Master Thesis for International Business & Management
Determinants of Subsidiary Centralization in the Central and
Eastern European Countries
By
Qingsong Zhou s1665251
First supervisor: Dr. F. Becker-Ritterspach Second supervisor: Prof.Dr. L. Karsten
University of Groningen
Faculty of Economic and Business
Landleven 5
9747 AD Groningen
Acknowledgements
I would firstly like to acknowledge Dr.F. Becker-Ritterspach and Pro.L.Karsten ’s understanding, support and provide valuable feedbacks for further research during the defense. Moreover, I would like to express my special thanks for Dr Gjalt de Jong, who gave the guidance when I began wrote the thesis. Special thanks also go to Thong, who answered my questions concerning the statistics.
I am grateful to my parents for their overwhelming love, financial support and great encouragement when I am living study far from home. Finally, I would like to thank my friends in Groningen who have shared the difficulties which arose during my stay here.
07 Jan, 2009
Abstract
This study uses multiple regression analysis to analyze those factors determining the level of centralization located in the higher organizational hierarchy in 25 MNCs. The secondary data is derived from the European Union’s fifth
Framework Project, Hofsteds’s study of national cultural distance and the Global Competitiveness Report (2001—2002). The sample consists of 212 subsidiaries in 4 Central Eastern European Countries (CEECs), including Poland, Hungary, Estonia and Slovakia. The empirical findings show that the level of centralization is associated with the characteristics of subsidiary, in particular the age of
subsidiary and the number of product lines are negatively related to headquarter and subsidiary centralization. On the other hand, none of the size of subsidiary, the extent of ownership of subsidiary, subsidiary environment and cultural distance is found to have significant effect on the level of centralization.
Table of contents
1. Introduction ... 5
2. Theoretical background ... 9
2.1 Factors on characteristics of subsidiary ... 10
2.2 Factors on the subsidiary Environment level ... 13
2.3Factor on the headquarter-subsidiary level ... 14
3. Theoretical model... 16
3.1 Hypotheses ... 16
3.1.1Factors on characteristics of subsidiary ... 16
3.1.2Factors on the subsidiary Environment level ... 19
3.1.3Factors on headquarter and subsidiary relationship ... 20
3.2 Framework ... 21
3.3 The empirical model ... 22
4. Methodology ... 24
4.1 Data ... 24
4.2 Variables and Measurement ... 25
4.2.1. Dependent variable ... 25
4.2.2. Independent variables ... 26
5. Results and discussion ... 32
5.1 results ... 32
5.2 Discussion ... 36
6. Conclusion, limitation and further research ... 39
7. References ... 41
1. Introduction
As the internationalization of the world marketplace, business organizations are growing, spreading and developing into multinational corporations (MNCs). Today the largest MNCs have activities in nearly all corners of the world.
Subsequently, the number of studies on the control mechanism within MNCs has increased considerably. The dispersed activities and different subsidiary contexts mean that headquarters may control subsidiaries embedded in different cultural, political, legal and educational systems across physical, i.e. geographical, distances and languages (Björkman, 2007). Different headquarters’ control mechanisms and different degrees of environment competition and subsidiary contexts make the headquarters’ control task more demanding. But how are foreign subsidiaries controlled? The existing research on the control includes formalization, centralization, output control, socialization, socialization
mechanisms and expatriate control (Björkman, 2007; Hedlund, 1981; Nohria & Ghoshal, 1994; Edwrads et al. 2002).
In this study I will only focus on the centralization. Because in the era of increasing competition, it is essential to assign appropriate decision-making autonomy to the subsidiaries to allow reactions to the change in local
environments to be timely, and without losing sight of the corporate strategic direction(Quester and Conduit, 1996). Besides, Gates and Egelhoff (1986) argued that centralization is an important aspect of the structure of the
Centralization is not an end in itself. Instead, it enables the organization to mobilize the local recourses to solve the various problems in the best possible way and reach the objectives they set. Although, the objectives are many, they ultimately relate to either to the maximization of profits (global profits in a MNC) or to the minimization of risk (Garnier, 1982). Other benefits arise from the centralizing a multinational company’s decision-making authority as resources can be pooled, and synergy among operations can be gained (Quester and Conduit, 1996). Adversaries of centralization believe that the lack of clear-cut responsibility at a subsidiary level can be detrimental. As decisions are made by headquarters, the local managers feel isolated and the perceived need for interaction disappears.
Previous researches have highlighted the trend in the level of centralization adopted by the organizations (Aylmer, 1970; Brandt and Hulbert, 1977). Factors influencing the level of centralization in MNCs can be categorized as relation to (1) the characteristics of parent company and (2) the characteristics of subsidiary and the relationship between headquarter and subsidiaries. Gate and
The aim of this study is to explore whether factors mentioned above have a impact on the level of centralization, and the research questions are:
1. How do the subsidiary characteristics affect the level of centralization?
2. How do characteristics of headquarter-subsidiary relationship affect the level of centralization?
3. How does the institutional environment of host country affect the level of centralization?
The empirical analysis of the paper is based on a survey carried out in 433 firms from five CEE countries under the European Union’s Fifth Framework Project “EU Integration and the Prospects for Catch-Up Development in Central and Eastern European countries (CEEC): The Determinants of the Productivity Gap”. Five countries are selected Hungary, Poland, Estonia and Slovakia in this study. There are several reasons why I want to investigate which factors determine the level of centralization in MNCs, in particular CEECs. First, the majority of studies about centralization deal with the development of subsidiaries and headquarters in advanced market economies, and few researches have been undertaken into the subsidiaries of MNCs that are operating in emerging and in transition
economies. Second, after the break down of socialist system in CEECs, the growing inflow of foreign direct investments into CEECs, and the transforming economy markets are economically fast-growing and structurally volatile
(Varblane et al, 2005). Third, the institutional environment of host countries and cultural distance are expected to have a large impact on the level of
The contribution of this paper is that I use cross-section data from 212
subsidiaries in 25 MNCs in 4 CEECs and study the effect of characteristics of subsidiary, institutional environment and relationship between headquarter-subsidiary on the level of centralization.
The remainder of the paper is organized as follows: Chapter2 presents
2. Theoretical background
Over the past two decades, research on the centralization has evolved from initially on the structural characteristics of MNCs, especially the management style, to the interdependence relationship between headquarter-subsidiary relationship and the subsidiary-specific issues. These researches show that the headquarter-subsidiary relationship still matters (for example. Birkinshaw et al, 2000, O’Donell 2000). Therefore, headquarter continues to exert considerable influence over the
organization as a whole. Goehle (1980) states that ‘one way to assess the decision-making authority of a subsidiary manager would be to consider the width of his “zone of discretion” in making programmed or basic decisions as well as
non-programmed or routine decisions’ (1980: 20). Goehle suggested that the width of the particular discretion zone is influenced by environmental factors of a particular
subsidiary surrounding and the corporation characteristics itself. The author also proposed that these effects can be classified into two groups of characteristics: corporate and environment factors. Later study by Hedlund (1981) on the autonomy of subsidiaries and the formalization of headquarters-subsidiary relationships in Swedish MNCs is also based on these two aspects. Besides, the interdependence relationship between headquarter and subsidiary also plays a critical role in
determining the authority of the local manager. In Björkman’s study (2007), this idea was used. Thus the study will be constructed based on the above points of view: the subsidiary characteristics, environment factors and the relationship between headquarter and subsidiary. In the following theoretical part, only some important factors influencing the level of centralization are reviewed.
Throughout the centralization literature, the terms of decentralization and
Centralization is considered an important aspect of the structure of the headquarter-subsidiary relationship in MNCs. Because on the one hand, headquarters can make the decisions concerning the subsidiary activities for subsidiaries; on the other hand, headquarters can leave a certain degree of decision-making authority to subsidiaries. On the opposite side of centralization, it is decentralization, which is often used as a synonym to autonomy (for example. Garnier, 1982). Decentralization refers to the notion of the headquarter
delegating decision-making power to the subsidiary, while autonomy contains a trace of subsidiary assertiveness (Björkman (2007). As a result, autonomy is bargaining process between the headquarters and the subsidiary (Taggart 1997, Taggart& Hood 1999).
2.1 Factors on characteristics of subsidiary
Once the terms of centralization, decentralization and autonomy are defined, this chapter is to review the existing literature related to the relationship between characteristics of subsidiary and centralization. By now the research on characteristics of subsidiary has been received a lot of attention (for example: Björkman,2007), and the degree of decision-making latitude allowed to
subsidiaries has been found to be influenced by a few of their characteristics (for example, Goehle, 1980; Hedlund, 1981; Garnier, 1982; Gates and Egehoff, 1986; Taggart and Hood, 1999; Björkman, 2003; Varblane et al., 2005): including age of subsidiary, size of subsidiary, the extent of ownership of subsidiary and the number of product lines, but the findings are rather mixed. Each of these variables are discussed below.
Age of subsidiary Age of subsidiary is one factor that has been considered
(Robock et al., 1973: 94). Moreover, Garnier (1982) reasoned that older
subsidiaries, having more experience, represent less risk to the MNC. Therefore, older subsidiaries can enjoy more autonomy than the younger ones. This
argument is supported by Gate & Egelhoff (1986) for financial centralization. However, there is little indication of any support for marketing or manufacturing centralization. But none of the MNC researchers found significant results
between subsidiary age and the level of centralization. Stopford and Wells (1972) maintained that some young subsidiaries appear to have more autonomy than others. While Ganier studied 42 subsidiaries in France and 102 in Mexico and found that age of affiliate has little direct influence on the degree of autonomy. Most of these findings have also been confirmed by recent research (for example, Taggart and Hood, 1999; Young and Tavares, 2004).
Size of subsidiary Size of subsidiary has been frequently considered another factor
little direct relationship or no relationship exists between size of subsidiary and the level of centralization. Ganier (1982) indicated that the absolute size of subsidiary has little direct influence on the degree of autonomy. Gate and Egelhoff (1986) examined centralization in fifty large MNCs from US, UK and Europe. The results suggested that the absolute size of a company might have a different influence on different types of centralization (e.g., marketing centralization, manufacturing centralization) within an organization.
Extent of ownership of subsidiary It has furthermore been argued that the extent of
Number of product lines Finally, the number of product lines is considered as
(internal) factor variable that may impact on the level of centralization (Kirpalani, Laroche and Darmon, 1988; Sim 1977; Garinier, 1982). Garnier(1982) argued that firms with many products tend to give more autonomy than those with few products. The underlying reasons are as follows: first, firms with many products, the dependence rate of each product is very low, therefore, the level of risk is lower, and these firms will tend to be more decentralized and give more
autonomy. Second, the whole group has many products, and the opportunities of some products will have been designed by foreign affiliates specifical to their market. These will be more difficult to control than standardized products
(Garnier, 1982). Besides, Gates and Egelhoff(1986) also examined the impact of the number of product lines on the degree of decision-making authority, and the results demonstrate that there is no relationship between them. This argument is supported by the Kirpalani et.al (1988).
2.2 Factors on the subsidiary Environment level
The other important aspect that can affect the level of centralization is subsidiary firm’s environment factors.Garnier (1982) emphasized that the actual degree of autonomy granted to a given affiliate is influenced by external environment factors. So environment factors of subsidiaries play a crucial role in the MNC research, especially the legal environment a subsidiary is located. In the following paragraphs I will review the relationship between institutional environment and the level of centralization.
Legal environment Garnier (1982) argued that the legal environment captures
effect, much less has been analyzed concerning its impact on the level of centralization.
Rosenzweig and Singh (1991) argued that the relative influence of home and host country institutional environments was dependent on a set of contextual, strategic, and structural variables. In addition, Kostova (1999) argued that organizational practices and routines were shaped by the institutional
environment, and the successful transfer of these practices and routines from the parent firm to the foreign subsidiary depends on the distance between the host and home environments. Xu and Shenkar (2002) asserted that institutional distance provides an alternative explanation for MNE behavior. The construct has been linked so far to two aspects of MNE operations: (1) the establishment of legitimacy in the host country (Kostova and Zaheer, 1999) and (2) the transfer of strategic orientations and organizational practices from the parent firm to the foreign subsidiary (Kostova, 1999). Kostova and Zaheer (1999) then proposed that the larger the institutional distance, the more difficult it is for the MNC to establish legitimacy in the host country and to transfer strategic routines to foreign subsidiaries. Garnier (1982) indicated that the greater the institutional environment differ, the less understandable the message issued by that
environment (coming from customers, suppliers, government officials), will be to the home executives; therefore, the larger the risk of misinterpreting these
messages, and the stronger the tendency to leave most decisions in the hands of local executives who can react immediately.
2.3Factor on the headquarter-subsidiary level
Cultural distance has frequently been used in international management studies (Björkman, 2007), and it is used to describe headquarter and subsidiary
Cultural distance Similar to the institutional environment, the amount of cultural
distance between headquarters and the subsidiary in question might be an important factor influencing the level of centralization.
Cultural distance refers to the degree to which the shared norms and values in one country differ from those in another country (Hofstede, 2001; Kogut and Singh, 1988). The larger cultural distance, the stronger tendency to leave the decision-making authority to local executives, because the larger cultural differences make it increasingly difficult to understand, control, and predict the behavior of others (Elsass and Veiga, 1994). Besides, according to Neal (1998), some of the most notable difficulties associated with cross-cultural interactions between headquarter and subsidiaries include those associated with
understanding and communication, particularly those associated with differences in perceptions of the same situation.
However, little previous studies on centralization are available in this field. Garnier(1982) found a positive, but not significant, correlation between
“perception of differences in executives’ attitudes and values” and the level of centralized decision-making. In the same way, the finding is supported by Wolf (1994), although again the results were not statistically significant. On the other hand, Hedlund (1981) indicated that with the increasing cultural distance
between the headquarters and a subsidiary, no effect is found on the
centralization. Besides, Björkman (2007) showed that the cultural distance had no effects on centralization. Even although he used different methods, the same findings were obtained.
In summary, the empirical analysis will take into account the factors from
3. Theoretical model
3.1 Hypotheses
The preceding chapter gives a guide in establishing the related hypotheses at three levels: (1) the characteristics of subsidiary, (2) the environment factors, (3) the relationship between headquarter and subsidiary. So hypotheses involving charactering the relationship between headquarter-subsidiary and related argument are derived as follows:
3.1.1Factors on characteristics of subsidiary
Age of subsidiary The research shows that age to be a determinant of level of
participate in a higher degree of decision-making. Later empirical studies by Ganier (1982) and Gate and Egelhoff (1986) also show that negative relationship between the age of subsidiaries and the level of centralization due to the older subsidiaries, having more experience, thus presenting less risk to the parent company and enjoying more autonomy. A recent empirical study by Taggart and Hood (1999) stresses that the degree of subsidiary autonomy positively correlates with its age. Finally, the most recent study by Young and Tavares (2004)
confirmed this argument with descriptive research. Consequently, the first hypothesis is derived as:
H1. The level of centralization is negatively correlated with the age of
subsidiary.
Size of subsidiary The impact of the size of subsidiary on the level of
centralization has produced mixed results. Some studies found positive
relationship between them (for example Child (1973a), Gate and Egelhoff (1986), Harzing (1999). Moreover, Garnier(1982) and Taggart & Hood(1999) found there was no effect onthe level of centralization.
Many studies examined headquarter vs. subsidiary influence on the decision-making in various domains of MNCs operation. Hedlund (1981) suggested that a large subsidiary would ceteris paribus have greater resources than a small
subsidiary autonomy until a certain size and afterwards the starts to lose autonomy again. The quadratic inverted-U model supports and extends
Hedlund’s (1981) proposition. In addition, Taggert and Hood (1999) tested distinct proxies for subsidiary size, namely employment and sales. Their empirical findings established a positive relationship between employment and autonomy which was not significant, and a negative relationship between sales and autonomy. Based on the above standpoint, so I propose the following hypothesis:
H2. The level of centralization
isnegatively correlated with size of
subsidiary.
Extent of ownership of subsidiary MNC researchers and organization theorists
support the negative direction between the extent of ownership and centralization. This argument is supported by Ganier(1982), due to the fact that the higher the extent of ownership in foreign subsidiaries, the lower is the subsidiaries’ legal dependence. Hence, there is tendency to decentralization, and the empirical findings confirmed that the percentage of the responding affiliate’s equity held by parent is significant and positively correlate with autonomy. The investigation of Gate and Egelhoff (1986) also substantiated for financial centralization. In
addition, the empirical researches by Alsegg (1971), Youssef (1975) Picard (1977) and Hedlund supported that the relationship of the extent of ownership of
subsidiaries positive correlated with the autonomy as well. However,
perspective between the ownership of subsidiaries and centralization, I firmly consider that if a subsidiary will have absolute equity majority in its unit, it has higher level of autonomy. So the third hypothesis is following:
H3: The level of centralization is negatively correlated with the extent of
ownership of subsidiary.
Number of product lines Little research has been done on the number of product lines and the level of centralization. Garnier (1982) argued that firms with many products tends to leave more decision-making authority to local executive than firms with few products, due to on the one hand, that firms with many products, dependence on each product is lower, therefore the level of risk is lower, and these firms will tend be more decentralized and are given more autonomy. On the other hand, if a whole group has many products, and the chance of some particular products will have been designed by foreign affiliates, it is difficult to control than standardized products, so more decision-making authorities are left to subsidiaries. Ganier(1982) conducted 42 subsidiaries in France and 102 in Mexico, the findings showed that the number of products is significant in both France and Mecixo. Based on Ganier’s arguments, I put forward the hypothesis:
H4: the level centralization is negatively correlated with the number of
product lines.
3.1.2Factors on the subsidiary Environment level
legal environment Organizational theorists seem to agree that an organization
environment, and successful transfer of these practices and routines from the parent firm to the foreign subsidiary depend on the distance between the host and home environments. Moreover, Rosenzweig and Singh (1991) argued that the relative influence of home and host country institutional environments was dependent on a set of contextual, strategic, and structural variables. The legal environment reflects the institutional distance. In my opinion, CEECs are embedded in less developed institutional systems with more corruption and less developed regulatory regime for copyright protection, so the institutional distance between the headquarter and subsidiary will become larger than those that in the developed institutional system countries are, hence the headquarter granted a higher degree of decision-making authority to the subsidiary manager. In other words, the larger institutional distance will make the subsidiaries have a higher control of the subsidiary, which means higher autonomy. Although the empirical findings by Ganier(1982) and Gate and Egelhoff (1986) showed that the environment factors have little impact over the level of centralization. In the same vein, Kirpalani (1988) supported this argument. However, I argue that in the less developed institutional environment of CEECs, there is a higher institutional distance between the developed home and host countries, so it is more difficult to transfer organizational practices, routines and a set of contextual, strategic and structural variable, therefore, logically the fifth hypothesis is derived as:
H5.The level of centralization is negatively correlated with the legal
environment of subsidiary.
3.1.3Factors on headquarter and subsidiary relationship
Cultural distance Björkman (2007) provided evidence that an increasing
cultural distance leads headquarter to choose for less involvement in subsidiary control. Since managers are usually not familiar, comfortable, or even in
cultural distance countries (Caves, 19996). Besides, Elsass and Veiga (1994) argued that the most distinct difficulties with increasing cultural distance include those associated with understanding, communication, controlling and predicting the behaviors of others. Moreover, Harizing(1999) indicated that cultural distance probably makes the achievement of shared values more difficult. Harzing’s
empirical finding use extensive data from over one hundred multinationals with headquarters in nine different countries and subsidiaries located in 22 nations, the results showed a significantly negative relationship between cultural distance and centralization.
On the contrary, regardless of different methods used in Björkman (2007)’s investigation in China and Finland, the conclusion demonstrated that cultural distance did not have an impact on centralization, and this findings is supported by Hedlund (1981). However, Garnier(1982) examined 42 subsidiaries in France and 102 in Mexico, the finding found a positive , but not significant correlation between attitudes and values and the level of centralization decision-making. Wolf(1994) also supported Gariner’s standpoint. However, in spite of different arguments concerning the effect on centralization, I argue that an increasing cultural distance reduces level of control between countries. In others word, the higher the culture distance is, the less control of a subsidiary, therefore:
H6. The level of centralization is negatively correlated with cultural distance.
3.2 Framework
Figure 1 overview of the thesis
3.3 The empirical model
Multiple regression analysis is performed to analyze the determinants of the level of centralization in the MNCs. The multiple regression is used for most research to estimate which factors have effect on the dependent variable. Data applied in the analysis is collected from 24 countries. These countries are listed in
Appendix Table 2. The multiple regression model estimates with the following specification:
Factors on subsidiary level 1. Size of subsidiary 2. Age of subsidiary
3. The number of product line 4. The extent of ownership of
subsidiary The level of Centralization Characteristic of Subsidiary Legal environment
6. Factor on HQ-Sub level
Cultural distance
Headquarter- subsidiary
(HQ-Sub) relationship
Ŷ = β0 +β1X1 + β2X2 +β3X3 + β4X4 + β5X5 +β6X6+ε
Where Ŷ = the level of centralization X1 = the age of subsidiary
X2 = the size of subsidiary
X3 = the number of product lines
X4 = the extent of ownership of subsidiary
X5 = the legal environment
X6 = the cultural distance between headquarter and subsidiary
β0 = intercept, the value of Y when all X values are 0
β1, 2, 3, 4, 5, 6, = the slope of the regression associated with Xi
4. Methodology
4.1 Data
The secondary data used to estimate the empirical model is derived from EU fifth project, Hofsteds’s study of national cultural distance and the Global
Competitiveness Report (2001—2002). Most of the analysis is based on the special questionnaire for Foreign Investment Enterprises (FIEs), which was undertaken in 2001—2002 under the EU Fifth Framework Project: “EU Integration and the Prospects for Catch-Up Development in (CEECs): The
Determinants of the Productivity Gap”. The project was carried out by the ‘Institut für Wirtschafsforschung Halle’. Questionnaire was sent out to mainly
manufacturing enterprises with foreign ownership in Poland, Hungary, Estonia, Slovakia and Slovenia. The response rate was 20.8% or 458 questionnaires. The biggest number of responses (33.4% of the total) came from Poland, followed by Hungary with 18.6%, Slovakia 17.0%, Estonia 15.9% and Slovenia 15.1%. The selection of subsidiary for this study is conducted in the following stages. First, one of contingency variables is cultural distance, but in Hofstede’s research that is not any available information about Slovenia, so Slovenia is excluded from the sample. Second, subsidiaries which don’t provide information about the country of origin of foreign investor and those whose headquarter are multinational (more than one headquarters) are also deleted because nationality is used to calculate cultural distance and institutional distance between headquarter-subsidiary. As far as the establishment year of subsidiary and the size of subsidiary are
there are 212 enterprises in this sample, 90 subsidiaries from Poland, 44 from Hungary, 42 from Slovakia and 36 from Estonia.
4.2 Variables and Measurement
4.2.1. Dependent variable
In the study the dependent variable ---- the level of centralization--- is measured according to the Gate and Egelhoff (1986). This method is widely used by other previous researchers (for example, Hedlund, 1988; Vachani, 1999; Varblane, 2005; and Johnston, 2005). Centralization refers to the decision-making located at higher level in the organizational hierarchy (Alexander1991, Child 1972/1973a, Pugh et al, 1968). The levels are ranked: (a) on your own only, (b) mainly on your own, (c) mainly by your foreign owner, (d) by your foreign owner only. The level of centralization is determined by 8 business functions, which are described in Table 1. These functions consist of two parts. The first is marketing decisions, and the second is financial decisions. The individual decision score is averaged together to construct centralization scale. The scores in the database are given to indicate the level of centralization, and it ranges from 0 to 1. 0 means
Table 1
Decisions used to measure level of centralization
Marketing decisions Financial decisions
Determining the product price Accounting and Finance of operations Market research Investment finance Distribution, sales After-sales services Advertisement Marketing
4.2.2. Independent variables
This study also measures 6 independent variables, and these variables belong to three categories. First, four variables are measured at the subsidiary level
(characteristics of subsidiary): size of subsidiary, age of subsidiary, the extent of ownership of subsidiary and the number of product lines. Second, at the
subsidiary environment level, legal environment is used to analyze the environment factor of the host country. Third, the relationship between headquarter and subsidiary, cultural distance is used to for the analysis. The measurement and the sources of the variables are described below:
1 Size of subsidiary: is measured as the number of employees in the subsidiary.
2. Age of subsidiary: is measured as the number of years since the subsidiary was established. Since it was undertaken during 2001—2002, so the age of subsidiary is calculated by subtracting 2001 from the year subsidiary was established.
3. The extent of ownership of subsidiary: is a dummy variable. I code the value of 50% of the local shareholders’ ownership of the subsidiary as 0, and the value of 100% percentage of the local shareholders’ of subsidiary as 1.
4. The number of product lines: is measured how many product lines the subsidiary currently has.
5. Legal environment: is measured as the institutional distance between two countries. Kostova (1999) argues that Institutional distance is a recently developed construct in strategic management literature that captures the differences between the institutional environments of two. It is based on three pillars of institutional environment, namely, the regulative, normative and
cognitive pillars as defined by Scott’s (1955). Whereas regulative pillar refers to the formal rules and regulations as sanctioned by a state (North, 1990).
Normative pillar refers to the legitimate means to pursue goals (Scott, 1995). Cognitive pillar, on the other hand, refers to the beliefs and value system of a society (Dimaggio & Powell, 1983). So it is reasonable to group normative and cognitive aspects of institutions into one concept because these two aspects of institutions are quite similar to one another (Scott, 1995). Together regulative and normative pillars define the differences in institutional environments between the home and host countries. These data are obtained by country-level indicators related to the institutional environment from the Global Competitiveness Report (2001—2002). The Global competitiveness Report series has evolved over the last three decades into the world’s most comprehensive and respected
choose the irregular payments in government procurement and irregular
payments in tax collection. These two indexes are likely to measure the overall corruption in a country. The first four indicators seem to capture the regulative aspects, whereas the latter two indicators are related to the normative aspects of a country’s legal environment. Every item is scored from 1 to 7, I measured the institutional distance by Kought and Singh (1988) index and formed a composite index for each headquarter-subsidiary country pair based on their deviations from one another on each of the six institution dimensions. The institutional distance for each headquarters-subsidiary country pair was calculated using the following equation:
Where ik refers to the institutional indicator (I) for country
k
,iu
refers to the institutional indicator (I
) for CEECs (u
), andV
i is the variance of indicatorI
.ID
k isinstitutional environment between
k
and CEECs countries.Using the above equation, these six items are averaged together to produce a new variable for legal environment. The higher score means the better legal environment a country has.
6. Cultural distance: is operated so as to reflect the differences between two countries in which headquarter and subsidiary are located. I also measured the cultural distance by Kogut and Singh (1988) index, which is based on Hofstede’s study of how values in workplace are influenced by culture. He analyzed a large database of employee valued scores collected by IBM between 1967 and 1973
covering 74 countries and regions (http://www.geerthofstede.com/).Hofstede (1980) found culture differed substantially on five dimensions which he labeled as power distance, individualism, masculinity, uncertainty avoidance and Confucian. Because there was not complete data in Confucian dimension, this dimension is excluded in this study.
Power distance refers to the extent to which people believe that power and status are distributed unequally and the extent to which they accept an unequal
distribution of power as the proper way for social systems to be organized (Hofstede, 1980). In organization, power distance reflects the amount of formal hierarchy, the degree of centralization, and the number of participation in
decision making (Newman and Nollen, 1996). In high power distance countries, there is a general belief that there should be a well-defined order in which everyone has a rightful place, while in the lower power distance countries all people expect to have equal right and opportunity to change their position in society ( Very et al. 1996). Hofstede (1980) found that Latin European, Latin American, Asian, and African countries generally had a larger power distance, while the Northern European and Anglo-Saxon countries are scored low on this dimension.
Individualism and its opposite refer to the degree to which a society stresses the role of individual or the group. Hofstede (1980) found that in individualism
societies the ties between individuals are loose: everyone is expected to look after him/herself and his/her immediate family. On the collective side, people from birth onwards are integrated into strong, cohesive in-groups, and extended families, which continue protecting them in exchange for unquestioning loyalty. Hofstede (1990) found a strong correlation between a country’s degree of
South Korea, Singapore, and Hong Kong are much more collectivistic than those of most Western countries, but their living standards are relatively high.
Masculinity and its opposite femininity refer to the distribution of roles between the genders. Very et al. (1996) emphasize that the traditional masculine social values are competitiveness, assertiveness, achievement, ambition and
acquisition money and other material possessions. On the other hand, feminine social values are nurturing, helping others, putting relationship with people before money, not showing off, and minding the quality of life. The women in feminine countries have the same dominant values as men have, while in the masculine countries they are somewhat assertive and competitive, so there is a gap between men’s values and women’s values (Hofstede, 1980). Japan, the US, and the Germanic countries are typical masculine countries, and the
Scandinavian countries and the Netherlands are predominantly feminine (Hofstede, 1980).
Uncertainty avoidance refers to the extent to which people are threatened by uncertain, unknown, or unstructured situations (Hofstede, 1980). In the low uncertainty avoidance countries, people tend to take risks and more tolerant of different opinions they are used to, and try to have as few rules as possible because they do not feel threatened by them. While people from high uncertainty avoidance countries, they feel uncomfortable with uncertainty and ambiguity, and they generally have a higher level of anxiety, which is manifest in greater
Similar to the legal environment variable, cultural distance is also calculated using the following equation:
Where Iij stands for the index for the ith cultural dimension and jth country, u
indicates the CEECs, Vi is the variance of the index of the ith dimension, and CDj
is cultural difference between J country and CEECs country.
The average score of these four dimensions also create a new variable to
5. Results and discussion
5.1 results
Table 2 shows the descriptive statistics for independent variables and the dependent variable. The numbers of observations of dependent variable, the level of centralization, are 212. The value ranges from 0 to 1, with an average of 0.3354, and the minimum and the maximum values are 0.00 and 0.89 respectively. The average value suggests that more than 0.3 out of 1 of the
Table 2 Descriptive statistics
N Minimu m Maximu m Mean Std. Deviati
on Variance Skewness Kurtosis Statisti
c Statistic Statistic Statistic Statistic Statistic
decision are made by headquarter. From table 2, it can be seen that the value of Skewness of the level of centralization is 0.167, with the value of Kurtosis 0.333 respectively. But 0.333 is not sufficiently different from 3, 0.167 is not near to zero, for the normal distribution the values of Skewness and Kurtosis should be around 0, and 3 respectively, so the dependent variable is not normal distribution. Therefore I use the square root to transform into new variable. As a result, the dependent variable presented in the histogram has a normal distribution (See appendix 1).
From table 2 we can see that the numbers of observations of the explanatory variables are a little differences between size of subsidiary and the number of product lines than others variables, because one of observations of them are extremely large, outliers are deleted to avoid the influence. In addition, the magnitudes of ages of subsidiary and size of subsidiary variable are substantially different sizes, so I transform them into new variables to ensure the estimated coefficients in the regression analysis interpretable (see Appendix 1, table 1, the descriptive statistics after transformation).
Tables 3 Correlation Matrix of variables (N=211) 1 2 3 4 5 6 7 1.size of subsidiary 2.ages of subsidiary .061 3.the extent of ownership of subsidiary -.114 -.045 4.number of product lines -.064 -.037 .000 5.cultural distance .036 -.058 -.081 -.057 6.Legal environment .048 .100 -.114 .038 .007 7.centralization .022 -.126 .117 -.131 .107 -.045
The results for multiple regression analysis are summarized in Table 4. According to my aforementioned hypotheses, all independent variables are expected to be negatively associated with the level of centralization. In fact, the results show that only H1, H3 and H5 are consistent with my hypotheses. However, the adjusted R-squared values are too low to be considered to have explanatory powers regarding the level of centralization (adjust R square=3.7%). In this study, the significant a level is set at a=0.10, H1 is confirmed as F-test and the T-test are significant, the p values are 0.033 and 0.040 respectively. The sign of coefficient is accordance with my expectation.
H3 I propose a negative relationship between the number of product lines and the level of centralization. This hypothesis is supported as expected the estimated coefficient is -.123 and p value =.073.
H4 I assume a negative association between the extent of ownership of subsidiary and the level of centralization. Significant results are found P=.090, but the sign of coefficient is out of expectation, so H4 is rejected.
H5 is rejected since it is not statistically significant, even although the legal environment is negatively related to level of centralization. However the p value of legal environment is .839.
H6 I posit cultural distance would be negatively related to the level of centralization, H6 is not supported since the analysis results indicate that the coefficient is positive and relationship is not significant, the p is higher than 0.10. For more information is illustrated in Table 4.
ANOVA(b)
Mod el Sum of Squares df Mean Square F Sig. 1 Regressi on .562 6 .094 2.338 .033(a) Residual 8.085 202 .040 Total 8.647 208Coefficients(a)
Mod el Unstandardized Coefficients Standardiz ed Coefficients t Sig. B Std. Error Beta 1 (Constant) .273 .092 2.973 .003 ages1 -.074 .036 -.145 -2.066 .040 Employees1 .015 .024 .042 .606 .545 number of product lines -5.71E-005 .000 -.123 -1.803 .073 the extent of ownership of subsidiary .103 .061 .118 1.705 .090 Legal environment -.001 .007 -.014 -.204 .839 cultural distance .012 .009 .096 1.385 .168a Dependent Variable: centralization
5.2 Discussion
global strategies and host country responsiveness. Moreover, it is posited that the determination of CEECs are heterogeneous and that the model requires more powerful explanatory variables. In addition, centralization may be influenced according to the origin of country of headquarter. Hedlund (1981) examined that US firms are more centralized compared to European and Asian firms. In this sample, there are more than 80% subsidiaries whose headquarter is European countries and Asian countries and only roughly 20 firms are from US. Therefore I find that subsidiaries in CEEC enjoy more autonomy.
However, this study has some worthwhile findings. First of all, I investigate how the subsidiary characteristics affect the level of centralization, and the
relationship produces a somewhat different picture. The effect of age of
subsidiary and the number of product lines are supported in the study, and the size of and the extent of ownership of subsidiary are rejected. However, few researches on MNCs found age of subsidiary is negative and significantly related to the level of centralization. My finding is consistent with Taggart and Hood (1999)’s their empirical study. Again, this study emphasizes Garnier’s (1982) argument that the older subsidiary, the more experience they have. The second Hypothesis is rejected because the relationship between the size of subsidiary and the level of centralization was in the positive direction and not significantly correlated. This is logic, because most previous empirical studies found
relationship and significant between the extent of ownership of subsidiary and the level of centralization. This argument is accordance with Männik et al. (2004)’s argument. In my opinion, it shares the same standpoint with size of subsidiary as mentioned in earlier. The larger the foreign ownership the bigger the
responsibility of parent company is to take care of the affiliate. Similar to the first hypothesis, the fourth hypothesis is confirmed the relationship between the number of product lines and the level of centralization. My finding is consistent with Garnier(1982)’s, firms with many products tends to give more authority to subsidiary itself, especially to some particular products are designed for different markets, it is harder to control. So parent firms leave more decision-making authority to the subsidiary. Therefore, the role of product lines is stressed in MNCs.
Secondly, I try to exam how the host countries’ legal environment affect the level of the centralization, however, there appears to be no relationship between them. The effect on centralization is not the characteristics of local environment,
instead the way in which the managers perceived. Despite that this hypothesis is not confirmed, it supports Garnier(1982)’s findings.
Finally, I describe how the characteristic of headquarter and subsidiary
6. Conclusion, limitation and further research
I conducted this is research on the CEECs context, attempting to explore the determinants of subsidiary centralization from three perspectives: the
characteristics of subsidiary, the legal environment of subsidiary and the relationship between headquarter –subsidiaries. My findings indicate from the subsidiary level, only the age of subsidiary and the number of product lines are supported while the others relationship are rejected. Moreover, taking into
account the host country institutional environment, it does not influence the level of centralization. Besides, the CEECs are embedded in the less developed legal system with important difference in the cultural value from their Western
headquarters, but the statistic results show that the national culture does not affect the level of centralization.
There are certain limitations to this study. First of all, the present study can not include the characteristics of parent firms, so the attempts to elaborate a model would incorporate more variables. Gate and Egelhoff (1986) argued that the centralization varies by the nationality of parent companies. They also identified the longer the MNC have had significant operation, the more decentralization of the parent-subsidiary relationship seem to be. These characteristics of parent company whether exert influence on centralization in CEECs, are still in question. Furthermore, different relationships may exist in different industries. For example, the Japanese electronic industry may show more correlation between
Overall, the study is an important addition to the existing literature. It presents some interesting and worthwhile findings, and in particular provides insights into the possible relationships between the contingency variables and the level of centralization in CEECs countries. The culture values in CEEC are different from the Western countries, and the national cultural distance can not affect the level of centralization. Therefore, parent headquarter use more expatriate control to facilitate the centralization of decision-making. Besides, cultural distance is addressed between headquarter and subsidiary for the first time in the transition economies. In a word, centralization is important aspect of structuring the
7. References
Alexander, J. A(1991). Adaptive Change in Corporate Control Praqctices.
Acedemy of Management Journal, Vol.34, Iss1, pp.162-193
Alsegg, R., (1971). Control relationships between American corporations and their European subsidiaries. AMA Research Study 107, American Management Association.
Aylmer, R. J. (1970). Who Makes Marketing Decisions in the Multinational Firm?
Journal of Marketing Vol.34 pp.25-30.
Becker-Ritterspach, F. A. A., (2006). The Hybridization of Local MNE
Production Systems: The Case of Subsidiaries in India. Labyrint Publications.
The Netherlands.
Birkinshaw, J., Hood, N, (2000). Characteristics of foreign subsidiaries in industry clusters. Journal of International Studies, First Quarter, 31, 1, pp.141.
Björkman, A., (2003). Subsidiary power and autonomy. 29th Annual EIBA Conference, Copenhagen, December 11–13, pp. 25.
Björkman. A (2007) Towards explaining the use of control mechanisms in
Foreign subsidiaries of MNCs. Swedish School of Economics and Business
Administration pp 1—77
Brandt, W. K., and J. M. Hulbert. (1977). Headquarters Guidance in Marketing Strategy in the Multinational Subsidiary. Columbia Journal of World Business. 11 pp.7-14.
Caves, R.E.(1996). Multinational Enterprise and Economic Analysis, Second edition, Cambridge, New York: Cambridge Universtiy Press
Child, J. (1972). Organization Structurer and Strategies of Control: A Replication of the Aston Study. Administrative Science Quarterly, Vol.17.Iss. 2 pp. 163-177
Child, J.(1973a). Strategies of Control and Organizational Behavior.
Administrative Science Quarterly. Vol.18, Iss. 1, pp 1--17
DiMaggio, P., & Powell, W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American
Sociological Review, Vol. 48,pp286-304.
Edwards, R. Ahmad, A, and Moss, S.(2002). Subsidiary Autonomy: the Case of Multinational Subsidiaries in Malaysia. Journal of International Business, Vol, 13, Iss.1, pp. 183--191
Elsass,P.M and Veiga, J.F, (1994). Acculturation in acquired organizations: a force-field perspective. Human Relations. Pp431-453
Garnier, G.H. (1982). Context and Decision Making Autonomy in the Foreign Affiliates of U.S. Multinational Corporations. Academy of Management Journal. Vol. 25, Iss. 4 pp.893—908
Goehle, D. G., 1980. Decision making in multinational corporations. Ann Arbor: UMI Research Press.
Harzing, A.W.(1999). Managing the Multinationals: An international Study of
Control Mechanisms. Cheltenham: Edward Elgar.
Hedlund,G.(1981) Autonomy of subsidiaries and formalization of headquarters– subsidiary relationships in Swedish MNCs. In L. Otterbeck, The management of
headquarters–subsidiary relationships in multinational corporations New
York: St Martin’s Press.
Hofstede, G. (1980). Culture’s consequences: International differences in
work-related values. Beverly Hills: Sage Publications.
Hofstede,G. (1990). Culture’s Consequences. Beverly Hills: Sage.
Hofstede, G. (2001), Culture’s Consequences: Comparing Values, Behaviors,
Institutions, and Organizations across Nations, second edition, Thousands
Oaks: Sage Publications
Johnston,S and Mengus,B (2007) Subsidiary size and the level of subsidiary autonomy in multinational corporations: a quadratic model investigation of Australian subsidiaries Journal of International Business Studies,Vol.38 pp787--801
Kallinikos, J., (1984). Control and influence relationships in multinational
corporations: The subsidiary’s viewpoint. Acta universitatis Upsaliensis,
Studia Oeconomiae negotiorum 19. Uppsala.
Kirpalani, V. H., M. Laroche, and R. Y. Darmon. (1988). “Role of Headquarter Control by Multinationals inInternational Advertising Decisions.” International
Kogut, B and Singh, H, (1988). The effect of national culture on the choice of entry mode. Journal of International Business Studies. Vol.19, PP411—432
Kostova,T, (1999). Transnational transfer of strategic organizational practices: A contextual perspective. Academy of Management Review vol. 24, pp308-324
Kostova, T and Zaheer ,S. (1999). Organizational Legitimacy under conditions of complexity: the case of the multinational enterprise. Academy of Management
Review Vol.24, pp64—81
Männik, K.; Hannula, H.; Varblane, U. (2004). Country, industry and firm size effects on foreign subsidiary strategy: an example of five CEE countries.
University of Tartu, Faculty of Economics and Business Administration, Working
Paper Series, Vol.27, pp1--63
Martinez, J. I. & Jarillo, J. C.(1989), The evolution of research on coordination mechanisms in multinational corporations, Journal of International Business
Studies, Vol.20Iss.3,pp 489-514.
Neal, M. (1998) The Culture Factor: Cross-National Management and the
Foreign Venture, Houndmills, UK:MacMillan Press.
Newman, K.L and Nollen,S.D (1996) Culture and congruence: the Fit between Management Practices and National Culture, Journal of International
Business Studies. Vol.27 Iss. 4. pp 753--779
Nohria, N. and Ghoshal, S. (1994). Differentiated Fit and Shared Values: Alternatives for Managing Headquarters-Subsidary Relations. Strategic
Management Journal, Vol. 15. Iss. 6 pp:491—502
O’Donnell S.H., (2000). Managing foreign subsidiaries: Agent of headquarters, or an interdependent network? Strategic Management Journal, Vol.21,iss 5, pp. 525 – 548
Picard, J., (1977). Factors of variance in multinational marketing control. In L.G Mattson and F. Widersheim-Paul (eds.), Recent Research on the
Internationalization of Business. Uppsala, Almqvist & Wilksel.
Peccei, R. and Warner, M., (1976). Decision making o a multinational firm.
Journal of General Management, pp. 66-71.
Pugh, D.S, Hickson, D,J, Hinings, C.R and Turner, C.(1968). Dimensions of Organization Structure. Administrative Science Quarterl, Vol.13, Iss. 1, pp68--105
Pugh, D.D, Hickson, C. Hinings, and C. Tuner. (1969). The Context of
Organization Structures. Administrative science Quarterly Vol. 14, Iss. pp91-l14
Quester P.G and Conduit. J (1996). Size and marketing centralization in
Multinational Companies. Journal of market focused management vol.pp.231--247
Robock, Stefan, H., & Simmonds, K.,(1973). International business and
multinational enterprises. Homewood, I11: Rechard D. Irwin, Inc.
Sim, A. B. (1977). Decentralized Management of Subsidiaries and Their Performance. Management International Review Vol.17, No.2, pp 45-51.
Scott, W. R. 1995. Institutions and organizations. Thousand Oaks, CA: Sage.
Stopford, J. and Wells, L. T. Jr, (1972). Managing the multinational enterprise:
organization of the firm and ownership of subsidiaries. New York: Basic
Books.
Taggart, J. H., (1997). Autonomy and procedural justice: a framework for
evaluating subsidiary strategy. Journal of International Business Studies, Vol. 28, Issue 1, pp. 51–77.
Taggart, J. H., and Hood, N., (1999). Determinants of autonomy in multinational corporation subsidiaries. European Management Journal, Vol. 17, No. 2, pp. 226-236.
Vachani, S., (1999). Global diversification’s effect on multinational subsidiaries’ autonomy. International Business Review, Vol.8, pp535-560.
Varblane, Männik AND Hannula (2005), Autonomy and Performance of
Foreign Subsidiaries in Transition Countries, Tartu Universtiy Press,
Very, P. Lubatkin, M and Calori, R (1996) A Cross-National Assessment of Acculturative Stress in Recent European Mergers, International Studies of
Management & Organization. Vol.26, Iss. 1 pp59--86
Welge, M. (1981). The effective design of headquarters–subsidiary relationships in multinational corporations. In L. Otterbeck, The management of
headquarters–subsidiary relationships in multinational corporations New
Wolf,J(1994): International personalmanagement . Kontext-Koordiantion-Erfolg, Wiesbaden: Gabler, MIR Edition.
Xu, D., & Shenkar, O. (2002). Institutional distance and the multinational enterprise. Academy of Management Review, 27: 608-618.
Youssef, S. M., (1975). Contextual factors influencing control strategy of
multinational corporations. Academic of Management Journal, Vol. 18, No. 1, pp. 136-143.
Young, S. and Tavares, A. T, (2004). Centralization and autonomy: back to the future. International Business Review, vol. 13, pp. 215-237.
Database of European Union’s fifth framework project
http://www.iwh-halle.de/projects/productivity-gap/field_work/Field_work.htm#WP4
Descriptive Statistics
Table 1
Table 2 countries included in the sample
1. Austria 2. Belgium
3. Canada 4. Czech Republic
5. Denmark 6. Estonia 7. Finland 8.France 9. Germany 10. Hungary 11. Israel 12. Italy 13. Japan 14.Luxenbourg 15. Netherlands 16. Norway 17 Poland 18 Slovakia
19 South Korea 20.Spain
21 Sweden 22.Switzerland
23.United Kingdom 24. United States
N Minimum Maximum Mean
Std. Deviation
Skewness
Kurtosis
Statistic Statistic Statistic Statistic Statistic Statistic Std. Error Statistic Std. Error
ages1 211 .00 2.18 .9567 .40589 .544 .167 1.618 .333
Employees1 212 1.11 4.19 2.3359 .60808 .229 .167 -.321 .333 centralization1 212 .00 .94 .5395 .21111 -.724 .167 .343 .333 Valid N
