| 29 May |
Is There Really a Magic Formula for Investing? |
Geoff Gannon asked:
One question almost every investor asks at some point is whether it is possible to achieve above market returns by selecting a diversified group of stocks according to some formula, rather than having to evaluate each stock from every angle.
There are obvious advantages to such a formulaic approach. For the individual, the amount of time and effort spent caring for his investments would be reduced, leaving more time for him to spend on more enjoyable and fulfilling tasks. For the institution, large sums of money could be deployed without having to rely upon the investing acumen of a single talented stock picker. Many of the proposed systems also offer the advantage of matching the inflow of investable funds with investment opportunities. An investor who follows no formula, and evaluates each stock from every angle, may often find himself holding cash. Historically, this has been a problem for some excellent stock pickers. So, there are real advantages to favoring a formulaic approach to investing if such an approach would yield returns similar to the returns a complete stock by stock analysis would yield.
Many investment writers have proposed at least one such formulaic approach during their lifetime. The most promising formulaic approaches have been articulated by three men: Benjamin Graham, David Dreman, and Joel Greenblatt. As each of these approaches appeals to logic and common sense, they are not unique to these three men. But, these are the three names with which these approaches are usually most closely associated; so, there is little need to draw upon sources beyond theirs.
Benjamin Graham wrote three books of consequence: “Security Analysis”, “The Intelligent Investor”, and “The Interpretation of Financial Statements”. Within each book, he hints at various workable approaches both in stocks and bonds; however, he is most explicit in his best known work, “The Intelligent Investor”. There, Graham discusses the purchase of shares for less than two – thirds of their net current asset value. The belief that this method would yield above market returns is supported on both empirical and logical grounds.
In fact, it currently enjoys far too much support to be practicable. Public companies rarely trade below their net current asset values. This is unlikely to change in the future. Buyout firms, unconventional money managers, and vulture investors now check such excessive bouts of public pessimism by taking large or controlling stakes in troubled companies. As a result, the investing public is less likely to indulge its pessimism as feverishly as it once did; for, many cheap stocks now have the silver lining of being takeover targets. As Graham’s net current asset value method is neither workable at present, nor is likely to prove workable in the future, we must set it aside.
David Dreman is known as a contrarian investor. In his case, it is an appropriate label, because of his keen interest in behavioral finance. However, in most cases the line separating the value investor from the contrarian investor is fuzzy at best. Dreman’s contrarian investing strategies are derived from three measures: price to earnings, price to cash flow, and price to book value. Of these measures, the price to earnings ratio is by far the most conspicuous. It is quoted nearly everywhere the share price is quoted. When inverted, the price to earnings ratio becomes the earnings yield. To put this another way, a stock’s earnings yield is “e” over “p”. Dreman describes the strategy of buying stocks trading at low prices relative to their earnings as the low P/E approach; but, he could have just as easily called it the high earnings yield approach.
Whatever you call it, this approach has proved effective in the past. A diversified group of low P/E stocks has usually outperformed both a diversified group of high P/E stocks and the market as a whole.
This fact suggests that investors have a very hard time quantifying the future prospects of most public companies. While they may be able to make correct qualitative comparisons between businesses, they have trouble assigning a price to these qualitative differences. This does not come as a surprise to anyone with much knowledge of human judgment (and misjudgment). I am sure there is some technical term for this deficiency, but I know it only as “checklist syndrome”. Within any mental model, one must both describe the variables and assign weights to these variables. Humans tend to have little difficulty describing the variables – that is, creating the checklist. However, they rarely have any clue as to the weight that ought to be given to each variable.
This is why you will sometimes hear analysts say something like: the factor that tipped the balance in favor of online sales this holiday season was high gas prices (yes, this is an actual paraphrase; but, I won’t attribute it, because publicly attaching such an inane argument to anyone’s name is just cruel). It is true that avoiding paying high prices at the pump is a possible motivating factor in a shopper’s decision to make online Christmas purchases. However, it is an immaterial factor. It is a mere pebble on the scales. This is the same kind of thinking that places far too much value on a stock’s future earnings growth and far too little value on a stock’s current earnings.
The other two contrarian methods: the low price to cash flow approach and the low price to book value approach work for the same reasons. They exploit the natural human tendency to see a false equality in the factors, and to run down a checklist. For instance, a stock that has a triple digit price to cash flow ratio, but is in all other respects an extraordinary business, will be judged favorably by a checklist approach. However, if great weight is assigned to present cash flows relative to the stock price, the stock will be judged unfavorably. This also illustrates the second strength of the three contrarian methods.
They heavily weight the known factors. Of course, they do not heavily weight all known factors. They only consider three easily quantifiable known factors. An excellent brand, a growing industry, a superb management team, etc. may also be known factors. However, they are not precisely quantifiable. I would argue that while these factors may not be quantifiable they are calculable; that is to say, while no exact value may be assigned to them, they are useful data that ought to be considered when evaluating an investment.
There is the possibility of a middle ground here. These three contrarian methods may be used as a screen. Then, the investor may apply his own active judgment to winnow the qualifying stocks down to a final portfolio. Personally, I do not believe this is an acceptable compromise. These three methods do not adequately model the diversity of great investments. Therefore, they must either exclude some of the best stocks or include too many of the worst stocks. It is wise to place great weight upon each of these measures; however, it is foolish disqualify any stock because of a single criterion (which is exactly what such a screen does).
Finally, there is Joel Greenblatt’s “magic formula”. This is the most interesting formulaic approach to investing, both because it does not subject stocks to any true/false tests and because it is a composite of the two most important readily quantifiable measures a stock has: earnings yield and return on capital. As you will recall, earnings yield is simply the inverse of the P/E ratio; so, a stock with a high earnings yield is simply a low P/E stock. Return on capital may be thought of as the number of pennies earned for each dollar invested in the business. The exact formula that Greenblatt uses is described in “The Little Book That Beats the Market”. However, the fo
rmula used is rather unimportant. Over large groups of stocks (which is what Greenblatt suggests the magic formula be used on) any differences between the various return on capital formulae will not have much affect on the performance of the portfolios constructed.
Greenblatt claims his magic formula may be used in two different ways: as an automated portfolio generation tool or as a screen. For an investor like you (that is, one with sufficient curiosity and commitment to frequent a site such as this) the latter use is the more appropriate one. The magic formula will serve you well as a screen. I would argue, however, that you needn’t limit yourself to stocks screened by the magic formula, if you have full confidence in your judgment regarding some other stock.
These four formulaic approaches (the three from Dreman and the one from Greenblatt) will likely yield returns greater than or equal to the returns you would obtain from an index fund. Therefore, you would do better to invest in your own basket of qualifying stocks than in the prefabricated market basket. If you want to be a passive investor, or believe yourself incapable of being an active investor, these formulaic approaches are your best bet.
In fact, if I were approached by an institution making long – term investments and using only a very small percentage of the fund for operating expenses, I would recommend an automated process derived from these four approaches. I would also recommend that 100% of the fund’s investable assets be put into equities, but that is a discussion for another day (in fact, it’s a discussion for Tuesday; my next podcast is devoted to the dangers of diversification). If, however, you believe you have what it takes to be an active investor, and that is truly what you wish to be, then, I would suggest you do not use these approaches for anything more than helping you generate some useful ideas.
If you choose this path, you need to be clear about what being an active investor entails. Read this next part very carefully (it is correct even though it may not appear to be): I have never found a screen that generates more than one buy order per hundred stocks returned. Even after I have narrowed the list of possible stocks down by a cursory review of the industry and the business itself, I have never found a method that can consistently generate more than one buy order per twenty – five annual reports read.
Here, I am citing my best past experiences. In my experience, most screens result in less than one buy order per three hundred stocks returned, and I usually read more like fifty to a hundred annual reports per buy order at a minimum. You may choose to invest in far more stocks than I do. Perhaps instead of limiting yourself to your five to twelve best ideas as I do, you might want to put money into your best twenty – five to thirty ideas. Do the math, and you’ll see that is still quite a bit of homework.
That’s why remaining a passive investor is the best bet for most people. The time and effort demanded of the active investor is simply too taxing. They have more important, more enjoyable things to do. If that’s true for you, the four formulaic approaches outlined above should guide you to above market returns.
One question almost every investor asks at some point is whether it is possible to achieve above market returns by selecting a diversified group of stocks according to some formula, rather than having to evaluate each stock from every angle.
There are obvious advantages to such a formulaic approach. For the individual, the amount of time and effort spent caring for his investments would be reduced, leaving more time for him to spend on more enjoyable and fulfilling tasks. For the institution, large sums of money could be deployed without having to rely upon the investing acumen of a single talented stock picker. Many of the proposed systems also offer the advantage of matching the inflow of investable funds with investment opportunities. An investor who follows no formula, and evaluates each stock from every angle, may often find himself holding cash. Historically, this has been a problem for some excellent stock pickers. So, there are real advantages to favoring a formulaic approach to investing if such an approach would yield returns similar to the returns a complete stock by stock analysis would yield.
Many investment writers have proposed at least one such formulaic approach during their lifetime. The most promising formulaic approaches have been articulated by three men: Benjamin Graham, David Dreman, and Joel Greenblatt. As each of these approaches appeals to logic and common sense, they are not unique to these three men. But, these are the three names with which these approaches are usually most closely associated; so, there is little need to draw upon sources beyond theirs.
Benjamin Graham wrote three books of consequence: “Security Analysis”, “The Intelligent Investor”, and “The Interpretation of Financial Statements”. Within each book, he hints at various workable approaches both in stocks and bonds; however, he is most explicit in his best known work, “The Intelligent Investor”. There, Graham discusses the purchase of shares for less than two – thirds of their net current asset value. The belief that this method would yield above market returns is supported on both empirical and logical grounds.
In fact, it currently enjoys far too much support to be practicable. Public companies rarely trade below their net current asset values. This is unlikely to change in the future. Buyout firms, unconventional money managers, and vulture investors now check such excessive bouts of public pessimism by taking large or controlling stakes in troubled companies. As a result, the investing public is less likely to indulge its pessimism as feverishly as it once did; for, many cheap stocks now have the silver lining of being takeover targets. As Graham’s net current asset value method is neither workable at present, nor is likely to prove workable in the future, we must set it aside.
David Dreman is known as a contrarian investor. In his case, it is an appropriate label, because of his keen interest in behavioral finance. However, in most cases the line separating the value investor from the contrarian investor is fuzzy at best. Dreman’s contrarian investing strategies are derived from three measures: price to earnings, price to cash flow, and price to book value. Of these measures, the price to earnings ratio is by far the most conspicuous. It is quoted nearly everywhere the share price is quoted. When inverted, the price to earnings ratio becomes the earnings yield. To put this another way, a stock’s earnings yield is “e” over “p”. Dreman describes the strategy of buying stocks trading at low prices relative to their earnings as the low P/E approach; but, he could have just as easily called it the high earnings yield approach.
Whatever you call it, this approach has proved effective in the past. A diversified group of low P/E stocks has usually outperformed both a diversified group of high P/E stocks and the market as a whole.
This fact suggests that investors have a very hard time quantifying the future prospects of most public companies. While they may be able to make correct qualitative comparisons between businesses, they have trouble assigning a price to these qualitative differences. This does not come as a surprise to anyone with much knowledge of human judgment (and misjudgment). I am sure there is some technical term for this deficiency, but I know it only as “checklist syndrome”. Within any mental model, one must both describe the variables and assign weights to these variables. Humans tend to have little difficulty describing the variables – that is, creating the checklist. However, they rarely have any clue as to the weight that ought to be given to each variable.
This is why you will sometimes hear analysts say something like: the factor that tipped the balance in favor of online sales this holiday season was high gas prices (yes, this is an actual paraphrase; but, I won’t attribute it, because publicly attaching such an inane argument to anyone’s name is just cruel). It is true that avoiding paying high prices at the pump is a possible motivating factor in a shopper’s decision to make online Christmas purchases. However, it is an immaterial factor. It is a mere pebble on the scales. This is the same kind of thinking that places far too much value on a stock’s future earnings growth and far too little value on a stock’s current earnings.
The other two contrarian methods: the low price to cash flow approach and the low price to book value approach work for the same reasons. They exploit the natural human tendency to see a false equality in the factors, and to run down a checklist. For instance, a stock that has a triple digit price to cash flow ratio, but is in all other respects an extraordinary business, will be judged favorably by a checklist approach. However, if great weight is assigned to present cash flows relative to the stock price, the stock will be judged unfavorably. This also illustrates the second strength of the three contrarian methods.
They heavily weight the known factors. Of course, they do not heavily weight all known factors. They only consider three easily quantifiable known factors. An excellent brand, a growing industry, a superb management team, etc. may also be known factors. However, they are not precisely quantifiable. I would argue that while these factors may not be quantifiable they are calculable; that is to say, while no exact value may be assigned to them, they are useful data that ought to be considered when evaluating an investment.
There is the possibility of a middle ground here. These three contrarian methods may be used as a screen. Then, the investor may apply his own active judgment to winnow the qualifying stocks down to a final portfolio. Personally, I do not believe this is an acceptable compromise. These three methods do not adequately model the diversity of great investments. Therefore, they must either exclude some of the best stocks or include too many of the worst stocks. It is wise to place great weight upon each of these measures; however, it is foolish disqualify any stock because of a single criterion (which is exactly what such a screen does).
Finally, there is Joel Greenblatt’s “magic formula”. This is the most interesting formulaic approach to investing, both because it does not subject stocks to any true/false tests and because it is a composite of the two most important readily quantifiable measures a stock has: earnings yield and return on capital. As you will recall, earnings yield is simply the inverse of the P/E ratio; so, a stock with a high earnings yield is simply a low P/E stock. Return on capital may be thought of as the number of pennies earned for each dollar invested in the business. The exact formula that Greenblatt uses is described in “The Little Book That Beats the Market”. However, the fo
rmula used is rather unimportant. Over large groups of stocks (which is what Greenblatt suggests the magic formula be used on) any differences between the various return on capital formulae will not have much affect on the performance of the portfolios constructed.
Greenblatt claims his magic formula may be used in two different ways: as an automated portfolio generation tool or as a screen. For an investor like you (that is, one with sufficient curiosity and commitment to frequent a site such as this) the latter use is the more appropriate one. The magic formula will serve you well as a screen. I would argue, however, that you needn’t limit yourself to stocks screened by the magic formula, if you have full confidence in your judgment regarding some other stock.
These four formulaic approaches (the three from Dreman and the one from Greenblatt) will likely yield returns greater than or equal to the returns you would obtain from an index fund. Therefore, you would do better to invest in your own basket of qualifying stocks than in the prefabricated market basket. If you want to be a passive investor, or believe yourself incapable of being an active investor, these formulaic approaches are your best bet.
In fact, if I were approached by an institution making long – term investments and using only a very small percentage of the fund for operating expenses, I would recommend an automated process derived from these four approaches. I would also recommend that 100% of the fund’s investable assets be put into equities, but that is a discussion for another day (in fact, it’s a discussion for Tuesday; my next podcast is devoted to the dangers of diversification). If, however, you believe you have what it takes to be an active investor, and that is truly what you wish to be, then, I would suggest you do not use these approaches for anything more than helping you generate some useful ideas.
If you choose this path, you need to be clear about what being an active investor entails. Read this next part very carefully (it is correct even though it may not appear to be): I have never found a screen that generates more than one buy order per hundred stocks returned. Even after I have narrowed the list of possible stocks down by a cursory review of the industry and the business itself, I have never found a method that can consistently generate more than one buy order per twenty – five annual reports read.
Here, I am citing my best past experiences. In my experience, most screens result in less than one buy order per three hundred stocks returned, and I usually read more like fifty to a hundred annual reports per buy order at a minimum. You may choose to invest in far more stocks than I do. Perhaps instead of limiting yourself to your five to twelve best ideas as I do, you might want to put money into your best twenty – five to thirty ideas. Do the math, and you’ll see that is still quite a bit of homework.
That’s why remaining a passive investor is the best bet for most people. The time and effort demanded of the active investor is simply too taxing. They have more important, more enjoyable things to do. If that’s true for you, the four formulaic approaches outlined above should guide you to above market returns.
