- Top of page
- THEORY AND HYPOTHESES
Manuscript Type: Empirical
Research Question/Issue: How and to what extent does national culture influence the composition and leadership structure of the boards of directors of multinational firms?
Research Findings/Insights: Societal norms about corporate structure are treated as components of national culture. Hofstede's measures of national culture were shown to predict the board composition and leadership structure of firms based in that culture. The hypotheses were tested with data on 399 multinational manufacturing firms based in 15 industrial countries. The results suggest that national culture can have strong effects on corporate governance and should be considered in any transnational study.
Theoretical/Academic Implications: The predictive accuracy of the culture variables provides strong support for the argument that norms embedded in a society's culture affect organizational structure, at least at the board level. The results of the study contribute to our understanding of institutional theory in explaining observed variations in corporate board composition and leadership structure across countries. By linking board composition to the cultural environment, institutional theory provides an explicit framework for analyzing variations in board structure across national boundaries.
Practitioner/Policy Implications: When considering board composition and leadership structure, it is important to consider national culture norms. The findings of the study also have important implications for multinational firms setting up boards for their subsidiaries in different countries.
- Top of page
- THEORY AND HYPOTHESES
The structure of corporate governance varies widely across developed economies, ranging from the two-tiered supervisory and management boards found in Germany, to the insider-dominated boards of Japan, to boards including both insiders and independent directors in countries such as the US (Aoki, 1990; Charkham, 1994; Shleifer and Vishny, 1997). Everywhere, corporate boards find themselves scrutinized by regulatory agencies, institutional investors, ordinary shareholders, and their lawyers; and boards are more and more being criticized for their decisions regarding executive pay, diversification, takeover offers, and any perceived negligence (Norburn, Boyd, Fox and Muth, 2000). Numerous critics now demand corporate governance reform with the aim of better protecting shareholders. While board crises, investor activism, and market globalization increasingly call for international corporate governance reform, it is important to understand how corporate governance structures vary across countries (Jenkinson and Mayer, 1992; Roe, 1993).
Previous research has perhaps paid inadequate attention to the factors explaining the wide variation in governance structures found in different countries (Dalton and Kesner, 1987; Shleifer and Vishny, 1997). Previous research has emphasized variations in the political and legal constraints on ownership and control, particularly how government regulations affect the ways companies are owned and controlled, and the processes by which ownership and control change (Prowse, 1990; Roe, 1994). Studies from the agency theory perspective, for example, have suggested that statutory requirements and ownership structure are the key elements explaining international variations in corporate governance structure (Hoshi, Kashyap and Scharfstein, 1990; La Porta, Lopez-de-Silanes and Shleifer, 1999).
One significant effect of national environmental differences is variation in corporate ownership structures (Charkham, 1994; La Porta et al., 1999). Jenkinson and Mayer (1992) distinguish between two broad categories of corporate ownership structure. In the first category are firms in continental Europe and Japan, in which the ownership of individual firms is often concentrated within a small number of directly related firms, banks, and families (see also Roe, 1993). This concentration results in the formation of consolidated industrial groups, and in this category, cross-shareholdings between firms are common. In the second category, which includes firms in Great Britain and the US, ownership is more dispersed among a large number of unrelated individual and institutional investors.
The two types of ownership structure are reflected in the composition of these firms' boards of directors. In countries marked by concentrated ownership, such as Germany, Japan, and Sweden, the directors of a corporation tend to mirror the company's most important long-term stakeholder relationships. Through their role as major providers of capital and their direct participation on client company boards, major banks affiliated with industrial groups have access to a considerable amount of privileged information. Their responses to virtually any aspect of their client companies' activities represent important signals to other corporate stakeholders (Kester, 1992). Conversely, outside directors of Anglo-American corporate boards tend to have some special expertise but usually no substantial capital or other commercial stake in the company (Roe, 1994).
Other national differences in board structure include the degree of nonmanagement employee representation on the board (as is legally required in Sweden and Germany); the percentage of management directors (insiders) on the board (managers dominate boards in Japan and usually in Great Britain, but make up a minority of directors in other European countries and in the US [Bacon and Brown 1977; Jenkinson and Mayer, 1992]); the use of board committees (common in the US and Canada, but rare in most European countries; see Bacon, 1986); and single-tiered or two-tiered board structures. The typical two-tiered structure includes both a supervisory board and a management board, as is legally required in Germany, Finland, Denmark, and The Netherlands (Schneider-Lenne, 1992; Charkham, 1994). Because of the incomparability of the single- and two-tiered systems, and because legal requirements for two-tiered systems typically minimize or eliminate variation in the structural variables of interest in this study, the analysis was restricted to firms with single-tier boards.
A country's social and cultural characteristics also have an important influence on governance structure (Friedland and Alford, 1990; Hofstede, 1991; Hickson and Pugh, 1995). Institutional theory (Meyer and Rowan, 1977), which links organizational structure to the institutional environment, was used to develop hypotheses concerning variations in the composition and leadership structure of multinational corporations' (MNCs') governing boards. The focus was on the cultural environment and the effects of Hofstede's (1980; 1991) national cultural dimensions were examined. Hypotheses were formulated and tested on a sample of 399 large, multinational, manufacturing firms in 15 industrial countries including Australia, Japan, and countries in North America and Western Europe. The findings confirm that national culture has a significant influence on corporate governance structure.
Patterns of corporate governance in different countries will first be reviewed to provide a background for the discussion. The implications of national culture for understanding variations in corporate governance structures will then be examined, and hypotheses linking national culture to board composition and leadership structure will be developed. After describing the methodology and reporting our empirical tests, we discuss the key findings and their implications for theory and practice.
THEORY AND HYPOTHESES
- Top of page
- THEORY AND HYPOTHESES
The board of directors has the legal authority and obligation to monitor and control corporate activities and to protect shareholder interests (Bacon and Brown, 1977; Fama and Jensen, 1983; Lorsch and MacIver, 1989). Its responsibilities include approving strategic decisions, reviewing corporate performance, and the evaluation, compensation, and retention of top management.
This discussion of board composition will be limited to the percentage of outside, or nonmanagement, directors. Leadership structure will refer to whether or not the board chair is also the chief executive officer (CEO). Composition is clearly related to the board's independence from management (Beatty and Zajac, 1994; Dalton, Daily, Ellstrand and Johnson, 1998), but leadership structure also influences board independence and the concentration of executive power (Harrison, Torres and Kukalis, 1988; Finkelstein and D'Aveni, 1994; Westphal and Zajac, 1997).
Institutional Norms and National Culture
Organizations can be viewed as social entities integrated into the institutional and value structures constituting the culture of a society (Smircich, 1983; Swidler, 1986). In this view, organizations and societies tend to reflect each other structurally (Inzerilli, 1981). In addition, national cultural differences influence the shape and functioning of organizations (Crozier, 1964; Landsberger, 1970; Lammers and Hickson, 1979; Child, 1981; Hamilton and Biggart, 1988; Hofstede, 1991; Scott, 1992; Hickson and Pugh, 1995).
Hofstede (1991) and Scott and Meyer (1994) argue that national culture includes shared beliefs about organizational models. The “fit” between cultural values and organizational structure may be viewed as arising from implicit models of appropriate organization arising from the culture (Hofstede, 1991). We prefer organizational arrangements that are consistent with our basic cultural perspective. Going back a step further, it could be argued that the values symbolized by the elements of an organizational structure must be congruent with cultural values if these structural elements are to be included in the society's implicit models (Hambrick and Brandon, 1988).
These implicit models of appropriate corporate structures are based on institutional norms (Meyer and Rowan, 1977). As such norms constitute a shared belief system characteristic of the particular society, they can be considered components of the society's culture. For instance, the German system of corporate governance has long emphasized cooperative relationships among banks, shareholders, boards, managers, and employees in the interests of labor peace and corporate efficiency (Mintz, 2005). A relatively strong emphasis on collectivism and avoiding uncertainty underlies the system (Buck and Shahrim, 2005). This accounts, at least in part, for the emphasis in Germany on collective responsibility in company management (Mintz, 2005).
National Culture and Corporate Governance Structure
Institutional theory identifies formal norms about organizational structure – including norms for board structure – as components of national culture. If a board's structure expresses the culture of a society, then it should be related to the cultural dimensions identified by Hofstede (1980) (also see Mintz, 2005). Hofstede (1980; 1991) has identified four major dimensions of national culture: uncertainty avoidance, individualism/collectivism, masculinity/ femininity, and power distance. Uncertainty avoidance is a lack of tolerance for ambiguity. Individualism versus collectivism refers to a need for getting ahead versus a need to belong. The masculinity/femininity dimension is Hofstede's code for a preference for domination versus cooperation in superior/subordinate relationships. Power distance describes a preference for, or tolerance of, inequality. Hofstede (1991) and Hickson and Pugh (1995) have linked these cultural dimensions to organizational behavior. They found that high power distance is associated with strong authority and steep hierarchies, and that uncertainty avoidance is associated with formalization.
Hofstede's (1980) framework has been criticized on both empirical and theoretical grounds (e.g., one time, single company data; dimensions derived from factor analysis). Nevertheless, on balance, Hofstede's framework has been largely validated (e.g., Sondergaard, 1994) and provides a reasonable representation of national cultural attributes (see also Hickson, 1996). In addition, it seems obvious that board structure is also influenced by differences in countries' economic, political and legal systems (e.g., Roe, 1993). This study, however, was restricted to cultural differences as a first step. Country-level economic development was also included as a control in the models.
Uncertainty Avoidance. The concept of uncertainty avoidance concerns response to unstructured and ambiguous contexts. In high uncertainty avoidance cultures, members rely on clear procedures, well-known strategies, and well-understood rules to reduce uncertainties and cope with their discomfort with unknown situations (Hofstede, 1980). In low uncertainty avoidance cultures, there is a greater tolerance for uncertainty. Members are relatively more at ease with unfamiliar situations, and presumably more tolerant of different ideas, approaches, and concepts.
The institutional logic suggests that firms seek legitimacy within a society by conforming to societal norms and values. Therefore, we expect that the societal norms, as reflected in the cultural dimensions, should shape the legitimacy of corporate governance structures, and encourage firms to conform to those structural norms that appear to be legitimate. For instance, in high uncertainty avoidance cultures, detailed controls are more likely to conform to the societal norms than vague or informal controls (Hofstede, 1980). In such a situation, a formalization organization structure provides legitimacy as it relies on rules, procedures, and records to limit discretion and monitor activities, conforming to the societal norm of high uncertainty avoidance (Hofstede, 1980; Hofstede, Nuijen, Ohayv and Sanders, 1990). Close control of subordinates is imposed by issuing detailed instructions and/or limiting their authority.
People (and organizations) in high uncertainty avoidance cultures are more sensitive to risk, and perceive a higher level of risk and uncertainty in a given situation than in low uncertainty avoidance cultures (Nooteboom, Berger and Nooderhaven, 1997). Higher uncertainty avoidance implies a greater need to seek out information to reduce ambiguity (Zaheer and Zaheer, 1997). In nations low in uncertainty avoidance (e.g., the US) there is less acceptance of rules and less conformity to the wishes of authority figures. The opposite is true in high uncertainty avoidance nations such as Germany and Japan (Brislin, 1993). Tosi and Greckhamer (2004), for instance, found that the desire to minimize uncertainty will be reflected in the level of performance risk written into CEO compensation plans: uncertainty avoidance is negatively related to the proportion of variable compensation in CEO compensation packages.
For a society with high uncertainty avoidance cultures, the societal norms would imply that a board of directors has more outside members to appear legitimate by incorporating a broader range of expertise for managing uncertainty in its formal governance structure. Researchers have argued that more diverse groups have more skills and abilities with which to solve complex problems (Hill, 1982; Jackson, 1992) and consequently have greater information processing capabilities (Hambrick and Mason, 1984; Dutton and Duncan, 1987; Haleblian and Finkelstein, 1993). As outside directors have more diverse expertise and access to more varied sources of information than insiders, in high uncertainty avoidance cultures, boards with a higher proportion of outside directors would reflect the underlying societal norm for dealing with unstructured and ambiguous situations. Therefore,
Hypothesis 1a: There is a positive relationship between uncertainty avoidance in a firm's cultural environment and its percentage of outside directors on the board.
Similarly, in high uncertainty avoidance cultures, societal norms also suggest a preference for a firm's leadership structure, such that one type of leadership structure might appear to be more legitimate than another. In particular, when a firm's CEO is also the chair of the board, such consolidation of the two most senior leadership positions could convey a clear message of an established locus of command at the top of the firm and unambiguous decision-making authority, sending reassuring signals to uncertainty-averse stakeholders in the society (Finkelstein and D'Aveni, 1994; Boyd, 1995). Therefore, such consolidation of leadership positions is likely to confer legitimacy in high uncertainty avoidance cultures.
Hypothesis 1b: There is a positive relationship between uncertainty avoidance in a firm's cultural environment and the probability that its CEO will also chair the board of directors.
Individualism/Collectivism. Individualism refers to the preferred level of individual freedom and opportunity. In organizational terms, individualist values have been linked to preferences for individual decision making over group consensus (Hofstede, 1980). In societies emphasizing collectivist values, interpersonal relationships and group affiliation are highly valued. The self is defined as a part of the group; one's group memberships are an important statement of identity and achievement. Concerns over group welfare, equality, and loyalty are prominent, as aggregate interests tend to prevail over autonomous ones (Hofstede, 1991; Very, Lubatkin, Calori and Veiga, 1997; Schuler and Rogovsky, 1998).
For a society with high-individualism cultures, societal norms would imply that a board of directors has more outside members to appear legitimate by representing a more diverse range of individual interests. As individualistic cultures are more concerned with self-interest, corporate boards that include more outside directors are likely to signal that the interests of different stakeholders of the society are advocated by the board, conforming to the societal norm of respecting each individual's concerns. Therefore,
Hypothesis 2a: There is a positive relationship between individualism in a firm's cultural environment and its percentage of outside directors on the board.
Similarly, in cultures high on individualism, societal norms suggest a preference for a consolidated leadership structure, as this structure allows for strong individual leadership and the highest level of individual freedom at the top of the firm, and thus confers legitimacy. For instance, the early work in administrative theory, which was largely developed based on firms embedded in high individualistic cultures, advocates that organizations should be headed by strong leaders who set the strategic direction, command the lower levels, and monitor orders throughout the organization so that strategies are implemented as directed (Barnard, 1938; Andrews, 1971). Therefore, a strong leader may be viewed as an organizational characteristic that confers legitimacy, sending a signal to stakeholders in a society that a firm has a clear sense of direction managed by a single individual (Salancik and Meindl, 1984). This signal would conform to the societal norm in highly individualistic cultures. Therefore,
Hypothesis 2b: There is a positive relationship between individualism in a firm's cultural environment and the probability that its CEO will also chair the board of directors.
Masculinity/Femininity. Competitiveness, assertiveness, ambition, and the need to acquire material possessions are normally considered stereotypical masculine values. A value orientation with more emphasis on caring for others is described as more feminine (Hofstede, 1980; Very et al., 1997). Masculine cultures favor managerial decisiveness and a performance orientation, with an emphasis on proactive competitive behavior; in feminine cultures, a more supportive social orientation prevails, accompanied by a strong concern for the preservation of existing relationships. Performance-contingent rewards, merit pay, and management by objectives are practices consistent with a “masculine” culture, while attention to interpersonal relationships and quality of work–life issues are consistent with a “feminine” culture (Jaeger, 1986; Hofstede, 1991; Schuler and Rogovsky, 1998). Therefore, in a society with high masculine cultures, a board of directors with more management members would more likely conform to the societal norm of decisive, performance-oriented corporate action, and thus appear to be legitimate.
Hypothesis 3a (H3a): There is a negative relationship between masculinity in a firm's cultural environment and its percentage of outside directors on the board.
Similarly, in cultures high on masculinity, societal norms imply a preference for a consolidated leadership structure, as this structure embodies and symbolizes the decisiveness, assertiveness, and accomplishments of the CEO (Finkelstein and D'Aveni, 1994; Boyd, 1995). Firms with consolidated CEO and chair positions thus provide a valuable symbol that puts greater emphasis on the individual holding both positions, which conforms to the societal norm emphasizing masculine values. Therefore,
Hypothesis 3b: There is a positive relationship between masculinity in a firm's cultural environment and the probability that its CEO will also chair the board of directors.
Power Distance. Power distance is “the extent to which the less powerful members of institutions and organizations … expect and accept that power is distributed unequally” (Hofstede, 1991: 28). According to Hofstede (1980; 1991), relatively large differences in power are accepted and tolerated more in some cultures than in others. High power distance cultures prefer strong authority and steep hierarchies in part because they help preserve the existing social order and its related distribution of power (Hofstede, 1980). In high power distance cultures, organizations tend to be centralized, with power concentrated in a few hands, and they exhibit large differences in authority, salary, and privileges between those at the top and at the bottom. In low power distance cultures, organizations are more decentralized; there is more consultation in decision making, and independent action by less-powerful actors is valued and encouraged (Hodgetts and Luthans, 1993).
It follows, therefore, that in high power distance countries, a board with fewer inside directors appears to be more legitimate; it conforms to the societal norm for a large power distance between CEOs and their subordinates by minimizing the participation of subordinates in the formal corporate governance structure. Managers in these cultures are likely to accept this difference in power and status as it reflects the societal norm, and organizations tend to maintain appropriate differentials between different levels to appear legitimate. Therefore,
Hypothesis 4a: There is a positive relationship between power distance in a firm's cultural environment and its percentage of outsider directors on the board.
Similarly, in a society with high power distance, a consolidated leadership structure where a CEO simultaneously chairs the board confers legitimacy by conforming to the societal norm for a large power distance. A consolidated chair and CEO position exemplifies high power distance as this structure represents a steeper hierarchy, a more dominant CEO role, and a higher concentration of power, and it gives the CEO greater stature and political influence over the board (Westphal and Zajac, 1994). When the two positions are separated, the power of the CEO is likely to be counterbalanced by the separate board chair, which would not be consistent with the societal norm for a large power distance. Therefore,
Hypothesis 4b: There is a positive relationship between power distance in a firm's cultural environment and the probability that its CEO will also chair the board of directors.
- Top of page
- THEORY AND HYPOTHESES
These hypotheses were tested on a sample of 399 multinational manufacturing firms based in 15 countries with various cultural environments. The study examined national culture variables, board composition, and CEO/Chair consolidation, controlling for firm size, performance, capital structure, ownership structure, and country-level economic development. The dataset was originally collected as part of a larger project which did not consider the effects of national culture. The sample was taken from the third edition of the Directory of Multinationals (Stafford and Purkis, 1989), which profiles the world's 450 largest industrial corporations with sales of over one billion US dollars and significant international operations as of 1987. Fifty-one of these 450 firms were eliminated either because of missing data or because they had two-tiered boards (which are legally required in Germany, Finland, Denmark, and the Netherlands). Most of the 399 remaining firms were based in Japan (59), Western Europe (121), or North America (193). Only 26 were based elsewhere.
The data on these firms' board structure and ownership in 1987 were obtained from the Directory of Multinationals and checked and supplemented using other sources such as Standard and Poor's Register of Corporations, Directors and Executives, Moody's International, and corporate annual reports. These sources plus the Fortune 500 and Fortune International 500 listings provided data for the independent and control variables.
The “outsider percentage” variable was defined as the percentage of directors on the board who were not full-time managers of the firm; the available data did not reveal additional relationships of outsiders to the firm (where the outsider had some business or personal relationship with the firm). The CEO/Chair consolidation was coded as 1 if the Chair and CEO positions were held by the same person in 1987, and 0 otherwise. The power distance, uncertainty avoidance, individualism/collectivism, and masculinity/femininity measures for the various societies were taken from Hofstede (1980).
The models focus on the relationship between national culture and board composition and leadership structure. But other firm-level variables such as firm size, debt, and prior performance may also affect board composition, and it was considered important to include these control variables in the empirical analysis (Tosi and Gomez-Mejia, 1989). Firm size was represented by the natural logarithm of the firm's average annual sales in 1986 and 1987. Debt was represented by the average debt ratio the firm reported for the two years (calculated as long-term debt divided by total assets). Prior performance was represented by each firm's profit margin in 1986, using a one-year lag to gauge the effects of poor performance on governance structure (Harrison et al., 1988; Hermalin and Weisbach, 1988). The data on these firm-level control variables were obtained from the Directory of Multinationals (Stafford and Purkis, 1989).
Because of data limitations, dummy variables rather than continuous variables were used to indicate ownership concentration. Ownership concentration was coded as 1 if a single shareholder owned 5 per cent or more of a firm's shares, and 0 otherwise (see O'Reilly, Main and Crystal, 1988; Tosi and Gomez-Mejia, 1989). The data were obtained from the Directory of Multinationals. Each country's level of economic development, measured with GDP per capita, was also controlled for. Data for 1987 were obtained from the World Bank.
Table 1 shows descriptive statistics and correlations for the study variables. On average, outside directors accounted for about 58 per cent of the board members. However, there were wide variations, ranging from an average of only 9 per cent for Japanese boards to 83 per cent for French boards. US and European firms with single-tier boards tended to have outsider-dominated boards, but in Japan and Great Britain, insiders tended to dominate the boards. The CEO chaired the board in about 62 per cent of the firms. The countries represented in the sample also varied substantially in terms of Hofstede's national culture dimensions. The sample firms averaged about US$4 billion in annual sales (shown in the table in log form). In terms of ownership structure, about 36 per cent had at least one large shareholder. It was the companies in Japan and Western European countries which tended to have large shareholders.
Table 1. Descriptive Statistics and Correlationsa
| 1. Outsider percentage||.58||.28||1.0|| || || || || || || || || |
| 2. CEO/Chair consolidation||.62||.49||.32*||1.0|| || || || || || || || |
| 3. Firm size (log)||1.40||.87||−.03||.03||1.0|| || || || || || || |
| 4. Prior performance||3.67||4.55||.01||.08||.02||1.0|| || || || || || |
| 5. Debt||.31||.18||.01||−.01||.03||−.28*||1.0|| || || || || |
| 6. Ownership concentration||.36||.48||−.28*||−.26*||−.06||−.07||.15*||1.0|| || || || |
| 7. Power distance||42.4||9.03||−.21*||−.04||.19*||−.16*||.11*||.19*||1.0|| || || |
| 8. Individualism||80.5||16.1||.63*||.40*||−.13*||.18*||−.15*||−.40*||−.61*||1.0|| || |
| 9. Masculinity||63.7||18.4||−.60*||−.06||.08||−.05||−.11*||−.02||.35*||−.45*||1.0|| |
|10. Uncertainty avoidance||54.0||20.2||−.41*||−.19*||.18*||−.19*||.12*||.26*||.89*||−.83*||.56*||1.0|
|11. GDP per capital||17.34||3.23||.15*||.04||.08||−.18*||.04||−.16*||.09||−.28*||.09||.27*|
The hypothesized relationships between national culture and board composition were tested using regression analysis. (Tobit analysis was also performed for board composition as it is a percentage variable, with almost identical results.) The possibility of multicollinearity among the independent variables was tested using variance inflation factors (VIFs). The maximum VIF obtained in any of the models was 2.06, which is below the rule-of-thumb cutoff of 10 for multiple regression models (Belsey, Kuh, and Welsch, 1984; Ryan, 1997). Logistic regression was used to test hypotheses concerning leadership position consolidation.
- Top of page
- THEORY AND HYPOTHESES
The findings are reported in Tables 2 and 3, where the dependent variables are the percentage of outside directors on the board (Table 2) and CEO/Chair consolidation (Table 3). Three models are reported in each table. The first model includes the control variables only: firm size, prior performance, capital structure, ownership concentration, and economic development. As power distance and uncertainty avoidance were found to be highly correlated in the sample (r = .89), they were entered into the models separately (models 2–3, 5–6).
Table 2. OLS Regressions Relating National Culture with the Percentage of Outside Directors on the Boarda
|Independent variables||Model 1||Model 2||Model 3|
| Firm size||−.02||−.001||.001|
| Prior performance||.002||−.001||−.001|
| Ownership concentration||−.16***||−.01||.01|
| GDP per capita||.01*||.03***||.03***|
|National culture|| || || |
| Uncertainty avoidance|| ||.01***|| |
| Individualism|| ||.02***||.01***|
| Masculinity|| ||−.01***||−.01***|
| Power distance|| || ||.01***|
|Degrees of freedom||5||8||8|
Table 3. Logistic Regressions Relating National Culture with Chair/CEO Position Consolidationa
|Independent variables||Model 4||Model 5||Model 6|
| Firm size||.04||.15||.13|
| Prior performance||.04||.04||.04|
| Ownership concentration||−1.12***||−.11||−.01|
| GDP per capita||.001||.13**||.17***|
|National culture|| || || |
| Uncertainty avoidance|| ||.05***|| |
| Individualism|| ||.12***||.11***|
| Masculinity|| ||.01||.03*|
| Power distance|| || ||.09***|
|Cases of CEO/Chair position|
|−2 Log likelihood||501.7||421.8||410.2|
|Degrees of freedom||5||8||8|
Table 2 gives the board composition regression findings. All four cultural variables were significant (models 2–3). Hypothesis 1a was supported by our data whereby uncertainty avoidance was positively associated with the proportion of outsiders on the board (t = 12.02; p < .001; model 2). Hypothesis 2a was supported whereby individualism was positively associated with the outsider percentage (t = 19.28; p < .001; model 2). Hypothesis 3a was supported whereby masculinity was negatively associated with the outsider percentage (t = 15.8; p < .001; model 2). And finally, Hypothesis 4a was supported whereby power distance was positively associated with the outsider percentage (t = 10.28; p < .001; model 3). The cultural effects were found to dominate; the adjusted R2 for both model 2 and model 3 with the cultural variables was greater than .70. This was significantly higher than in model 1, representing only firm- and country-level controls, which had an R2 of less than .10.
Table 3 shows the logistic regression estimates for the CEO/Chair consolidation models. In models 5 and 6, all four cultural variables were significant as predicted. Hypothesis 1b was supported by our data whereby uncertainty avoidance was positively associated with a consolidated chair/CEO leadership structure (χ2 = 15.56; p < .001; model 5). Hypothesis 2b was supported whereby individualism was positively associated with a consolidated chair/CEO leadership structure (χ2 = 54.5; p < .001; model 5). Hypothesis 3b was supported whereby masculinity was positively associated with a consolidated chair/CEO leadership structure (χ2 = 6.47; p < .05; model 6). And finally, Hypothesis 4b was supported whereby power distance was positively associated with a consolidated chair/CEO leadership structure (χ2 = 24.2; p < .001; model 6).
Several significant effects were also observed for the control variables. The level of a country's economic development, as shown by its GDP per capita, was positively correlated with the percentage of outsiders on boards of directors and the extent of leadership position consolidation. Ownership concentration was negatively correlated with these variables, but the effects disappeared when cultural variables were added to the models. It is reasonable that national culture should influence ownership structure, and this would be a promising area for future research.
- Top of page
- THEORY AND HYPOTHESES
The observed success of the culture variables in explaining the corporate governance structures provides strong support for the argument that institutional norms, as embedded in a society's culture, affect organizational structure, at least at the board level. The cultural dimensions of uncertainty avoidance, individualism/collectivism, masculinity/femininity, and power distance have significant power to predict the composition and leadership structure of corporate boards in different countries. Firms based in uncertainty avoiding cultures tend to have more outside directors on their boards and tend to consolidate the CEO and chair positions. Firms based in societies that value higher levels of individual freedom tend to have a higher percentage of outside directors on their boards, and also consolidate the leadership positions. Firms based in societies that value personal dominance (masculinity) tend to have fewer outside directors, and also to consolidate the leadership positions. Firms based in societies that prefer high power distances are more likely to have a single leader as both board chair and CEO and fewer insiders on the board. Overall, these findings confirm our hypotheses about the relationship between national culture and board structure.
The results of the study contribute to our understanding of institutional theory in explaining the variations in corporate board composition and leadership structure across countries. Institutional structure norms were considered as components of national culture and were applied to infer the implications of these norms for board structure from national culture dimensions. This approach was highly successful in predicting the composition and leadership structure of MNC boards. By linking board composition to the national cultural environment, institutional theory has thus been shown to provide an explicit framework for analyzing variations in board structure across national boundaries.
These findings have important theoretical implications for corporate governance. The institutional theory perspective suggests that firms seek legitimacy within a society by conforming to societal norms and values, and the findings of this study confirm that corporate governance structures reflect prevailing societal cultural norms, and thus confer organizational legitimacy. This institutional logic extends the extant corporate governance research, and challenges the predominant agency theory logic for corporate governance. This study thus opens up several new research questions. First, future research should explore the possibility that the agency theory relationship might be culturally embedded; that is, the nature of the agency relationship may vary with the culture of a society. For example, research on Japanese corporate governance has suggested that the agency relationship needs to be modified in cultures high on collectivism (Crossland and Hambrick, 2007). Second, the findings of this study call attention to legitimacy motivations for corporate governance structure, as predicted by institutional logic, in addition to the efficiency- and monitoring-based motivations predicted by agency logic, consistent with DiMaggio and Powell's (1983) perspective that structural conformity is driven by institutional forces that are unrelated to the need for efficiency. Future research is needed to explore to what extent the two theoretical perspectives are complementary to each other in predicting corporate governance structures, and the boundary conditions for these relationships.
In addition, the findings demonstrate that the institutional environment of an MNC's home country can powerfully influence its corporate governance structure, and by extension, we conjecture, other aspects of organizational structure. The relationship demonstrated between Hofstede's national culture variables and board structure carries a clear message for students of international business: National culture may strongly influence organizational structure and should be considered in any international structural study. While the effects of national culture on other organizational characteristics were not addressed in this study, the findings suggest that transnational research addressing strategy, process, and other aspects of organizations may benefit from the consideration of national culture. A few studies have already taken this approach (see Ghoshal and Bartlett, 1990; Doz and Prahalad, 1991; Nelson, 1993; Westney, 1993). Hofstede's measures appear to capture cultural elements relevant to the shape and functioning of organizations.
The study also has managerial implications for understanding corporate governance in different countries. Corporate governance structures can be vehicles for firms to seek legitimacy, and thus conform to societal norms, as reflected in the cultural dimensions of the society. Pfeffer (1972: 219) argued that corporate boards “are used as if they were instruments with which to deal with the environment.” While Pfeffer's (1972) focus was on managing interdependencies with other organizations, this study extends Pfeffer's perspective, using institutional logic, to suggest that corporate boards can also be used to gain and maintain legitimacy by managing the firm's image in the institutional environment in which the firm is embedded. This would require a fundamentally different logic from the prevailing agency perspective. For instance, when considering board composition and board leadership structure, it is important according to institutional logic to consider what types of structures would conform to societal norms so that organizations can gain legitimacy by adopting structures consistent with these norms. The different dimensions of national culture, as a reflection of societal norms, would provide a clue as to the types of governance structures that appear to be legitimate for organizations to adopt. For instance, in countries valuing uncertainty avoidance, a corporate board with more outside directors and a consolidated CEO/chair structure might confer more legitimacy, thus encouraging the adoption of this type of governance structure for firms based in these cultures.
Limitations and Future Research
As this study focused on the insider/outsider composition and leadership consolidation of boards of directors, it was not able to examine the two-tiered boards in countries such as Germany.1 Second, while classifying board members as insiders or outsiders is common in such studies, a finer-grained measure differentiating among different types of outsiders (e.g., those with significant ownership in the firm, those with family ties to the firm, those having business relationships with the firm, former firm executives, and those who are independent of the firm's management) is clearly needed for future studies. Future research should also examine the extent to which national culture is related to the use of portfolios or bundles of board components (e.g., Finkelstein and D'Aveni, 1994; Boyd, 1995; Rediker and Seth, 1995) in different countries. The data applied in this study were appropriate for hypothesis testing, but it would nonetheless be more desirable to test for culture effects with more recent corporate data. Similarly, this study relied on Hofstede's (1980; 1991) national culture assessments. Research with more recent measures of cultural norms – either updated measures of Hofstede's dimensions or alternative measures – would be desirable.
Although the models successfully explained board composition and leadership structure, they addressed only the influence of national culture. They might be improved by including a broader array of national legal, political, and economic variables. Of course, the theory linking variations in these environment characteristics to variations in board composition must be developed. Extending the models in this way should be a top priority for research in this area.
It should be noted that the sample in this study consisted entirely of large MNCs operating in a number of countries. One would expect the effects of the home country culture on MNCs to be weaker than on domestic firms. Thus, the results represent somewhat conservative tests of the culture hypotheses. The findings, however, are consistent with recent studies, which have shown that even with the globalization of markets, home country institutional characteristics still play important roles for multinational firms (e.g., Norburn et al., 2000). The extent to which these models might be useful for explaining board composition in smaller MNCs and in firms which are not MNCs, and in explaining other aspects of organizational structure, are also important issues for future research.
It would also be very interesting to learn if there is a systematic relationship between board structure measures and subsequent firm performance. Past performance was controlled for in this analysis to gauge the effects of poor performance on governance structure. Subsequent firm performance was not considered, but this is a promising area for future research. Finally, board processes, such as the use of board committees, constitute another significant area deserving attention in subsequent intercountry research on corporate governance.
Jiatao (J. T.) Li is Professor and Head of the Department of Management of Organizations, and Director of Hang Lung Center for Organizational Research, Hong Kong University of Science and Technology. He received his PhD in strategy and international management from the University of Texas, and was previously with McKinsey & Company in Hong Kong. His current research interests are in the areas of strategy, organization theory, and entrepreneurship, with a focus on issues related to global firms and those from emerging economies.
J. Richard Harrison is an Associate Professor of organizations, strategy, and international management at the University of Texas at Dallas. His research interests include organizational culture, corporate governance, and computer simulations of organizational processes. He recently completed a book with Glenn R. Carroll, Culture and Demography in Organizations (Princeton University Press, 2006). He received his PhD in business from Stanford University.