UNIVERSITY OF AMSTERDAM
FACULTY OF ECONOMICS AND BUSINESS
Master Thesis
Supervisor: Dr. Ilir Haxhi
Second Reader: M.Sc. Francesca Ciulli
The impact of Corporate Governance on Wage Inequality across Countries and
the Moderating effect of National Culture
Student
FELIX TRAPP – 10825584 28.06.2015
Statement of originality
This document is written by Student Felix Trapp who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
1
Abstract
Extensive research has been conducted about possible factors influencing wage inequality across countries. However, the literature has been neglecting the role of Corporate Governance (CG) in this context so far. Therefore, this thesis aims to investigate how CG may contribute to cross-country differences in wage inequality. Moreover, we investigate the moderating effect of national culture. We argue that central aspects of CG and cultural dimensions are likely to predominate in a national environment and that a central issue of CG is in whose interest the company operates, and in turn, will play an essential role for the allocation of production returns among a firm's stakeholders. We analyze CG and its cross-country differences along with certain mechanisms of CG that may influence wage inequality. Furthermore, we hypothesize that a moderating effect exists, as culture influences both CG and wage inequality.
Using a sample of 41 countries, our empirical analysis indicates that central characteristics of CG influence wage inequality. First, our study reveals significant empirical support for the protection of minority shareholder rights, and to a smaller degree for the role of the stock market. Second, we find no empirical support for takeover activities or the moderating effect of national culture. Our findings on CG contribute to the ongoing academic debate on inequality in the way that it may help to explain why some nations are constantly having a problem to reach a greater degree of wage and economic equality. In addition, the associated managerial practices of CG can increase our understanding of wage inequality across countries, as for managers these findings may imply that the CG system they are operating within has stronger impacts on their wage and reward system than the cultural context. More precisely, the structure of incentives, and therefore wage inequality, may have to be adjusted to the prevailing CG context.
2 Table of Contents
Abstract...1
1. Introduction...4
2. Literature Review...8
2.1. Factors of Wage Inequality...8
2.2. Corporate Governance...12
2.2.1. Corporate Governance and Wages...16
2.3. National Culture...18
2.3.1. National Culture and Wages...20
2.3.2. National Culture and Corporate Governance...21
3. Theoretical Framework...23
3.1. The Influence of Corporate Governance on Wage Inequality...23
3.2. The Moderating effect of National Culture...29
3.2.1. Power Distance...32
3.2.2. Masculinity...33
4. Data and Method...36
4.1. Sample and Data Collection...36
4.2. Dependent Variable...37
4.3. Independent Variable...38
4.3.1. Protection of Minority Shareholder Rights...38
4.3.2. The Role of the Stock Market...39
4.3.3. Takeover Activity...39
4.4. Moderating Variable...40
4.5. Control Variable...41
4.6. Method and Model Specification...43
5. Results and Analysis...47
5.1. Bivariate Analysis...47
5.2. Regression Analysis...50
6. Discussion...57
6.1. Findings and Academic Relevance...57
6.2. Limitations and Future Research...63
7. Conclusion....66
3 List of Figures
Figure 1. The two main models of Corporate Governance...13
Figure 2. Corporate Governance and HRM System Approaches and Outcomes...17
Figure 3. Conceptual Model...35
Figure 4. Overview Results Hypotheses...63
List of Tables Table 1. Variables, Definitions, and Sources...42
Table 2. Representation of Models used; Model 1-5...45
Table 3. Representation of Models used; Model 6-11...46
Table 4. Collinearity Statistics...48
Table 5. Correlation Analysis...50
Table 6. Regression Results; Model 1-5...55
4
1. Introduction
The literature shows that income disparity has increased significantly in the last decades (Autor et al., 2008; Jaumotte et al., 2013; OECD, 2011). The variance of income increased not only between social groups such as genders, ethnical groups, educational groups or professional classes, but also within those specific groups (Juhn et al., 1993). Common ways to explain income differences are for example through demographic, political, environmental, and economic factors or skill-biased technological change and globalization (Autor et al., 2008; Kaasa, 2005). However, several studies argue that not all findings for the respective explanations are consistent with the empirical data (Kaase, 2005; Snower, 1999; Card & DiNardo, 2002).
In this context, recent research suggests that central aspects of CG may influence wage inequality, a factor the literature has been neglecting so far (Sjöberg, 2009). The institutional theory argues that organizational practices are under social influence and pressure and that a specific institutional context tends to support the implementation of certain practices and depress others (DiMaggio & Powell, 1983, 1991; Scott, 2001; Brewster et. al., 2008). Based on this, there is a variety of literature suggesting “that particular patterns of corporate governance are likely to predominate in national contexts reflecting the embedded nature of practices” (Brewster et al., 2008: 323).
Despite the fact that CG practices vary across institutional environments, the literature identifies two main models of CG, which are namely the Anglo-American and Continental-European model (Weimer & Pape, 1999; Aguilera & Jackson, 2003, 2010; Hall & Soskice, 2001). Studies argue that the behavior of a firm is influenced by the requirement of the dominant stakeholders (Konzelmann et al., 2006) and that organizational practices are likely to differ between the Anglo-American and Continental-European model (Gospel & Pendleton, 2003; Aguilera & Jackson, 2003). One of these organizational practices comprises
5
the human resource management (HRM), which is responsible for the management of rewards and accordingly for wages (Storey, 2001; Brewster, 1999; Dunlop & Galenson, 1978; Gerhart & Fang, 2005; Almond et al., 2003; Konzelmann et al., 2006). Consequently, CG is fundamentally “a question of in whose interest corporations are run and, as a result, will have important consequences for how the returns from production are distributed among the parties with a stake in the corporation” (Sjöberg, 2009: 520).
Furthermore, the literature argues that national culture, as an important aspect of international management, influences both CG and wages. On the one hand, central aspects of CG should be consistent with the existing cultural environment in a nation (Roland, 2004; Breuer & Salzmann, 2012). In addition, Bebschuk and Roe (1999) argue that the cultural background of a society had set the initial environment when CG systems, practices and structures were first established, which is in line with the argument of “path dependency”. For instance, the study of Licht et al. (2005) discovers a relationship between minority shareholders rights and national culture. The research of De Jong and Semenov (2006) finds significant connections between Hofstede’s cultural values and the protection of minority shareholders, corporate control, and the ownership structure. On the other hand, culture influences HRM practices and therefore wages (Aycan, 2005; Chiang, 2005; Gerhart & Fang, 2005; Dowling et al., 1999). For instance, social theorists suggest a connection between the cultural background of a country and the allocation of resources (Flanagan & Rayner, 1988; Leacock, 1978). Hence, depending on the national culture, a society may exhibit an equal distribution of compensation between employees or a broader wage dispersion (Aycan, 2005; Black, 2001; Chiang, 2005). In sum, economists should include cultural characteristics in their analyses of wage dispersion and CG (Mushinski & Pickering, 2000; Breuer & Salzmann, 2012).
6
Based on the outlined relationships above and due to the limited number of literature on the influence of CG on wage inequality, this study aims to gain more insights and shed more light on this relationship. Moreover, we test the moderating effect of national culture. No previous study has considered the possible moderating role of national culture onto this relationship. Therefore, to fill this gap in the literature, we investigate a two folded research
question. First, to what extent central aspects of Corporate Governance influence wage inequality across countries and second, to what extent national culture can moderate this relationship.
In this context, a central issue of CG is in whose interest the company operates, and in turn, will play an essential role for the allocation of production returns among a firm's stakeholders. We argue that cross-country differences in CG may influence wage inequality, as central aspects and mechanisms of CG influence managerial behavior concerning HRM practices. Hence, through the different prioritizations of the stakeholders, different outcomes across countries are expected. Thus, we suggest that the degree of shareholder pressure influences the structure of rewards and incentives, and therefore the wage distribution. Based on the assumption that the main intention of most shareholders is profit maximization, managers are pressured to prioritize this objective. This focus on shareholder value will lead to strong stock markets, frequent corporate restructuring through high takeover activities, and dispersed ownership, which may influence wage inequality (Sjöberg, 2009).
Furthermore, we argue that depending on the cultural characteristics of a country, a positive or negative moderating effect exists, because some societies demonstrate cultural values that favor wage inequality or equality and some cultural values tend to be associated with aspects of the Anglo-American or Continental-European CG model. More precisely, the cultural dimensions power distance and masculinity lead to an unequal distribution of wages between employees (Aycan, 2005; Black, 2001). In addition, societies emphasizing these two
7
dimensions tend to exhibit central aspects of the Anglo-American CG model (Breuer & Salzmann, 2012). Hence, based on the arguments about the relationship between culture and CG and the relationship between culture and wages, we argue that higher levels of masculinity and power distance further enhance the positive relationship between central aspects associated with the Anglo-American CG model and wage inequality.
In order to answer the research question, we test the proposed hypotheses with a multiple linear regression analysis. Based on a sample of 41 countries, we investigate the influence of protection of minority shareholder rights, the role of the stock market, and takeover activities, which are central aspects of CG, on wage inequality. Moreover, our analysis examines the moderation effect of the cultural dimensions power distance and masculinity. We obtain the data for inequality from the “The Standardized World Income Inequality Database”. To capture the moderating effect of national culture, we use the framework of Hofstede (1980a). Hofstede’s framework is frequently incorporated in cross-cultural studies because it demonstrates “clarity, parsimony, and resonance with managers” (Kirkman et al., 2006: 286). After controlling for wage centralization, unemployment, and GDP per capita, we find significant empirical support for the protection of minority shareholder rights, and to a smaller degree for the role of the stock market. However, we find no support for takeover activities or the moderating effect of culture.
This study contributes to the ongoing debate on inequality in several ways. If certain societies demonstrate CG characteristics and cultural aspects that encourage wage differences, these findings can help to explain why some nations are constantly having problems to reach a greater degree of wage and economic equality. For instance, countries exhibiting high income inequality are more vulnerable to political instability and experience higher mortality rates, and reduced social cohesion (Kawachi et al., 1996). For multinational corporations this could imply that they have to adjust their HRM practices and hence, the wage and reward
8
system, depending on the CG System and cultural context they are operating within. For example, high inequality in the wage dispersion as in the home country is likely to be declined in the host country.
This thesis is structured as follows. First, we discuss the relevant literature about wage inequality, corporate governance, national culture, and their relationships. In a next step, we outline the theoretical framework with the development of the hypotheses in section 3. Subsequently, in section 4, we illustrate the sample, the variables, and the method that are used to carry out this research, followed by the statistical analysis in section 5. Afterwards, we discuss the interpretation, academic implications, the limitations of the results, and future research suggestions in section 6. At last, section 7 contains conclusions about the conducted research and the managerial implications.
2. Literature Review
In this section, we illustrate the different relationships that exist between the variables included in the research question of this thesis.
2.1. Factors of Wage Inequality
To bring clarity to the topic of inequality, it is important to know that the literature uses the following expressions as earning inequality, wage inequality, wage distribution, income disparity, income differences, or income inequality in a similar way, indicating a similar phenomenon. In this context, wage inequality is an important part of income inequality, but income additionally includes e.g. earnings from investments, such as interest. Wage inequality is work related and derives from the income of employment. The reason why these terms are often used in a similar context is because they are interrelated, e.g. wage inequality indicates income inequality (Rueda & Pontusson, 2000; Bakis & Polat, 2015; Autor et al., 2008; Gottschalk, 1997). In general, there is a rich stream of research regarding the
9
different factors influencing income differences. To date, no universal theory including all hypothetical determinants is established. Most of the published papers on this topic focus either on one single factor or a limited number of factors. Therefore, this section illustrates important factors to get insights into the complexity of this topic.
In general, an interesting finding in the literature relates to the fact that income inequality is stable within a certain nation and over time, while differing significantly across nations (Li et al., 1998). A potential factor influencing this inequality could be the economic development of a country. In this context, Kuznets (1955) suggests that the relationship between GDP (Gross Domestic Product) and income inequality is shaped like an inverted U. Kuznets (1955) argues that as the GDP grows, inequality will increase at the beginning and then decrease. Based on the data for industrial countries, the author illustrates that income inequality increases with the industrial development, reaches a peak, and then declines with further development. Other researches support this relationship, while using different sets of data (Weede & Tiefenbach, 1981; Nielsen & Alderson, 1997). Contradicting, Ram (1997) identifies in his study about developed countries from 1951-1992 an un-inverted U shaped relationship between GDP and income inequality. Hence, as the GDP grows, the inequality decreases at the beginning and then increases. Therefore, the influence of a country’s wealth or economic development on income inequality is not clear.
Since the 1970s, a broadly accepted explanation for the development of income inequality relates to the development of wages and is based on a simple “supply and demand model” of the labor market and on the idea of a skill mismatch. This skill mismatch is caused by a faster increasing demand for better-educated workers than the supply for them, and hence, leading to an increase of their wages (Howell, 2002; Morris & Western, 1999). Simultaneously, the wages for employees at the lower end of the wage allocation decline compared to the wages of more trained and educated employees through a decreasing demand
10
for low-skilled labor (Morris & Western, 1999; Howell, 2002). In this context, the reasons for these demand changes and the emerging mismatch is not clear. Most of the research has concentrated on three interrelated causes, which are namely skill-biased technological change, globalization (international competition), and deindustrialization (Sjöberg, 2009).
First, the skill-biased technological change, in particular the rise of microcomputers, is responsible for a higher productivity of skilled labor compared to the productivity of less-skilled labor, which in turn, leads to a broader wage dispersion (Berman et al., 1998; Wolff, 2002; Autor et al., 1998). Second, researchers investigating globalization suggest that the growing amount of less skill-intensive products coming from nations exhibiting low wages raise the supply of unskilled workers in relation to trained and educated workers, which in turn, leads to forces pressuring the wages of unskilled labor downwards (Wood, 1995). Leamer (1996) suggests that not the amount of trade is important, but the decreasing prices for less skill-intensive products. In this context, trade with less skill-intensive nations impacts the prices in the home country and consequently the wage dispersion, independent from the amount of trade. Third, the deindustrialization hypothesis in general suggests that the increasing wage dispersion is the result of the growing high-wage service sector and the decreasing importance of the low-wage production and manufacturing sector, leading to higher compensations for skilled workers (Burtless, 1995).
Besides these findings supporting the idea of a mismatch between supply and demand and the causes for it, there are several studies, which argue that certain factors are not in line with the empirical data (Card & DiNardo, 2002; Snower, 1999). For instance, studies illustrate that the increasing wage inequality in the OECD nations is influenced to a larger extent through declining payments for the less trained and educated workers than through the rising remuneration for the trained and educated labor forces. Therefore, studies should consider factors like minimum wages, as they influence the wages of less skilled workers. In
11
addition, the development of wage inequality in the USA is not fully consistent with the rate of technological change (Card & DiNardo, 2002; Snower, 1999). Even though the supply and demand model illustrates useful insights into the topic of wage inequality, due to the inconsistencies in the empirical support, other aspects as political, social, and labor-market institutions should be considered. Regarding to this view, different institutions may impact wages and therefore the inequality (Sjöberg, 2009).
The institutional explanations for the distribution of wages have mainly focused on union density, the policy of the government, and wage-bargaining coordination and centralization (Sjöberg, 2009). Referring to union density, Freeman (1980) argues that the wage-setting is usually more homogeneous in unionized organizations, and equal wage dispersion is in general a goal of labor unions. Nevertheless, nations with a high union density may exhibit a higher wage inequality, if union membership is related to a wage premium, and workers with high wages tend to be involved in a union.
Additionally, Garret and Way (1999) argue that the size of the government sector may influence wage inequality. A possible explanation could be that public-sector unions tend to favor wage solidarity more than private ones. In addition, they are also under more political pressure emphasizing egalitarian wage dispersion.
Furthermore, several studies have illustrated that the wage dispersion is likely to be more compressed in nations with a wage-bargaining system that is considered as centralized compared to nations exhibiting a wage-bargaining system that is less centralized. Centralization changes the power of different parties in the wage negotiation and setting, which in turn, makes wage disparities more apparent and visible. Thus, wage-bargaining centralization leads to an empowerment of the low-wage unions, and in turn, to a stronger demand for a redistributive wage structure (Oskarsson, 2005; Iversen, 1996; Rueda & Pontusson, 2000; Wallerstein, 1999; Pontusson et al., 2002).
12 2.2. Corporate Governance
Finding a clear and universal definition of CG is a difficult task due to the diversity in the literature on this topic. According to Aoki (2000: 11) CG in general concerns “the structure of rights and responsibilities among the parties with a stake in the firm” or following Aguilera and Jackson (2010: 487) it is “the study of power and influence over decision making within the corporation”. The definition of Blair (1995: 3) is probably one of the most
comprehensive ones, CG concerns “the whole set of legal, cultural and institutional arrangements that determine what publicity traded corporations can do, who controls them, how that control is exercised and how the risks and returns from activities they undertake are allocated”. Despite the fact that CG practices vary across institutional environments, the
literature identifies two main models or systems of CG, which are namely the American and Continental-European model (Aguilera & Jackson, 2003, 2010). The Anglo-American model is described in terms of strong shareholder rights, short-term equity finance, effective markets for capital control, dispersed ownership, and flexible labor markets, whereas the Continental-European model is described in terms of weak shareholder rights, long-term debt financing, concentrated ownership (block holder), inflexible labor markets, and ineffective markets for capital control (Cernat, 2004; Aguilera & Jackson, 2003; 2010). Figure 1 illustrates the characterizations of the two models.
13
Figure 1: The two main models of Corporate Governance
Source: Cernat (2004) and Rhodes and van Apeldoorn (1997), cited in Sjöberg (2009: 522).
We outline central aspects of the two main CG systems to better understand their differences. For example, one fundamental aspect of the Anglo-American CG system is the concept of shareholder value. Based on this concept, firms simultaneously serve best their owners and the society, if they aim to maximize the profits for their shareholders (Parkinson, 2003; Lazonick & O’Sullivan, 2000; Fligstein & Brantley, 1992). In this context, firms are viewed as a set of assets with the main goal to increase short term wins for the shareholders. Furthermore, CG is mainly seen as a principal-agent problem due to the separation of
Dimensions Anglo-American Continental-European Labor-Related
Co-operation between social partners
Conflictual/minimal Extensive at national level
Labor Organizations Fragmented and weak Strong and centralized
Labor Market Flexibility Low internal flexibility, high external flexibility
High internal flexibility, low external flexibility
Employee influence Limited Extensive (e.g. working councils and/or co-determination
Capital-Related
Ownership Structure Dispersed Concentrated (bank/other corporations major shareholder) Role of Bank Financing Minimal role in corporate
ownership
Important both in corporate finance and control
Protection of Minority Shareholders
High Low
Market for Corporate Control
Hostile takeovers a
“correction mechanism” for management
Takeovers restricted
Role of Stock Market Strong role in corporate finance
14
shareholder (owner) and management (control) and the potential different interests of these two parties (Shleifer &Vishny, 1997). For instance, hostile takeovers are used to approach this problem. If a company thinks that they can increase the performance and profits with a different management, they may initiate bids against these companies (Roe, 2003; Froud et al., 2000). An additional way to deal with the principal-agent problem concerns the usage of profit related compensation, usually in the structure of stock-purchase plans and stock options (Gospel & Pendleton, 2003). In addition, developed capital and stock markets are fundamental in the Anglo-American CG system, because the prevailing option for dissatisfied shareholders is “exit” rather than “voice”. Furthermore, in this CG system, households finance the corporate bond and equity market, which in turn, leads to a relative high dispersion in the share ownership (Gospel & Pendleton, 2003; Aguilera & Jackson, 2003).
Contrasting to the focus on shareholder value, the Continental-European model of CG is also known as the “stakeholder” model because it takes the interests of all stakeholders into account and not only the interests of shareholders (Parkinson, 2003; Jackson, 2001; Aguilera & Jackson, 2003, 2010). In this CG model, the main stakeholders are frequently part of the boards, and occasionally of the management. Furthermore, banks are crucial financial organizations, which are converting deposits from private persons into loans available to firms (Gospel & Pendleton, 2003; Allen & Gale, 2000). Due to the strong connection between firms and banks, the relevance of capital markets decreases and the reliance by corporations on debt increases. Ownership tends to be more concentrated due to the fact that finance providers of long term debt usually have significant equity stakes (Moerland, 1995b). The tight connection between corporations and financiers through ownership or debt promotes a long term perspective, as opposed to short term earning increases in the shareholder model (Rhodes & van Apeldoorn, 1997). Additionally, the less liquid stock markets increase costs for an “exit” strategy of the shareholders, which in turn, encourages shareholders to achieve
15
“voice” in the management by holding an appropriate large stock (Moerland, 1995b).
Moreover, takeovers are more complicated in this CG system due to the relatively small relevance of the stock market (Moerland, 1995a).
Another important aspect of the CG system concerns labor. Referring to the Anglo-American CG system, labor is usually not known for demonstrating a long term involvement in the corporation. Hence, “exit” is the primary way for labor to demonstrate their dissatisfaction with the current working situation. The firm is usually viewed as an exclusive association of shareholders, while the workforce is seen as the “outsider”, whose rights are regulated by contract, which results in high mobility on the side of the workers (Rhodes & van Apeldoorn, 1997; Carr & Tomkins, 1998; Cernat, 2004). Furthermore, the short term orientation of firms could decrease their willingness to implement long term career plans for their workforce. This could lead to negative effects on employee’s decision about investing in company related education and training when the return of these investments is combined with high risk. As a result, the training and education systems are more likely to be fragmented in the Anglo-American CG system (Hall & Soskice, 2001).
Contrasting, the relevance of labor in the Continental-European CG system is observable not only on the “macro” but also on the “micro” level. Referring to the macro level, this CG system exhibits powerful and important labor unions, and a rather centralized wage-bargaining system. Concerning the micro level, the Continental-European CG system usually exhibits developed firm-labor structures of information transfer and collaboration. Even though the board is accountable for the decisions and actions of the company, it is common that they have discussions with employees before key decisions. This may lead to a more powerful position for workers in reorganizations and takeovers (Hall & Soskice, 2001; Aguilera & Jackson, 2003; Jackson, 2001; Gospel & Pendleton, 2003). Additionally, Gospel and Pendleton (2003) argue that the participation of financiers with a long term interest in the
16
corporation eases a better understanding for the importance of human capital and the concerns of employees. Furthermore, Hall and Soskice (2001) suggest that the closer relationship between the employees and the management of the corporation and the lower pressure through the financial sector are crucial for the higher occurrence of long term work tenures and firm related training in the Continental- European CG system.
2.2.1 Corporate Governance and Wages
The research on a direct relationship between CG and wages is very limited. However, based on the crucial differences between the two CG models, it becomes apparent that the interest in whose the firm acts and the management of organizational practices like HRM differ, depending on central aspects of the CG systems and the influence of the dominant stakeholders (Osterman, 1999; Aguilera & Jackson, 2003; Konzelmann et al., 2006). In this context, HRM is responsible for the management of rewards and accordingly for wages (Storey, 2001; Brewster, 1999; Dunlop & Galenson, 1978; Gerhart & Fang, 2005; Almond et al., 2003; Konzelmann et al., 2006). Hence, through the different prioritizations of the stakeholders, different outcomes across countries are expected (Konzelmann et al., 2006; Aguilera & Jackson, 2003). In companies where the shareholders (external stakeholder) are powerful and dominant, the managers are under pressure to prioritize their requirements and interests, which in turn, makes it more difficult to take the interests of internal stakeholders like employees into account. For example, emphasizing shareholder interests pressures the management to favor short term profit increases, obtained through cost cutting in the form of e.g. workforce downsizing (Konzelmann et al., 2006). The following Figure 2 illustrates the relationship between CG and HRM.
17
Figure 2. Corporate Governance and HRM System Approaches and Outcomes
Source: Konzelmann et al. (2006: 551).
Therefore, the idea for the existence of the relationship between CG and wage dispersion is based on the fact that a central issue of CG is in whose interest the company operates, and in turn, will play an essential role for the allocation of production returns among a firm's stakeholders. In sum, it may be possible that countries exhibiting central aspects of the Anglo-American CG system demonstrate higher wage inequality than countries exhibiting central aspects of the Continental-European CG system.
18 2.3. National Culture
Before we investigate the relationship between culture and wages and the relationship between culture and CG, we have to define culture in general and illustrate the framework to operationalise culture. Studies and cross-cultural research have defined the term culture numerous times and in different ways. The emergence of a variety of definitions relates to the difficulty to capture culture and its values and dimensions. Hofstede (1984) argues that it is necessary to study culture extensively to understand it completely due to its nature of being part of all aspects of a country. According to the research of Dodor and Rana (2007: 77), culture is the “whole complex of distinctive spiritual, material, intellectual and emotional
features that characterize a society or social group”. The study of Ball et al. (2010: 138) defines culture as the “sum total of beliefs, rules, techniques, institutions, and artifacts that characterize human population”. Hofstede (1984: 82) comes to the conclusion that culture is the “collective programming of the mind which distinguishes the members of one group or
society from those of another”.
Although culture is complex and can vary to a large extent between nations, it is possible that culture can have basic similarities (Hofstede, 1984). Therefore, culture is conceptualized and measured through several value dimensions, which are considered to be at the appropriate level between generality and detail (Hofstede, 1984; Aycan, 2005). In addition, these dimensions are able to create a connection between “individual, organizational and societal level phenomena” (Aycan, 2005: 1085). Up to now, there are five important cross-cultural studies that have been conducted to conceptualize culture, which are namely (1) Hofstede's (1980a, 2003) cultural dimensions; (2) the GLOBE Project (House et al., 2004); (3) Smith et al. (2002) research on event management; (4) the cultural theory (Schwartz, 1999); and the (5) World Values Survey (Inglehart et al., 2004).
19
According to Kirkman et al. (2006: 286), Hofstede’s cultural dimensions are frequently incorporated in cross-cultural studies because they demonstrate “clarity, parsimony, and resonance with managers”. Yet, studies criticize Hofstede’s cultural
dimensions for several reasons (Schwartz, 1994; McSweeney, 2002; Smith, 2002; Shenkar, 2001). The main points of criticism concern that the alteration of culture over periods of time is not captured, the reduction of culture to five dimensions, limiting the study to IBM, and ignoring the heterogeneity culture may have within a society (Sivakumar & Nakata, 2001; Kirkman et al., 2006). However, Tang & Koveos (2008) argue that the influence of Hofstede’s cultural dimensions might have been far greater than the influence of other
cross-cultural studies and research projects.
To set up his cultural framework, Hofstede (1980a) used data from over 116,000 questionnaires from 40 countries at IBM between 1967 and 1969 and again between 1971 and 1973. Hofstede later extended the available data for additional countries. These 40 countries are classified along the initial four dimensions power distance, masculinity, uncertainty avoidance, and individualism. Hofstede (1980b: 45) defines power distance (PD) as “'the extent to which a society accepts the fact that power in institutions and organizations is distributed unequally”. The second dimension concerns masculinity (MAS) and femininity
(FEM), with MAS expressing “the extent to which the dominant values in society are ‘masculine’ - that is, assertiveness, the acquisition of money and things, and not caring for others, the quality of life, or people” (Hofstede, 1980b: 46) and FEM capturing the opposite
of MAS. The third dimension captures uncertainty avoidance (UA), which “indicates the extent to which a society feels threatened by uncertain and ambiguous situations and tries to avoid these situations by providing greater career stability, establishing more formal rules, not tolerating deviant ideas and behaviors, and believing in absolute truths and the attainment of expertise” (Hofstede, 1980b: 45). Finally, the fourth dimension concerns individualism (IND)
20
and collectivism (COL). Hofstede (1980b: 45) defines IND as “a loosely knit social framework in which people are supposed to take care of themselves and of their immediate families only”, while COL “is characterized by a tight social framework in which people
distinguish between in-groups and out-groups, they expect their in-group to look after them, and in exchange for that they feel they owe absolute loyalty to it”. These dimensions can take values between 0 and 120. A higher value indicates a society with a higher level of power distance, higher masculinity, higher uncertainty avoidance, and higher individualism (Hostede, 1980a).
2.3.1 National Culture and Wages
Since the publication, Hofstede’s framework has inspired “thousands of empirical studies” (Kirkman et al., 2006: 285) and has “been widely used in cross-cultural research in
the past three decades” (Tang & Koveos, 2008: 1060). In particular, the framework of Hofstede has been used in relation to work-related attitudes, change management, decision-making, conflict management, reward allocation, human resource management, entrepreneurship, leadership, negotiations, social networks, group behavior relating to processes and personality, joint venture characteristics and performance, entry modes, alliance formation, foreign direct investment, social outcomes, research and innovation development, and organizational justice and motivation (Kirkman et al., 2006). From these various relations, the relationship between national culture and HRM is important to explain wage inequality across countries.
In general, HRM practices, which include the management of rewards and therefore wages, are influenced by national culture, and in turn, vary between different countries (Aycan, 2005; Schuler & Rogovsky, 1998; Storey, 2001; Newman & Nollen, 1996). “All these studies provide support for the proposition that national culture is a significant explanatory factor for cross-country differences in HR practices and policies” (Schuler &
21
Rogovsky, 1998: 164). Cross-cultural studies suggest that the compensation preferences of workers are culturally bound, which in turn, leads to the need to align HRM practices to cultural differences (Redding, 1994; Pennings, 1993; Hofstede, 1980a; Adler et al., 1986). Researchers argue that through the impact of culture on these practices the firm behavior relating to decisions about rewards for their workers is affected (Segalla et al., 2006; Schuler & Rogovsky, 1998). Previous studies emphasize for instance that in collective societies, companies are more likely to employ reward systems on a team basis and favor an equal allocation of rewards between workers. In contrast, in societies emphasizing individualism, the compensation of workers is based on their performance (Aycan, 2005). Furthermore, in societies demonstrating high power distance, it is likely that workers in a similar position receive different wages, because individuals and not their work are valued (Aycan, 2005). In addition, countries with high levels of power distance and masculinity tend to exhibit an extensively dispersed wage system (Black, 2001). Therefore, national culture impacts the allocation rules for distributing rewards or compensation of employees (Segalla et al., 2006).
In sum, depending on the national culture, a society exhibits for example an equal distribution of compensation between employees or a broader wage dispersion (Aycan, 2005; Black, 2001; Chiang, 2005). These findings lead to a potential relationship between cultural dimensions of a nation and wage inequality. Different cultural values favor either inequality or equality.
2.3.2 National Culture and Corporate Governance
As already mentioned, national culture does not only influence wages, it is also identified as an important factor for differences in CG. Therefore, it is necessary to investigate CG in relation to its embeddedness in different national and social environments (Breuer & Salzmann, 2012; Granovetter, 1985). For example, certain aspects of a nation are able to explain more of the variance in CG (39 % to 73 %) than observable company aspects (4 % to
22
22 %) (Doidge et al., 2007). One of those country characteristics is national culture, which is a crucial factor for explaining why economies differ in central aspects of CG (Licht, 2001).
This idea is based on the institutional model of Williamsons (2000). In his work, Williamson (2000) makes a distinction between four levels of social institutions. Level 1 considers informal institutions, such as cultural values and norms, Level 2 considers the formal institutional environment e.g. property rights and law, Level 3 considers CG systems, practices and structures, and Level 4 considers company behavior and practices. According to this, culture as Level 1 has an impact on institutions, which in turn, affect CG (Williamson, 2000; Breuer & Salzmann, 2012). Based on this argument, studies have investigated the relationship between dimensions of national culture and central aspects of CG systems.
For instance, the empirical results in the study of Licht et al. (2005) indicate support for the influence of culture on the protection of minority shareholder rights. Furthermore, Li and Harrison (2008) find support for the influence of Hofstede’s cultural values on the composition of boards. The research of De Jong and Semenov (2006) reveals connections between Hofstede’s cultural values and the protection of minority shareholder rights, corporate control, and the dispersion in the share ownership. In addition, the study of Haxhi and van Ees (2010) illustrates that culture is essential for understanding the worldwide diffusion of CG practices. In sum, central aspects of CG should be consistent with the existing cultural environment in a nation (Roland, 2004).
23
3. Theoretical Framework
The relationship between CG and wage inequality is relatively unexplored in the existing literature. Up to date, only one study has been conducted on this topic. Additionally, no previous study has considered the possible moderating role of national culture onto this relationship. However, based on the institutional theory, we argue that organizational practices, such as HRM, are under social influence and pressure, and that a specific institutional context tends to support the implementation of certain practices and depress others (DiMaggio & Powell, 1983, 1991; Scott, 2001). According to North (1991), the institutional environment consists of formal and informal constraints. In this context, the CG system of a nation is viewed as formal, whereas the national culture is categorized as informal. Yet, both aspects affect the dispersion of wages across countries, and the cultural background of a society had set the initial environment when CG systems, practices and structures were first established (Bebschuk & Roe, 1999; Aycan, 2005; Sjöberg, 2009; Chiang, 2005). Based on this background and the connection between the different variables, we first investigate the general relationship between CG and wage inequality, before we outline certain CG mechanisms that may influence wage inequality. We then examine the moderating effect of national culture onto this relationship.
3.1. The influence of Corporate Governance on Wage Inequality
In general, there is a stream of literature suggesting that due to the embeddedness of practices, certain characteristics of CG dominate within a national environment (Brewster et al., 2008). Studies argue that the behavior of a firm is influenced by the requirement of the dominant stakeholders (Konzelmann et al., 2006) and that organizational practices are likely to differ between the Anglo-American and Continental-European model (Gospel & Pendleton, 2003; Aguilera & Jackson, 2003). One of these organizational practices is HRM, which is responsible for the management of rewards and accordingly for wages (Storey, 2001;
24
Brewster, 1999; Dunlop & Galenson, 1978; Gerhart & Fang, 2005; Almond et al., 2003; Konzelmann et al., 2006).
In companies where the shareholders are powerful and dominant, the managers are under pressure to prioritize their requirements and interests, which in turn, makes it more difficult to take the interests of internal stakeholders as employees into account. For example, emphasizing shareholder interests pressures the management to favor short term profit maximization, which is obtained through cost cutting in forms of e.g. workforce downsizing (Konzelmann et al., 2006). Hence, the Anglo-American CG model pressures managers to rationalize their workforce and to prevent investments, e.g. training, with unsecure returns (Gospel & Pendleton, 2003), whereas the Continental-European CG model not only prioritizes the shareholder value maximization, but also considers the interests of the employees (Jackson, 2001; Aguilera & Jackson, 2003; Parkinson, 2003).
We argue that a central issue of CG is in whose interest the company operates, and in turn, will play an essential role for the allocation of production returns among a firm's stakeholders (Osterman, 1999; Sjöberg, 2009). In this context, we further argue that the structure of rewards and incentives, and therefore wage inequality, is influenced by the degree of shareholder pressure. Based on the assumption that the main intention of most shareholders is profit maximization, managers are pressured to prioritize this objective. This focus on shareholder value will lead to strong stock markets, frequent corporate restructuring through high takeover activities, and dispersed ownership, which may influence wage inequality (Sjöberg, 2009). This implies that the wage inequality could be higher in the Anglo-American CG system than in the Continental-European CG system, because of the associated managerial practices.
For example, Sjöberg (2009) argues that there is empirical evidence that CG and the associated managerial practices can increase our understanding of wage inequality. In this
25
context, CG and its cross-country differences along with certain mechanisms of CG may influence wage inequality. “The empirical assessment indicates that central aspects of these
institutions, such as the role of the stock market in channeling capital to corporations, the extent of mergers and acquisitions, and protection of minority shareholders are all related to cross-national differences and trends in earnings inequality” (Sjöberg, 2009: 519). Therefore, we outline possible mechanisms that play a role in this relationship.
As mentioned before, the literature frequently distinguishes between the Anglo-American and the Continental-European CG system. In this context, the argument developed in this section is not about separating nations between the two systems but rather that these systems of CG are applied in an analytical and efficient way to illustrate central characteristics of CG. However, we also like to mention, that these characteristics capture the same construct, which is CG and therefore have to be seen in the same context, as the mechanisms of CG that may influence wage inequality are intertwined.
The literature argues that CG in the Anglo-American system puts executive management under pressure to favor shareholders interests instead of the interests of other stakeholders, such as labor. In contrast, the Continental-European CG system takes the interests of labor into account (Gospel & Pendleton, 2003; Aguilera & Jackson, 2003; Parkinson, 2003). Based on the shareholder value model, the Anglo-American CG system exhibits a more narrow financial view of the company and hence, managers are under pressure to focus on the maximization of short term profits. The resulting management behavior tends to be associated with the breaking of working contracts, which offer long term career development opportunities to workers (Capelli, 1999; Osterman, 1999). Furthermore, since the 1980s, this behavior is viewed as a central aspect influencing the decrease of the redistributive nature of a company’s inner wage structure, which leads to decreasing concerns about a higher remuneration connected with more complex work or seniority (Osterman,
26
1999). In line with Sjöberg (2009), we argue that this may result in higher wage inequality, because companies no longer stick to a certain plan for promoting and hiring employees. In addition, employees are not able to rely on routine promotions and increasing wages within the same company anymore.
Another aspect of the Anglo-American CG system is the problem of “skill poaching”, resulting from the pressure on companies from e.g. shareholders to avoid long term employee expenses. Poaching makes the employment of trained workers of other companies vital (Capelli, 1999). Capelli (1999) argues that in the case where traditional reward systems, which are characterized by e.g. promotion ladders, are hardly present, the most evident approach to deal with this problem is to compensate important workers to stay in the company. Consequently, the resulting competition between companies to get trained employees may lead to higher wages for skilled workers and intensify the wage dispersion (Sjöberg, 2009).
As mentioned before, labor is usually not known for demonstrating a long term involvement in the corporation in the Anglo-American CG system. Hence, “exit” is the primary way for labor to demonstrate their dissatisfaction with the current working situation. The firm is usually viewed as an exclusive association of shareholders, and the workforce is seen as the “outsider” whose rights are regulated by contract, which results in high mobility
on the side of the workers (Carr & Tomkins, 1998; Cernat, 2004; Rhodes & van Apeldoorn; 1997).
Contrasting, in the Continental-European CG system interest of labor is taken into account, leading to e.g. a more powerful position of employees in a corporate reorganization. Furthermore, Gospel and Pendleton (2003) argue that the participation of financiers with a long term interest in the corporation eases a better understanding for the importance of human capital and the concerns of employees. In addition, the closer relationship between the
27
employees and the management of the corporation and the lower financial pressure through the financial sector are crucial for the higher occurrence of long term work tenures and firm related training in the Continental-European CG system (Hall & Soskice, 2001). Therefore, we argue that the greater influence of employees in a company e.g. through work councils and the general consideration of the interests of all stakeholders will lead to lower wage inequality in the Continental-European CG system, and in turn, to a higher wage inequality in the Anglo-American CG system.
Corresponding with Sjöberg (2009) and in line with the argumentation above concerning dominant shareholders, management is under pressure to prioritize their requirements, which may lead to higher wage inequality. To capture this aspect of CG, we hypothesize:
H1: A strong protection of minority shareholder rights positively influences wage inequality.
Furthermore, companies emphasizing narrow financial ambitions like changes of stock prices are usually characterized by investments with shorter payback times, which result in the avoidance of fixed long term expenses with unsecure profits, e.g. training and similar types of human resource improvements (Gospel & Pendleton, 2003; Carr & Tomkins, 1998; Hall & Soskice, 2001). Existing research also indicates that companies have to bear parts of investment costs in more general skills and not only in firm-specific skills (OECD, 2003). In this case, Black et al. (2007) argue that the equity market may have negative influence on the willingness of companies to pay for these expenses, since the return is to a large extent exposed to employee exit. Therefore, the extent of training offered by the company seems to be lower in the Anglo-American CG system, leading to a higher necessity for employees to acquire the knowledge and skills they think will be expected by the companies on their own. Getting access to these skills and knowledge may depend on resources, such as capital, that are not equally spread in a country. Accordingly, in line with Sjöberg (2009), we argue that
28
people with a higher income have the needed resources to undertake private investments in e.g. training, leading to increasing wage inequality in a society.
Additionally, using market-based transactions is another central aspect of the Anglo-American CG system, which supports the employment of market-type mechanisms to get key workers and additionally to ensure the loyalty of the workforce (Gospel & Pendleton, 2005; Pendleton et al., 1998). This could influence the wage allocation for the upper and the intermediate levels of wage earners. As outlined before, the principal-agent problem prevails in the Anglo-American CG system, leading to shareholders and investors implementing incentive based compensation systems to achieve managerial interests, which are similar to their objectives. The alignment of managerial remuneration with firm performance, e.g. through stock options and profit related compensation, could lead to higher compensations for the employees at the top of the company, which in turn, could increase managerial/non-managerial compensation differentials (Sjöberg, 2009).
Corresponding with Sjöberg (2009) and in line with the argumentation above, high pressure from the stock market and managerial practices concerned with the movement of stock prices may lead to higher wage inequality. To capture this aspect of CG, we hypothesize:
H2: A strong role of the stock market positively influences wage inequality.
Furthermore, the Anglo-American CG system is characterized by a frequent number of corporate restructuring in the type of takeover activities, which will lead to winners and losers in the workforce. In this context, the research of Froud et al. (2000) suggests a decline in the amount of workers in companies being part of a restructuring. However, there is also support suggesting raising remuneration for the employees who stay. Additionally, the transfer of work places from one company to another that takes place when dividing the companies into
29
different business units could decentralize the process of wage setting and therefore further influence the variability of wage (Sjöberg, 2009).
The restructuring of a company could augment the dependence upon contingent workers, e.g. consultants and temporary workers, which all have a different compensation. In this context, several studies suggest significant pay differentials between these contingent workers. The literature argues that contingent workers get a lower payment than permanent members of the workforce, while contingent workers in “high demand” are usually rewarded with a higher payment than long term workers (Capelli, 1999). Companies could use externals to maintain the company’s internal wage system stable during times of labor shortage.
Employees in short supply are able to “demand wages above the wages of those currently employed, which in turn, could be translated into pay raises for other workers if the new worker is integrated into the internal labor market structure rather than kept outside as a temporary worker or independent contractor” (Sjöberg, 2009: 525). Based on this, we argue that a higher dependence upon contingent workers could lead to increasing wage inequality.
Corresponding with Sjöberg (2009) and in line with the argumentation above, the managerial practices associated with higher takeover activities may lead to higher wage inequality. To capture this aspect of CG, we hypothesize:
H3: High takeover activities positively influence wage inequality.
3.2. The moderating role of National Culture
The existing literature argues that national culture significantly influences both CG and wage dispersion, and therefore, we hypothesize that a moderating effect may exist. For instance, on the one hand, depending on the national culture, a society exhibits an equal distribution of compensation between employees or a broader wage dispersion (Aycan, 2005; Black, 2001; Chiang, 2005). Hence, different cultural values can either be in favor of
30
inequality or equality. On the other hand, research also suggests that there is a significant connection between central aspects of CG and national culture (Licht et al., 2005; De Jong & Semenov, 2006; Li & Harrison, 2008; Roland, 2004). Based on this and in order to hypothesize the moderating effect, we first illustrate the general relationship between national culture and CG, followed by the relationship between the cultural dimensions of Hofstede and wages.
In general, the idea that culture impacts CG is based on the already mentioned framework of Williamson (2000) and the influence from Level 1 on level 2 and 3. According to this, culture as Level 1 has an impact on institutions, which in turn, affect CG (Williamson, 2000). Moreover, the influence of culture on CG can be illustrated in two ways. For instance, culture may encourage interest groups and policy makers to favor certain CG agreements over others. In addition, culture also could impede CG changes, which are not aligned with existing value preferences in a society (Breuer & Salzmann, 2012). Put it differently, Bebschuk and Roe (1999) argue that the cultural background of a society had set the initial environment when CG systems, practices, and structures were first established, which is in line with the argument of “path dependency”. In this context, the culture of a society may be viewed as the “mother” in the development of CG systems (Breuer & Salzmann, 2012).
The findings of Breuer and Salzmann (2012) indicate that societies in favor of the cultural dimensions harmony, egalitarianism, and embeddedness are more likely to exhibit the Continental-European CG system, whereas societies favoring the cultural dimensions of mastery, hierarchy, and autonomy are more likely to exhibit the Anglo-American CG system. These findings can be translated into the cultural values of Hofstede, as high power distance is similar to high levels of hierarchy, high masculinity is similar to high levels of mastery, and high individualism is similar to high levels of autonomy (Breuer & Salzmann, 2012). Hence, we argue that countries with high power distance and masculinity tend to exhibit central
31
aspects of the Anglo-American CG system, whereas in a country with low power distance and femininity the Continental-European CG system is more likely.
For example, masculinity is associated with values like influence and ambition, which correspond with the idea of influence coming from takeover activities (Breuer & Salzmann, 2012). Furthermore, the empirical results in the study of Licht et al. (2005) indicate support for the influence of culture on the protection of minority shareholders. Li and Harrison (2008) find support for the influence of Hofstede’s cultural dimensions on the composition of boards. The research of De Jong and Semenov (2006) reveals connections between the cultural values of Hofstede and the protection of minority shareholders, corporate control, and the dispersion in the share ownership. In addition, the study of Haxhi and van Ees (2010) illustrate that culture is essential for understanding the worldwide diffusion of CG practices. In sum, “Culture can be seen as a main determinant for the prevalent features of corporate governance
systems” (Breuer & Salzmann, 2012: 391).
To capture national culture we use the cultural framework of Hofstede (1980a), because of its broad acceptance in the literature (Kirkman et al., 2006). As already mentioned, there is no previous literature about the use of national culture as a moderator onto this relationship and the literature in general on this topic is not well established. Hence, this thesis investigates the cultural dimensions power distance and masculinity, as they are the ones with the most support from the existing research regarding the influence on wage inequality and CG. In addition, masculinity is correlated with individualism and therefore it captures a similar dimension. Furthermore, uncertainty avoidance is not identified to have a significant impact on wages or CG and hence, is not included in the analysis (Aycan, 2005; Black, 2001; Chiang, 2005; De Jong & Semenov, 2006; Li & Harrison, 2008; Licht et al., 2005; Breuer & Salzmann, 2012). Based on the relationship between culture and CG, we
32
further investigate the relationship between the cultural dimensions of Hofstede and wage inequality in order to hypothesize a moderating effect.
3.2.1. Power Distance
Cross-cultural studies suggest that the compensation preferences of workers are culturally bound, which in turn, leads to the need to align HRM practices to cultural differences (Redding, 1994; Pennings, 1993; Hofstede, 1980a; Adler et al., 1986). The literature argues that through the influence of national culture on these practices the firm behavior relating to decisions about rewards for their workers is affected (Segalla et al., 2006; Schuler & Rogovsky, 1998). Hence, national culture impacts the allocation rules for distributing rewards or compensation of employees (Segalla et al., 2006). In this context, power distance demonstrates the degree to which less powerful individuals of an organization or a society agree to an unequal allocation of wealth and power within the organization or the society. Hofstede argues that inequalities in wealth and power are accepted in some societies, while other societies attempt to reach higher levels of equality. In the organizational environment, PD expresses the degree of autocratic management behavior and the centralization of authority (Hofstede, 1980a, 1984, 1994).
Variations in PD not only influence the interaction between employees, they also affect the employee-employer relationship in general. More precisely, PD affects how superiors and subordinates work together. In the existing literature, several studies (Gatley et al., 1996; Hofstede, 1984) argue that in societies with low PD, authority is supposed to be less concentrated and the interaction between subordinates and superiors tends to be more equal, leading to a smaller amount of status differences. Furthermore, Gatley et al. (1996) argue that the wage dispersion between the executives and non-executives is relatively small, with a small degree of status and qualification differences. In addition, all members of the workforce are viewed as equal and superiors frequently discuss with subordinates about decisions and
33
future activities (Gatley et al., 1996). In contrast, societies with a high PD exhibit a very unequal relationship between superiors and subordinates, where the latter are more powerful and show their authority (Hofstede, 1984). Consequently, wages and status differ to a large extent between employees from top and lower levels (Gatley et al., 1996). Based on that, we argue that if individuals in a society with high levels of power distance are willing to accept high power inequalities and favor a distinct hierarchical order, then a society may be more likely to exhibit a higher wage inequality.
For example, studies illustrate that companies in high power distance societies exhibit a extensively dispersed wage system, as individuals and not their work are valued (Aycan, 2005; Black, 2001). In addition, countries with high power distance tend to exhibit central aspects of the Anglo-American CG system (Breuer & Salzmann, 2012). Hence, based on the arguments about the relationship between culture and CG and the relationship between culture and wages, we hypothesize that higher levels of power distance further enhance the positive relationship between central aspects associated with the Anglo-American CG system and wage inequality:
H1a: Ceteris paribus, high power distance positively moderates the relationship between a strong protection of minority shareholder rights and wage inequality.
H2a: Ceteris paribus, high power distance positively moderates the relationship between a strong role of the stock market and wage inequality.
H3a: Ceteris paribus, high power distance positively moderates the relationship between high takeover activities and wage inequality.
3.2.3. Masculinity
In line with the argumentation above, masculinity as a dimension of national culture also influences HRM practices, and hence, the firm behavior relating to decisions about
34
rewards for their workers is affected (Segalla et al., 2006; Schuler & Rogovsky, 1998). On the one hand, masculine societies are described in terms of success, competition, and achievement where people try to outperform each other. On the other hand, feminine societies tend to favor values like relationships with others instead of monetary achievements, quality of life, and caring (Hofstede 1984, 1994). Therefore, in societies emphasizing masculinity, individuals want to be acknowledged for their performance and increase their income, whereas feminine societies place higher value on the satisfaction of social needs than on individual achievements (Sliburyte, 2005).
The study of Dartey-Baah (2013: 42) suggests that “this type of culture tends to give the utmost respect and admiration to the successful achiever, who fulfills his ambition and demonstrates assertiveness and willingness to take risks in order to achieve goals”. Therefore, the ideals of managers in more masculine societies are independence, self-realization, and leadership, whereas in feminine societies solidarity among persons and social welfare is more central than personal self-realization (Gatley et al., 1996; Hofstede, 1984). Hofstede (1984) argues that people in countries with high levels of masculinity are motivated by success and accomplishment, whereas individuals in feminine countries are motivated by relationships. Consequently, interests in material gains, obtaining wealth and money, and financial arrangements are more likely to be favored in masculine societies (Hofstede, 1980b; Gomez-Mejia & Welbourne, 1991), whereas feminine societies tend to emphasize personal relationships and social needs instead of material gains and self-realization (Schuler & Rogovsky, 1998). Hence, we argue that if people in a masculine society are more likely to favor attributes like competitiveness, achievement, assertiveness, making money, and ambition, this society may be more likely to exhibit greater wage inequality, because people in feminine societies are more likely to emphasize values like a high quality of life, relationships instead of money, and overall welfare.
35
For example, high levels of masculinity result in an extensively dispersed wage system within companies (Black, 2001). In addition, masculine societies tend to exhibit central aspects of the Anglo-American CG system (Breuer & Salzmann, 2012). Hence, based on the arguments about the relationship between culture and CG and the relationship between culture and wages, we hypothesize that higher levels of masculinity further enhance the positive relationship between central aspects associated with the Anglo-American CG system and wage inequality:
H1b: Ceteris paribus, masculinity positively moderates the relationship between a strong protection of minority shareholder rights and wage inequality.
H2b: Ceteris paribus, masculinity positively moderates the relationship between a strong role of the stock market and wage inequality.
H3b: Ceteris paribus, masculinity positively moderates the relationship between high takeover activities and wage inequality.
The developed hypotheses result in the conceptual model as illustrated in Figure 3.
