Fundamental Analysis, Behavioral Finance and Technical Analysis on the Stock Market

Table of Contents

List of Abbreviations

List of Figures

List of Tables

1. Introduction

2. Information Efficiency on the Stock Market
2.1. Efficient Market Hypothesis
2.1.1. Definition and Theoretical Assumptions
2.1.2. Efficiency Forms
2.1.3. Classification of Fundamental Analysis, Behavioral Finance and Technical Analysis
2.2. Empirical Studies
2.3 Preliminary Conclusion

3. Fundamental Analysis
3.1. Definition and Premises
3.2. Company Evaluation Methods
3.2.1. Separate Evaluation Methods
3.2.2. Overall Evaluation Methods
3.2.2.1. Present Value Methods
3.2.2.1.1. Capitalized Earnings Value Approaches
3.2.2.1.2. Discounted Cash Flow Approaches
3.2.2.2. Market Multiples
3.2.2.3. Real Options
3.3. Empirical Studies
3.4. SWOT Analysis
3.5. Preliminary Conclusion

4. Behavioral Finance
4.1. Definition and Premises
4.2. Anomalies
4.2.1. Anomalies Concerning the Investors Behavior
4.2.1.1. Information Perception
4.2.1.2. Information Treatment
4.2.1.3. Decision Making
4.2.2. Stock Market Anomalies
4.3. Sentiment Indicators
4.4. Empirical Studies
4.5. SWOT Analysis
4.5. Preliminary Conclusion

5. Technical Analysis
5.1. Definition and Premises
5.2. Analysis Methods
5.2.1. Chart Analysis
5.2.1.1. Basic Concepts
5.2.1.2. Formation Analysis
5.2.2. Indicator Analysis
5.2.2.1. Trend-Following Indicators
5.2.2.2. Oscillators
5.3. Empirical Studies
5.4. SWOT Analysis
5.5. Preliminary Conclusion

6. Synthesis Capabilities of Fundamental Analysis, Behavioral Finance andTechnical Analysis
6.1. Objective and Procedure
6.2. Theoretical Synthesis Capabilities
6.3. Practical synthesis capabilities using the example of the DAX Performance Index
6.3.1. Practical Application of Fundamental Analysis, Behavioral Finance and Technical Analysis
6.3.1.1. Intrinsic Value
6.3.1.2. Sentiment Indicators
6.3.1.3. Technical Indicators
6.3.2. Practical Synthesis Capabilities
6.4. Preliminary Conclusion

7. Final Conclusion and Outlook

Appendix

Bibliography

List of Abbreviations

illustration not visible in this excerpt

List of Figures

Figure 1: Overview of the Different Company Evaluation Methods

Figure 2: Systematization of the Discounted Cash Flow Approaches

Figure 3: The Three Methods of the Comparative Company Approach

Figure 4: Anomalies Concerning the Investors’ Behavior and Capital Market Anomalies

Figure 5: The Three Phases of the Investment Process

Figure 6: Prospect Theory’s Value Function

Figure 7: Chart Analysis Applied to the BMW AG

Figure 8: Indicator Analysis Applied to the BMW AG

Figure 9: DAX Performance Index: Actual Prices versus Intrinsic Values

Figure 10: DAX Sentiment Index

Figure 11: DAX Performance Index: Actual Prices versus RSI and MACD

List of Tables

Table 1: Evaluation of the Relevance and Usefulness of the Different Fundamental Analysis Approaches

Table 2: Information Perception Anomalies

Table 3: Information Treatment Anomalies

Table 4: Decision Making Anomalies

Table 5: Stock Market Anomalies

Table 6: Stock Splits of DAX Performance Index Members during 1-1-2004 - 10-31- 2008

Table 7: Index Memberships of the DAX Performance Index during 1-1-2004 - 10- 31-2008

Table 8: Distribution of Analyst Recommendations Employed within the Empirical Study

Table 9: DAX Performance Index: Intrinsic Value versus Actual Prices (January 2004 - December 2006)

Table 10: DAX Performance Index: Intrinsic Value versus Actual Prices (January 2007 - October 2008)

1. Introduction

During the last eleven months, the stock market has been characterized by crashes and a historically high volatility. In comparison to their peaks at the end of 2007, the Dow Jones Industrial Average Index and the DAX performance index have dropped almost 50% in value. The newspaper headlines around the world documented the extent of the stock market decrease and this historically high volatility:

“Wave of profit warnings expected – European groups face earnings fall out 40%” (Financial Times 2008a), “Porsche intentions baffle markets – Hedge funds scramble to assess damage” (Financial Times 2008c). Stock market analysts questioned whether “equity markets reached fair value?” (Financial Times 2008b) and partially drew the conclusion that “We’re getting to a point now with valuations where shares are incredibly low – we haven’t seen these levels since the 1930s.” (Financial Times 2008d)

Considering these statements, the following two questions are raised: What is the “fair” value of a stock? And by which analysis concepts could stock market investors be enabled to evaluate, if stocks are worth buying or if they should be sold?

In order to be able to answer these questions, three general analysis concepts have been developed: fundamental analysis, behavioral finance and technical analysis. Historically considered, fundamental and technical analyses have always competed, often leading to advocates that ideologically judge either a fundamental analysis or technical analysis to be the one and only analyzing concept. Behavioral finance is a relatively new scientific approach to explain psychological anomalies on the stock market, but is also more and more often considered to be able to compete with both fundamental and technical analyses.

Still, do these analysis concepts really compete in practice or could they actually supplement each other with their respective strengths?

Taking the turbulent stock market phases as well as these unanswered questions about fundamental analysis, behavioral finance and the technical analysis into consideration, this thesis ultimately pursues two general objectives:

Firstly, fundamental analysis, behavioral finance and technical analysis should be scientifically examined in terms of their premises, analysis approaches, empirical evidences as well as strengths and weaknesses.

Secondly, it should be examined as to whether the fundamental analysis, behavioral finance and technical analysis have theoretical and practical synthesis capabilities that could be used for developing a synthesis concept. The synthesis concept should combine the respective strengths and eliminate the respective weaknesses of each of the three analysis concepts.

Taking these two objectives into consideration, the thesis follows the following procedure: the second chapter deals with the question of how efficient the stock market is. Considering that the predictive capabilities of the three analysis concepts are based on different efficiency assumptions, the stock market efficiency must be discussed first.

Afterwards, the fundamental analysis (see Chapter 3.), behavioral finance (see Chapter 4.) and technical analysis (see Chapter 5.) are examined in detail. The procedure required to guarantee an objective evaluation of the respective analysis concept always remains the same: at first, the definitions as well as the premises of the respective analysis concept are explained. Afterwards the most important analysis approaches are examined. In addition to that, empirical studies should prove if, and by which approaches, the analysis concept is able to predict future stock prices. In order to be able to develop a synthesis concept, each analysis concept is evaluated by a SWOT analysis, pursuing the objective of determining the respective strengths, weaknesses, opportunities and threats of the analysis concept being considered. At the end of each chapter, a preliminary conclusion is drawn, enabling the reader to follow the most important insights of each chapter.

Based on the previous SWOT analyses, the sixth chapter examines the synthesis capabilities of the fundamental analysis, behavioral finance and technical analysis. In a first step, the synthesis capabilities are theoretically analyzed (see Section 6.2.). Based on the theoretical consideration, the synthesis capabilities are also practically examined in a second step (see Section 6.3.). A broad empirical study using the example of the DAX performance index analyzes the predictive capabilities of the three analysis concepts. By evaluating the theoretical as well as the practical synthesis capabilities, a general conclusion is drawn about the synthesis capabilities (see Section 6.4.).

At the end of this thesis, a final conclusion as well as an outlook is developed, dealing with the realizations about the respective analysis concept as well as the cognitions and future opportunities of the developed synthesis concept (see Chapter 7.).

2. Information Efficiency on the Stock Market

2.1. Efficient Market Hypothesis

2.1.1. Definition and Theoretical Assumptions

Based on the assumptions of the expected utility theory (Morgenstern/von Neumann 1944) and the concept of rational expectations (Muth 1961), Fama developed the Efficient Market Hypothesis (EMH), which is currently one of the cornerstones of important neoclassical capital market models such as CAPM and APT (Cesar 1996, Schäfer and Vater 2002).

The EMH comes to the conclusion that “a market in which prices fully reflect available information is called efficient” (Fama 1970, p.383). “Fully reflect” means that “prices adjust rapidly and unbiased to new information” (Gonedes 1976, p. 612). Prices, therefore, only alter when new information is received by rational investors (Peters 1996). Information that is known by rational investors has already been processed by those and is therefore correctly reflected by the actual stock prices (Widdel 1996). Thus, future stock price movements are only effected by new fundamental information.

By definition, new information is surprising and unpredictable for the investor, making future information and thus, future stock prices, follow a random walk (Dressendörfer 1999). On the one hand, random walk means that future stock price changes are independent from previous stock price changes. On the other hand, random walk also means that future stock price changes are distributed normally (Dressendörfer 1999). According to Malkiel (1999, p.24), random walk therefore “means that short-run changes in stock prices cannot be predicted. Investment advisory services, earnings predictions, and complicated chart patterns are useless.” But is this conclusion really indicative of reality for all of the three analysis concepts?

To be able to answer this question, the EMH is differentiated by its three efficiency forms in the following (Section 2.1.2.). Based on that differentiation, the three analysis concepts – fundamental analysis, behavioral finance and technical analysis – are classified in terms of their respective efficiency assumptions (Section 2.1.3.)

2.1.2. Efficiency Forms

According to Roberts, Fama classified the concept of information efficiency in three different forms of efficiency (Sapusek 1998). The efficiency forms differ in terms of their respective available information, whereby the stronger efficiency form contains the respective weaker efficiency form (Shleifer 2000):

- Weak form of information efficiency:
The weak form of information efficiency is present if all of the information about the previous stock prices is contained in the actual stock market prices at any time.
- Semi-strong form of information efficiency:
The semi-strong form of information efficiency is present if all public
information is contained in the actual stock market prices at any time.
- Strong form of information efficiency:
The strong form of information efficiency is present if all possible information – which includes insider information – is contained in the actual stock market prices at any given time.

2.1.3. Classification of Fundamental Analysis, Behavioral Finance and Technical Analysis

The distinction between the three forms of information efficiency is particularly important in terms of the applicability of the three analysis concepts – fundamental analysis, behavioral finance and technical analysis – which differ in their assumptions of their underlying information efficiency form (Sapusek 1998):

- Technical Analysis:

This approach is based on the idea of analyzing historical stock prices in order to be able to draw conclusions on the future stock price movements (Brunnermeier 2001). Thus, technical analysis is based on an information efficiency assumption that is at least weaker than the weak form of information efficiency (Aronson 2007, Dornbusch 1998).

Otherwise, the investor would not be able to generate steady abnormal returns by applying technical analysis, because all of the information about the historical stock prices is already contained in the actual stock prices (Aronson 2007, Menz 2004, Niquet 1997).

- Fundamental Analysis:

This approach tries to generate an abnormal return by analyzing fundamental factors of a company to be able to draw a comparison between the theoretically justified fair value of a company’s stock and the actual stock market price (Damodaran 2006). These fundamental factors are derived by analyzing public information – e.g. annual reports, macroeconomic indicators, etc.

Unlike the technical approach, the fundamental analysis is able to generate an abnormal return, even if the weak form of information efficiency exists (Sapusek 1998). However, fundamental analysis requires at least one weaker form than the semi-strong form of information efficiency (Dornbusch 1998). Otherwise, the investor would not be able to generate steady abnormal returns by applying the fundamental analysis, because all public information is already correctly processed by the stock market and therefore, correctly reflected by the actual stock market prices (Haugen 1999).

- Behavioral Finance:

The behavioral finance approach is based on the idea that investors do not behave rationally, but rather recurrently irrational (Shleifer 2000).

In contrast to this, the EMH is based on the expected utility theory as well as the concept of rational expectations (Ellenrieder 2001). Thus, behavioral finance denies all three forms of information efficiency, because it denies the general concept of the EMH. This is due to the fact that it is based on completely different assumptions about the investors’ behavior – e.g. irrationalities, like reacting biased and timely lagging in accepting new information (Frankfurter 2007).

2.2. Empirical Studies

During recent decades, the EMH has been widely analyzed on stock markets – qualitatively as well as quantitatively – following the objective of being able to draw a conclusion whether or not the EMH is present. Particularly the quantitative empirical studies have had the objective of determining which form of information efficiency goes along with stock price movements, in reality.

With reference to the qualitative arguments against the EMH, the joint-hypothesis-problem (Kosfeld 1996) as well as the information-paradox (Sommer 1999) must be mentioned.

The joint-hypothesis problem was recognized by Fama and describes the problem that (1991, p.1575): “Market efficiency per se is not testable. It must be tested jointly with some model of equilibrium, an asset pricing model. As a result, when we find anomalous evidence on the behavior of returns, the way it should be split between market inefficiency or a bad model of market equilibrium is ambiguous.” The information paradox introduced by Grossmann/Stiglitz (1980) state that – unlike the EMH – the information-generating process creates real costs. Due to the fact that actual stock market prices reflect all available information within an information-efficient market, nobody would be willing to make an effort to receive costly information. That wouldn’t result in any advantages in comparison to simply using the actual stock market price. Yet, if all investors behave that way, the stock market would no longer be efficient, due to the fact that the information generating process – as an essential requirement for an efficient market – would no longer exist (Spremann 2006). Therefore, Grossmann/Stiglitz assume that markets have to be information-efficient to an extent to which costs – caused by the information generating process and the following evaluation of that information – have to at least be covered (Lo and MacKinlay 1999).

Concerning the quantitative analysis of the EMH, the empirical studies refer to the respective information efficiency form (Section 2.1.2.). The most important analysis tools and their respective analysis results are listed in the following:

- Weak form of information efficiency:

The existence of the weak form of information efficiency can be tested by analyzing the predictability of future stock returns on the basis of historical stock prices (Hruby 1991).

Only if historical stock prices do not show any stochastic independence (i.e. correlation of zero) with future stock prices, can the weak form of information efficiency be assumed to exist. Otherwise the random walk characteristic of future stock prices cannot be considered as fulfilled.

On the one hand, the independence can be analyzed by autocorrelation, spectral and run tests, on the other hand, by filter techniques (Hoffmann 2001). Particularly the autocorrelation, spectral and run tests over the last thirty years have shown that historical stock prices are at least weakly correlated with future stock prices (Sapusek 1998).¹

The empirical results of filter technique concepts – which analyze if mechanically triggered trading strategies are able to beat a simple buy-and-hold strategy – support the assumption that the random walk characteristic cannot be assumed to be existent on the stock market (Dressendörfer 1999).²

- Semi-strong form of information efficiency:

The existence of the semi-strong form of information efficiency is tested by event studies, which analyze if actual stock market prices react immediately to new information (Sapusek 1998).

The event studies prove that stock prices react to new information, but those reactions can take place biased and late (Shleifer 2000). Therefore, stock prices mostly reflect new information, but do not always “adjust rapidly and unbiased to new information”, like Gonedes (1976, p.612) defined the efficiency criteria of “fully reflected” by Fama (1970, p.383). Thus, the semi-strong form of information-efficiency cannot be confirmed in general.³

- Strong form of information efficiency:

The existence of the strong form of information efficiency is tested by analyzing if insiders are able to generate an excess return on the stock market by using non-public information (Sapusek 1998). Empirical studies have shown that private information enables insiders to generate an excess return in comparison to investors who do not have access to that private information (Mörsch 2005). If that would not be the case, laws for generating transparency in terms of insider deals – like the German § 15a WpHG – would not be necessary (Spremann 2006). Science agrees that the strong form of information efficiency is not existent in reality (Mörsch 2005).⁴

2.3 Preliminary Conclusion

Under consideration of the controversial results of the empirical studies, it can be summarized that the EMH can be neither fully rejected nor can it be fully confirmed (Hruby 1991). Thus, stock analysis concepts like technical analysis, fundamental analysis and behavioral finance cannot generally be considered worthless (Dressendörfer 1998).

The joint-hypothesis-problem and particularly the theory of information paradox support the idea that these analysis concepts are necessary to achieve at least a certain level of information efficiency on the stock market (Albrecht and Maurer 2005).

So, which of the three efficiency forms exists on the stock market?

At least the science seems to agree on the rejection of the strong form of information efficiency, because insiders are proven to be able to generate an excess return by using their private information.

With regard to the semi-strong form of information efficiency, science agrees that in general, stock prices react to new public information, but it is possible that these reactions can take place in a biased and late manner. That leads us to the conclusion that the semi-strong form generally cannot be confirmed over time.

The existence of the weak form of information efficiency is also doubtful, due to the fact that empirical studies have shown that the random walk characteristic of stock prices can be denied by autocorrelation, spectral and run tests. In addition to that, filter-techniques underline that mechanically triggered trading strategies seem to be at least temporarily able to generate an excess return, in comparison to simple buy-and-hold strategies.

With reference to an experimental study by Huber et al. (2006), the conclusion as to which of the three information efficiency forms is present on the stock market, cannot be given a general, final judgment, but rather is also dependent on the respective stock market cycle. This is due to the fact that influencing factors on information efficiency– like psychological factors and the application frequency of different analysis concepts – change over time with different stock market cycles (Shleifer 2000).

3. Fundamental Analysis

3.1. Definition and Premises

The term “fundamental analysis” is widely used in capital market analysis and therefore describes a wide range of fundamentally-driven analysis concepts (Cesar 1996). In the following, the term “fundamental analysis” encompasses the most important fundamentally-driven analysis concepts for determining the value of a company – expressed by the price of its stocks (Gantenbein and Spremann 2005). Fundamental analysis is based on the premise that the actual stock market price fluctuates around its intrinsic value over time (Brigham and Houston 1998).

The intrinsic value is defined as the fair value of a company’s stock. It is determined by analyzing which and to which extent fundamental factors have an influence on a company’s value (Beike and Schlütz 2005, Mattern 2005). Finally, fundamentally-driven concepts – no matter which concept is considered – all have the same objective: comparing the calculated intrinsic value with the actual stock market price to be able to draw a conclusion, if the analyzed stock is undervalued or overvalued (Brigham and Houston 1998).

On the one hand, a stock is assumed to be undervalued if the intrinsic value lies under the actual stock market price (McLeavy and Solnik 2003). On the other hand, a stock is respectively assumed to be overvalued if the actual stock market price exceeds the intrinsic value of the stock (McLeavy and Solnik 2003).

The concept of intrinsic value implies that market participants are assumed to be rational in terms of buying a company’s stock, because of the fundamental value of that company (Mattern 2005).

Nevertheless, fundamental analysis as a whole depends on the assumption that actual stock market prices do not always correctly reflect the real fundamental strength of a company. This is due to the fact that actual stock prices are also influenced by temporarily occurring anomalies, resulting in divergences between the intrinsic value and the actual stock market price in the short run (Franke and Hax 2003).

Those temporarily occurring anomalies are necessary for the successful application of fundamentally-driven analysis concepts. Otherwise, fundamental analysis would not be able to generate any advantage by determining the intrinsic value, because if at least a semi-efficient stock market is to be assumed – implying that actual stock market prices adjust rapidly and unbiased to new public fundamental information (Gonedes 1976) – the actual stock market price would already be equal to the intrinsic value of a stock. Therefore, fundamental analysis assumes that the market price of a stock ultimately follows its intrinsic value, but can vary from its intrinsic value in the short run (Bohl and Siklos 2001, Shiller 1981).

The fundamental value of a company – expressed by its intrinsic value – can be measured differently, leading to different fundamentally-driven analysis concepts that are explained in the following (Section 3.2.).

3.2. Company Evaluation Methods

With regard to the different premises of the most important value drivers as well as different motives for evaluating a company (Borowicz 2005), fundamentally-driven analysis methods are differentiated by different methods and approaches, shown in Figure 1: illustration not visible in this excerpt

Figure 1: Overview of the Different Company Evaluation Methods (Coenenberg and Schultze 2002)

Depending on whether a company is considered by its separate assets and debts or as an interdependent complex, the analysis methods are subdivided into separate evaluation and overall evaluation methods (Dehmel and Hommel 2008). Due to its higher frequency of application and importance in practice (Gantenbein and Gehrig 2007), the following explanations of the basics as well as the most important advantages and disadvantages of the respective analysis methods are focused on the overall evaluation methods.

The overall evaluation methods are subdivided into present value methods, market multiples and real options, due to its different premises of how to evaluate a company fundamentally. With regard to its higher practical (Gantenbein and Gehrig 2007) and scientific (Prokop and Zimmermann 2002) relevance, the present value methods will be explained and discussed in detail and therefore, be subdivided into capitalized earnings value and discounted cash flow approaches.

3.2.1. Separate Evaluation Methods

The separate evaluation methods assume that the value of a company can be calculated by summing up the separately evaluated assets (a), subtracted by the nominal sum of debts (d) of the company (Deter et al. 2005a). Thus, the enterprise value (ev) can generally be expressed by the formula:ev=a−d.

The two most important separate evaluation methods are the net asset value approach and the liquidation value approach. They differ in terms of their respective going-concern premises (Drukarcyk and Schüler 2007). The net asset value approach assumes the continuation of the company, whereas the liquidation approach implies the liquidation of the company (Dehmel and Hommel 2008).

The net asset value of a company reflects the amount which would have been paid if a company had been identically reproduced (Brösel and Matschke 2007). It is calculated by summing up all operationally necessary assets of a company by its reproduction prices and then subtracting the sum of all debts by its nominal values. The result is added to the value of the sum of the entire operationally non-necessary assets of the company by its liquidation prices (Deter et al. 2005a).

The net asset value approach can be differentiated by two different concepts – the partial reproduction and the overall reproduction approach. The partial reproduction approach is limited on the evaluation of tangible assets, whereas the overall reproduction approach also takes estimated values for intangible assets into consideration (Dehmel and Homel 2008).

The liquidation value approach is applied in case of restructuring or termination of a company. It is calculated by summing up the company’s assets according to its liquidation prices and subtracting the sum of debts by it respective nominal prices afterwards. The result has to be settled with the costs caused by the liquidation process of the assets. (Deter et al. 2005a).

Within the main context of fundamentally-driven analysis concepts, the net asset value approach, as well as the liquidation value approach, currently has the function of checking up the plausibility of values, calculated by overall evaluation methods (Kames 1999). Therefore, it fulfills the function of a lower value limit (Helbling 2007) or a value of correction (Dehmel and Homel 2008).

Unlike the overall valuation methods, the separate valuation methods just evaluate the actual value of a company’s assets. That goes along with the disadvantage of ignoring future cash flows, generated by the interaction of different assets (Deter at al. 2005a). In addition to that, the evaluation of intangible assets is difficult and imprecise if reproduction or liquidation prices are used. Therefore, the separate valuation methods are currently mainly used either for companies, whose business is predominantly based on tangible assets, or to check up on whether the results of the overall valuation methods are plausible by delivering a lower value limit for the price paid for a company (Kames 1999).

3.2.2. Overall Evaluation Methods

Unlike the separate evaluation methods, the overall evaluation methods evaluate a company as a whole, by taking the interdependencies between separate assets and debts into consideration (Borowicz 2005). Therefore, “overall evaluation” means that assets and debts are evaluated in context, due to the fact that the company is evaluated as a whole (Ballwieser 2007). The overall evaluation methods can be subdivided into present value methods (Section 3.2.2.1.), market multiples (Section 3.2.2.2.) and real options (Section 3.2.2.3.), which are explained in the following.

3.2.2.1. Present Value Methods

Present value methods are based on the premises of the future-oriented capital budgeting approach (Deter et al. 2005a, Drukarczyk and Schüler 2007). This leads to the conclusion that the value of a company is represented by the present value of all net-incomes of the investor (De Fusco et al. 2001, Helbling 2001).

Depending on how these net-incomes are defined, present value methods are subdivided into earnings and cash flow approaches (Beike and Schlütz 2005). Therefore, a differentiation must be maintained between the capitalized earnings value and discounted cash flow approaches (Borowicz 2005).

In terms of the structure of the method for calculating the respective income figure, it doesn’t matter whether earnings or cash-flows are calculated. In both approaches, the evaluation period is subdivided into two time periods – t and n (Borowicz 2005, Streitferdt 2008).

The first period extends over a time period of three to five years (t) (Streitferdt 2008) and calculates the income on the basis of the companies business plans (Borowicz 2005) or a concrete estimation of the respective evaluator (Henselmann 2002). The second period encompasses the subsequent years up to eternity (n) (Streitferdt 2008) and calculates a residual income, which is calculated by deriving steadily achievable values of the first evaluation period and extrapolating those to eternity (Henselmann 2002).

In addition to different income figures, the two approaches also differ in terms of their discounting factors (Borowicz 2005). The discounting factor of the earnings approach is calculated by adding an estimated risk premium to a risk free rate, whereas the discounting factors of the cash-flow approaches vary from the expected return of shareholders (equity-approach) to the expected return of shareholders as well as creditors (entity-approach) (Bruns and Steiner 2007, Hachmeister 1994).

In comparison to other fundamentally-driven analysis concepts – such as overall evaluation methods and market multiples – the present value methods have the advantage of taking the time value of money explicitly into consideration (Dehmel and Hommel 2008). In addition to that, the consideration of earnings or cash flows better reflects the interdependent value of a company than the overall evaluation methods (Deter et al. 2005a) and is less susceptible to market-driven misinterpretations, like the approach of market multiples (Damodaran 2006).

With regard to the disadvantages of the present value approaches, one can argue that particularly the residual value of a company is often characterized by uncertainty, yet has a share in overall value of the company from approximately 50 to 80 percent (Henselmann 2002, Hoke 2002).

In the following, the capitalized earnings value approach as well as the different discounted cash flow approaches are explained and evaluated in terms of their advantages and disadvantages.

3.2.2.1.1. Capitalized Earnings Value Approaches

Capitalized earnings value approaches calculate the value of a company by discounting and capitalizing the future sustainable earnings of a company (Borowicz 2005). Moxter (1994) defines the capitalized earnings value as the sum of financial and non-financial values of a company.

With regard to their practical application, the reference figure for describing the sustainable success of a company can vary, depending on the preferences of the evaluator. On the one hand, balance sheet-driven success figures like EBT can be used (Borowicz 2005). On the other hand, the application of adjusted and more cash-flow oriented success figures, such as EBIT and EBITDA, are also possible (Deter et al. 2005a).

The discounting factor is based on the idea of an alternative comparable investment opportunity (Deter et al. 2005a) and reflects the costs of equity (Dehmel and Hommel 2007). The costs of equity are calculated by two steps:

In the first step, a risk free rate of return is determined – usually by considering the return of a risk-free rate, such as a triple A-rated government bond (Borowicz 2005).

In a second step, a risk premium is added for meeting the requirement of considering individual risk drivers, like operative, capital structure and industry sector risk factors (Borowicz 2005).

In comparison with the discounted cash flow approaches, the capitalized earnings approach has two disadvantages:

The capitalized earnings approach falls back on an imprecisely defined discounting factor, which can lead to different results in practice, due to the respective definition of the applied discounting factor (Deter et al. 2005a, Dehmel and Hommel 2008). Also, balance sheet-driven success figures are more susceptible to manipulations than cash-flow oriented figures (Behringer 2007, Deter et al 2005a).

According to Borowicz (2005), as well as Drukarczyk and Schüler (2007), the capitalized earnings value approach can be considered as the German version of the DCF equity approach, due to its similarities.

3.2.2.1.2. Discounted Cash Flow Approaches

Discounted cash flow approaches calculate the value of a company by discounting its future cash flows (Borowicz 2005). The discounted cash flow approach is differentiated by three methods: weighted average cost of capital (WACC), adjusted present value, and the equity methods: illustration not visible in this excerpt

Figure 2: Systematization of the Discounted Cash Flow Approaches (Steiner and Bruns 2007)

Figure 2 illustrates that the three methods differ in terms of the definition of their cash flows and their different discounting factor, which has influence on the treatment of the company’s tax shield (Bruns and Steiner 2007), explained in the following:

• WACC method: illustration not visible in this excerpt

WACC WACC

The WACC method is based on the assumption that the company is first evaluated under the point of view of all providers of capital – equity as well as debts (Bruns and Steiner 2007). Therefore, the cash flow is expressed by the Free Cash Flow (FCF), which still contains the cash flows obtained by creditors (Deter et al. 2005b).

The free cash flows are discounted by the WACC, which expresses the expected return of shareholders as well as creditors (Copeland et al. 2002, Vettiger and Volkart 2002). The WACC is calculated by the formula: illustration not visible in this excerpt

The expected return of the shareholders (iCAPM) as well as the expected return of the creditors, multiplied by the company’s tax shield (id*(1−tr)), is illustration not visible in this excerpt

In a second step, the residual value of the company (RV) is also discounted by the WACC.

The sum of WACC-discounted free cash flows and residual value expresses the value of the company under the point of view of both, the shareholders and the creditors. Therefore, the value of debts has to be subtracted afterwards to ensure that the value of equity of the company remains (Deter et al. 2005).

- APV method: illustration not visible in this excerpt

The APV can be considered as the modified version of the WACC, dividing the value of the company into its separate value drivers (Drukarczyk and Schiller 2007, Krolle 2001).

In the first step, the company is assumed to be fully self-financed, so that the operative value of the company can be considered separately from the value of the company’s capital structure (Deter et al. 2005, Krolle 2001). Therefore, the free cash flows are not discounted by the WACC, but only by the expected return of the shareholders (iCAPM). Analogous thereto is that the residual value of the company is also discounted by the shareholders’ expected return (Vettiger and Volkart 2002).

In a second step, the value of the tax shield is calculated by the sum of the interest rate for debts (id), the amount of debts (dt) and the company’s tax rate (tr), discounted by the interest rate for debts.

In order to achieve the value of equity, the sum of CAPM-discounted free cash flows and residual value, as well as the value of the tax shield, are subtracted by the value of debts at the end.

In comparison to the WACC method, the APV has the advantage of generating transparency in terms of the respective value drivers of the company (Steiner and Wallmeier 1999).

- Equity method: illustration not visible in this excerpt

In contrast to the entity methods, the equity method only considers the cash flows obtained by the shareholders – the so called Flows to Equity (FTE)

(Damodaran 2002, Deter et al. 2005).

In analogy to the applied cash flow definition, the flows to equity are only discounted by the expected return of the shareholders (iCAPM) in a first step. In a second step, the residual value of the company (RV) is determined and also discounted by the expected return of the shareholders (Vettiger and Volkart 2002).

The sum of CAPM-discounted flows to equity and residual value already reflects the company’s value of equity. Unlike the WACC and APC methods, the equity method already considers the debts of the company in its cash flow definition, such that the distraction of debts at the end is omitted (Drukarczyk and Schüler 2007).

With regards to the advantages and disadvantages of the discounted cash flow concept as a whole, the following conclusions can be determined:

Unlike balance-sheet driven success figures, cash-flows are less susceptible to manipulations (Behringer 2007, Deter et al 2005a). This explains its superiority in comparison to its most similar evaluation method – the capitalized earnings value approach. On the other hand, the estimation of future cash flows can be complex in practice, due to the large number of influencing factors on the estimated cash flow (Damodaran 2001).

The application of the CAPM has the advantage of a consistent structure for determining the shareholders expected return (Hachmeister 1994), but goes along with the disadvantages of the CAPM (Müller and Röder 2001, Vettiger and Volkart 2002), as well as its complex application in practice, e.g. the application of consistently determined betas by different evaluators (Ballwieser 1995, Damodaran 2001).

Nevertheless, the discounted cash flow methods have been established as the most frequently used analysis methods in practice (Bömelburg et al. 1994, Hillers et al. 1999), which supports their advantages in comparison to other fundamentally-driven analysis concepts.

3.2.2.2. Market Multiples

The application of market multiples is based on the premise, that the value of a company can be derived by comparing a predefined performance figure of the evaluated company with the value of that performance figure of a comparable company (Coenenberg and Schultze 2002).

Therefore, the formula for calculating the fair value of the evaluated company is illustration not visible in this excerpt

Whereby:

p(e)= searched price of the evaluated company

fe = performance figure of the evaluated company
p(c) = kwon price of the comparable company

fc= performance figure of the comparable company

mi= multiple.

The multiple can refer to profit-orientated, earnings-orientated (e.g. EBITDA, EBIT), cash-flow-orientated (sales, cash-flows) or other-oriented (customers, visitors) parameters, that are adequate to measure the value of the company (Bruns and Steiner 2007, Seppelfricke 1999).

The multiple is determined by the comparative company approach, which is differentiated by its three sub-methods, differing in terms of its derivation methods of the multiple:

illustration not visible in this excerpt

Figure 3: The Three Methods of the Comparative Company Approach (Dehmel and Homel 2008)

The recent acquisition method derives the multiple from recently executed transactions, whereas the initial public offering approach calculates the multiple based on recently paid prices for companies that have gone public (Dehmel and Hommel 2008). In contrast to that, the similar public company method takes each comparable publicly-quoted company into consideration (Bausch 2000).

If its possible, the respective multiple is not only derived by taking a single comparable company into consideration, but by building a peer group that is as similar as possible to the evaluated company in terms of its industry sector, growth rate and risk structure (Drukarczyk and Schüler 2007). Otherwise, the expressiveness of the used multiple has to be severely challenged.

In addition to that, there are some other arguments that put the concept of multiples into question: The market price of the company’s peer group builds the basis for evaluating the considered company. That could lead to a misinterpretation of the company’s real value, if the entire peer group is wrongly priced by the stock market (Fleischer 1999). In hindsight, this was the case during the new economy phase at the end of the 20th century (Perkins and Perkins 1999), where the heavy use of market multiples at least supported the process of wrong stock pricing (Hoffmann 2001), due to the fact that the peer group itself was wrongly priced by the market. Furthermore, multiples belong to the group of static analysis concepts, meaning that future growth rates are not taken into consideration (Deter et al. 2005a). That could have a serious impact on the comparability of companies if the considered companies clearly differ in their predicted future growth rates (Dehmel and Hommel 2008).

Taking these aspects into consideration, the IDW (2007) advises that the concept of multiples should only be used for checking the plausibility of results of other more detailed and dynamic concepts, such as the discounted cash flow or the capitalized earnings value approach (Aders et al. 2000, Auge-Dickhut and Moser 2003).

Nevertheless, Kames’ (1999) examinations of the application frequency of multiples – which proved that 72% of all financial analysts always use the concept of multiples when a company has to be evaluated – indicates that the concept of multiples has its advantages. This is particularly due to the fact that multiples are persuasive in terms of a simple, fast and inexpensive application for determining at least a broad price range for a potentially fair value of a company (Deter et al. 2005a, Krolle et al. 2005, Stock 2001).

3.2.2.3. Real Options

The concept of real options is based on the premise that management’s flexibility to react to modified circumstances has a value itself, which should be taken into consideration when a company is evaluated (Crasselt and Tomaszewski 1999). Brown and Reilly (2006, p.979) are convinced that “conventional net present value calculations ignore the benefits of flexibility and may therefore undervalue projects that allow companies to react rapidly to changing circumstances.”

This is due to the assumption of present value concepts, which states that actions by the company’s management are not flexible but static (Antikarov et al. 2003). This leads to the conclusion that present value concepts are based on the assumption of a symmetrically distributed future company value, whereas the concept of real options is based on an asymmetrical distribution assumption (Allen et al. 2006). Unlike real options, present value concepts therefore underestimate the value of a company (Antikarov et al 2003).

On the basis of the concept of financial options, the flexibility of the management’s decisions is measured by real options – e.g. options to expand, options to abandon, timing options, production options etc. (Allen et al. 2006, Deter et al. 2005a). Those real options always have a value, so that the value of a company is composed by a static value – analogous to the present value concepts – and the sum of the options’ values (Antikarov et al. 2003).

[...]

---

¹ Autocorrelation and run tests on the German stock market (1961-1972) show that the random walk hypothesis cannot be applied to the German stock market (Reiß 1974). Autocorrelation and spectral tests on chosen stock indices prove the random walk hypothesis to be incorrect (Granger and Morgenstern 1970). Autocorrelation tests on the New York Stock Exchange reject the random walk hypothesis (Kinney and Rozeff 1976) and the assumption of normal distribution (Fiellitz and Greene 1977).

² Alexander (1961 and 1964) proves the superiority of filter techniques in comparison to a simple buy-and-hold strategy, using the example of the Dow Jones and Standard and Poor’s. Blume and Fama (1966) criticize Alexander’s results and show them to be wrong, if transaction costs are considered. >Filter techniques on the Austrian stock market (1965-1974) prove that historical stock prices and future stock prices are dependent and correlated (Uhlir 1979). Bertoneche (1979) analyzes six European stock markets and draws the conclusion, that the application of filter techniques is able to beat a simple buy-and-hold strategy.

³ Waud (1970) confirms the semi-efficient form by analyzing the effect of interest rate changes on the stock market. Fama (1991) and Möller (1985) consider the semi-strong form of information efficiency to be proven by event studies on the German stock market. Berry and Howe (1994) show that stock market prices react in a biased and late manner to new fundamental information.

⁴ Meulbroek (1992) draws the conclusion that insiders are able to generate an excess return on the American stock market. Schmidt and Wulff (1993) confirm Meulbroek’s statement for the German stock market.