As I was meandering through the local book store called Half Price Books, an independent bookseller, I came upon my discovery. There it was, sitting about eye level, a copy of the 1973 version of the "Intelligent Investor" by Benjamin Graham with the original book jacket cover with the appearance that it had never been opened. They had placed a value of $7.99 on it and I scooped it up. The jacket had a price of $30, but the value… well, most of you know the rest.
With that great investment in hand, I went home and began to re-read the book from the great investor and want to share some of the gems or ideas picked up through my reading. To many of you, these will be reminders from a great mind.
Graham begins Chapter 12 about Per-Share Earnings, by cautioning investors to not settle for using a single year’s earnings in the research of a potential investment, but to look at the bigger or longer picture of 7 to 10 years. He goes on to state that prudence is required when looking at earnings because of potential manipulations such as special onetime charges. His example of Alcoa is still appropriate.
In more recent times, we can look at Alcoa (AA) and see similar patterns. For instance, in 2007, the average price for Alcoa was approximately $38. Reported earnings per share were $2.95; however, after stock options, pensions, goodwill impairments or other various earnings expenses, the diluted eps was closer to $1.74. This is a huge discrepancy. Further, consider how this affects the P/E. Using the average price of $38.00, note the P/E comparisons:
What might have been reported as a value stock with a P/E of 12.88 actually had a more accurate P/E of 21.84 and was probably overvalued. Consider the effect of this type of disparity when you are calculating long-term earnings growth. Also, consider the noise coming from Wall Street. I did not read how the reporting of these earnings hit the media, but you can imagine reporting a great number of $2.95 that was up from the previous year’s number of $2.47. In actuality, the diluted earnings of 2006 was only different by .01, so the earnings should have more accurately been described as going from $2.46 in 2006 to an actual earnings of $1.74. Not quite as exciting.
Let’s take an example a little further. But first, note that Graham had his own idea about using an average of earnings. In a 10-year period, such as shown below, he would average the first three years (2002, 2003 and 2004 and average the later three years (2009, 2010 and 2011) to determine his long-term growth note in our example how this affects the growth percentage. Our stock has the following reported earnings per share and, for the sake of argument, we will say that these are diluted or are more accurate.
Numbers like these can lead you to the following conclusions:
1. The 10-year growth is 10.04%
2. The Graham 10-year growth is 3.10%
3. The 5-year growth is 11.22%
4. The 3-year growth is 15.17%
5. The 1-year growth is 41.61%
So, which do you use? Discounted earnings or discounted cash flow have their problems, but you can see how your assumptions can accentuate the problem. It’s imperative that investors consider the quality of the earnings and determine how pensions, research and development, interest expense and other items affect the quality of the earnings you are investigating. The demonstration shown of the various earnings growth and the effect on the P/E are enormous and must be measured.
Robert Arnott, Financial Analyst Journal editor believes that earnings are manipulated such that nearly 25% of earnings “are fictitious.” That number may be high and it may be low, depending on the company that you are investigating, but it clearly demonstrates Graham’s point of exploring the quality of the earnings. While I’m not a huge fan of S&P analysis, they do provide a great service in displaying what they call “core earnings.” S&P reports show both reported earnings and their core earnings. Core earnings, or least the objective of them, are to exclude the various expenses mentioned and provide the investor with a more accurate or diluted earnings number.
While I fully believe it is incumbent that investors determine the accuracy of the earnings themselves, if you are not prepared to do the research yourself, one might at least use caution and check out these adjusted earnings numbers. They are free at many locations. Warren Buffett commended S&P for providing these numbers in an open letter. Graham’s says, “The more seriously investors take the per-share earnings figures as published, the more necessary it is for them to be on their guard against accounting factors of one kind and another that may impair the true comparability of the numbers.”
Good advice from a great man. That $7.99 is already paying off.