Readability of Quarterly Reports: Do companies mislead investors?

Table of Content

1. Introduction
1.1 Aim of this Work
1.2 Structure

2. Reporting on Financial Markets
2.1 Introduction
2.2 Purpose and Functions of Reports
2.3 Typical Content of Reports
2.3.1 Letter to the Shareholders
2.4 Specifics of Reports at Frankfurt Stock Exchange Prime Market
2.4.1 The Frankfurt Stock Exchange
2.4.2 Annual Reports at Frankfurt Stock Exchange Prime Market
2.4.3 Quarterly Reports at Frankfurt Stock Exchange Prime Market
2.5 Market Functioning
2.5.1 Introduction
2.5.2 Efficient Markets Hypothesis
2.5.3 Incomplete Revelation Hypothesis

3. Stakeholder Theory
3.1 Historical Origins
3.2 Stakeholder Definition
3.3 Stakeholder Groups
3.4 Stakeholder Concept
3.5 Relationship Between Shareholders and Managers
3.5.1 Agency Theory
3.5.2 Obfuscation Hypothesis
3.5.3 Shareholder Influence and Independence

4. Readability Research
4.1 Concept of Readability
4.2 Measures of Readability
4.2.1 Introduction
4.2.2 Historical Overview of Readability Research
4.2.3 Gray and Leary’s Readability Study (1935)
4.2.4 Dale-Chall Formula (1948)
4.2.5 Gunning-Fog Index (1952)
4.2.6 Flesch Index (1948)
4.3 Limitations of Readability Measures
4.3.1 Introduction
4.3.2 Bruce’s Categorization
4.3.3 Other Critics
4.3.4 Alternatives
4.3.4.1 Usability Testing
4.3.4.2 Cloze Procedure
4.3.4.3 Coh-Metrix
4.4 Readability of Corporate Reports
4.4.1 Introduction
4.4.2 Annual Report Readability
4.4.2.1 Pashalian and Crissy (1950)
4.4.2.2 Soper and Dolphin (1964)
4.4.2.3 Barnett and Leoffler (1979)
4.4.2.4 Still (1982)
4.4.3 Studies on Performance Influence on Readability
4.4.3.1 Introduction
4.4.3.2 Adelberg (1979)
4.4.3.3 Courtis (1986)
4.4.3.4 Jones (1988)
4.4.3.5 Baker and Kare (1992)
4.4.3.6 Smith and Taffler (1992)
4.4.3.7 Subramanian, Isley and Blackwell (1993)
4.4.3.8 Courtis (1995)
4.4.3.9 Rutherford (2003)
4.4.3.10 Smith, Jamil, Johari and Ahmad (2006)
4.4.3.11 Li (2008)
4.4.3.12 Hossain/Siddiquee (2008)
4.4.3.13 Conclusio

5. Research Design
5.1 Hypothesis Development
5.1.1 Introduction
5.1.2 Hypothesis 1 – Level of Difficulty
5.1.3 Hypothesis 2 – Bilingual Reports
5.1.4 Hypothesis 3 – Company Size
5.1.5 Hypothesis 4 – Shareholder Independence
5.1.6 Hypothesis 5 – Share Price Volatility
5.1.7 Hypothesis 6 – Financial Performance
5.2 Methodology
5.2.1 Research Objects
5.2.2 Sample
5.2.3 Statistical Methods

6. Results
6.1 Descriptive Statistics
6.1.1 Introduction
6.1.2 Main Variables
6.1.3 Readability Scores
6.1.3.1 Adjusted Flesch Score in German
6.1.3.2 Adjusted Flesch Score in English
6.1.3.3 Raw Flesch Score in German
6.1.3.4 Raw Flesch Score in English
6.2 Hypothesis Testing
6.2.1 Introduction
6.2.2 Testing Hypothesis 1 – Level of Difficulty
6.2.2.1 Testing Hypothesis
6.2.2.2 Testing Hypothesis
6.2.3 Testing Hypothesis 2 – Bilingual Reports
6.2.4 Testing Hypothesis 3 – Company Size
6.2.5 Testing Hypothesis 4 – Shareholder Independence
6.2.6 Testing Hypothesis 5 – Share Price Volatility
6.2.7 Testing Hypothesis 6 – Financial Performance
6.2.7.1 Testing Hypothesis
6.2.7.2 Testing Hypothesis
6.2.8 Summary Table of Hypotheses Testing
6.3 Discussion
6.4 Limitations

7. Conclusio

References and Sources

List of Figures

Figure 1: Typical components of annual/quarterly reports

Figure 2: Typical content of a letter to the shareholders

Figure 3: Typical stakeholders of a company

Figure 4: Factors that influence reading difficulty

Figure 5: Exemplary excerpt of words contained in Dale's list

Figure 6: Readability-score.com analysis form

Figure 7: Superfluous dots in a letter to the shareholders

Figure 8: Distribution adjusted Flesch score in German

Figure 9: Distribution adjusted Flesch score in English

Figure 10: Distribution raw Flesch score in German

Figure 11: Distribution raw Flesch score in English

Figure 12: Readability correlation in bilingual quarterly reports

List of Tables

Table 1: BvDEP Independence Indicator classification

Table 2: List of style factors affecting readability by Gray and Leary

Table 3: Dale-Chall formula correction table

Table 4: Interpretation of the Fog index

Table 5: Flesch' reading ease scale

Table 6: Key questions unanswered by readability formulas

Table 7: Research studies on the relationship between performance and readability of corporate annual reports

Table 8: Descriptive statistics of the sample

Table 9: Summary statistics adjusted Flesch score in German

Table 10: Summary statistics adjusted Flesch score in English

Table 11: Summary statistics raw Flesch score in German

Table 12: Summary statistics raw Flesch score in English

Table 13: Readability by industries

Table 14: Test on industry difference with confidence intervals

Table 15: Regression analysis adjusted Flesch score German

Table 16: Regression analysis adjusted Flesch score English

Table 17: Regression analysis raw Flesch score German

Table 18: Regression analysis raw Flesch score English

Table 19: Summary of hypotheses testing

1. Introduction

1.1 Aim of this Work

Financial reports can be regarded as the primary means of communication between a company’s management and its shareholders. The reports also address all other kinds of stakeholders like employees, suppliers, customers, competitors, governments, potential investors, bond holders and, in a broad sense, the entire society.^[1] Generally speaking, financial reports concern “any group or individual who can affect or is affected by the achievement of the organization’s objectives.”^[2]

The reports aim to keep stakeholders informed about the business history, financial status and future expectations.^[3] Typically, both narrative information and quantitative data are provided in firms’ reports through the inclusion of financial statements, notes, discussion of operations and letters to the shareholders.^[4] Investors to a great extent base their risk assessment and investment decision on these published reports. Therefore, they rely on the usefulness, accuracy and understandability of the “intelligence” provided.^[5] Bloomfield cites a vivid example in which he compares financial reporting with the provision of raw data – it still depends on the investor himself to extract meaningful statistics out of the data and draw reasonable conclusions. This of course takes time and effort and is by no means a trivial task. Hence, the process is said to be costly. Still, the traditional Efficient Markets Hypothesis (EMH) leads to the conclusion that no matter how obscurely financial reports are presented or how costly their interpretation is, market prices fully reflect all available information. In contrast to the EMH, the Incomplete Revelation Hypothesis (IRH) states that investment decisions indeed depend on the extent of costs required to extract statistics from public data. It is argued that additional costs decrease trading interest. Low trading interest then leads to a lower revelation in market prices.^[6] Moreover, and in line with the IRH, Bloomfield states that by various means managers could actively hamper investors to uncover information that should not be uncovered to keep stock prices high.^[7] The IRH therefore implies that market reactions to negative business results can be reduced by providing information that is costly to analyze and interpret.^[8]

One possible way of obscuring corporate information when results are negative, or of being forthcoming in disclosing information when results are good, is to adjust the reports’ readability which can influence understandability as a consequence.^[9] There is a broad set of readability formulas that allow defining a level of text difficulty based on linguistic factors like sentence length as well as word length and characteristics.^[10] Careful application of these formulas allows providing an estimation of the probable success a target group will have in reading and understanding a certain text.^[11] Various studies with different variables and samples have already shown that there are tendencies that weak financial performance induces poor readability of a report.^[12] Still, it is doubtful whether reports are written to be difficult on purpose or it is just more demanding to explain why losses have occurred.^[13] Moreover, numerous studies clearly disprove any significant relationship between performance and readability.^[14] Additionally, the influence of a company’s shareholder structure and the resulting dependence or independence of the management board on its financial report’s readability is a topic that has not been examined adequately.

Most of the research conducted in the field of readability focuses on reports in English either as first or second language. Little attention has been paid to German reports. Besides of that it has always been the annual report being used as sample basis. Nonetheless, reports that are published throughout a business year, quarterly reports, should not be underestimated.

Therefore, the concrete aim of this study is to focus on the readability of letters to the shareholders of bilingual (German and English) quarterly reports of listed companies at Frankfurt Stock Exchange.

The underlying research question of this work shall be formulated as follows:

“What influences readability of letters to the shareholders of quarterly reports of publicly listed companies on the Frankfurt Stock Exchange?”

This study contributes to existing literature in various ways. On the one hand, it is among the very few works that consider the influence of performance on the readability of German reports and, in addition, sets the focus on quarterly reports. On the other hand, it takes into account the dependence or independence of managers of their shareholders measured by the company’s shareholder structure, industry, share price volatility and the size of firms. Moreover, it is examined whether there is a tendency that reports in different languages tend to have the same relative level of difficulty. In general, the study also examines and questions research methods of prior studies in the field of report readability.

1.2 Structure

Chapter 2 gives the reader background on reporting habits and typical elements of content. The current legal and practical reporting situation at Frankfurt Stock Exchange is then discussed to become aware of what can actually be expected from published reports. Moreover, the Efficient Markets Hypothesis in contrast to the Incomplete Revelation Hypothesis is discussed to provide a theoretical basis why managers could offer (in)complete information in reports.

In Chapter 3 the stakeholder theory is reviewed to gain understanding to whom company reports are addressed. Moreover, the status of shareholders is examined drawing on principal-agent theory and focusing on the question of how dependent or independent managers are from shareholders. The concept of obfuscation in reports is also discussed.

Chapter 4 explores the theoretical foundations of the empirical study and deeply reviews and compares existing literature and studies conducted in the field of readability. A special focus is set on prior research on readability of (annual) reports.

Then, in Chapter 5 , the hypotheses are formulated and the research design for the empirical part is discussed. This also includes the presentation of the sample and the utilized variables. Moreover, the relevant statistical research methods are described.

Finally, in Chapter 6, various statistical analyses are conducted. The empirical results are then described, interpreted and compared with previous research. Besides, potential further fields of research are discussed. Moreover, the limitations of sample and methodology are discussed and evaluated.

In Chapter 7 , a final conclusion is provided.

2. Reporting on Financial Markets

2.1 Introduction

Business participants, investors and all other types of stakeholders are familiar with the concept of annual reports. It is essentially “a financial and narrative report” of the firms’ business activities during the previous year. The motivation to publish those reports is not solely based on a legal obligation since they also offer a suitable means for stakeholder communication.^[15] Moreover, depending on local legislation, many listed companies are required to issue quarterly reports. These reports are often unaudited, less extensive and have very different levels of comprehensiveness. They are mainly designed to update existing, not prospective, shareholders.^[16]

The following chapters elaborate on the functions, content and importance of both annual and quarterly reports. For reasons of simplicity and consistency with already existing literature, the term reports is used instead of annual and quarterly reports throughout this work with the exception of those sections where one type is meant specifically.

2.2 Purpose and Functions of Reports

Courtis provides a comprehensible definition of what (annual) reports are:^[17]

“[…] a formal communication document comprising quantitative information, narratives, photographs and graphs. It seeks to inform shareholders, creditors and others […]. It is essentially a response to mandatory disclosure requirements […]. It is also a medium for voluntary disclosures perceived to produce net corporate benefits.”

One can conclude from this quote that reports cannot be reduced to their sole existence due to legal obligations, but should be seen as a means of communication with stakeholders that allows assessing the risk-return ratio of an asset prior to an investment decision.^[18]

Abstractly spoken, reports can be described as a “costly process of information gathering and processing” .^[19] Broken down, reports merely serve the following functions:^[20]

- Information
- Self information
- Information of shareholders, creditors and other stakeholders
- Limited due to various measurement and assessment regulations
- Limited due to obfuscation^[21]

- Protection of creditors
- Business risk of unsuccessful operation^[22]
- Risk of being ranked second after shareholders in terms of payments^[23]

- Justification
- Based on separation of ownership and control^[24]
- Documentation and evidence

- Determination of income
- Based on balance sheet and income statement
- Can be basis for tax calculation
- Highly dependent on legal codes

- Assessment of payments (e.g. dividends)
- Influenced by determined income and dividend policy
- Regulated by capital maintenance rules

2.3 Typical Content of Reports

Jurisdiction usually requires companies to publish reports that contain at least a balance sheet and an income statement. Moreover, depending on specific law, a broad range of extensions and explanations is required.^[25]

The factors influencing the concrete arrangement and content of reports are various. Typically, legal obligations, country, industry, firm size, habits and other factors are seen to play a crucial role. A rough separation of content into business reporting and financial reporting can be found in most reports.^[26]

illustration not visible in this excerpt

2.3.1 Letter to the Shareholders

Since the letters to the shareholders form the sample basis for the present study, common characteristics of these letters are discussed in more detail in the following.

Typically, the letter to the shareholders^[27] is said to be the “most widely read and understood section of the report.”^[28] Correspondingly, (potential) investors also base their investment decisions to a considerable extent on the content of the letters to the shareholders. The predictive strength of these letters concerning future results and possible failures has already been shown by studies. Analysts’ forecasts and share prices are also affected by how letters to the shareholders are composed.^[29] Critical voices see them as uncovered advertising measure.^[30] McConnell et al. even describe them as “documents with little, if any, substantive content” .^[31]

Since publishing a letter to the shareholder is not mandatory, the concrete arrangement of content depends on the individual firm. Nevertheless, elements like introduction, revenues, finance and future outlook seem to be present frequently. Consequently, there exists a broad range of recommendations how to structure letters to the shareholders. One example is given below:^[32]

- Introduction

- Sales and earnings

- Sales, revenues and earnings results plus comparisons and explanations

- Financial health

- Discussion of balance sheet, cash flow and capital structure

- Explanation of changes

- Dividends, stock splits, share repurchases etc.

- Achievements

- Progress made on strategic goals

- Link of operating performance and future results

- Changes in management or board of directors

- Expression of appreciation

- Outlook

- Future goals

- Industry trends and market expectations

- Closing

- Acknowledges for stakeholders

- Date and signatures

The structure recommended by The Heights is now applied to a randomly selected letter:

illustration not visible in this excerpt

Figure 2: Typical content of a letter to the shareholders (Netflix, Inc. 2009, p.1. and The Height)

Letters to the shareholders are mostly not audited and not verified against the data that they refer to. Therefore, they can also be used as an instrument of impression management. Good results are usually described as the result of good internal strategy and investment behavior, whereas negative results are often attributed to external economic factors or other unforeseen events.^[33] Abrahamson and Amir speak of the “relative freedom in choosing the information reported in the president’s letter and the lack of restrictions on the way the information is reported” .^[34]

2.4 Specifics of Reports at Frankfurt Stock Exchange Prime Market

2.4.1 The Frankfurt Stock Exchange

The very early history of Frankfurt Stock Exchange (FSE) dates back until 1150. The year 1585 is widely known as the natal hour FSE.^[35]

Nowadays FSE is an important player in the world’s financial markets. The traded securities are divided into three main levels of transparency, so-called segments :^[36]

- Prime Market – A market regulated by the European Union that is mainly intended for companies that position themselves against an international investor base. Companies in this segment have to fulfill high requirements of transparency such as:

- Quarterly reporting in German and English
- IFRS/IAS or US-GAAP
- Publication of a company calendar
- At least one analysts’ conference per year
- Ad-hoc disclosures in German and English

The Prime Market of FSE forms the basis of the sample that is used throughout the empirical part of this study.

- General Standards – A market regulated by the European Union that follows the minimum legal requirements. The requirements of transparency in this segment are lower and cover:

- Ad-hoc disclosures
- IFRS/IAS or US-GAAP
- Publication of one interim report

- Open Market – The majority of traded securities in this segment are non-German. The market is not regulated by the European Union but by the stock exchange (regulated unofficial market).

- Entry Standard – A market that offers an alternative capital allocation for small and medium sized companies. It is not regulated by European Union transparency law but by the stock exchange. It forms part of the Open Market but has higher standards of transparency.

2.4.2 Annual Reports at Frankfurt Stock Exchange Prime Market

Section IV./2. of the directives of FSE regulates the publication of a Jahresfinanzbericht of companies listed on the Prime Market. § 50 (1) thereby refers to § 37v (2) WpHG which determines the minimum requirements of such a report:^[37]

- An audited annual report according to national law of the company’s country of residence
- Management report
- Declaration according to German Commercial Code (HGB) that the report has been prepared to the best of knowledge and judgment (so-called Bilanzeid )
- Proof on the entitlement of the auditor

The directive of FSE requires publishing both in German and in English. Companies with residence outside of Germany are allowed to provide annual reports solely in English. The reports have to be submitted four months after the respective period at the latest.^[38] No regulatory obligation exists that claims the publication of a letter to the shareholders.

In general, all other regulatory laws remain unaffected. Exceptions granted by the Federal Financial Supervisory Authority (Bundesanstalt für Finanzdienstleistungsaufsicht) are valid, unless they contradict with basic ideas ( “tragende Gedanken” ) of the Prime Market.^[39]

2.4.3 Quarterly Reports at Frankfurt Stock Exchange Prime Market

The directives of FSE regulate the publication of quarterly reports in § 51. Semi-annual reports (end of second quarter) underlie separate rules. Companies have to deliver quarterly reports that, according to §37w, comprise shortened financial statements and an interim management report.^[40]

Quarterly reports must date at the end of the first and third quarters and have to be submitted within two months (three months if the residence lies outside of the European Economic Area). Regarding the language, the same rules apply as for annual reports.^[41]

Quarterly reports can be audited on a voluntary basis. Different from annual and semi-annual reports, the auditors can be assigned by the management itself.^[42]

2.5 Market Functioning

2.5.1 Introduction

Reports, as the primary means of communication between firms and shareholders, are crucial when it comes to investment decisions and the functioning of the capital market. The elementary role of this market is the allocation of ownership and resources in terms of capital. Investors base their investment decisions on the information that is provided. Nevertheless, it has to be scrutinized whether markets are “efficient” and “fully reflect all available information”.^[43] Two hypotheses on market efficiency that are important for the course of this study are discussed below. Due to conflicting theoretical and empirical results, no conclusion on the accuracy of one model can be provided.^[44]

2.5.2 Efficient Markets Hypothesis

The Efficient Markets Hypothesis represents a prime area of research in capital market theory.^[45] The central statement of the hypothesis is that security markets are highly efficient in providing information about individual stocks and their sum.^[46]

Efficiency in this respect is subject to the following three conditions:^[47]

- No transaction costs are involved.
- All information available is obtainable without cost.
- An agreement exists that all information available is reflected by market prices.

As soon as new, market-relevant information arises, it is immediately reflected in the prices of securities. This happens by fast spreading of information without time delay. As a result, it is not possible to make advantage of buying or selling securities that are not priced correctly.^[48]

The hypothesis is historically based on the model of random walk which states that all price changes of securities are random in respect to their starting point. This also implies that future stock prices cannot be predicted since they are solely depending on information that is available in future. Past development also does not allow drawing conclusions about future expectations. Therefore, a completely inexperienced investor that buys stocks randomly would have the same rate of return as an expert.^[49]

Based on the works of Fama , the following classification of information efficiency for capital markets can be set up:^[50]

- Weak-form information efficient capital markets – current market prices reflect all information on past price developments (poor empirical validation)
- Semi-strong-form information efficient capital markets – current market prices reflect all current publicly available information (empirical validation)
- Strong-form information efficient capital markets – current market prices reflect all publicly available and private information (falsified)

Despite its popularity gained in the 1970s, the Efficient Markets Hypothesis nowadays underlies criticism among scholars.^[51] Having reviewed a broad range of previous studies, Heun concludes that the existence of strong-form information efficient capital markets has to be denied. For semi-strong and weak form information efficient markets the empirical validations point out the practical limitations of the theory. The semi-strong form is empirically best proven. Moreover, the model of random walk is empirically not validated.^[52]

Bloomfield summarizes “The EMH [Efficient Markets Hypothesis] has been highly influential among academics, but practitioners and regulators appear unconvinced.”^[53]

2.5.3 Incomplete Revelation Hypothesis

The Incomplete Revelation Hypothesis developed by Bloomfield is theoretically based on the noisy expectations model by Grossman and Stiglitz .^[54] Rational agents can freely decide to collect information on the value of an asset. If they decide to do so and then trade the asset, they are confronted with noise traders that trade randomly. These noise traders are the reason why market prices do not reveal information completely. For example, if a noise trader sells extensively, the resulting drop in market prices does not come from negative company news that is obtained by informed traders, but from the random behavior of noise traders. Consequently, the market heads to an equilibrium between traders that collect information (at cost) and traders that decide not to do so. Informed trading only results in limited gains since the amount of gain decreases, the more informed traders are active. Therefore, it depends on a cost-benefit ratio whether to search for and analyze information or not. Bloomfield , instead of referring to the number of information-collecting traders, speaks of trading interest . It is defined as the willingness to trade, if the market price deviates from the value that is identified based on available information and its statistics.^[55]

The practical link of this equilibrium is described by Bloomfield by referring to the difference between data and statistics:^[56]

“Data are ink spots on sheets of paper […] Statistics are the useful facts extracted from that data, such as earnings figures and financial ratios. While public data are often free, it takes time and effort to extract statistics […].”

The Incomplete Revelation Hypothesis subsequently states that the more costly it is to extract statistics about securities on capital markets, the less those statistics induce price changes. It is argued that information extraction costs decrease trading interest. Low trading interest then leads to a lower revelation in market prices. In contrast, the Efficient Markets Hypothesis does not consider the reaction of prices based on the extraction costs of statistics due to its disregard of transaction costs.^[57] Concerning share price volatility, the Incomplete Revelation Hypothesis does not yet provide an explanatory problem-solving approach.

Seen globally, Bloomfield’s position reflects an intermediary position between supporters and critics of the Efficient Markets Hypothesis. On the one hand, inefficiencies are tolerated; on the other hand, market players act rationally.^[58] Most importantly, the model considers the effect of demand and supply of financial information on prices.^[59]

Practical implications of the Incomplete Revelation Hypothesis on reporting mentioned by Bloomfield target managers’ desire to keep stock prices at a high level or increase them. Therefore, they partially make decisions that reduce the ease of uncovering and analyzing information which mitigate the influence on share prices. Typical decisions would be:^[60]

- Promotion of accounting methods that improve popular statistics like earnings-per-share and debt-equity ratio and ensconce expenses and liabilities in footnotes
- Presentation of extraordinary gains as operating gains and ongoing expenses as extraordinary expenses
- Announcement of pro forma performance figures that are compared to strategically chosen previous figures
- Generation of hidden reserves to keep up positive results in tensed times

Essential for this study is the obfuscation of information by adjusting the readability of corporate disclosures. The obfuscation hypothesis is discussed in chapter 3.5.2.

3.Stakeholder Theory

3.1 Historical Origins

The modern interpretation of the term stakeholder dates back until 1963 when it was used internally at the Stanford Research Institute (SRI) and referred to the fact that shareholders are not the only interest group of companies.^[61] Stakeholders were seen as “those groups without whose support the organization would cease to exist” by the SRI.^[62]

Notwithstanding the above, the topic itself was already considered earlier. Actually, already Adam Smith (1759 and 1776) provides a framework of considering the necessity of a coalition between economic and ethical interests. According to Smith , this symbiosis results in the best functioning of society.^[63] Interestingly, Smith is sometimes misunderstood as a pure proponent of shareholder value based on his concept of the invisible hand ( The Wealth of Nations ). Yet this is just one side of the coin of Smith’s work. Especially in the light of his work Theory of Moral Sentiments , which sets a focus on morality, justice and beneficence, this strict point of view has to be denied.^[64]

Wentges names Berle and Means (1932), Dodd (1932), Barnard (1938) and March and Simon (1958) as scholars of the 20th century that provide essential conceptual input for later work. The key message in their works is the need for balancing the interests of various groups.^[65]

Nevertheless, it is Freeman in the 1980s who initiates contemporary discussion and research amongst academics and the link to strategic management.^[66]

3.2 Stakeholder Definition

Besides the early definition from 1963 mentioned above, the term stakeholder is often depicted in various other ways. The list of different definitions is nearly endless and leads to unnecessary confusion.^[67]

Freeman and Reed (1983) propose to separately define stakeholders in a wider and narrower sense:^[68]

“Any identifiable group or individual who can affect the achievement of an organization’s objectives or who is affected by the achievements of an organization’s objective.” (definition in the wide sense)

“Any identifiable group or individual on which the organization is dependent for its continued survival.” (definition in the narrow sense)

Freeman (1984) states a very similar, often cited, definition shortly thereafter:^[69]

“A stakeholder in an organization is (by definition) any group or individual who can affect or is affected by the achievement of the organization’s objectives.”

Clarkson (1995) draws a separating line between primary and secondary stakeholders and defines as follows:^[70]

“ A primary stakeholder group is one without whose continuing participation the corporation cannot survive as a going concern. Primary stakeholder groups typically are comprised of shareholders and investors, employees, customers, and suppliers […]”

“Secondary stakeholder groups are defined as those who influence or affect, or are influenced or affected by, the corporation, but they are not engaged in transactions with the corporation and are not essential for its survival.”

Post et al. (2002) state:^[71]

“The stakeholders in a corporation are the individuals and constituencies that contribute, either voluntarily or involuntarily, to its wealth-creating capacity and activities, and that are therefore its potential beneficiaries and/or risk bearers.”

Along with the definitions of stakeholders mentioned above, one can set up a list of potential resources that can be at stake for the stakeholders:^[72]

- Capital resources – equity, liabilities, real estates, workforce, natural resources
- Trust – social acceptance and good working environment
- Information and know-how

3.3 Stakeholder Groups

Despite the existence of both narrow and broad definitions and classifications of stakeholders, some groups tend to be involved by most of the definitions. Among them shareholders naturally occupy a special and central position.

illustration not visible in this excerpt

Figure 3: Typical stakeholders of a company (Parmar/Freeman et al. 2010, p.406.)

3.4 Stakeholder Concept

The stakeholder concept is a concept of analysis of strategic management. It is based on the insight that the majority of companies are no longer solely in the possession of a few individuals. Instead, they have to be interpreted as quasi-public institutions and form a place of exchange of conflicting and complementary interests of various internal and external stakeholders.^[73] These institutions form a coalition between stakeholders that have financial investments and those who do not. Stakeholders with no financial investment also play a crucial role in the link between corporate strategy and finance.^[74] By the pragmatic approach of focusing (solely) on the needs of their stakeholders, companies can be most effective in accomplishing their strategic objectives.^[75]

The key questions that are addressed by the stakeholder concept in connection with strategic management are:^[76]

- How can value be created in rapidly changing societies?
- What is the relation between capitalism and ethics?
- What is managers‘ mindset in the tradeoff and relation between profit and ethics?

3.5 Relationship Between Shareholders and Managers

3.5.1 Agency Theory

In the construct of a company, shareholders, managers and especially their cooperation occupy a central role in value creation. From a theoretical perspective the relationship between these two stakeholders can be well described with the agency theory^[77]. This theory ties in with the stakeholder theory by considering the situation when cooperating parties have different goals and delegation is involved.^[78] Consequently, the relationship between shareholders and managers represents a situation of “separation of ownership and control” that is based on contractual arrangements.^[79] The shareholder as the principal decides to delegate managerial tasks to his agent, namely the manager.^[80] Nevertheless, agency theory must not be reduced to shareholders and managers. It has to be interpreted in a broad context and encompasses relationships that are based on delegation and representation:^[81]

“Essentially all contractual arrangements, as between employer and employee or the state and the governed, for example, contain important elements of agency.”

The aim of agency theory is to resolve two main problems that can arise due to the existence of information asymmetry and conflictual, opportunistic behaviour. These can be conflicting interests between principal and agent as well as the difficulty and expense to screen the actions of the agent on the one hand. On the other hand, the two parties’ risk preferences need not to be equal which in turn leads to divergent preference actions.^[82] Due to the fact that both attempt to maximize their utility, it is reasonable to assume that the agent does not always act in the very best interest of his principal. By using incentive measures and monitoring, these divergences can be partly reduced. Nevertheless, this induces costs and does not align all actions of the agent.^[83] The total costs of agency comprise of:^[84]

- Monitoring expenditures – controlling the agent’s behavior through rules, policies and restrictions
- Bonding expenditures – contractual costs of the agent
- Residual loss – costs that arise from the difference between the agent’s decision and the decision that would maximize the principal’s welfare (opportunity costs)

3.5.2 Obfuscation Hypothesis

Based on the concept of information asymmetry between managers and investors, the obfuscation hypothesis describes a situation where the former underscore their successes and obfuscate their failures. The hypothesis finds a special application in readability research and questions whether disclosures with good results are written more readably than disclosures with negative results.^[85] Obfuscation “obscures the intended message, or confuses, distracts or perplexes readers, leaving them bewildered or muddled.”^[86] Consequently, reports with poor readability decrease investors’ motivation to dig deeper and investigate more precisely.^[87] Intentional obfuscation typically arises through the usage of:^[88]

- Esoteric or obscurantist vocabulary
- Jargon
- Irrelevant information
- Complex and long sentences
- High variability in reading ease
- Overlapping and spurious argumentation

3.5.3 Shareholder Influence and Independence

The degree of effectiveness of monitoring (expenditures) in agency theory depends on how much influence shareholders can exert on the management or how independent managers are from their principals. While this conclusion tends to be rather trivial, measuring or qualifying the degree of independence is certainly not.

A first practical attempt to solve this issue is made by the Bureau van Dijk with its BvDEP Independence Indicator. It aims “to characterise the degree of independence of a company with regard to its shareholders.”^[89]

The indicator is solely based on information about ownership structure and therefore does not question how shareholders are organized or which alliances exist between them. Small-scaled investors^[90], which are typically stated aggregately, are not considered in the calculation since they are assumed to have no effective controlling power over a company.^[91]

To fully understand the definitions of the main groups of independence, it is necessary to get acquainted with the different types of ownership that are used by the indicator:^[92]

- Direct ownership – A owns a certain percentage of B
- Indirect ownership – A owns B through another entity
- Total ownership – A owns of B through direct and indirect ownership
- Calculated total ownership (control) – Concept which follows the IFRS concept of control “and calculates a total percentage by making the sum of the direct percentages of all minority direct shareholders controlled by the same controlling company.”

This independence indicator knows five main levels of independence ranging from “A” to “D” with “U” reserved for unknown levels of independence. “A” refers to situations where managers are relatively independent from shareholders and can expect little influence, whereas “D” represents a situation of low independence from shareholders.^[93]

illustration not visible in this excerpt

Table 1: BvDEP Independence Indicator classification (Bureau van Dijk 2008, pp.15.)

4. Readability Research

4.1 Concept of Readability

The concept of readability has been described by a number of authors. Dale and Chall characterize readability as:

“[…] the sum total (including interactions) of all those elements within a given piece of printed material that affects the success, which a group of readers have with it. The success is the extent to which they understand it, read it at optimum speed, and find it interesting.”^[94]

In simple terms, readability describes the ease with which a text can be read and understood.^[95] The underlying premise that readability also reflects understandability has substantiated a broad range of studies. Some researches even treat readability and understandability as synonyms.^[96] Others argue in favor of a clear linguistic separation.^[97] Smith and Taffler mark that they are “[…] different concepts [...] and that synonymous treatment […] is not justified.”^[98] It is generally said that readability focuses on the problem of matching text and reader and therefore explains the effectiveness of written communication.^[99]

According to Dale and Chall, the variables that influence the readability of a passage are an accumulation of comprehension, interest and fluency. Comprehension describes the reader’s capability to understand, while interest refers to the level of motivation to continue reading a certain text. Fluency determines the capacity to read at the most favorable speed.^[100] Reader’s comprehension in this respect is defined trough elements like content, style, format and organization.^[101]

An alternative, more qualitative definition of readability is provided by Lesikar. If one can understand the intended message of a text on the first reading attempt, the text is said to be readable. Besides of that, the effort of the reader is minimal and the intended message reaches the reader free of error.^[102]

Still, neither the definitions by Dale and Chall nor by Lesikar provide clear characteristics that allow differentiating between a readable and an unreadable text.

A comprehensive overview of what potentially influences reading difficulty is given by Oakland and Lane:

illustration not visible in this excerpt

Figure 4: Factors that influence reading difficulty (Oakland/Lane 2004, p.248.)

4.2 Measures of Readability

4.2.1 Introduction

"[…] readability formulas do not define readability. They measure factors which can correlate with difficulty […].”^[103]

Transforming a text into a single indicator that describes its readability is a challenging task which requires simplification and concentration on a few characteristics. It depends on the underlying formula’s ability to consider elements of writing that are related to comprehension (content, style, format and organization) how successful and useful a given result is. In this respect it is only the style element that can be measured in a practical way.^[104] One should always keep in mind Robert Gunning’s quote:^[105]

“No formula will guarantee that you write well. Nonsense written simply is still nonsense.”

Typically, formulas that consider two text-related variables of style are used: The semantic variable word length, for example, is regarded as an indicator of the speed of recognition during the reading process. Sentence length, a syntactic variable, is linked to the memory span of the reader. Although it is sentence complexity that induces reading difficulty, sentence length may be used as a proxy. However, for both word and sentence length it is important to consider that they do not cause difficulty or readability, but solely represent good indices.^[106]

Most of the readability measures attempt to define the minimum reading age to understand a certain text or the difficulty that is typical for a certain level of education. Some are originally also intended to be used for children.^[107] Today the list of readability formulas is nearly endless. Already by 1982, Kare counted more than 200 of them.^[108] An overview of the most influential measures that can be applied for adults is provided below.

4.2.2 Historical Overview of Readability Research

Around 2,900 years ago the Talmudists counted both the number of word appearance as well as individual ideas in their texts. Their intention was to compare the appearance of these factors in an unusual and usual sense.^[109]

By the end of the 19th century the Russian scientist Nikolai Aleksandrovich Rubakin pursued a word frequency study based on thousands of texts written by soldiers, artisans and agriculturists. He finally drew up a list of 1,500 words that seemed easy enough to be understood by the majority of the population. Rubakin concluded that the main reasons why people cannot read texts properly are the usage of uncommon vocabulary and long sentences.^[110]

Sherman’s publication about sentence study of English literature from 1893 is very influential for later readability formulas. He notices the effect of progressive shortening of sentences throughout the history of literature. Early writers like Geoffrey Chaucer (14th century) write with an average sentence length of 50 words, whereas William Ellery Channing (19th century) has an average sentence length of around 24 words. His equation of sentence length and readability, which was also confirmed by later studies, is of key importance for the usage of sentence length in most readability formulas that emerged later on. Moreover, Sherman discovers that some writers have a quite constant length of sentences which supports the trend to use samples instead of the entire work. Most importantly Sherman concludes that it is essential to recognize the reader and his abilities during the process of writing. Many studies that are later conducted agree with this notion.^[111]

4.2.3 Gray and Leary’s Readability Study (1935)

Gray and Leary use experts’ opinion to gain insights which factors influence the readability of books. They conduct a survey with cooperation of librarians, teachers and publishers and come up with 228 potential characteristics of reading ease that can be grouped into content, style, format and organization. Since all basic elements cannot be measured except of style, they focus on 64 variables of style that can be counted reliably.^[112] To specify if the factors are connected to reading ease, the authors use a reading comprehension test that is pursued with around 1,000 participants. 17 out of this 64 variables show a significant correlation (>0.35) between variable change and perceived change in textual difficulty.^[113]

illustration not visible in this excerpt

Table 2: List of style factors affecting readability by Gray and Leary (Lekfi 1987, p.267.)

Aware of the fact that correlation coefficients between 0.35 and 0.52 are not extraordinarily high, Gray and Leary search for elements that are highly predictive but not related to each other to increase explanatory power. The finally come up with a formula that uses five of the above mentioned variables: numbers 1, 5, 8, 15 and 17. Adding another variable would only lead to a minimally increased accuracy but to more complexity and more difficult application.^[114]

It is not the formula built by Gray and Leare itself that highly influences later research, but their stimulation for other academics to find better formulas based on their data.^[115]

4.2.4 Dale-Chall Formula (1948)

The formula developed by Dale and Chall estimates readability of English texts in terms of grades of school based on a set of 3,000 words that they consider to be best known by 8-year-old children in the United States. This list is based on 30 rules that decide whether a word is familiar for a child or not.^[116] It is primarily seen as a list of words that have high correlation with reading difficulty, but does not represent a set of the most important words for children or adults.^[117] Moreover, the formula considers sentence length as the second factor of readability.^[118]

illustration not visible in this excerpt

Figure 5: Exemplary excerpt of words contained in Dale's list (DuBay 2004, p.85)

The formula itself is defined as follows:^[119]

Raw score = (0.1579 * percentage of unfamiliar words) + (0.0496 * average number of words per sentence)

The resulting raw score can be transferred to the corrected US grade levels by the help of the following table:

illustration not visible in this excerpt

Table 3: Dale-Chall formula correction table (DuBay 2004, p.81.)

[...]

^[1] Baker/Kare (1992), p.1.

^[2] Freeman (1984), p.46.

^[3] Courtis (1995), p.4.

^[4] Subramanian et al. (1993), p.49.

^[5] Courtis (1995), p.4.

^[6] Bloomfield (2002), pp.233.

^[7] Bloomfield (2002), p.238.

^[8] Bloomfield (2008), p.249.

^[9] Li (2008), p.221.

^[10] Tekfi (1987), p.262.

^[11] Courtis (1986), p.285.

^[12] Adelberg (1979), Jones (1988), Baker and Kare (1992), Subramanian et al. (1993) and Li (2008)

^[13] Bloomfield (2008), p.249.

^[14] Courtis (1986), Rutherford (2003) and Smith et al. (2006)

^[15] Stittle (2003), pp.3.

^[16] Downes/Goodman (2003), p.118.

^[17] Courtis (1995), p.4.

^[18] Courtis (1995), p.4.

^[19] Strieder (2003), p.34.

^[20] Strieder (2003), pp.35.

^[21] Bloomfield (2008), p.249.

^[22] Elliot/Elliot (2006), p.127.

^[23] Elliot/Elliot (2006), p.127.

^[24] Korn (1997), p.63.

^[25] Strieder (2003), p.25.

^[26] Stolowy/Lebas (2006), p.103.

^[27] often referred to as “letter from the president/CEO”, “letter to the shareholder”, “chairman’s statement/address” etc.

^[28] Balata/Breton (2005), p.5.

^[29] Hooghiemstra (2010), pp.2.

^[30] Smith/Taffler (2000), p.624.

^[31] McConnel et al. (1986), p.66.

^[32] The Heights (2013), slightly adapted

^[33] Hooghiemstra (2010) , p.2.

^[34] Abrahamson/Amir (1996), p.1159.

^[35] Website Frankfurter Wertpapierbörse (2013)

^[36] Website Frankfurter Wertpapierbörse (2013)

^[37] Börsenordnung für die Frankfurter Wertpapierbörse (2013), pp.36. and WpHG (2013)

^[38] Börsenordnung für die Frankfurter Wertpapierbörse (2013), p.37.

^[39] Börsenordnung für die Frankfurter Wertpapierbörse (2013), p.37.

^[40] Börsenordnung für die Frankfurter Wertpapierbörse (2013), p.37.

^[41] Börsenordnung für die Frankfurter Wertpapierbörse (2013), pp.37.

^[42] Börsenordnung für die Frankfurter Wertpapierbörse (2013), p.38. and KPMG (2013)

^[43] Fama (1970), p.383.

^[44] Heiden (2006), p.54.

^[45] Heun (2007), p.93.

^[46] Malkiel (2003), p.59.

^[47] Fama (1970), p.387.

^[48] Malkiel (2003), p.59.

^[49] Malkiel (2003), pp.59.

^[50] Fama (1970), p.414., Shleifer (2000), p.6. and Scheufele/Haas (2008), p.27.

^[51] Shleifer (2000), pp.9.

^[52] Heun (2007), p.95.

^[53] Bloomfield (2002), p.233.

^[54] Bloomfield (2002), p.234. and Heiden (2006), p.56.

^[55] Bloomfield (2002), pp.234.

^[56] Bloomfield (2002), p234.

^[57] Bloomfield (2002), pp.235.

^[58] Heiden (2006), p.56.

^[59] Heiden (2006), p.58.

^[60] Bloomfield (2002), p.238.

^[61] Parmar/Freeman et al. (2010), p.405.

^[62] Freeman/Reed (1983), p.89.

^[63] Andriof et al. (2002), p.10.

^[64] Brown/Forster (2013), pp.301.

^[65] Wentges (2002), pp.88.

^[66] Parmar/Freeman et al. (2010), pp.405.

^[67] Miles (2012), p.285.

^[68] Freeman/Reed (1983), p.91.

^[69] Freeman (1984), p.46.

^[70] Clarkson (1995), pp.106.

^[71] Post et al. (2002), p.19.

^[72] Figge/Schaltegger (2000), p.11.

^[73] Figge/Schaltegger (2000), p.11.

^[74] Cornell/Shapiro (1987), p.5.

^[75] Freeman (1999), p.234.

^[76] Parmar/Freeman et al. (2010), pp.405.

^[77] often referred to as “principal-agent problem”

^[78] Jensen/Meckling (1976), p.308.

^[79] Jensen/Meckling (1976), p.309.

^[80] Jensen/Mecking (1976), p.310.

^[81] Ross (1973), p.134.

^[82] Eisenhardt (1989), pp.58.

^[83] Jensen/Meckling (1976), p.308.

^[84] Jensen/Meckling (1976), p.308.

^[85] Courtis (1998), pp.461.

^[86] Courtis (2004), p.292.

^[87] Courtis (1998), pp.461.

^[88] Courtis (2004), p.292.

^[89] Bureau van Dijk (2008), p.15.

^[90] often referred to as “public investors”, “private investors”, “other individuals” or “unnamed shareholders”

^[91] Bureau van Dijk (2008), p.15.

^[92] Bureau van Dijk (2008), p.6.

^[93] Bureau van Dijk (2008), p.15.

^[94] Dale/Chall (1949), p.23.

^[95] Oakland/Lane (2004), p.244.

^[96] Jones (1988), pp.297.

^[97] Smith/Taffler (1992), pp.84.

^[98] Smith/Taffler (1992), p.91.

^[99] Lewis et al. (1986), pp.199.

^[100] Dale/Chall (1948), pp.13.

^[101] Courtis (1986), p.285.

^[102] Mohamad/Rahman (2006), p.35.

^[103] Dreyer (1984), p.335.

^[104] Courtis (1986), p.285.

^[105] Gunning (1969), p.12.

^[106] Courtis (1986), pp.285.

^[107] Tekfi (1987), p.267.

^[108] DuBay (2004), p.42. Crossley et al. (2007) only count more than 50

^[109] Tekfi (1987), p.263.

^[110] Tekfi (1987), p.263.

^[111] Tekfi (1987), p.264.

^[112] DuBay (2004), pp.40.

^[113] Tekfi (1987), p.267.

^[114] DuBay (2004), p.42.

^[115] DuBay (2004), p.42.

^[116] Lewis et al. (1986), p.200.

^[117] DuBay (2004), p.84.

^[118] Lewis et al. (1986), p.200.

^[119] Lewis et al. (1986), p.200.