Delineating Publicly Listed Family and Nonfamily Controlled Firms: An Approach for Capital Market Research in Australia*

Abstract (Summary)

Recent capital market research evidence suggests that a large proportion of public companies worldwide are characterized by controlling stockholders who are more often families, usually the founder(s) or their descendants. There has been considerable debate on whether "family" firms can indeed be accurately delineated from nonfamily firms given the diversity and abundance of family business definitions in the literature. This paper provides a robust definition of family business for the purposes of capital market research. Using an accounting-based definition of family business, the paper outlines a four-step procedure that provides validation for identifying family controlled companies listed on the Australian Stock Exchange. A significant feature of the research methodology was reliance on data collected from the Australian Securities and Investments Commission. Having access to the corporate regulator's restricted data enabled the researchers to establish important links between directors and their private related entities.

Introduction

Family businesses are difficult to define given the diversity of definitions in the literature, which, no doubt, has a direct impact on the validity of studies. The literature demonstrates the problematic nature of the term "family business," with most definitions constructed to suit the specific needs of the researcher (Wortman 1994). Although there is consensus that a family business is one in which family members have substantial ownership interest and exercise control of an enterprises' operating and financing decisions, Shanker and Astrachan (1996) highlight the need to reach consensus on the definition of family business.

This paper examines definitions of family business in the literature and identifies the difficulties in using "ownership" as a criterion for establishing the existence of family firms. Analogous problems encountered by accountants in using the "ownership" criterion to establish the relationship between a parent company and the companies in which it has an ownership interest (e.g., for determining whether or not a company should be consolidated), are also considered. In addition, this paper explores the manner in which accounting practitioners currently apply the "control" criterion in establishing the existence of a parent/subsidiary relationship, and indeed, the usefulness of adopting this criterion for establishing the existence of a family business. Accordingly, an objective of this paper is to outline a definition of family business sufficiently robust for capital market research purposes.

The paper begins with a review of the family business literature, particularly focusing on definitional issues followed by an examination of agency theory and the development of a theoretical framework for a definition of family business. Procedures for development of the operational definition are also described, followed by implications and recommendations for further research.

Review of the Literature

An increasing interest in the family business literature has focused on the role of family businesses in international economies, and in many instances, evidence suggests that these firms are emerging as substantial contributors to the gross domestic product (GDP) (Smyrnios, Romano, and Tanewski 1997; Shanker and Astrachan 1996; Sharma, Chrisman, and Chua 1996). However, this evidence has been the subject of debate, primarily because inappropriate definitions of family business result in failure to properly delineate these enterprises from nonfamily firms. Thus, the actual contributions made by family businesses are not clearly separated from those of other businesses. It is further argued that many of the sources of statistics quoted in the family business literature are not sourced empirically (Shanker and Astrachan, 1996).

Sharma, Chrisman, and Chua (1996) report no less than 34 different definitions, of which 20 make reference to an element of ownership interest as a factor that determines the existence of a family enterprise, whereas nine definitions make some reference to influence and/ or control as determining factors. The importance of these factors is acknowledged by Astrachan, Klein, and Smyrnios (2002), who argue that an enterprise can be directly or indirectly influenced through a family's power, experience, and culture. However, although these studies make a contribution to the family business literature in the context of proprietary firms, they do not adequately articulate the meaning of their criteria in an operational context for capital market research, which primarily deals with publicly listed companies.

To appreciate the magnitude of the definitional problems confronting family business researchers, Shanker and Astrachan (1996) examine the potential outcomes of applying a range of different definitions. They argue that "based on these definitions, the number of family firms in the US can range from 4.1 million to 20.3 million firms, employ 19.8 million to 77.2 million individuals, and provide 12 percent to 49 percent of the GDP of the US" (p. 4). These statistics serve to illustrate that a generally accepted definition of family business is an essential and much-needed element of empirical research in this area.

The literature identifies three dimensions in which the different definitions of family business can be classified. These three dimensions are briefly outlined.


Degree of Ownership and Management by Family Members

This dimension requires that family members have a minimum level of ownership and be involved in management of the business. A predominant focus on ownership as a key attribute is not unreasonable to expect, given that the literature has identified ownership as a critical variable and an essential part of the production function of the firm (see Gallo and Villaseca 1998; Jensen and Meckling 1976). Major contributors to this dimension include Barnes and Hershon (1976), Lansberg, Perrow, and Rogolsky (1988), and Litz (1995), among many others.
Barnes and Hershon (1976), and Lansberg, Perrow, and Rogolsky's (1988) critical insights into ownership and control are worth noting because of their direct relevance to the current study.

Lansberg, Perrow, and Rogolsky (1988) argue "much of the confusion about the extent of family control derives from the fact that mechanisms used by families to exert their influence over management (voting, trusts, foundations, holding companies) are deliberately designed to keep the identities of shareholders hidden" (p. 3). This point highlights the hazardous problem of using ownership as a sole criterion for defining a family business. As will be outlined later, this problem of ownership led to the development of a concept of "control" in AASB1024-Consolidated Financial Accounts (Australian Accounting Standards Board 1992), a specific Australian accounting standard designed to identify controlling interests in firms.

Although Lansberg, Perrow, and Rogolsky (1988) were concerned with the legal aspects of control, Barnes and Hershon (1976) recognized that dominant individuals are at play in the family business, and these individuals have significant influence in the decision-making process. Indeed without the existence of dominant individuals influencing the dynamics of the family business, the chances of some family businesses surviving are significantly diminished.

Chua, Chrisman, and Sharma (1999) similarly acknowledge that the essence of a family business is a vision developed by a dominant coalition that shapes and pursues the vision in a way that it is "potentially sustainable across generations of the family" (p. 25). Thus, the appearance of dominance by specific individuals is regarded as an important factor associated with firm value. Hence, McConaughy (1994) observes that the identity of the owner-manager is more important than the level of ownership (p. 5), whereas Berglof and von Thadden (1999) maintain that most firms (even listed) in the world have a dominating owner, and in most cases a family or the state holds such a dominant stake.

These views are consistent with evidence provided by 27 accounting practitioners randomly selected to partake in an interview for the purposes of this study. It is observed that most businesses owned by a family or a group of families have at least one dominant individual steering the course of the business. The participation of a dominant individual in the family business is an important aspect of the current study, as the definition of family business is based on "control," which in turn is defined as the capacity to dominate decision making.

The importance of dominance with respect to the performance and success of firms is also acknowledged in the literature. Neun and Santerre (1980) find that the existence of dominant shareholders increases the value of the firm. Similarly, Zeckhauser and Pound (1990) reveal that the existence of dominant shareholders increases the value of the firm because they have a more effective ability to monitor firm performance. These studies indicate that larger shareholders have the propensity to act as monitors and thereby reduce agency costs.

Familial Context

Definitions within this dimension focus on control of the business passing to younger members of the family over time (e.g., Churchill and Hatten 1987; Ward 1987), and authors (e.g., Beckhard and Dyer 1983; Davis 1983) have relied on a systems approach to identify family from nonfamily control. The systems approach primarily makes an attempt to determine the unique characteristics and relations between the family and business systems. Other definitions within this dimension propose that a diverse range of attributes are simultaneously required to define a family business, including ownership, involvement in day-to-day management of the business, generational transfers and links, family influence, and family control.

Listed Family Firms

A limited number of studies have provided definitions of family firms in the context of capital market research. One of the earliest was Burch (1972), who examined the top 300 companies based on the 1965 Fortune 500 list. He found that families controlled more than 47 percent of these top 300 companies. Burch's definition of family business falls within Handler's (1989) classification of "multiple conditions." He used the following two criteria to define a family business: (1) Between 4 percent to 5 percent or more of the voting stock must be held by a family or group of families or one (affluent) individual; and (2) family representation on the board over time.

More recent studies have been conducted by Jetha (1993) and McConaughy (1994). Jetha, who defines a family business where at least one member of the current senior management is linked (at least to second-generation level) to the founding family, found that 37 percent of the largest publicly held companies are family-run operations. Similarly, McConaughy (1994) found that 21 percent of publicly held companies on the Business Week 1000 list were family businesses. He defined a family business as one in which the chief executive officer (CEO), president, or chairman is a descendant of the founding family.

Control was used as the criterion to identify both the ownership and the voting rights of listed firms in a study of the ownership structure of the 250 largest listed firms on the Paris stock exchange (Blondel, Rowell, and Van der Heyden 2002). The researchers found that approximately 57 percent of the 250 largest firms on the Paris bourse were patrimonial companies, representing some 35 percent of the overall market capitalization. Following closely on Blondel, Rowell, and Van der Heyden's (2002) research design, Klein and Blondel (2002) provide similar ownership structure results in Germany. They established that patrimonial firms comprise approximately 51 percent of the 250 largest listed firms in Germany, but only represent 19 percent of the overall market capitalization of listed companies.

Anderson and Reeb (2003) investigated the relationship between founding family ownership and firm performance of the top 500 Standard and Poors (S&P) companies in the United States between 1992 and 1999. They found that approximately 35 percent of the firms on the S&P 500 list were family controlled. Consistent with prior studies, the researchers used fractional equity ownership as a measure of control by the founding family and/or the presence of family members on the board of directors. Additionally, for older firms, that is, where the firm has expanded over time to include relatives several generations after the founder and the family names of such relatives may be different from the founder, Anderson and Reeb (2003) examined corporate histories using several sources of data to resolve decendancy issues. This is of interest to the current study because the collective ownership by family members is critical to understanding the dynamics of family control. Thus, being able to trace family holdings over time is an important operational issue which establishes the presence of a familial link.
This review identifies a number of common themes, which form the basis of a definition of family business for the purposes of capital market research. These themes are as follows:
* ownership of shares by an individual or related individuals, or by another entity in which an individual or related individual has/had an interest (directly or indirectly);
* control of a company by an individual or individuals, exercisabie through various means including share ownership and board membership;
* concentration of ownership restricted to a relatively small group of individuals either directly or indirectly;
* continuity of control by an individual or group of related individuals; and
* dominance of decision-making by an individual.

Toward a Capital Market Definition of Family Business

This study focuses on definitions and concepts that not only have commonality in the literature, but also have the support of authoritative bodies. These include the various bodies representing the accounting profession and corporate regulators, such as CPA Australia, the Institute of Chartered Accountants in Australia, and the Australian Securities and Investments Commission. The rationale for this approach is that a definition of family business based on concepts and definitions that already have independent authoritative support, has the propensity to attract credibility and ultimately general acceptance among scholars and practitioners. In this respect, the definition of control, as embodied in Australian Accounting Standard AASB1024-Consolidated Accounts (hereafter AASB1024, superseded recently by AASB127: Consolidated and Separate Financial Statements), has particular appeal, as it highlights dominance of decision-making by one individual. For instance, AASB1024 defines control as "the capacity of an entity to dominate decision-making directly or indirectly, in relation to the financial and operating decision of another entity." The definition of entity in the standard includes a person, whereas capacity to dominate and the exercise of dominance by an individual in relation to operating and financing decisions is recognized as a critical feature in family firms. More importantly, this capacity to dominate is often required to ensure the survival of the family firm (e.g., Chua, Chrisman, and Sharma 1999; McConaughy 1994).

The issue of dominance by an individual in family firms has also been given considerable support by accounting practitioners, as indicated by the results of a survey of technical (accounting) specialists and practitioners undertaken as part of this study. Results show that the importance of dominance in family firms was rated highly significant by 96 percent of respondents, and almost all respondents rated "ownership," "control," and "management structure" as highly appropriate attributes of family firms.

There are also other potential reasons why use of the "control" criterion in defining a family firm has theoretical appeal. For example, several contributors to the family business literature (for example, Fiegener et al. 1994; Dönckels and Fröhlich 1991; Dreux 1990; Lansberg, Perrow, and Rogolsky 1988; Alcorn 1982) identify "control" as an important element in defining the family firm. There are also parallels in the literature relating to the definition of family firms and the genesis of AASB1024 as an accounting standard. For example, comments by Lansberg, Perrow, and Rogolsky (1988) suggest that difficulties in defining family firms, based on ownership, rests upon the various mechanisms used by families to keep the identities of shareholders hidden (p. 3). This inability to determine the relationship between two entities (parent and subsidiary) based on an ownership test is one of the primary reasons for introducing AASB1024 in Australia. Thus the definitions developed in AASB1024, which provide the framework for identifying controlling interests, can also have application in the family firm context.

Hence there are persuasive arguments for using "control" as one of the important criteria for defining a family firm. In determining the existence of a family business, this study adopts in part, the control criterion promulgated by both AASB1024 and International Accounting Standard IAS27: Consolidated and Separate Financial Statements (hereafter IAS27), for preparing consolidated financial statements of an entity. It also takes cognizance of the evidence in the literature, which indicates that control and ownership of family businesses are nondiffuse and are typically characterized by an ownership structure in which shares are closely held by family members Qensen and Meckling 1976). Thus, a family business is defined as an entity controlled by a private individual, directly or indirecdy, in conjunction with close family members. Moreover, control is broadly defined as the capacity to dominate decision-making. A detailed description of this definition of family business is provided in the following discussion, together with a framework for operationalizing various aspects of the definition for the purposes of capital market research.

Method

An archive-based case study method was selected to support the primary objective of determining whether family and nonfamily firms listed on the Australian Stock Exchange (ASX) can be delineated with the guidance of AASB 1024/ IAS27 and their concomitant definitions of "control," "closely held," and the "capacity to dominate decision-making."

The data gathering and analysis procedures were divided into four steps.

First, an examination of financial statements and other corporate details using various databases was undertaken to determine whether it was possible to access and link information relating the themes of ownership, control, and continuity of control and dominance.

Second, several procedures were undertaken to validate the delineation process.

The third step involved the final compilation of categories of family and nonfamily controlled firms.

The final step (not covered in this paper) involved the application of the definition of family business to all qualifying initial public offerings (IPOs) that listed between January 1987 and December 1999.

The analysis began with a cross-sectional qualitative examination of the population (N = 2,022) of companies registered with the ASX for the period ending June 30, 1998. All publicly available information from the ASX on the population of companies were electronically downloaded and recorded into spreadsheets. The listed companies were categorized into active and delisted, and specific criteria such as ownership concentration, number of shareholders, paid up capital, among other information, were examined.

Where company information was not available from the ASX, a complementary database (i.e., Bloomberg's) was used to complement the initial data source. In addition, Australian Securities and Investments Commission (ASIC) annual financial statement files were also used to validate initial data sources, to examine related party disclosures, to assess continuity of control for a period of two years after the initial year ending June 30, 1998 (thus a total observation period of three years), and to validate the ASX data generally. Figure 1 outlines the procedures used in the initial stages of the study.

As the definition of family business is based on control by an individual (in conjunction with related parties) and continued control, relevant files of listed companies were examined, and information that provided evidence of the following factors was noted: evidence of existing control, history of control, continuity of control (ASIC data), and related party relationships. Table 1 illustrates the specific criteria examined and used to differentiate family from nonfamily controlled firms. In addition to these criteria, only companies that exhibited the following characteristics were categorized within the family controlled group:
* The existence of a dominant shareholder who is identified as a founding member involved in the management of the company and has a direct interest of greater than 20 percent of voting shares;
* The dominant shareholder is the CEO or key member of the board (that is, managing director or chairperson);
* The dominant shareholder continues to be the dominant shareholder and member of the board during the observation period, that is, in 1998 plus two years;
* At least one other related party is a member of the board; and
* The dominant shareholder, in conjunction with other related parties, holds greater than 30 percent of the voting shares in the company directly or indirectly, after the IPO listing.
The initial differentiation revealed that of the 1,214 active companies on the ASX at June 30, 1998, 197 (16.23 percent) were family controlled and 1,017 (83.77 percent) were nonfamily controlled. However, details of companies not falling directly within the dichotomous groups required a further comprehensive analysis for evidence of ownership, control, and dominance. This was undertaken for approximately 60 companies using detailed ASIC data including relevant statutory lodgements where relevant (for example, Form 316, listed within in the regulations of the Corporations Law, which requires companies to annually disclose ownership interests).

This process provided a validation and confirmed that information relating to company ownership, control, and dominance was in fact accessible. That is, if not initially during the first level of searching (via a variety of databases), then almost always after second, third, fourth, etc., level searches using the ASIC database. The validation procedure allowed links to be established between directors, their related interests, and their director-related entities. In this respect the current study was unique given that the authors had access to data that was not publicly available and were thus able to make significant links between directors and their director-related entities.

In order to provide independent verification and validation of the criteria used in determining the definition of family business, a questionnaire was developed and distributed to all 32 members of the Emerging Accounting and Auditing Issues Group (EAAIG), a group comprising technical directors of the major Chartered Accounting firms, and financial and technical specialists from industry and regulatory bodies. The standardized instrument comprised four short questions covering the fundamental attributes of family business viz; control, ownership, and dominance. The intent was to solicit the views of technical specialists who would have in-depth knowledge and experience to comment on a wide variety of issues relating to businesses in general.

A short presentation and discussion (of 20-minute duration) outlining the objectives of the research was given by the first author during a meeting in 2000. 17 members present of the 19, completed and returned their questionnaires during the meeting. The balance of the membership (13 members) was canvassed through various means including telephone calls, mail, e-mail, or facsimile. Eight further responses were received and there were no additional follow-up procedures conducted. In all, 27 replies out of a possible membership of 32 were received.

Results indicate that overwhelmingly, (96 percent) members in the group support the notions of "dominance" (by one or more individuals) as being a highly significant factor in family business. All (100 percent) respondents indicated that "ownership" and "control" were important, whereas 93 percent stated that "management structure" was an important attribute.
Following these procedures, it was established that of the 1,214 active companies listed on the ASX as at June 30, 1998, 211 (17.1 percent) were family controlled and 1,012 (82.9 percent) were nonfamily controlled companies.

Operational Definition of Family Business

A family business is defined as an entity controlled by a private individual in conjunction with close family members with continuity of control evident. The terms "control" and "close family members" are both critical components in operationalizing the definition of family business. Control is defined as the capacity of an entity to dominate decisionmaking, directly or indirectly, in relation to the financial and operating policies of another entity so as to enable that other entity to operate with it in pursuing the objective of the controlling entity (AASB1024). Control is also defined as the power to govern the financial and operating policies of an enterprise (IAS27).
There are several indicators that support the existence of control of an entity. These indicators include: the holding of a majority ownership interest and associated voting rights either singularly or in conjunction with close family members; the capacity to dominate the composition of the board of directors or governing body of another entity; the capacity to appoint or remove all or a majority of the directors or governing members of another entity; and the capacity to control the casting of the majority of the votes cast at a meeting of the board of directors or governing board of another entity (AASB1024, 7[xvi(a) to (e)]). As continuity of control is a significant factor in defining a family firm, the presence of control by individuals was examined over several periods subsequent to the initial delineation date in 1998.

A second important aspect to the definition relates to the meaning of "close family members." In this regard the study relies on definitions from Australian and International Accounting Standards to link family members. For instance, "close family members" means "close family members of the family of an individual... that may be expected to influence or be influenced by, that person in their dealings with the enterprise" (International Accounting Standard IAS24 1995-2001, 3[0.851). In the context of Australian Accounting Standards, close family members are persons included within the definition of "director-related" entities, a defined term in Australian Accounting Standard AASB1017 (Australian Accounting Standards Board 1997) Related Party Disclosures (hereafter AASB1017, and recently superseded by AASB124). Director-related entities are defined in relation to particular directors, as "the spouses of such directors, relatives of such directors or spouses, and any other entity under the joint or several control or significant influence of such directors, spouses or relatives" (AASB1017, 9.1). In addition, the term "relative" was defined in the Australian Corporations Law 1991 (Commonwealth Government of Australia 1991) as "spouse, parent or remoter lineal ancestor, son, daughter or remoter issue, or brother or sister of the person" (S5.1).

Discussion

Against a background of definitional diversity in the literature, this study provides a theoretical and practical basis for developing a definition of family business for capital market research purposes. By combining common themes (within existing definitions), with authoritative definitions and explanations in accounting standards and legislation, an operational definition of family business is derived.

This outcome adds a new dimension to the family business, accounting, and finance literatures and has the potential to contribute significantly to future research in these fields. For instance, it may assist in addressing the urgent need for definitional consensus (Shanker and Astrachan 1996; Wortman 1994) by providing the foundation from which a generally accepted definition of family business can be developed. This outcome may in turn provide valuable information that could assist in validating statistics (Shanker and Astrachan 1996) and other data in existing or future studies that draw comparisons between family and nonfamily businesses.

Perhaps one of the more significant contributions of this paper is the ability to delineate family from nonfamily businesses for the purposes of capital market research. Recent studies (for example, Anderson and Reeb 2003) have shown that family businesses are emerging as significant contributors not only to GDP but also to capital markets generally. It is evident from the literature that ownership and control are significant variables that influence managerial incentives and thus impact firm performance. Jensen and Meckling (1976) and Fama and Jensen (1983) have posited that the dynamics underlying family relationships reduce agency costs and improve efficiency. Extant research (for example, Anderson and Reeb 2003; Ang, Cole, and Lin 2000; McConaughy 1994; McConnell and Servaes 1990; Wruck 1989; Morck, Scleifer, and Vishny 1988) has also found an association between concentrated ownership and control, and firm performance. Given these findings and the evidence that family businesses are typically characterized by higher levels of ownership and control, studies which examine the performance of listed enterprises would expect to find differences in the performance levels of family firms compared to nonfamily firms ceteris paribus. In this regard, the ability to accurately differentiate between family and nonfamily businesses is of importance in further advancing knowledge in this area.

One example of the potential usefulness of an operational definition of family business in capital market studies is the examination and delineation of IPO firms to determine the significance of ownership and control on both underpricing and performance. A study of this nature, which delineates family and nonfamily enterprises in a widely acceptable manner, would be useful in understanding the phenomena, particularly if it can be established that the level of underpricing is different between family and nonfamily groupings. It would also be useful to establish whether factors known to influence the level of underpricing are mediated by family control.

Overall, the main strength of this research is the rigorous process by which family controlled firms have been delineated and validated. The researchers had access to detailed restricted ASIC data, which included relevant statutory lodgements such as Form 316 requiring companies to annually disclose ownership interests, documents related to initial subscriber and substantial interests, correspondence to ASIC from company directors, prospectus documents, and other relevant background information. Over a three-year period, the researchers were able to gather enough evidence from the corporate regulator to validate the delineation of family controlled firms by establishing the nature of control for the overwhelming majority of companies listed on the ASX in 1998,2 by ascertaining the history of each firm's control, by determining the continuity of each firm's control, and by establishing the firm's related party relationships. Thus, through, this process the researchers were able to establish important links between directors and their private related entities.