-term gradual increases.

Whew – not a single parenthesis! (it just about killed me)

5 replies on “Uncertainty in Stock Statistics”

Shows how much your writing has improved over the past 3 years. Nice clear and concise (not like the one from yesterday ;))

Nicely done! Great information. My take on the markets is it’s a game of probabilities. You have to be right more often than wrong (and you will be wrong sometimes). You also have to make money more often than lose money (and you will lose money sometimes). The trick it to develop a process or strategy or discipline that keeps you focused that tries to enhance your probability of success. I think Warren Buffet said it best “Successful investing requires a sound framework that prevents emotion from corroding that framework.” Passive investing is a strategy that statistically should work about 70% of the time. Picking stocks with research means you have to win more than 70% of the time.

It’s emotions and willpower that will determine investment success. Some systems are better equipped to handle this; the do-it-yourself investor can bet on a certain strategy, but they need to bet on one that, as Buffett said (and pointed out), “prevents emotion from corroding that framework.”

Your post is timely, today I posted an article that talks about how I use p/e ratio within my criteria (system).

Excellent food for thought Mr. Cheap.

Leave a Reply

Your email address will not be published. Required fields are marked *