The Security Analysis Problem
Thomas E. Berghage
The goal of investors has not changed over the years even though markets have. Investors would like to find some place where they can put their money to work and have some expectation that when they retrieve their funds in the future they will have greater purchasing power than when they initially invested them. To accomplish this, a number of things must happen, none of which will occur with perfect certainty.
First of all, investors must have a stable way to measure the value of their investments; when they make it and when they retrieve it. The traditional way of measuring this value is with a metric called, “money,” and unfortunately, it is anything but stable. For traders involved in the quick turnover of assets the fluctuation in the value of money is not much of a problem, except during extremely volatile times such as occurred in Germany at the end of World War II. For long-term investors, however, the stability of a currency is a significant problem and needs to be taken into consideration. Periods of prolonged inflation or deflation can alter the value of assets, impact appreciation potential, and change the attractiveness of investing. The central banks around the world have as their primary objective the maintaining of stability of their local currency. Without stable money a country’s economic activity in the world markets suffers. The cost of doing business in countries around the world is directly related to the volatility of the local currency.
Adjusting the value of investments to take into consideration the future fluctuations in currencies, however, is not the topic of this book. It is a very interesting subject and worthy of considerable study and maybe even a future book, but is not the topic to be addressed here. I want to focus here on the more important topic of how to evaluate and select investment opportunities (stocks).
You will note that the authors have suggested that there are two aspects to this problem, evaluation and selection. As I pointed out in my earlier book, Beyond Human Comprehension, being able to detect or find a good investment is one thing, and being able to act on it is an entirely different problem. Detection and action are two completely different aspects of the investment process that I will deal with in Chapter 7.
The first and most important thing the reader should understand is the underlying premise of this book: there is a fundamental difference between Financial Analysis and Security Analysis. Financial Analysis has as its objective the understanding of corporate structure and operations. Security Analysis is focused on identifying investments that will hopefully increase the investor’s future purchasing power.
The identification and evaluation of stock investment opportunities have changed little since Benjamin Graham and David Dodd (now considered the fathers of financial analysis) published their now famous book, Security Analysis: Principles and Technique back in 1934. The goal then as it is today is to identify operational and financial measures that will give the investor clues about the future prospects of a potential investment. You have to understand, however, that this goal is not the driving force in the education, training or even the evaluation of financial analysts. Currently, financial analysis, as it is being taught in most universities, is focused on trying to understand how companies operate and function within the economic environment. Little or no attempt is made to forecast how a company’s stock will perform in the market place. It is just assumed that if a company is well structured and operating efficiently that its stock will do well in the trading markets. This assumption is not necessary correct. Performance in the market place is governed by a number of factors unrelated to the performance of the company. The measures of corporate performance that financial analysts evaluate, may or may not be related to how a stock (a piece of paper traded in the market place) performs. This disconnect between what is taught and practiced by financial analysts, and what is sought by investors is a significant problem. Financial analysis, as it is currently being taught, is probably better at forecasting future dividend yields than it is at forecasting future stock appreciation.
What we need to do is make a distinction between Financial Analysis and Security Analysis. As you will see as you read further in this book there is a real difference between the two. Benjamin Graham and David Dodd mis- titled their book, Security Analysis; it is probably more aptly titled, Financial Analysis because most of the operational and financial measures described in the book relate to corporate performance rather than security performance, and as anyone who has studied the market knows, there is a big difference between how a company performs and how its stock performs in the market.
Figure 3-1
The Holy Grail Of Security Analysis
What every investor would like to find is a measure, or group of measures, that when plotted against future total returns, forms a straight monotonic line as shown in Figure 3-1. That given the value “X” (your forecasting measure) you have perfect knowledge of the value “Y” (Stock appreciation over some period of time) This is the holy grail of investment analysis as opposed to financial analysis. Unfortunately, in security analysis this kind of precise relationship does not exist. This is God’s knowledge and is not available to we poor mortals. The best we can possibly hope for is something similar to the less than perfect relationship shown in Figure 3-2, where there is some significant and understandable relationship between the selected independent measures (the operational and financial data) and the dependent measure (the future total return on investment). In statistics the preciseness of this relationship is described by using what is known as the correlation coefficient or the square of the coefficient, the coefficient of determination. The coefficient of determination tells you what proportion of the variation in “Y” is accounted for by knowledge of “X.” For a more detailed description of these measures the reader is referred to any basic statistical text or our previous book (Berghage and Berghage, 2002).
Figure 3-2
The Best We Can Hope For
The problem with current financial analysis is that most of the operational and financial measures used have little or no predictive value with regard to future total return on investment (appreciation plus dividends). It is no wonder that the academic community has embraced the Efficient Market Hypothesis, The Capital Asset Pricing Model (CAPM), and Modern Portfolio Theory (MPT). Harry Markowitz, Bill Sharp, Eugene Fama, and others have all recognized the lack of predictability in the markets and have resorted to structuring portfolios based on stock inter-correlations and random price movements. The premise in these theories is that the markets are efficient and that sustained excess returns are not possible. I would like to suggest that their assertion be qualified by saying that, “sustained excess returns are not possible using current financial data and analysis techniques”.
Richard Thaler and his Behavioral Finance colleagues at the University of Chicago suggest that the problem with the above theories and models is the assumption that the investor/analyst is a rational individual. Anyone that has done even a casual evaluation of investor performance knows that the rational investor does not exist. Backed by years of psychological research, the Behavioral Finance community has demonstrated that humans are poor observers, are emotional, are overconfident in their abilities, over and under-react to events, and are generally the weakest link in the investment decision process (see Chapter 6).
Regardless of whether the problem is in the predictability of the data or in the humans making the decisions, the fact remains that the investment portfolios actively managed by professionally trained analysts and portfolio managers under-perform the simple (S&P 500) market cap weighted index most of the time. Because of this consistent underperformance, more and more institutions and individual investors are moving assets into index funds, a trend that is not healthy for the U.S. or world economies or the free enterprise system.
The Financial Analyst’s Handbook Volume I, Edited by Sumner Levine (1975) states that “The primary objective of any security analysis is the determination of future earning power because earning power is the source of cash flow to the investor (interest or dividends). Financial analysts are evaluated on their ability to forecast future corporate earnings. For example, the Institutional Investor All-America Research Team rankings and the StarMine SmartEstimate are based partly on past accuracy of earnings forecasts. The ratings of analysts published annually in the Wall Street Journal are based entirely on earnings forecast accuracy.
Every financial analyst in the world has been taught that a company’s stock appreciates in value because of the underlying earnings growth of the company. In the last five or ten years, however, analysts have had to change this view somewhat because a number of the new technology companies have elected to forgo earnings for company growth or increasing market share which, in the long run, may be better for investors. It is now apparent that even companies with earnings need to be viewed differently depending on when and how earnings are derived and reported. Analysts have now begun hedging their bets a bit by talking about the quality of earnings. They point out that not all earnings are the same and that their meaningfulness is dependent on a number of other factors. Analysts are now saying that the earnings, or lack there of, reported by accountants are sometimes misleading and not truly representative of the future performance of the company. The problem gets even worse when it comes to international stocks where the accounting rules for deriving earnings are different.
The relationship between a company’s actual performance and its stock performance is nebulous at best. The disconnect between company performance and stock performance became painfully evident during the run up of the .com companies in the late 1990s and led Dr. Greenspan, head of the Federal Reserve, to declare that investors were suffering from “irrational exuberance.” Most of the accounting numbers reported by companies in their quarterly and annual reports and the multitude of their government filings are provided to help understand what is going on in the company, and have little to do with what the company’s stock is going to do over the short or intermediate term. There is predictive information in traditional accounting data; it is just not in the isolated financial figures. I will discuss this in Chapter 9, but for right now, we need to focus on what is currently being used to forecast market performance.
Most investors’ look to financial analysts to tell them which companies to invest in to increase their wealth, and most wealth enhancement is accomplished by stock price appreciation rather than interest or dividends. If investors are interested in stock appreciation and analysts are forecasting corporate earnings, it is important then to take a look at the relationship between earnings and stock price appreciation to determine if what the analysts are doing will help the investor achieve his/her objective of investment appreciation.
In Figure 3-3 I have plotted the relationship between corporate earning growth and the appreciation in stock price over one year. The data in this figure were obtained from the Ford Investor Service, Inc. database for the years 1999 through 2001. What becomes immediately apparent is that there is no meaningful relationship between earnings growth and price appreciation. The authors have looked at other time periods, but the same lack of predictability holds. This creates a bit of a conundrum, because financial analysts have been telling us that earning growth is the most important variable for investors to look at in deciding which companies to invest in, but earnings growth does not appear to help the investor find stocks that are going to appreciate in value, at least not over a one year time horizon.
Figure 3-3
The Concurrent Relationship Between Earnings
Growth And Total Return (Years 1999-2001)
Now lets be totally fair; Levine (1975) also points out that financial analysis is a forward looking process and focused on the future rather than the past. The growth in earnings that has occurred over some historical period or even a concurrent period such as that shown in Figure 3-3 is not indicative of future earnings or price appreciation. To look out into the future, accountants and financial analysts make some assumptions about future earnings and interest rates, plug them into an equation and calculate a discounted cash flow or discounted dividend flow. These discounted figures are supposed to be indicative of the company’s future earning power and thus related to future stock appreciation. Given the importance and weight given to these discounted figures it is instructive to take a look at how they relate to future stock price appreciation. In Figure 3-4 I have plotted Ford Investor Services’ “Intrinsic Value” (based on a proprietary discounted dividend model) and stock price appreciation. Again it is apparent that the financial measures used are not very helpful in forecasting future stock appreciation
Figure 3-4
Relationship Between Discounted Future
Earnings And Total Return (Years 1999-2001)
There are many that suggest that any analysis based on earnings is open to question because of the ability of accounting firms and management to manipulate the figures. In light of what has happened at Enron and other major corporations using off balance sheet financing and other creative account practices it is not unreasonable to question the validity of these earnings figures. The head of Deloitte Touche, one of the largest accounting firms in the country, has said that the accounting practices and financial dealings of major corporations such as Enron are so complex that it is virtually impossible to understand what is going on in the corporation by reviewing the corporation’s financial statements. Even corporate insiders often don’t understand the market ramifications of the actions taken by the corporation. The Chief Executive Officer of Enron was buying Enron stock in the open market right up until three or four months before the corporation declared bankruptcy and he lost over $100 million on the transactions.
Congress and the governmental regulatory bodies are now considering adding more reporting requirements for corporations and expanding the amount of information that must be made available to analysts and the investing public. Certainly, having accurate, truthful accounting information is important, but it is only part of the problem. We have already exceeded the ability of humans to assimilate the information currently available and now we are talking about adding more. The next worst thing, after not having enough information for a human decision maker, is having too much information. Humans get overwhelmed and bogged down rather quickly as the amount of input that they have to deal with increases. Maybe what we should be doing is spending a little more time developing new analysis techniques that can extract the predictive information that investors need. It appears that the way current accounting information is being interpreted is not very helpful.
The one figure that is the least likely to be tainted by these complex financial dealings and “Hollywood” accounting is the top line revenue figure, and even here one must be careful because of the ability of accountants to manipulate the time when revenues are recognized for the financial statements. One generally assumes that companies are in business to produce and sell products and services, so you would naturally think that the more one produced and sold would be good and would be an important indicator of investors. If this were true you would expect revenue growth to be highly correlated with the performance of a company’s stock. Well throw that logic out the window! The correlation between revenue growth and stock appreciation is close to zero. It is obvious that the amount of sales/revenue generated by a corporation, in and of itself, is not what investors are looking for. Sure, increasing sales is important, but what the corporation does with the money is at least as important and perhaps even more so. Figure 3-5 indicates that despite the fact that we have eliminated a lot of the accounting manipulations by using the revenue figure we are still at a lost to find a good predictor of future stock performance.
Figure 3-5
The Relationship Between Concurrent Growth
Of Sales And Total Return (Years 1999-2001)
Financial Ratios
If none of these basic financial figures are related to a company’s stock performance in the market, what is an analyst to do? Early on financial analysts recognized that isolated financial figures, in and of them-selves, were not very useful so they started using first order relational information in the form of ratios. Anyone who has even casually studied the market or looked at a stock for investing has run across these ratios. The Price to Earnings,(P/E), Price to Book Value (P/BV), Price to Cash Flow (P/CF), Debt to Equity (D/Eq), Current Earnings to Current Liabilities (CE/CL) and many more are popular tools of the analyst profession. I could go through each of the ratios covered in popular security analysis textbooks and show you what the relationship of each of them is to the performance in the market place but it would be a waste of paper and ink. The story is the same for all of them; there is little or no relationship between the ratios and stock appreciation. Lets look at two of the more popular ratios just to make the point. Figure 3-6 shows the relationship between the stock price to earnings ratio and the twelve-month performance of stock price. The Figure speaks for itself.
Figure 3-6
Relationship Between Price/Earnings Ratio
And Stock Total Return (Years 1999-2001)
Perhaps one of the best first order relationships is the one between price and sales (P/S). This ratio has been discussed at length by Kenneth Fisher (1984) and again reviewed by James O’Shaughnessy (1997). Despite the positive reviews by both of these analysts, the use of the ratio by itself is of little value in predicting future price appreciation in the market. Figure 3-7 shows the relationship between the P/S ratio and twelve-month price appreciation.
Relationship Between The Price/Sales
Ratio And Stock Total Return (Years 1999-2001)
Now don’t get me wrong, I am not opposed to traditional financial analysis. I am from Missouri and you have to show me that it works. I will be the first ones to admit I am wrong and jump on board with support if someone can demonstrate that any of the traditional financial measures predict future security performance. I need some type of hard evidence to support its use in security selection. So far what I have seen suggests that it is being applied inappropriately and lacks what is known as “predictive validity.”
Some have questioned and criticized our use of a 12-month time horizon, suggesting that traditional financial measures have greater predictability for longer time horizons. I am not opposed to this notion, given the experience of Warren Buffet and others that use infinitely long time horizons, but I have to see the evidence. If you can show me a well designed quantitative study (one that controls for survivorship) that uses 3, 5, or 10-year time horizons to evaluate traditional financial measures I will willingly embrace the results. In the author’s view, however, forecasting anything out beyond one or two years is fraught with uncertainty and tenuous at best.
Finding additional information is not the problem. There is now more information available to financial analysts than they can possibly mentally digest. The question is how do you handle all of the information? The only option an individual has had up until now is to eliminate some of the information items and pare down the analysis task to some manageable size. The problem is, which pieces of information do you eliminate, and which pieces are important or even critical in the analysis task? Which bits of information interact with each other and produce an emergent pattern that relates to future market performance? Unfortunately the studies needed to answer these questions and make these decisions have not been done. Financial analysts have been too busy calculating the discounted cash flow of some undetermined future earnings stream. They are applying rather precise mathematics to some numbers of questionable value, based on some rather tenuous assumptions. In defense of the analysis community, it hasn’t been until just recently that they have had the tools necessary to conduct the sophisticated multivariate, nonlinear analysis needed, and most business programs don’t teach students how to use these analytical techniques.
The search for emergent patterns is just now becoming possible with some of the new Artificial Intelligence technologies being developed. The use and application of these new technologies, however, are going to require a change in mind set for the analyst community. Instead of thinking about eliminating data items and restricting their analysis they need to seek out additional information, items that seem just peripherally related to the problem, but may interact with other items to produce critical forecasting patterns. In place of a eugenic type mind set that tries to screen out diversity, the new analysis techniques will seek out unique data items that provide a different prospective for problem solution. We will finally come to recognize that diversity is good, and should be sought out for its unique contribution in the solution of previously intractable problems.