by Dirk Beerbaum

July 2015

Submitted for the degree of Doctor of Business Administration

Faculty of Business, Economic and Law

Surrey Business School

© Dirk Beerbaum

Copyright © 2015 Dirk Beerbaum

Books on Demand GmbH

ISBN: 978-3-7412-4345-5

I Table of Contents

• II Abstract
• III Acknowledgements
• IV List of Figures
• V List of Tables
• VI Abbreviations
• 1 Problem statement and purpose
  • 1.1 Increased implication of Corporate Governance
  • 1.2 Increased relevance and complexity of Corporate Governance Reporting
  • 1.3 Corporate Governance Reporting for Financial Institutions
  • 1.4 Framework of research
  • 1.5 Research objectives
  • 1.6 Overview of the structure of the thesis
• 2 Literature Review
  • 2.1 Definitions and scope
    • 2.1.1 Corporate Governance
    • 2.1.2 Corporate Governance Reporting
    • 2.1.3 XBRL and Taxonomy
  • 2.2 Theories on Corporate Governance
    • 2.2.1 Principal-Agent theory
    • 2.2.2 Transaction-cost theory
    • 2.2.3 Stewardship theory
    • 2.2.4 Property-Rights theory
    • 2.2.5 Stakeholder theory
    • 2.2.6 Institutional theory
  • 2.3 Theories on Corporate Governance Reporting
    • 2.3.1 Full disclosure theorem
    • 2.3.2 Discretionary disclosure theorem
    • 2.3.3 Quality of disclosures and taxonomies
    • 2.3.4 Relationship between Corporate Governance and Corporate Reporting
  • 2.4 Empirical studies on Corporate Governance
    • 2.4.1 First Generation Studies
    • 2.4.2 Second Generation Studies
    • 2.4.3 Third-Generation Studies
  • 2.5 XBRL and taxonomy
    • 2.5.1 Theories about XBRL adoption
    • 2.5.2 XBRL impacts and issues
    • 2.5.3 Selected empirical studies about XBRL
    • 2.5.4 XBRL for Corporate Governance Reporting
    • 2.5.5 Geographical diffusion of XBRL
    • 2.5.6 XBRL specification requirements
    • 2.5.7 Catalogue of corporate reporting taxonomies
    • 2.5.8 Technical considerations for building up an XBRL taxonomy
  • 2.6 Conceptual Framework
  • 2.7 Research questions and objectives
• 3 Research Methodology and Design
  • 3.1 Philosophical underpinnings
  • 3.2 Research methods
    • 3.2.1 Deduction versus Induction
    • 3.2.2 Mono versus Mixed-Methods
  • 3.3 Research design and strategy
    • 3.3.1 Research approach
    • 3.3.2 Research paradigm
    • 3.3.3 Research methodology
    • 3.3.4 Research process
    • 3.3.5 Research variables
    • 3.3.6 Empirical content analysis
    • 3.3.7 Determining the number of factors
    • 3.3.8 Data sampling
  • 3.4 Verification of the Methodology
    • 3.4.1 Reliability
    • 3.4.2 Validity
    • 3.4.3 Generalizability
• 4 Study results
• 4.1 Overview of the results
  • 4.1.1 Unit of analysis
  • 4.1.2 Category and sub-category development
  • 4.1.3 Test of Intercoder Reliability
  • 4.1.4 Test of compliance
  • 4.1.5 Correlations of main categories
  • 4.1.6 Correspondence Analysis
• 4.2 Inductive development of categories
  • 4.2.1 Results of pre-study
  • 4.2.2 Extraction of Corporate Governance Disclosures
  • 4.3 Integration of qualitative and quantitative results
  • 4.3.1 Taxonomy
  • 4.3.2 Validation of XBRL Taxonomy
• 5 Discussion of the findings
  • 5.1 Corporate Governance Reporting taxonomy - summarized findings
  • 5.2 Corporate Governance Reporting according to each main category
    • 5.2.1 Main Category 1: Internal Governance/Board organization
    • 5.2.2 Main Category 2: Shareholder Protection
    • 5.2.3 Main Category 3: Audit and Accounting
    • 5.2.4 Main Category 4: External Governance
    • 5.2.5 Main Category 5: Compliance
• 6 Conclusions and direction for future research
  • 6.1 Main Findings
  • 6.2 Contributions of this research
    • 6.2.1 Empirical contribution
    • 6.2.2 Theoretical contribution
    • 6.2.3 Practical contributions
    • 6.2.4 Contribution to regulation
  • 6.3 Limitations and directions for future research
    • 6.3.1 Impact of XBRL on content of financial information with regard to IFRS filers
    • 6.3.2 Better data, better decisions
    • 6.3.3 Global Taxonomy
• 7 Reflective diary
  • 7.1 Learning the Basics
  • 7.2 Starting the DBA Thesis
  • 7.3 Research and Writing
• 8 Appendices
• APPENDIX A: “Guidance on good practices in corporate governance disclosure”
• APPENDIX B: Codebook
• APPENDIX C: Factor Extraction
• APPENDIX D: Results of Latent Semantic Analysis
• APPENDIX E: Project Plan
• 9 REFERENCES

I Declaration of Originality

This thesis and the work to which it refers are the results of my own efforts. Any ideas, data, images or text resulting from the work of others (whether published or unpublished) are fully identified as such within the work and attributed to their originator in the text, bibliography or in footnotes. This thesis has not been submitted in whole or in part for any other academic degree or professional qualification. I agree that the University has the right to submit my work to the plagiarism detection service Turnitin UK for originality checks. Whether or not drafts have been so assessed, the University reserves the right to require an electronic version of the final document (as submitted) for assessment as above.

Dirk Beerbaum

Dirk Beerbaum, July 2015

II Abstract

Corporate Governance failures of large corporations such as Lehman Brothers have evoked a plethora of calls for changes in Corporate Governance Principles. Despite the fact that after the financial crisis, a number of transnational institutions have called for wide-scale changes related to Corporate Governance Principles by further protecting shareholders’ rights, it has been proven that theoretical extension and reform of Corporate Governance Principles is not enough, because it does not ensure successful implementation. According to a report from Isaksson (2009), corporate disclosures were not in line with OECD principles on good Corporate Governance. Moreover, according to Isaksson (2009), the financial crisis can also be traced back to incomplete Corporate Governance disclosures.

One major implementation topic involves Corporate Governance Reporting, which is very complex for multinational companies, as it involves statutory compliance with multiple rules and laws, coping with heterogeneous “Systems of Corporate Governance” (Weimer and Pape, 1999; Leuz et al., 2003; Tylecote and Visintin, 2007) and additionally the compilation of codes (Cadbury, 1993) and industry-specific standards (Mach et al., 2006; Bebchuk and Spamann, 2010). Extensible Business Reporting Language (XBRL) is assumed as a benchmark related to interactive business disclosures (Chen and Sun, 2009; Debreceny et al., 2010; Alles and Piechocki, 2012). Does XBRL have the potential to help to reduce the complexity of companies’ Corporate Governance Reporting (Alles and Piechocki, 2012)? In the academic literature, there are many articles which conclude that it is possible for XBRL to enhance transparency and improve Corporate Governance for financial reporting (Abdullah et al., 2009; Roohani et al., 2010; Alles and Piechocki, 2012; Müller-Wickop et al., 2013): therefore, an additional question is whether the application of XBRL to the non-financial reporting of Corporate Governance can also contribute to enhanced transparency.

XBRL will become mandatory based on the legal requirement of the Security and Exchange Commission (SEC) for Foreign Private Issuers (FPI) that are listed on the NYSE (Debreceny et al., 2010). Investors demand the extension of XBRL-enabled reporting to other financial reports, including CGR (Arnold et al., 2012), because within the peer group of listed Foreign Private Issuers on the New York Stock Exchange, comprising Financial Institutions, there is no adequate Corporate Governance framework to reflect Corporate Governance Reporting. Therefore, in this doctoral thesis, a taxonomy on Corporate Governance will be developed which can be put into practice for Corporate Governance Reporting.

III Acknowledgements

This thesis is grounded on my personal research and many years of experience as Vice President of Corporate Disclosures working for several banks in Germany and as a consultant in the area of Finance Transformation, Disclosure Management and Business Process Reengineering.

One of the main triggers for this thesis was my direct experience of the financial crisis, as it showed that many principles and assumptions that were believed to be “carved in stone” were challenged overnight. Investors lost confidence as Corporate Governance turned out to be very poor in reality, although companies consistently declared in their disclosures that they followed good Corporate Governance guidelines.

“The foundation of any structure in Corporate Governance is disclosure”: this quote from Sir Adrian Cadbury (Iskander, 2000, p. vi) can be regarded as the guiding principle of this thesis, as it emphasizes the important role of disclosure and transparency for Corporate Governance (Iskander, 2000).

The finalization of this thesis could not have been realized without the backing, the encouragement and the comments from my supervisors Professor Jean Chen, University of Southampton, Dr Laura A. Costanzo, University of Surrey and Dr Dimitrios Gounopoulos, University of Sussex, to whom I am very thankful. Furthermore, I am also very thankful to my wife Nora Beerbaum and my family for their support and commitment during the research work.

IV List of Figures

Figure 1.1: Overview of the structure of the thesis

Figure 2.1: Components of financial reporting

Figure 2.2: Components of CGR

Figure 2.3: XBRL Taxonomy breakdown

Figure 2.4: CG theories

Figure 2.5: Classification of CG literature

Figure 2.6: XBRL Financial reporting framework

Figure 2.7: The effect of XBRL on the financial supply chain

Figure 2.8: Different determinants of Taxonomies

Figure 2.9: US Financial Reporting Taxonomy

Figure 2.10: Synopsis of CG theories

Figure 3.1: Overview of deductive and inductive research methods

Figure 3.2: Exploratory design - Taxonomy development model

Figure 3.3: Deductive-inductive determination of category development

Figure 3.4: Research process

Figure 3.5: Country distribution of sample

Figure 3.6: Sample distribution- IFRS vs. US – GAAP

Figure 4.1: Boxplot total compliance score

Figure 4.2: ISAR CG Code and level of compliance

Figure 4.3: Scree plot

Figure 4.4: Taxonomy Matrix

Figure E.: Project Plan and Costings

V List of Tables

Table 2.1: Empirical CGR studies based on national CG codes

Table 2.2: Empirical CGR Studies based on international CG codes

Table 2.3: Overview of existing empirical studies about XBRL

Table 2.4: Overview of existing empirical studies about XBRL

Table 2.5: Overview of existing empirical studies about XBRL

Table 2.6: Research approach in previous empirical XBRL studies

Table 2.7: Country-related application and diffusion of XBRL

Table 2.7.1: Attributes of CG for CGR to ensure good CG based on theoretical prepositions

Table 2.7.2: Attributes of CG for CGR to ensure good CG based on theoretical Prepositions

Table 3.1: Decision model

Table 3.2: Four major types of mixed-method design

Table 3.3: Overall context between research objectives, hypotheses and the methods

Table 3.4: Research variables

Table 3.5: Sample selection for empirical study

Table 3.6: Market capitalization of largest financial institutions

Table 4.3.1: Codes of CG applicable for sample

Table 4.3.2: Codes of CG applicable for sample

Table 4.3.3: Codes of CG applicable for sample

Table 4.3.4: Codes of CG applicable for sample

Table 4.4: Intercoder reliability

Table 4.5: Compliance with ISAR

Table 4.6: Correlations of main categories

Table 4.7.1: CG Disclosures: CG Report, Item 16G and word frequency

Table 4.7.2: CG Disclosures: CG Report, Item 16G and word frequency

Table 4.8: Assessment of data suitability

Table 4.9: KMO and Bartlett’s Test

Table 4.10.1: Correlation matrix

Table 4.10.2: Correlation matrix

Table 4.10.3: Correlation matrix

Table 4.11.1: Communalities

Table 4.11.2: Communalities

Table 4.12: Results of principal component analysis

Table 4.13: Decomposition of CGR disclosures

Table 4.14.1: Code reporting elements

Table 4.14.2: Code reporting elements

Table 4.14.3: Code reporting elements

Table 4.14.4: Code reporting elements

Table 4.14.5: Code reporting elements

Table 4.14.6: Code reporting elements

Table 4.14.7: Code reporting elements

Table 4.14.8: Code reporting elements

Table 4.15.1: Development of main categories

Table 4.16: Clustering

Table 4.17: Results of integration of qualitative with quantitative research

Table 4.18.1.1: Taxonomy of CGR – Audit and Accounting

Table 4.18.1.2: Taxonomy of CGR – Audit and Accounting

Table 4.18.2.1: Taxonomy of CGR – Shareholder Protection

Table 4.18.2.2: Taxonomy of CGR – Shareholder Protection

Table 4.18.2.3: Taxonomy of CGR – Shareholder Protection

Table 4.18.2.4: Taxonomy of CGR – Shareholder Protection

Table 4.18.3.1: Taxonomy of CGR – External Governance

Table 4.18.3.2 Taxonomy of CGR – External Governance

Table 4.18.4.1: Taxonomy of CGR – Internal Governance

Table 4.18.4.2: Taxonomy of CGR – Internal Governance

Table 4.18.5: Taxonomy of CGR – Shareholder Protection

Table A.1: “Guidance on good practices in corporate governance disclosure”

Table B.1: Factor extraction without length adjusted

Table B.2: Results of LSA

VI Abbreviations

1 Problem statement and purpose

1.1 Increased implication of Corporate Governance

CG has developed into a dominant critical business issue in the last decade and has been controversially discussed in public policy debates and in the academic literature for various reasons. Although the issue of CG has been discussed since the early days of corporations, CG soon becomes a matter of public concern whenever giant corporate failures and scandals break out. Several such shocking high profile corporate scandals and the bankruptcy of several large companies in the western hemisphere at the beginning of the twenty-first century have stimulated public discussion on good governance (Nix and Chen, 2013).

Owing to these corporate accounting fraud scandals, particularly Enron in 2001 and WorldCom in 2002, CG reforms were enacted in several countries, constituting main new laws, regulations and guidelines: the US Sarbanes-Oxley Act (2002), the OECD Guidelines of CG and their implications for Non-OECD Countries (2004), the EU Transparative Directive (2005), the EU Transparency Law for Corporate Executives (2006) and the US NYSE CG Standards (2008).

When Lehman Brothers had to announce its bankruptcy on September 15, 2008, resulting from huge losses in the subprime mortgage market and fuelled by the global financial crisis, it became obvious that prior reforms to CG (Akinbami, 2010) had failed (Cioffi, 2010). Many academics and business professionals share the opinion that the turmoil required a re-examination of CG (Akinbami, 2010; Ramos et al., 2012; Fridson, 2013; McNulty et al., 2013) and a search for opportunities to improve CG (Baker and Anderson, 2010). Companies are often criticized for only following a box-ticking approach to CGR without disclosing the true reality of their governance (PwC, 2013).

According to a report by Isaksson (2009), corporate disclosures, specifically those related to risk management and remuneration, did not comply with OECD principles on good CG and the financial crisis can be traced back to incomplete CG disclosures.

To be compliant with the principles of CG issued in 2004 by the OECD (OECD, 2004), corporations should provide information to the investor about CG that represents a realistic view of the financial and non-financial position of the firm’s governance. The OECD therefore demands that a CG framework should be set up to “ensure that timely and accurate disclosure is made on all material matters regarding the corporation, including the financial situation, performance, ownership, and governance of the company” (OECD, 2004, p. 22). These OECD objectives become obvious in the rapidly increasing concept that CG’s key task is to “create accountability and transparency for stakeholders and shareholders”.

The role of CG in enhancing accountability and transparency can also be traced back to the definition of the Association of Chartered Accountants (ACCA), which is a member of the British Association of Authorised Public Accountants.

ACCA recently developed a framework in response to the reaction to the financial crisis (Solomon, 2011), which defines three main purposes of CG. The objective of CG should be to ensure that:

• “the board, as representatives of the organisation’s owners, protects resources and allocates them to make planned progress towards the organisation’s defined purpose” (ACCA, 2009, p. 4);
• “those governing and managing an organization account appropriately to its stakeholders” (ACCA, 2009, p. 4);
• “shareholders and, where appropriate, other stakeholders, can and do hold boards to account” (ACCA, 2009, p. 4).

Inherent in all three of these factors is the requirement to transparently disclose the CG structure, processes and issues faced by the company. Organisations need to demonstrate their authentic commitment to these values with the aim to create and keep investors, stakeholders, and society as a whole confident.

Reconciliation on global capital markets is founded on sophisticated communication networks between companies and investors and there exist information intermediaries which serve as mediators between these key actors (von Westarp et al., 1999).

Shareholders request transparency in corporate reporting (AKEU, 2010), as each investment decision requires the investor to be enabled to assess the risk and return of their investment. Studies show that investors are committed to paying additional margin for good CG (Newell and Wilson, 2002). This has also been confirmed by Patel and Dallas (2002). Academic empirical studies from Gompers et al. (2003) for the US capital market and Drobetz et al. (2004) for the German capital market confirm that the value of a company is strongly influenced by its CG. Resulting from that, companies, including financial institutions, are induced to improve their CGR.

Stakeholders such as debtors or suppliers to the corporation require the same quality of information about the company’s CG as shareholders to be in a position to evaluate their financial relationship to the corporation (Velamuri and Venkataraman, 2005). According to a user-centric model developed by Bovee et al. (2003), information quality is influenced by four essential attributes: accessibility, interpretability, relevance, and integrity. The model defines accessibility in terms of the usefulness of the information and interpretability in terms of its clarity and meaningfulness. Relevance implies that the information provided to the shareholders and stakeholders is of interest in the given context. Integrity means that the information is believed to be “free from defects” (Bovee et al., p. →). This model is extrinsic, as it focuses on the user’s perception of the given information; however, information quality is also dependent upon intrinsic quality. Even for institutional investors with their own research teams, it is a challenge to assess the intrinsic quality of CG and perform peer analysis based on low quality reporting about CG (Bushee and Noe, 2000). Reporting about CG becomes an important information source to eliminate asymmetries between the internal and external available information demanded by shareholders and stakeholders. Due to increasing information search requirements from international investors, the impact and the implications of CG disclosures of corporations keep on growing (Cadbury, 1999).

Corporations should be committed to high quality CGR and it should be in their interest to follow high international minimum quality standards on CGR, which underpins the rapid worldwide publication and implementation of codes on best-practice CG. According to Aguilera and Jackson (2003), companies have significant flexibility in implementing CG bundles and could internationalize their CG practice.

However, the question arises as to how investors can perceive that companies are really committed to high quality CG standards, focusing on the information disclosure of CGR and additionally on how investors can make efficient comparisons between different corporations related to the implementation of high standards for CG.

1.2 Increased relevance and complexity of Corporate Governance Reporting

“The foundation of any structure in Corporate Governance is disclosure.” This quote from a foreword in a World Bank report (Iskander, 2000, p. vi) written by Sir Adrian Cadbury, who developed the Cadbury Code in the United Kingdom, underlines the important role played by disclosures in relation to CG. During the last twenty years, a technological wave of development has taken place, supported by the widespread diffusion of the internet. With regard to corporate disclosures, this trend has also generated the creation of Extensible Business Reporting Language (XBRL), which many accounting experts expect to revolutionize corporate disclosures, since it enables corporate disclosures to be aggregated, transferred, analysed faster and in more detail, and enhances transparency (Bovee et al., 2003; Alles and Piechocki, 2012; Mihaela, 2013). Despite this positive impact of XBRL, which is confirmed empirically (Peng et al., 2011; Yoon et al., 2011), and considering theoretical studies (Christopher, 2007; Lester, 2007; Pinsker, 2007), to date attempts have not been made to apply XBRL to non-financial corporate disclosures, including Corporate Governance Reporting (CGR).

Disclosures are a means of communicating performance and governance (Healy and Palepu, 2001). There are several reasons why CG systems and the quality of public disclosures have become increasingly important to business. Sustainability is becoming an ever more critical business issue and stakeholders have started to pay more attention to the content of the report and the way in which it is presented. The global financial crisis pinpointed the accountability of CG and expectations around transparency have increased. Instead of the release of CG details or policies following a passive role, comprehensive and proactive disclosures from forward-thinking organizations are demanded. Healy and Palepu (2001) assume that two different external publication policies related to companies can be identified. The first involves ongoing communication based on an integrated CG policy. Contrary to this approach, there are corporations which pursue isolated or single issue-based communication. However, there are also organisations which recognize the benefits of increasing disclosures on CG as a clear signal for higher transparency.

The quality and quantity of external disclosures directly impact the transparency of an organization’s CG systems, thereby revealing its strengths and weaknesses. Such transparency has grown in importance for stakeholders, shareholders and management for a number of reasons. Accountable CGR and expectations around transparency have resulted in more comprehensive disclosures and more disclosures on a voluntary basis from larger global organizations, in contrast to the publication of CG details or policies in a less transparent retrospective approach.

Nevertheless, there are still many corporations that only disclose what represent the minimum requirements. At the same time, investors regard greater transparency in company reports as a signal and an equal result of more sound CG. In the experience of the author, the current systems, processes and culture behind the financial reporting embedded in the reporting calendar are major obstacles to a proactive CG environment.

CGR has become very complex for multinational companies in recent years, as it involves statutory compliance with multiple rules and laws, coping with heterogeneous and even non-unitary systems of CG. Additionally, CGR also involves compliance with codes and industry-specific standards. The increasing complexity makes it progressively harder to achieve the goal of ensuring transparency and reducing information asymmetries between investors and management (Möllers and Kernchen, 2011).

The increased complexity means that annual reports including non-financial information are greatly increasing in terms of quantity of information without providing improved quality of disclosures (Paredes, 2003). Paredes (2003, p. 416) uses the metaphor “blinded by the light” to illustrate this development, as people are exposed to too much information, and therefore reach information overload and become increasingly confused. This issue has been reflected in organizational development, accounting and information science (Eppler and Mengis, 2004) and the debate over a substantial disclosure reform with the aim to solve the issue of information overload is ongoing and is discussed by several Accounting Standard Boards with different jurisdictions, including the IASB (Heffes, 2014) and the SEC (Lynch, 2013), and in several academic articles and books (Casey Jr, 1980; Stocks and Harrell, 1995; Speier et al., 1999; Möllers and Kernchen, 2011).

Several articles analyze the impact of cross-border transactions on CG (Bris and Cabolis, 2002; Martynova and Renneboog, 2008; Erel et al., 2012). Improvements in CG resulting from cross-border transactions – the so called spillover effects – are explained with enhanced regulatory regulations from the bidder compared to the target (Martynova and Renneboog, 2008). Cross-border transactions for financial institutions during the 1990s led to very large financial institutions with multiple types of financial services in multiple nations (Berger et al., 2000).

“Disclosure overload” (Johnson, 1992, p. 101) is a term that refers to “information overload” in connection with the publication of financial reports such as annual reports for year-end. “Information overload” (Grether et al., 1986, p. 277) is an issue that has been discussed in various contexts within academic research for more than fifty years. This paradigm mainly addresses the effect of variation in information on individual decision-making. Researchers came to the conclusion that the quality of the decision is proportional to the information perceived, but only to a limited point (Eppler and Mengis, 2004). For management research, the main focus for the theory of information overload has been accounting (Schick et al., 1990). Information overload is associated with excessive information, which prevents someone from making a decision.

The regulators have reacted to cases of corporate corruption, including Enron and WorldCom, with stricter regulatory requirements (Hermalin and Weisbach, 2012). The Sarbanes-Oxley Act (SOX) requires that the management performs an evaluation of its internal control system for financial reporting and additional disclosures related to CG.

The following are additional reasons for the increasing complexity of CGR for financial institutions:

• Increasing pressure from the regulatory side on capital requirements (Basel II and Basel III);
• The fierce global competition has been followed by an increase in cross-border transactions, so that transparency has become even more critical in the eyes of the shareholders and stakeholders (Luo and Tung, 2007; Bris et al., 2008);
• Globalization of financial markets requires increased financial market communication and transparency;
• Increasing public pressure on what is perceived as good CG practice as financial institutions are on the spot due to the financial crisis.

1.3 Corporate Governance Reporting for Financial Institutions

Financial institutions have specific extended needs with regard to CGR. According to Adman and Mehran (2003), financial institutions face additional CG issues, as CG for financial institutions differs from that of non-financial or industrial companies. According to Mehran (2011), financial institutions deal with several more stakeholders than non-financial firms. This results from the fact that financial institutions have more stakeholders which do not have legal relationships but might enter into such depending upon the CGR. Additionally, the business is more complex and the rate of change is much higher within financial institutions compared to non-financial firms (Mehran et al., 2011).

Although many studies exist on CG and industrial companies, the number of studies that address CG and reporting for financial institutions has historically been low (Levine, 2004; Caprio et al., 2007, Macey, 2003). However, this changed considerably after the financial crisis, as financial institutions’ CG became the topic of many academic works (Erkens et al., 2012, Faleye, 2010; Minton et al., 2014). One main area of focus was the risk-taking of banks. According to Shapiro, the SEC chairman, “many of the problems leading to our economic crisis can be laid at the door of poor CG. Too many boards failed in their primary function of diligently overseeing management. As a result, too many managers took on too much risk and made decisions that were too focused on the short-term” (Grabar et al., 2010, p. 90).

The main research question asks whether common attributes of CG for financial institutions during the financial crisis can be identified which contributed to high risk-taking behaviour. As part of these studies, the following attributes were considered: the relative proportion of independent directors (Minton et al., 2014), the board size (Faleye and Krishnan, 2010), the experience of the executives (Fernandes and Fich, 2013), the risk committee (Ellul and Yerramilli, 2013), executive compensation (Cheng et al., 2010) and institutional ownership (Laeven and Levine, 2009). According to Laeven and Leveine (2009) and Ellul and Yerramilli (2013), financial institutions with existing institutional investors have higher risk appetite.

Financial institutions are defined in compliance with Berger et al. (1999) and Achleitner (2014) as companies working in the financial service industry: Financial services comprise products and services offered by financial institutions, which are also banks and insurance firms. Those products consist of credit-related loans, advisory and money management for individuals as well as investment banking-related products such as sales and trading or money markets targeted at institutional clients. Among the organizations that have such products in their portfolios are banks, insurance companies, private business clients, consumer banks, investment funds, cooperative banks and savings and loans companies.

Furthermore, banks usually consist of 90% debt, compared to about 40% on average for non-financial companies, because stakeholders have a much more pivotal implication in banks compared to non-financial companies. Above all, banks take over a crucial role related to the function of the financial system, so that insolvency might cause systematic failure of the system. As a consequence, banks become regulated in different ways from non-financial companies and governments even become stakeholders in the banks. Further differentiators between non-financial and financial companies include the usage of leverage. For non-financial companies, it is a source of financing, while for financial companies it is basically a source of revenue generation.

Additionally, the business models and the business in which banks operate are extremely complex and are subject to dramatic shifts (Mehran et al., 2011). According to Levine (2004, p. 2), “In banking, loan quality is not readily observable and can be hidden for long periods. Moreover, banks can alter the risk composition of their assets more quickly than most non-financial industries and banks can readily hide problems by extending loans to clients that cannot service previous debt obligations.” The growth in complexity is also, according to Mehran et al. (2011), attributable to the fact that financial institutions have grown in size and expanded into other businesses, while new competitors in the “shadow banking” sectors have been created and have intensified competition. According to Levine (2004), another fact that changes governance mechanisms for financial companies is that governments are often stakeholders of banks even in countries that intervene very little.

1.4 Framework of research

XBRL has evolved as the benchmark for electronic disclosures (Cohen et al., 2005; Chen and Sun, 2009; Debreceny et al., 2010; Alles and Piechocki, 2012; Liu et al., 2014). There are many articles within the literature which conclude that XBRL has the capability to enhance clarity and improve CG for financial reporting (Roohani et al., 2010; Alles and Piechocki, 2012; Müller-Wickop et al., 2013). Can XBRL contribute to enhancing transparency when applied to CGR? In the academic literature, there is very little research about the usage of XBRL for CGR (Urdari et al.; Alles and Debreceny, 2012). This not only represents a research gap, but also has practical implications, which justify the present research, as investors demand an extension of XBRL-enabled reporting to CGR (Arnold et al., 2012).

Non-financial reporting consists of reporting on environmental, social and governance aspects; however, the terminology is used in different contexts: risk valuation, socially responsible investing or corporate responsibility (Bassen and Kovacs, 2008). Within the literature about non-financial reporting, it is stated that the main issue was that the companies could not submit standardized tags in XBRL and therefore only a manual mapping could enable a direct comparability. As a result, the non-standardized reporting elements for non-financial reporting needed to be replaced by a taxonomy. The Global Reporting Initiative started to develop a sustainability framework, which was translated into an XBRL taxonomy (Moreira and da Silva, 2013).

The universal objective of the research is to focus on the disclosures related to stock-listed financial institutions within the scope of CG. Applying an empirical study, this thesis intends to contribute to the academic stream on CG through the systematization of CGR of financial institutions by means of a taxonomy, which can be used for the purpose of XBRL filings. XBRL filings are interactive data enabling automatic processing of business information by software (Debreceny et. al., 2009).

The development of this taxonomy has been undertaken for several different reasons. First, according to the academic literature on XBRL, the non-availability and the incomplete knowledge about non-financial taxonomy development is a major obstacle to the transformation of XBRL to narrative information, including CGR. Secondly, the development of a taxonomy also enables the classification of CGR to better understand similarities and differences. Thirdly, the quality of XBRL is inevitably connected to the design of an adequate taxonomy.

In recent years, CGR for multinational companies has been characterised by increasing complexity. Could XBRL potentially help to reduce the complexity of companies’ CGR (Alles and Piechocki, 2012)?

Moreover, according to Beattie et al. (2004), business reporting models should be expanded to “serve the changing information needs of the market” (Beattie et al., 2004, p. 28) and disclose the “information required for enhanced corporate transparency and accountability” (Beattie et al., 2004, p. 28). Can XBRL become an enabler of the enhanced corporate transparency and accountability requirements brought by the fluctuating requirements of the market on corporate information? XBRL, according to Piechocki et al. (2009, p. 224), is regarded as an enabler of additional transparency, and the contribution of this thesis is to consider this assumption for CGR.

Since the 1990s, a magnitude of literature has been published about CG, which can be categorized into three generations of studies: national country-by-country studies, international studies and performance studies.

As many companies have published fraudulent reports in recent years (Enron, WorldCom and others), the accountability, transparency and reliability of CGR has been questioned by market participants. This thesis intends to make a contribution to the academic literature with regard to CG to increase the transparency and reliability of CGR (Piechocki et al., 2009).

A sequential exploratory mixed model, with a qualitative and then a quantitative approach, will be applied. This thesis is based on the institutional and principal-agent theory, as the theoretical prepositions help to understand a corporation’s disclosure policy to both its shareholders and stakeholders.

This thesis addresses a real working issue and is grounded on my personal research and many years of experience as Head of Corporate Disclosures working for several banks in Germany and as a consultant in the area of Finance Transformation, Disclosure Management and Business Process Reengineering. I have conducted empirical research regarding CGR concepts and the design of XBRL taxonomies. This provides me with broad practical as well as theoretical knowledge for the development of a CG taxonomy which is relevant for financial institutions.

This thesis intends to close research gaps with regard to the following main topics.

The main motivation for this thesis is that in the academic literature there is very little research into the usage of XBRL for CGR (Urdari et al.; Alles and Debreceny, 2012). Despite the existence of several taxonomies with regard to IFRS, U.S. GAAP, Sustainability Reporting (GRI) and Integrated Scoreboard (IS-FESG), to the researcher’s knowledge there is no CGR taxonomy that could be used for financial institutions.

Moreover, within academic CG research, a global theory on governance remains to be developed, so that the existing theory has limitations in explaining CG mechanisms, including the CGR reporting of global corporations (Carver et al., 2005). According to Williams and Aguilera (2008), disclosure research faces different challenges, the most severe of which is that there is disagreement about a unifying theory.

According to Roohani and Furusho (2009), the development of further taxonomies is expected due to the lack of existing taxonomies. The chosen topic is a practical use case and it intends to close existing gaps within the practical research field of CGR.

The following is a short summary of existing taxonomies: Several institutions have developed XBRL-enabled taxonomies for application within companies; however, CG is not part of their scope. The IASB in London focuses on the development of an XBRL taxonomy for its IFRS accounting standards. The FASB follows the same approach for the US GAAP XBRL taxonomy, as the focus is on the legally required Financial Reporting (Graham et al., 2005). The Global Reporting Initiative (GRI), which further develops reporting on sustainability, has also issued an XBRL-enabled taxonomy. Although there are reporting elements covering disclosures on CG within the GRI XBRL taxonomy, these elements do not cover the full reporting spectrum of CG, as the main objective for GRI is the measurement of sustainability and the development of key performance indicators (Butler et al., 2011). Due to this inadequacy, the development of an XBRL-enabled CGR taxonomy follows the objective to support investors, auditors and analysts to obtain information about CG more quickly and simply.

This thesis intends to answer the following research question: Can XBRL be applied to CGR by developing a taxonomy for financial institutions as foreign private issuers?

The answer can be used to formulate several hypotheses or prepositions about the issues involved in developing a taxonomy for non-financial disclosures, which fulfils the XBRL specifications. In order to answer the question, several steps have been followed which lead to a framework for the development of an XBRL-enabled taxonomy.

1.5 Research objectives

The overall objective of this dissertation is to analyse in detail the disclosures of stock-listed financial institutions relating to CG. To better explain corporate reporting disclosure transparency, a globalizing taxonomy will be developed using an inductive approach.

The DBA thesis therefore pursues the following two main objectives, which contain practical as well as theoretical aspects:

1. To develop a taxonomy that allows a holistic coverage of the spectrum of CGR of Financial Institutions (Roohani et al., 2009), to cut through the increasing complexity (Paredes, 2003) and to identify common reporting elements (Cicon et al., 2012), which could also bring about a cohesive world theory with regard to corporate governance (Schiehll et al., 2014). This research project shall provide the basic framework for a voluntary implementation of XBRL-enabled CGR. Therefore, the taxonomy implies a classification scheme (Liu, 2013) that provides an implementation-oriented framework for financial institutions planning to convert to XBRL reporting, by providing main categories and reporting elements. The CG literature lacks studies on basic CGR elements and how they are reported (Stiglbauer, 2010b). The majority of the literature is focused on explaining CG divergences of core elements with country-specific de-jure requirements and the underlying embedded systems (Barker, 2010). Very few studies focus on XBRL for CGR (Roohani et al., 2010), although XBRL is regarded as a promoter of transparency (Müller-Wickop et al., 2013), which is seen as a main deficit of financial reporting within the financial crisis (Isaksson, 2009). XBRL embedded in CGR could also have the potential to enhance monitoring of external stakeholders, investors, auditors, customers and suppliers, as the comparability of CGR could be improved. As CGR provided in a non-computer-readable format requires manual conversion by financial intermediaries such as rating agencies, financial data providers or analysts, XBRL has the potential to diminish or even abolish data conversion steps (Alles and Piechocki, 2012). Data conversion also bears the risk of misinterpretation and human errors, while XBRL enables automated processing of disclosures (Piechocki et al., 2009).
2. To identify a bundle of determinants for CGR. Considering voluntary disclosure literature, there are two basic determinants within financial reporting, which are either mandatory or voluntary determinants (Coffee Jr, 1984). This study aims to derive a multitude of explanatory factors, which is a prerequisite for XBRL-reporting preparation (Piechocki et al., 2009) and should also provide more transparency within the wide landscape of CGR. In line with taxonomies for Financial Statement Reporting such as IFRS or US-GAAP XBRL (Kurt and David, 2003), requirements, including preferences and definitions, will be analysed to identify whether there are common characteristics, which can be summarized as de-jure, codex/principle and common practice/industry standards. If, in the context of the research project, it becomes apparent that they are not sufficient, they will be extended further.

As taxonomy development plays an important role with regard to XBRL implementation, this thesis also aims to enhance the understanding of which methods can be used to create or engineer a taxonomy, considering a scientific approach, a method of inquiry which is based on practical evidence. Additionally, it addresses the practical issue that although many corporations are willing to enhance transparency and accountability, they are hesitant towards voluntary implementation of additional disclosure practices such as XBRL exceeding legal requirements. The study intends to close gaps with regard to academic studies on Corporate Reporting. The need for research is based on the existing literature. There are studies which conclude that converting complex narrative, including CGR disclosures, into XBRL also implies additional benefits with regard to information access (Arnold et al., 2012).

In line with the segregation proposed by Bushman et al. (2004), the researcher will focus on CGR as one aspect of corporate reporting besides financial disclosure.

According to Kaplan (2011), it should be an objective for any accounting scholar to learn to understand the upcoming drivers for the application of accounting. Moreover, any accounting scholar should understand how advances and improvements in analytical tools in accounting can be implemented to improve the practice of external reporting (Kaplan, 2011). The analytical tools also refer to taxonomy engineering, for which contemporary exploratory factor analysis is applied, adapted to the bespoke CGR framework.

The quality of a taxonomy can be assessed using the criteria of completeness and interoperability (Zhu and Wu, 2010). The taxonomy prepared in Japan is too rudimentary with regard to the criteria of completeness to satisfy the needs of multinational FPI in the financial service sector. The Japanese taxonomy is too general, as it does not cover most of the CGR requirements. As a result, its interoperability, or in other words the degree of comparability, is low, as companies that follow this taxonomy will not be in a position to disclose in a way that makes them comparable to other corporations. Currently, based on the performed literature review, empirical studies about the implementation of XBRL for the CG Report for corporations listed on the Tokyo Stock Exchange are not documented in academic journals. However, in light of an article by Mitsui in 2011 about the Japanese experience, the author concludes that the way XBRL was provided by Japanese companies revealed that there were many errors.

1.6 Overview of the structure of the thesis

The increasing implications of CG, not only caused by the financial crisis, are described within the first chapter, “Problem Statement and Purpose”. The background of CG from an institutional perspective, which, despite its growing implication and diffusion, is still new within the academic disciplines, is explained. This section is superseded by disclosures on CG and the explanation of increasing associated complexity from a company perspective. Finally, the purpose of the DBA and the research problem with expected limitations are revealed.

In the second chapter, as part of a literature review, CG and disclosures on CG are defined and differentiated, including taxonomy and XBRL, as these three represent the main topics within this thesis.

As part of this literature review, theoretical underpinnings to explain the issue of CG are explored. As the research aims to analyse empirical international data on CGR, the theoretical part includes a wide spectrum from principal-agent, transaction-cost, stewardship, property-rights, stakeholder and institutional theory. This section completes a critical assessment of the underlying theories. The theoretical propositions are followed by an analysis and overview of empirical studies on CGR.

For XBRL adoption, theoretical explanatory patterns are revealed, as XBRL still represents a recently developed technology within Financial Reporting. In the consecutive Chapter Three, the design of the research and methodology and how this is embedded in the overall study are explained. The following chapter gives an overview of the main research concepts and explains which concept was chosen for this thesis. As part of the data collection, annual reports are analysed using content analysis. In the next step, the collected data are further analysed by applying factor analysis. Data analysis and results are elaborated in the following chapters. Based on the theoretical and empirical studies, attributes of CGR are formulated. Finally, a reflexive diary is included.

Figure 1.1: Overview of the structure of the thesis

(Source: author, 2015)

2 Literature Review