Michael Mauboussin – Market Myths And Market Reality
Michael Mauboussin is the author of The Success Equation: Untangling Skill and Luck in Business, Sports, and Investing (Harvard Business Review Press, 2012), Think Twice: Harnessing the Power of Counterintuition (Harvard Business Press, 2009) and More Than You Know: Finding Financial Wisdom in Unconventional Places-Updated and Expanded (New York: Columbia Business School Publishing, 2008). More Than You Know was named one of “The 100 Best Business Books of All Time” by 800-CEO-READ, one of the best business books by BusinessWeek (2006) and best economics book by Strategy+Business (2006). He is also co-author, with Alfred Rappaport, of Expectations Investing: Reading Stock Prices for Better Returns (Harvard Business School Press, 2001).
Visit his site at: michaelmauboussin.com/
- Exploration of market myths and market reality draw a picture of an economically sound stock market.
- Popularly used valuation techniques offer the advantage of simplicity at the expense of accuracy.
- A value-based model incorporates good theory and good practice and should be the primary tool in capital allocation for companies and investors alike.
This text is based on a presentation given at the annual National Investor Relations Institute (NIRI) conference in June 1997. The session was originally entitled “Valuation: A to Z.” We thank the NIRI for allowing us to reproduce this speech.
Good morning. My objective today is to walk through, very logically, why we think value-based analysis is a powerful tool for both investors and the corporations. We’ll approach the issue in three different ways. First, we’ll talk about stock market myths and stock market reality. Next, we will evaluate valuation techniques, weighing the pluses and minuses of each. Finally, we will lay out the case for a value-based model.
Michael Mauboussin - Market Myths And Market Reality
Let’s go right to the first section—stock market myths and reality. I would like to preface these comments by emphasizing that I work on the sell side. Everyday I deal with companies and investors. So I have to be pragmatic, and must try to understand what it is that really matters. My objective in this analysis is to be practical, as well as value-added, for the investors.
The first popular myth is that earnings per share matter. Why would anyone be of that opinion? First, earnings are widely cited in the financial press, it is what companies talk about—earnings are the common language of Wall Street. Second, earnings are the result of audited financial statements—there is authority behind the numbers. Finally, stock prices change, and in some cases significantly, based on earnings per share reports. This morning we woke up with the unfortunate news that Cabletron is going to miss the consensus EPS for the quarter by 15%: the stock tumbled by almost 30%. The fact that earnings will be below current expectations and the stock is down suggest a high correlation between the two events.
We suggest that earnings are really a proxy for what is going on, and that earnings per share figures, in and of themselves, are to be used with a lot of caution. In order to develop this point we used data from the food industry. (See Figure 5.) We took the fiscal 1994 numbers for a handful of major packaged food companies, started with the reported earnings per share number, and sorted through those items that differentiated the reported numbers from the cash numbers. We all know that because of varying, acceptable accounting standards, amortization of goodwill and other sundry issues, reported earnings per share and cash earnings can be two very different things. So we made the necessary adjustments to show the “cash” earnings for every dollar of reported earnings. In 1994, for every dollar of reported earnings Kellogg delivered $1.39 in cash earnings; ConAgra generated over $2.08 in cash earnings. So when we are talking about earnings, and applying multiples to those earnings, it doesn’t seem to make sense to use comparable multiples on two business that have vastly different economics, no matter what their reported numbers suggest. Earnings per share, then, are a proxy for cash, but only a proxy, for cash.
Now we go to the next level of reasoning: earnings themselves may not matter, it is really earnings growth that everyone cares about. Why would we think that? First, earnings growth tends to be held up as an absolute good. In almost every organization, managers are drilled to believe that growth is good. Very rarely do managers ever consider that growth can be bad, and I am going to demonstrate in a moment why EPS growth can be bad. Second, investors appear to be after growth. You hear it all the time: the faster the growth rate, the better. Finally, in most cases executive compensation is tied to some growth metric. Before I go on to explain why I don’t think that earnings per share growth, in and of itself, is important, let’s do a mental exercise:
Let’s pretend I am an unlimited source of capital. Everyone in this room can come to me and I will give you all the money you want. There is just one catch: the expected return on the capital I give you, not the explicit return, but the implicit return, is 10%. Now, let’s say you can go out and reinvest the money I give you at 8%. I give you capital with a 10% “cost” and you reinvest it at an 8% return. Now here is the question. What is your earnings per share growth going to be for this venture? The answer is, It is going to be anything you want it to be! All you have to do is keep coming back to me for more money, reinvest it, and you will generate earnings growth at whatever rate you choose. Clearly, the more rapidly you grow earnings the worse off you are, because you are investing below the required rate of return.
That may make sense, but how does it apply to the real world? The answer is that the primary source of capital for most companies is retained earnings. (Roughly 75% of investments are funded internally.) Retained earnings have an implied opportunity cost—that is what shareholders could earn on that capital if they invested it in other ventures of similar risk—but most managers think of retained earnings as “free” money. Hence, managers who take cash generated by the business and reinvest it at below appropriate rates of return in order to drive earnings per share are doing their shareholders injustice. Mind you, investing in projects that do not earn the costs of capital may drive earnings per share growth, but will clearly destroy value. This type of investing goes on every day in corporations around the world.
Here are some other illustrations in the last few months that demonstrate the earnings/value dichotomy. Tyco International—which has been a very good performing stock, by the way—announced in March it was acquiring a company called ADT. Tyco is in the building supply business, ADT is a leading home security company. It came in as a white knight. The acquisition is structured as stock-for-stock—formally called a pooling of interest. It is a mathematical truism when a high P/E company buys a low P/E company, EPS for the acquiring company go up. This point is axiomatic. This was also the case for Tyco and ADT. The day of the announcement, ADT stock was up 18%—no big surprise. Tyco suggested that the deal should result in EPS accretion of about 8%. The stock, however, declined 3% in a flat overall stock market. So earnings per share went up, the stock went down, and the company witnessed a 4 P/E point multiple contraction.
This underscores the important point that P/E multiples are not a determinant of value, but rather a function of value. On Wall Street, the typical valuation formula is EPS x P/E = value. We just talked about earnings and how they can be misleading. Now we go to this thing called the P/E multiple. We are asking the P/E multiple to reflect growth, capital intensity, risk, quality of management, and competitive advantage. We are heaping a lot of responsibility on one number and we argue that it is practically impossible to know what that right number is. So we try to break value down into components. We believe that value is determined by the present value of a stream of future cash flows, and the P/E falls out of that equation.
Here’s another good illustration. A few days after the TYCO/ADT news, Coca-Cola Enterprises announced that its earnings would be roughly 10% lower than the consensus because of some noncash accounting issues. No cash impact—earnings down 10%, but no cash impact. The stock price was unchanged in a flat overall market. What happened? The P/E multiple went up. P/E is not driving the value, the value is driving the P/E; and you really have to keep that relationship straight.
Here is the next market myth: EPS growth drives valuation. Companies say that if we could just get our EPS growth rate from 12% to 14% we would get a higher P/E. Maybe yes, maybe no. Figure 9 is a chart for the food industry that helps clarify the EPS growth/valuation debate. We have done similar correlations for 25 industries around the world, and the results have been remarkably consistent. On the X axis, we show projected earnings per share growth. We use consensus estimates, in this case Value Line. On the Y axis, we show enterprise value to invested capital—a fancy way of saying price-to-book. For those familiar with Stern Stewart’s work, this is relationship is “scaled” market-value-added (MVA). There is a whopping 1% correlation between EPS growth and valuation. If you employ return on equity as your independent variable, the R-squared rises to about 25%. If you use return on invested capital less cost of capital spread on the X axis, the R-squared goes to 75%. Let’s go through that again. The correlation for earnings per share growth, 1%; for return on equity, 25%; and for return on invested capital, 75%. Does anyone think that the market doesn’t get it? You can be assured that the market gets it. We did this analysis for the food industry going back 15 years. What we found was that return on invested capital versus cost of capital spread explains valuation well over the whole period. Companies moved up or down the regression line as their prospects got better or worse, but the relationship held true.
The next myth is that the market is very short-term orientated. This one is a favorite of the business press, and is a theory in which many CEOs indulge. What is the evidence for “short-termism?” First, stocks react to quarterly earnings per share. Next, money mangers are evaluated every 90 days so they have to run fast to make sure their portfolio is well positioned for the short term. Finally, there is a disproportionate focus on next quarter’s earnings per share. This is particularly endemic to the sell side.
How are stock prices are really set? We will go back to the food industry to gain a perspective on the answer. While the food industry may be somewhat unique in its stability and visibility, let me suggest that this is analysis that is relevant in most sectors.
What we did was quite simple: accepting the basic premise that the value of a business is the present value of future cash flows, we estimated what percent of the value of these stock prices is attributable to cash that will be recognized after five years. (See Figure 11.) The answer is 75% for Campbell Soup and 76% for CPC International. These examples are not unique: roughly 70% of the stock market value of the food industry is attributable to cash that is going to be received beyond five years.
How can we rationalize such a long-term perspective? To address that question, we can hold the “Coca-Cola auction.” This auction is a proof of the market’s extended view via some basic armchair logic.
We start the auction process by recognizing that Coca-Cola is one of the bestknown companies in the world. Its return on capital in 1996 was 37%, its global market share is 45%, it has an impressive presence in emerging markets. In short, Coca-Cola is a wonderful franchise, and is clearly a strong value-creating company, any way you want to define it.
Now let’s pretend for a moment that I was given the right to auction the company, and that everyone here had the resources to buy the whole business. Rather than auction off the company for a given amount, I will auction of the “rights” to future value creation, measured in years. Who would be willing to bet, with their own money, that KO will have returns above its cost of capital over the next five years? How about 10 years? How about 15 years? How about 20 years? Here is the point I am trying to make: portfolio managers want to make money so they buy things they think are going up and they sell things they think are going down. It is nothing personal. Now think about the forecast horizon, or the “short-termism” argument in this context. If the market pushes the horizon in short enough, you are certainly going to have self-interested, motivated, intelligent people say, “I’m willing to bet that KO could earn above its cost of capital for the next (say) seven years and I’m going to buy the stock.” The process of setting of stock prices, which is really an auction, assures that the market will look out well into the future when appropriate.
The next canard is that the market is “unsophisticated” or in some instances “irrational.” Allen Greenspan and Warren Buffett have recently contributed to this myth. Sumner Redstone, the CEO of Viacom, recently said that his stock price is depressed because the stock market is irrational; people don’t get it. All these portfolio managers, who are out to make money in a competitive setting, are collectively irrational. That is a difficult one to buy.
What does happen is a company makes an acquisition, tells the world it’s a great strategic deal, and is shocked when the stock is down 2% or 3%. Now in some cases the market doesn’t get it because the market doesn’t have the appropriate information. In that case, it is incumbent on the company to provide the relevant information so as to allow investors to make the right decision. Collectively, the market gets it a lot more times than it doesn’t get it.
Here are two examples of the power of group. First, if you put lots of people together and you ask them to guess a “commodity” number, and their errors are independent, the result will be something very close to the actual price. This illustration was demonstrated very clearly by Jack Treynor. Let me give you a very specific illustration. If I handed everyone in this room a form, asked you to estimate IBM’s assets at year-end 1989 (I would assume most of you wouldn’t know the exact number), collected them, and took a mean of the responses, that mean would be within 5-10% of the true number. What happens, so long as the guesses are independent, is that the low ballers and high ballers cancel out, and the mean comes out close to the true number. I do this exercise every year in my class at Columbia Business School. Without fail, we come within 5%, and that is without any information being conveyed.
The second example we use, called the “Academy Awards,” is a little more fun. We get a group together about this size and we give them a ballot before the Academy Awards are announced. The ballot has two sides. On the first side are top six categories for the Academy Awards—Best Picture, Best Actor, that type of thing. The second side has more remote categories—things like Best Cinematography, Best Screen Play, and so forth. We ask each participant to chip in one dollar to play and ask them to pick the winners for each category. What we see consistently is that the consensus guess for every category does better than any single human being. In the spring of 1997 we had 125 people play this game: the consensus got 11 out of 12 right, the best human got 9 out of 12. So, the market mechanism is much more powerful than individuals, and that is something that should be respected by everyone that is in this business.
Let’s leave this section by suggesting that there are three things that we really want to dwell on. The first is cash flow, the second is some measure of risk, and the third is some sort of time horizon. These are the basics required to value any financial asset, including bonds, options and commercial real estate. For stocks, of course, these variables are based on expectations.
See full PDF below.