HomeChapter 1: The Introduction to This Study1.6 The Research Approach of This Study

1.6 The Research Approach of This Study

Although there is an enormous body of prescriptive literature available on how directors should govern corporations and while there is a growing interest in the formal independence of boards from practitioners as well as academics, there are several factors that challenge the research agenda on corporate boards of directors. Zahra and Pearce (1989) conclude that there is still a pressing need to document what boards actually do. Judge (1989) and Judge and Zeithaml (1992) report that researchers simply do not know what boards' roles are in decision making. The factors that challenge the research agenda on boards of directors are discussed in more detail in the following paragraphs.

 

The Availability of Data

The limited availability of meaningful data makes a study on the formal independence of boards a challenging endeavor. Boards of directors are hard to study, there is a scarcity of data, directors are difficult to approach, are always busy and often conduct their business behind closed boardroom doors. Although stringent disclosure regulations provide detailed information on board practices in annual reports and proxy statements of listed corporations in the US and the UK, information on board practices in these and other countries is still rarely systematically collected for comparative international research purposes. Most databases concentrate on financial data of corporations.

When available, corporate governance data is fragmented or difficult to access. The SEC Edgar database - for example - has reported on individual corporations in the format of separate proxy statements and Forms 10K in the US. The database is not designed for comparative, longitudinal research purposes. Fortunately, executive search firms have taken the lead to provide their clientele with detailed information on board practices. Material from Heidrick and Struggles, Korn Ferry International and Spencer Stuart has been used in this study to collect detailed information on corporate governance practices in listed corporations in the US (between 1981 and 1997) and the UK (between 1992 and 1997). Data on board attributes of boards in these countries are presented in chapter six and chapter seven of this research.

The disclosure of board practices is much less developed in (continental) European countries. Due to limited board disclosure regulations in the Netherlands - certainly compared to disclosure standards in the US and the UK - this study relied on a questionnaire sent to fifty chairmen of the largest Dutch corporations in 1996 and fifty chairmen in 1997 to reveal information on the Dutch two-tier board model. The surveys resulted in positive response rates of 60 percent in 1996 (n=30) and 64 percent in 1997 (n=32).

In addition to the surveys, information on board composition and board structure has been culled from 1987-1997 annual financial reports of one hundred of the country’s largest listed corporations. These corporations also verified the information on the composition and the structure of their management and supervisory boards. The results of the surveys are presented in chapter eight of this study. The database of the research project in the Netherlands currently contains information on some 14,000 management and supervisory board positions. This study also initiated a survey, sent to all leading stock exchanges in Europe, North- and South-America and Asia in 1997. Stock exchanges were asked to indicate the latest developments in listings rules, codes of best practices and other self-regulatory initiatives in their financial regions. The results of this survey are presented in chapter five of this study.

 

The Selection of Countries

The corporate governance literature is strongly dominated by research on one-tier corporate boards in the US and the UK. Second in prevalence, but much more less available, are studies on corporate governance models in Germany and Japan. With the exception of a few studies, comparative research on governance systems and board models in (continental) European countries is hardly existent (Charkham, 1994; ICA, 1995; Pic, 1995; Maassen and van den Bosch, 1999a). Although some progress has recently been made to understand the differences between the formal independence of one-tier and two-tier boards, the body of knowledge in this particular field of interest is still limited.

Demb and Neubauer (1992a, 1992b) were among the first scholars to publish a comparative study on corporate boards in the US and Europe. Sheridan and Kendall (1992) reported on governance systems in several European countries. Charkham (1994) wrote an influential book on corporate governance in five countries. More recently, the International Capital Markets Group (1995) and Pic (1995) systematically reviewed corporate governance issues from an international perspective. The limited number of comparative studies can be explained by the relative importance of economies in the US, the UK, Japan and Germany. Another reason could be the availability of data and research resources and the unfamiliarity of scholars with corporate governance systems in Europe and other financial regions. In addition to the analysis of board models in the US and the UK, the limited body of research on international corporate governance has stimulated this research project to explore developments in two-tier boards in the Netherlands - a country previously ignored in the internationally oriented corporate governance literature.    

 

The Domination of Anglo-Saxon Perspectives of Corporate Governance Research

The third challenge relates to the magnitude of Anglo-Saxon views of corporate governance in the literature. The existing body of knowledge is dominated by researchers who concentrate on the formal independence of one-tier boards in the US and the UK and who most often apply a shareholder perspective on corporate governance. These studies often concentrate on quantitative research methods to understand the relationship between attributes of boards and firm performance. Placed in the context of the one-tier corporate board model, this stream of research assumes that the structure of the board determines board behavior and therefore leads to changes in organizational performance (Judge, 1989).

Interestingly, Pearce and Zahra (1992:417) conclude that the “ . . . web of associations among organizational variables is so complex that unidirectional, causal relationships can not be claimed and defended reliably.” In addition, Pettigrew (1992a:178) suggests that “ . . . progress in . . . the study of boards and their directors, has not been helped by over-ambitious attempts to link independent variables such as board composition to outcome variables such as board and firm performance. The research agenda here need not be guided just by studies testing the relative explanatory power of the agency theory or theories of managerial hegemony.” Pettigrew (1992a:178) concludes that the task is perhaps a simpler one to “ . . . redress the overwhelmingly prescriptive bias in this literature, and (to, eds.) begin to provide some basic descriptive findings about boards and their directors.”

 

The Case Study Approach of This Research

In response to these observations, this study relies on a research method that concentrates on the exploration and classification of corporate governance models, i.e. a case study approach. According to Yin (1989:14), “ . . . the distinctive need for case studies arises out of the desire to understand complex social phenomena . . . real life interventions that are too complex for the survey or experimental strategies.” More specifically, Thurman (1990) indicates the following characteristics of case studies that make the use of this research design particularly useful for this study:

 

  • the phenomena are examined in a natural setting;
  • data are collected by multiple means;
  • one or few entities (person, group, organization or country) are examined;
  • the complexity of the unit is studied intensively;
  • case studies are more suitable for the exploration, classification and hypothesis development stages of the knowledge building process; the investigator should have a receptive attitude toward exploration;
  • no experimental controls or manipulation are involved;
  • the investigator may not specify the set of independent and dependent variables in advance;
  • the results derived depend heavily on the integrative powers of the investigator;
  • changes in site selection and data collection methods could take place as the investigator develops new hypotheses;
  • case research is useful in the study of “why” and “how” questions because these deal with operational links to be traced over time rather than with frequency or incidence;
  • the focus is on contemporary events.

Sources: based on Benbasat (1987), quoted in Thurman (1990:54).

 

The complexity of board independence, the limited body of knowledge on one-tier boards and the fact that the organization and the composition of two-tier boards are relatively unexplored in the literature suggest that there is an ideal match between the characteristics of case study research and the complexity of the central research questions of this study (Eisenhardt, 1989b). As indicated above, case studies are suitable for the exploration, clarification, classification and hypothesis development stages of the knowledge building process (Yin, 1989). In this study, the formal independence of one-tier and two-tier boards is explored in more detail both theoretically as well as empirically in the US, the UK and the Netherlands.

The case studies in this research are based on the description and the analyses of board models in these countries. The three studies in this study focus on contemporary events while supporting data for the description of board models is collected by multiple means and the web of associations between board structure and firm performance is understood to be too complex to predefine causal relationships. To explore differences between one-tier and two-tier boards, this research - for example - relies on a combination of previously published board indexes, surveys, archival records and direct observations. It is also important to understand that research findings of this study are method bound.

The research findings can only be generalized to the corporations that are included in this study and the findings should primarily be used to clarify and to compare the empirical results with the theoretical framework of this research. As such, the theory developed in part I of this research is used as a template to interpret and to understand the empirical results of this study in parts II and III of this study (Yin, 1989).

          

The Reliability of Data

Yin (1989) indicates that the goal of a reliability test is to minimize errors and biases in a study. Essential to the reliability of research findings is the accuracy of the data. When possible, this research relies on primary data sources. Information on corporate boards in the US and the UK is culled from publications from executive search firms. The data in these publications is mainly derived from proxy statements and annual reports. This research project did not influence nor control the research methods used by executive search firms that collect information on boards in the US and the UK. As part of the research tradition in these firms, corporations are generally requested to verify the information on their boards before the information is published by executive search firms.

This research project directly controlled the accuracy and reliability of data on corporate boards in the Netherlands. Annual reports and surveys have been used as a primary source of information. To assure the accuracy of data, a summary of findings was sent to the investor relation departments of corporations for verification in 1998. These findings were based on annual reports published between 1987 and 1998. When necessary, adjustments were made to the data set. The fact that the information would be published in booklets and that these would be sent to directors of corporations involved, acted as a strong incentive for representatives of investor relation departments of listed corporations to verify the accuracy of the project’s data files.


Related news items:
Newer news items:
Older news items:

 
You may also be interested in these articles:

PDF Download

Interested in the PDF version of the study? Click here to download.

Maassen, G.F. (2002). An International Comparison of Corporate Governance Models. A Study on the Formal Independence and Convergence of One-Tier and Two-Tier Corporate Boards of Directors in the United States of America, the United Kingdom and the Netherlands.

Maassen, G.F. (2002). An International Comparison of Corporate Governance Models. A Study on the Formal Independence and Convergence of One-Tier and Two-Tier Corporate Boards of Directors in the United States of America, the United Kingdom and the Netherlands. Amsterdam: Spencer Stuart Executive Search.