MASTER THESIS “Control Level over Subsidiaries in Central and Eastern Europe”

MASTER THESIS

“Control Level over Subsidiaries in Central and

Eastern Europe”

Subsidiary Point of View

Gjorgji Keckarovski

s

1621327

E-mail: G.Keckarovski@student.rug.nl

August 2007

University of Groningen,

Faculty of Economics, Management and Organization

Supervisors:

Table of contents

Abstract……….… 3

1. Introduction……….…. 4

1.1 Problem Statement

... 4

1.2 Research Objective and Research Questions

……….… 5

1.3 Significance of the Research

... 6

1.4 Structure of the Thesis

... 8

2. Literature Review and Hypotheses... 9

2.1 Agency Theory

... 9

2.2 Control Theory (Reference to MNC)

... 10

2.3 Subsidiary’s Characteristics

... 12

2.4 Subsidiary’s Environment

... 15

2.5. CEE Region

...18

2.6. Hypotheses

... 20

3. Methodology... 24

... 24

... 25

... 30

... 35

4. Conclusions and Limitations... 41

Abstract

1. Introduction

1.1 Problem Statement

The topic of the multinational corporations has been more then widely discussed in the past decades. Organizing the relations between the headquarters and the subsidiaries is only one of the many aspects of the multinationality studied throughout the years. One of the biggest contemporary obstacles for the MNC is how to manage the complexity of foreign subsidiaries within miscellaneous external environments and internal competen-cies (O’Donnell, 2000).

An unambiguous fact about this is that MNC are companies that own affiliates in differ-ent countries for what they encounter challenges of preserving their integrity and pro-gram by managing the outspread units. Dispersing their operations outside the home country brings significant amount of uncertainty to the parent company. Uncertain con-ditions put a lot of pressure on governing the transnational operations (Manolis, Ny-gaard and Stillerud, 1997). In order to secure the accomplishment of the company’s goals parents institute control over their subunits (Chang and Taylor, 1999; Paik and Sohn, 2003). The importance of creating internal control mechanisms was also under-lined by Hennart (1993,) with the emphasis on overcoming the knowledge asymmetries between the headquarters and the subsidiary.

1.2 Research Objective and Research Questions

The region of Central and Eastern Europe has been part of many debates in the eco-nomic circles recently. Its attractiveness and specificity come from the undergone envi-ronmental shift that has led to stable economic growth and to openness towards the world market. These changes have made the CEE market closer to the Western ones in terms of liberalizing the economy, successful privatization, democratization, creation of new institutions, amalgamation to the transnational organizations like EU, NATO, WTO etcetera (Nieminen, 1999). All of these contributed to massive entry of multina-tional corporations in these countries.

Although all of the above mentioned attributes were recognized on the international level still the CEE economies are to some extent regulated. This and the lack of stable institutional base (not in all CEE countries) create fairly unfamiliar area for the MNC. In this sense, subsequent to the entry Multinationals face insufficient information, lack of established networks, improper legislative and regulations, slow and long-lasting bu-reaucratic procedures, corruption and so on ( Dikova and Witteloostuijn, 2005).

Even under these circumstances MNCs do not hesitate to start their operation in the CEE countries. They have played a huge part in shaping these economies and bringing them closer to the globalization process.

In this paper I would like to shed a light on subsidiary’s specific characteristics and its environment while assessing their influence on levels of control in the headquarters-subsidiary relations in the following countries: Slovenia, Slovakia, Poland, Estonia and Hungary. Having in mind the apparent differences between the countries it is expected that there will be differences in the levels of control as well. The ultimate goal would be to discover where the possible variations in the control level come from and what they are dictated by. This at the end brings me to the Main research question:

What are the levels of control over the subsidiaries in Central and Eastern Europe?

In order to entirely answer the research question, several sub-questions were formu-lated.

Sub-questions:

1. In what way subsidiary’s specific characteristics influence the levels of control? 2. In what way subsidiary’s environment influences the levels of control?

3. To which extend is this influence important and what it is determined by?

1.3 Significance of the Research

In all of these analyses subsidiary’s point of view was a little bit bypassed although pio-neer efforts were made by Van Den Bulcke (1984), Egelhof (1984) and others. Related topics examining the subsidiaries’ strategies can be found in the articles of Jarillo and Martinez (1990) and White and Poynter (1984) but none of them really discovers how subsidiaries on the basis on their own capabilities and characteristics of their home envi-ronment fight for their autonomy. The control issues was mostly comprised only one-sided, showing why and under what conditions the headquarters decide to use diverse degree of control.

As it was noted before, subsidiary has its own qualities that can support its desire for being less controlled. These qualities come from the subsidiaries’ own resources and capabilities as well as from the environment they operate in. Some scholars captured this moment in their findings still treating the both factors separately. While some (Egelhoff and Gates, 1986; Kokko and Kravtsova, 2006; Chang and Taylor, 1999; Bow-man, Duncan and Weir, 2000) speak of the subsidiaries’ characteristic (age, size, num-ber of employees, business lines, local management etc.) as an influential aspect of de-centralization, others (Geppert and Williams, 2006; Pont and Noboa, 2003; Drogendijk, 2005) think that the support from subsidiaries’ environment (local institutions, culture, language, rules and regulations, trade unions etc.) can enhance their autonomy.

1.4 Structure of the Thesis

The thesis will be organized in the following order. After the introductory part that dis-covers the motives for this research, its significance and the main research questions, comes the chapter with the theoretical background that is concluded with the develop-ment of the hypothesis. The next chapter, methodology part, captures the explanation of the data set, presentation of the results and discussion of the results. Conclusions, limi-tations and suggestions for future research are comprised in the fourth chapter. This the-sis will be completed with the citation of the consulted literature and the appendices.

2. Literature Review and Hypotheses

2.1 Agency Theory

The main focus of the theory is the relations between subjects where the benefits for one of the party (principal) are dependent on the actions taken by the other party. The key assumption here is that the agents in some situation might be working for the benefits of their own rather then for completing the companies’ aims, therefore reducing their own risk and increase the risk for the principal. This comes from the self-interested and risk-aversive type of the human nature (Chang and Taylor, 1999) and it can normally lead to conflicts of interests between the two sides which initiate the control issue (Fong and Tosi, 2007).

This existing problem is usually solved by contractual agreements in which certain level of autonomy accompanied by previously defined duties is delegated to the agent. Not-ing, even after the agreement the principal could not fully control the agent because of the possibility of opportunistic behavior by the later (Pavlou, Liang and Xue, 2007). Apart from the fact that their goals are partially in the hands of the agents, whenever there is an agency asymmetry the principals would try to input control over the agents in order to mitigate the eventual agency costs (Chang and Taylor, 1999).

com-pany and its subsidiary function. Reasons like culturally, politically, economically, geo-graphically distant conditions can create an uncertain surrounding causing control prob-lems within the MNC. This makes it a good field for analyzing the crossroad in which Multinational finds itself, with the subsidiary’s bargaining power on one side, and ef-forts not to surrender all the decision rights to the subunit on the other side. It puts the Multinational in a position of having to choose some degree of control that can make the whole machinery work properly (Nohria and Ghoshal, 1994).

2.2 Control Theory (Reference to MNC)

The interdependence between HQ and subsidiary on each other’s resources can be en-couragement for practicing control. When the subsidiary’s work relies mostly on its parent’s resources the level of control is likely to be high. In the opposite situation, when the parent is greatly dependant on the subsidiary’s resources, it will be really dif-ficult for it to regain high control power over the dispersed units. The influence can be sustained through building of so called control mechanisms (Doz and Prahalad, 1981). The theory has said a lot about all kinds of control mechanisms in the MNC and the conditions under which they are introduced in the companies. I will not go into depths of explaining each of the mechanisms separately because that is not part of my observa-tions at the moment, instead I will give a short outline of the types and reasons why these mechanisms are implemented. Another motive for doing it this way is that in most of the cases these mechanisms are linked to the parent companies as a party that estab-lishes them and in the thesis the focus is on the subsidiary strength to contrast them. Organizational control has been defined in number of ways each of them capturing the essence of regulating and administrating the activities within the company so the ulti-mate results will be in accordance with the company’s goals and work in favor of the interest groups (Ferner, 2000). Further on, regarding the multinational relations, control means controlling the output and behavior of the subunit by use of authority in shapes

of formal and informal mechanisms (Marschan and Ghauri, 1998;Jarillo and Martinez,

standardiza-tions, and financial reports, managing the budget etc. while the focus is on monitoring the work of the subsidiary and practicing direct output control. Unlike formal, informal mechanisms are mainly used for shaping the intra-firm personal relations and corporate culture while trying to route stuff’s behavior.

Distinguishing between control mechanisms has led to formulating a broad palette of control types some of them overlapping in the theoretical meaning, still used differently by the various authors. Therefore one of the existing divisions speaks of cultural, behav-ioral and output control (Chang and Taylor, 1999). By behavbehav-ioral control one can un-derstand the implementation of parent company individuals in the leading positions in the subsidiary or supervising the activities and the behavior of the subsidiary (Chung, Gibbons and Schoch, 2000). This can be also famous as staffing control even though some researchers make difference between the two of them (Egelhoff, 1994). Output control is about assessing the quality and the quantity of the output, while cultural con-trol means creation of the shared values chain between the subsidiary and the parent company. The later could be found in some literature as social control where the goals and interests of the managers of both parent and subsidiary are united (Gupta and Go-vindarajan, 1991; Ferner, 2000).

Control studies also give several implications of the trends in using control within the Multinationals. This has become a field of opposing arguments among the authors that treat control in their studies. On one side there are Bartlett and Ghoshal (1989) who claim that the old hierarchical model with one single center of decision making power is becoming obsolete and is being replaced by new one. The new model is based more on decentralized and independent centers, leaning on the informal rather then formal con-trol mechanisms. On the other side Jarillo and Martinez (1989) speak of the evolution-ary patterns in controlling the subsidievolution-ary and claim that the various forms of mecha-nisms are supplementing each other rather then substituting. In this way they are more efficient in dealing with challenges that come not only from the subsidiary but also from its environment. Moreover the practice has proven that headquarters are keen on using combination of all reliable mechanisms especially because most of them are interre-lated.

interests between the HQ and the subsidiary by appointing suitable managers in the subunit i.e. personnel who would be less entrepreneurial and would follow company’s orders. Another feature of these mechanisms is that they give signals to the subsidiaries of the power and seriousness that HQ possesses. The opportunity of selective approach when it comes to delegation of autonomy is also in this group. This allows the parent company to judge upon the level of autonomy of its subsidiary. It is applicable in situa-tions when the subunit holds critical knowledge of a certain process and it is better the decision making to be left on the subsidiary’s management. But when the subsidiary encounters losses in the working practice, having control mechanisms enable the Multi-national to shift the power between the HQ and the subsidiary, thus taking over the de-cision making.

Controlling the subsidiary is a reality in the multinational corporations. Above were ex-plained various kinds of control types, mechanisms through which this control can be implemented and the necessity of creating these mechanisms. What follows is the theo-retical background on how subsidiary can resist the pressure from above and what its qualities are.

2.3 Subsidiary’s Characteristics

The state of interdependence is defined as a situation in which actions and results from those actions of one party are strongly determined of actions and results of another party and vice versa (O’Donnell 2000). The resource dependence can be reversible process meaning that both parties share mutual interests and obligations for what it is in their favor to create a system of shared values. This would mean engaging in system of common goals and joint help in achieving those goals (Forsgren, Holm and Johanson, 2005). Interdependence is also known as a serious linkage between the parent company and its subunit. The more the subsidiary depends on its parent the more attention will the headquarters get from the subunit’s management. Therefore, if the subunit has cre-ated enough resources to rely mostly on its own performance than the control becomes major issue for the parent (Dyer and Singh, 1998). It is so, because interdependence does not only imply relying on each other’s resources but it also allows both parties to interfere and sometimes obstruct the actions of the other when they have confronting ideas (Forsgren, Holm and Johanson, 2005).

This can create fairly awkward vacuum in the headquarters –subsidiary relation because on one hand the parent will pull the strains to hold on more power over the subunit be-cause the subunit’s resources are essential to him, and on the other hand the subunit will feel strong enough to start making independent decisions. Over the time subsidiary characteristics and capabilities evolve to a level when it reaches higher scales of sover-eignty and this position entitles it to ask for more autonomy (Drogendijk, 2005). During the process of internationalization company assets, human capital and capabilities can be found in the subsidiary hence making it extremely difficult for the headquarters to secure very high control levels (Bartlett and Ghoshal, 1989; Ghauri 1992; Doz and Pra-halad, 1981). As the support from the parent company declines due to the compact base of built up resources, the subunit takes control over its own strategy and planning (Birkinshaw and Hood, 1998; Doz and Prahalad, 1981). The bottom line is that the sub-sidiaries have enough assets to cease acting only as a tool of the parent company and start exercising significant level of independence (Birkinshaw, 2001).

same Multinational. Endowed by unique resources and capabilities subsidiary can not only impact the outcome of its own actions but also the one of the remaining units of the company. Being in this position it gives the subsidiary power to shape the decisions in the MNC that in other situations would be beyond its competence. Reasonably, one can assume that when the subsidiary is of a higher importance to its parent and other sub-sidiaries of the MNC it would apply more autonomy i.e. would be less controlled by its parent (Forsgren, Holm and Johanson, 2005).

Gaining more autonomy could enliven innovativeness and efficiency of, in some cases, underutilized capacities of the subsidiaries. Eventually this could be beneficial for both parent and subsidiary. The innovative aspect of subsidiary’s maturation was even rec-ognized as a world trend by Birkinshaw (2001). In the article she distinguishes between three eras in the evolutionary progress of the multinational corporations. Paternalism, where the entire decision making process took place in the HQ and the subunits acted as distributors of the final products. This era was eventually over because of the moderni-zation of the foreign markets and strengthening of the subsidiaries. Next era was called expansionism. These were Multinationals’ initiative attempts to transfer R&D responsi-bilities to some of the units, but not to all of them, in order to test the scientific capacity of the foreign market. Current era is called liberalism where some of the key decisions were largely decentralized; subunits started introducing their own product lines and were more involved in the corporations.

free-dom should be granted to the subsidiary and how much supervision should be main-tained facilitates the control gap.

Geppert and Williams (2006) supported the view of subsidiary obtaining autonomy while leaning on its specific characteristics. They described this situation via the strate-gic position of the subunit framed by the skilled personnel, high educated local man-agement and the huge scope of R&D activities. Second factor in these authors’ observa-tion was the subunit’s performance. Studies on the level of centralizaobserva-tion in the MNC and the influential factors date back in the beginning of the 1980s. Among the scholars who broke new ground in this field are Van Den Bulcke (1984) and Egelhoff (1984).

According to them, in general, there are certain features that subsidiary possesses that

can make a difference when it comes to the degree of its independency even though

none of both authors undermines the importance of the environment. Briefly, they speak

of subsidiary size, age, performance, number of products lines, and interaction with other subunits or outside actors, and conclude that not all of these characteristics have

equal relevance in determining the subunit’s autonomy. I will refer more

comprehen-sively to these authors’ works in the part of hypotheses that follows further on.

2.4 Subsidiary’s Environment

When speaking of the subsidiary’s environment some authors make a distinction be-tween subsidiary’s corporate environment and subsidiary’s business system. The former includes ownership, administrative and business relation of the subunit inside the Mul-tinational, whereas the later comprises all the other interactions of the subsidiary with its surrounding outside of the company. Both types of environments represent forces that can shape the MNC’s decisions in regard to the subunit (Forsgren, Holm and Jo-hanson, 2005). Here on when mentioning the subsidiary’s environment I will refer to the links that it has with the home national system (institutions) and the influences that can come from this connection.

en-vironments they operate in because of what they create specific competencies and capa-bilities. Societal, cultural, political and economical are just number of pressures to which these working units are exposed to and it is natural that they will have profound influence over the subsidiary resilience to fully accept the parent’s directives (O’Donnell 2000).

Local institutions whether they are governmental or non-governmental have powers to obstruct subsidiary’s actions if they find that the MNC’s and subsidiary’s activities are vicious for the general welfare in the country (Drogendijk, 2005). The margins of the subsidiaries operations are in a big way defined by the local cultural, social, political conditions as well as by the existing legislative. These limits in some cases can force the subsidiary to derail from the company’s plans at the same time making the headquarters to give up on the strict demands they have for the subsidiary (Rosenzweig and Singh, 1991).

When the subsidiaries work in more compound environment they usually request for grater autonomy. This comes from a reason that high complex surrounding demands more specific knowledge and therefore more sovereignty in the decision-making proc-ess (Nohria and Goshal, 1994). Some of the resources that evolve at the subsidiary level are strongly embedded in the local environment. It requires specific knowledge and ca-pabilities for their further development and that puts the local managers ahead of the HQ managers since they possess this knowledge (Birkinshaw, 2001). Local impedi-ments which MNC are facing can inject a lot of confidence in the subsidiary’s manage-ment since they have freedom of interpreting the local rules in their favor. Subunits sometimes present these rules as constraints in their work in that way giving them legal justification for possible failures or reluctance in taking some actions. This can be even more encouraged by eventual lack of thorough headquarters’ knowledge of the envi-ronment (Ferner, Almond, Clark, Colling, Edwards, Holden and Muller-Camen, 2004). Some authors even go that far stating that MNC could never entirely control the sub-sidiary because the outcome will be strongly linked to the environments of the detached units (Pont and Noboa, 2003).

well. In order to get this approval subunits have to adjust to the local practices that in some occasions might be contrasting those of its parent. The stronger local institutions are the higher will be requirements for complementarities between national and subsidi-ary patterns. Sometimes the development level of the host country, more specifically the education level, market constellation, infrastructure etc. are mentioned as important country features. In other cases the government policy towards Multinationals is high-lighted as a crucial factor in empowering the subsidiary’s side (Kokko and Kravtsova, 2006).

The issue of culture has been brought up by several scholars especially in the manage-ment theory. These authors classified culture as a trigger of differences within the or-ganizations. The fundamental point about this is that countries differ from one another in the cultural aspects of living which finally reflects on the organizations working in these countries and it has an influence over the management in the companies. This is understandable because management and the organization can not be exempted from the cultural impact the environment has over these players. Cultural dissimilarities are able to create gap in the relations between the HQ and subsidiary particularly if they are not ready to make the necessary adjustments (Myloni, Harzing and Mirza, 2004).

The host country national business system (NBS) and its institutions enhance subsidiary position in the Multinational. According to the institutional theory the NBS and institu-tional constellation take part in shaping the companies activities in a way that resembles the national patterns. These patterns could sometimes diverge from the MNC policy and depending on the cohesiveness of the NBS can ether force modifications in the MNC or can give the subunit weak bargaining position (Gooderham, Nordhaug and Ringdal 1999).

importance of the country of origin of the parent company in framing the company poli-cies and setting the working ambient but then again there is the host country of the sub-sidiary that can have just as much influence. Finally, one can presume a broad range of control levels in the subsidiaries from different countries due to the variations in the characteristics of their NBS.

2.5. CEE Region

In the 1990s, lasting for almost a decade, deep transitional processes took place in the CEE region comprising dozen of countries. Main characteristics of these changes were replacement of the old socialist system with the new capitalist one, liberalization of the economy, massive privatization, building of new institutions, etc. Eventually, at the dawn of the new millennium most of these countries could confidently conclude that the transition was finished and to a big extent, successful (Kocenda, 2001). Throughout the whole time they were strongly monitored and at the same time helped by transnational institutions such as EU, NATO and WTO. Today they are rightful members of these institutions. Five countries included in my research (Slovenia. Slovakia, Estonia, Poland and Hungary) are all part of the CEE picture.

One of the most influential growth initiatives was provided by the opening of the mar-ket. It led to attracting numerous Multinationals that found different motives for coming to these countries. MNC were not settling there only to sale their products but also to open new production facilities. Ether way they flooded the whole region with invest-ments making it frequent destinations for FDI. This process played a significant role in upbringing CEE economies (Kurz and Wittke, 1998; Varblane, Mannik and Hannula, 2005).

statements on two grounds. First, the increased international trade between the countries in the CEE region and second, applications for EU membership, that obliged them to fulfilling same criteria, that can prompt institutional convergence.

Even though the patterns that were taken by the CEE countries resemble each other, there are some authors who confront the arguments of high convergence inside CEE region by giving their views on the level of divergence. Referring to the liberalization of

the market, Bohle and Greskovits (20071) write that the CEE countries, particularly the

ones included in this research, reacted differently to the challenge. While some were completely devoted to market forces and high levels of liberalization (Estonia), others were swinging towards the social security (Hungary, Poland and Slovakia) and Slovenia well balanced between these extremes. In regards to the integration in the transnational organizations Bandelj (2003) argues that there is a difference between accepting the rules and regulations of these organizations and their factual implementations. The rea-sons for that are variances in interpretation of these regulations and their application on a country level. The scenarios in these regards proved to be different among the CEE countries. Adding to the arguments against the CEE convergence, Marginson and Meardi (2006), claim that even on macroeconomic level there were obvious inconsis-tencies between these countries. Macroeconomic stability showed that Estonia and Slo-venia are ahead followed by Slovakia and at the end Poland and Hungary, who did not succeed to make their core institutions as dominant factors in the system as they were in the rest of the states.

In other article of Bohle and Greskovits (20072_{) we can see the CEE countries placed in}

The conclusion at the end could be that CEE countries shared and still share many common features during and after the transition period. But there were also some seri-ous differences that could be detected. Prevalence of convergence or divergence in their patterns was not evident although it is clear that due to the significant amount of dis-similarities CEE region should not be referred to as a uniformed region. Different con-ditions would reflect differently on the actors in each of the CEE countries including the MNC and subsidiaries as such.

2.6. Hypotheses

In the previous chapter were given several theoretical views in regard to the degree of control in the subsidiaries as well as theories related to the region of Central and Eastern Europe. It was stated that subsidiaries would be more or less controlled depending on two crucial factors, the endowment of specific characteristic and the enhancement they can get from the host environment. Another important argument was given in favor of the expected fluctuations in the degree of control among the CEE subsidiaries due to the significant differences between the countries themselves. Some of commonly repeated subsidiaries’ features, taught to be important determinants of the level of control in the subunit, were subsidiary’s age, size, number of business lines and importance to the par-ent company. For the purpose of this research all of the before noted characteristics were translated into hypotheses accompanied by the last hypothesis which was derived from the institutional influence of the subunit’s environment. All in all there are five hypotheses which will be presented in the remaining of this chapter.

Subsidiary’s Age (as part of the MNC). Drogendijk (2005) relates the length of the

have. When the subsidiary is full-grown it is entitled to have more responsibilities which according to some life cycle are given to him by its parent. The next step is to make the subunit accountable for his activities by unleashing more autonomy. As sub-sidiary becomes more familiar with its tasks and the market it loads specific knowledge and creates its own networks which represents the evolutionary path of the subunit and earns him decentralization (Tavares and Young, 2004). To same conclusion but from other reasons have come Egelhoff and Gates (1984) explaining that the “older” subsidi-aries need to be granted grater degree of autonomy because they have been around for a while, have bigger experience and thus being less risky for the MNC. They find strong correlations between the age of the subsidiary and the level of centralization. Accord-ingly, I would argue that:

Hypothesis 1: Subsidiaries that have been working for a long time within the Multina-tional will experience lower levels of control then those that have been part of the MNC for shorter period.

Subsidiary’s Size. The findings about the influence of the subsidiary’s size over the

H2a: The larger the size of the subsidiary the more controlled it will be from its parent. H2b: The larger the size of the subsidiary the less controlled it will be from its parent.

Subsidiary’s Importance. Subsidiaries’ importance has been interpreted differently by

various authors. For Johnston (2005), subsidiary importance is represented by its size, whereas for Varblane, Mannik and Hannula (2005) importance equals power. This power was absorbed through accumulating specific capabilities and reinforces subsidi-ary’s autonomy within the MNC. Consequently, they conclude that more important sub-sidiaries are less controlled. Another group of authors defines the importance of the subsidiary as intensity of interactions between the subunit and other subunits of the MNC as well as relations with other actors of the subsidiary’s own business network. This approach allows more concrete overview of the subsidiary’s participation in the MNC activities and at the same time was followed by several scholars. That gives me enough reasons to stick to it in my subsequent research. Chang and Taylor (1999) found that when the subsidiary’s transactions with its parent or other sister-subsidiaries are more frequent in that case the parent will like to have more control over the unit. It was confirmed by Egelhoff and Gates (1984) as well as Cray (1984). They based their as-sumption on other authors’ surveys so I will have my postulation based on theirs.

H3: When the subsidiary has undisrupted and often transaction with the parent com-pany or other affiliates inside the same Multinational the HQ control is likely to be high.

Subsidiary’s Business Lines. The link between the subsidiary’s business lines and the

H4a: The greater is the number of business lines in the subsidiary, the more controlled it will be.

H4b: The greater is the number of business lines in the subsidiary, the less controlled it will be.

Subsidiary’s Environment. The surrounding in which the subsidiary is located can be

complicated in terms of the various aspects it has. Manolis, Nygaard and Stillerund (1997) write about those aspects, in particular naming three of them. Those are the mar-ket context, the regulatory context by which they mean political and institutional con-straints and the social context i.e. cultural differences. The significance of the National Business Systems (NBS) in guiding the Multinationals’ practices was underlined by Geppert and Williams (2006). In their view they leave the culture influence aside and concentrate on institutional presence which in this case comprises political, economical and social institutions. These institutions can be highly integrated and stable therefore establishing rules that cannot be evaded by the MNC and as such have deep impact on Multinational’s practices in the host countries. Forsgren, Holm and Johanson (2005) propose that when the subsidiary is deeply embedded in its environment it leaves a little room for the HQ to import high level of control in it. Subsidiaries that have more deci-sion-making freedom are known to be under great influence of the distinctive conditions in their home country. Strong home institutions will be favorable for those that like to be less dependent on their parents and have greater autonomy in the decision making process (Van Den Bulcke, 1984). Varblane, Mannik and Hannula (2005) add to these claims by writing that the strong coordination with the local structure is significant source of power for the subunits. Following the above implications I would assume that:

3. Methodology

3.1 Sample and Data

The data set used in the following research analysis was taken from the “CEEC subsidi-ary database”, the generation of which was financially supported by the EU in the RTD project "EU Integration and the Prospects for Catch-Up Development in CEECs - The Determinants of the Productivity Gap". It was constructed on the bases of returned questionnaires that were sent to foreign owned companies in five CEE countries: Slove-nia, Slovakia, EstoSlove-nia, Poland and Hungary, during the period of year 2001-2002. The original data set is consisted of 458 subsidiaries operating mostly in manufacturing in-dustries, than services and only one subsidiary comes from each construction industry and energetics. The ownership of these companies is largely versatile with European countries (old EU members) as most frequent owners, than follows a number of US owners and at the end are parent companies from some Asian countries such as Japan, Hong Kong etc.

that are part of this analysis were excluded as well. These changes deducted mine re-search sample to a total of 373 subsidiaries. For the distribution of the subsidiaries by countries see Table 1 below.

Table 1: Number of subsidiaries by country Slovenia 54 14.48% Poland 123 32.98% Hungary 66 17.69% Slovakia 67 17.96% Estonia 63 16.89% Total 373 100.00% Author’s calculation

Another data source that was used for the purpose of testing the fifth hypothesis was the website of the U.S. Department of State. This data source was chosen due to its in-volvement in publishing annual reports on the investment climate in large number of

countries including thefive CEE countries observed in the thesis. The prime purpose of

the so called Investment Climate Statements is to inform U.S. investors about the for-eign environment they could invest in by sharing information on the countries’ market openness, policy towards foreign investors, legislatives etc. Specificity of these invest-ment stateinvest-ments is in incorporating country indicators such as efficiency of the legal system, level of education, labor unions, corruption and others. These indicators will be later used for defining the countries’ institutional influence (Dikova and Witteloostuijn, 2005; Geppert and Williams, 2006) as part of the last hypothesis in my research.

3.2 Defining the variables

The analysis of the data set comprises five independent variables and one dependent

Independent variables

Subsidiary’s age (as part of the MNC) – as the name of the variable implies, subsidi-ary’s age was measured through the number of years the enterprise has been a part of the Multinational (Kokko and Kravtsova, 2006). Even though some authors relate the age of the subsidiary with the number of years since its founding (Egelhoff and Gates, 1984) that measure gives the subunits different starting position and it is associated with factors preceding the process of multinationality. Additionally it crosses with the factor trust that is important for the subunits that have been working for the longer period of time within the MNC. Hence, for testing the first hypothesis question number 4 (Q4) – What is the year of registration of your company as a foreign investment enter-prise or announcement of foreign owner to the Firm Registration Office? -was taken into consideration. The number of years of the company with the status of a sub-sidiary, just as the firms indicated in their answers, was estimated. Additionally a dummy variable was created to distinguish the “old” from the “young” subsidiaries. Consequently an average (8 years) of the total number of years was calculated and com-panies above that age fall in the group of “old” subsidiaries, while firms below that age fall in the group of “young” subsidiaries.

Subsidiary’s size – it was quite straightforward choosing the variable for measuring the size of the subsidiary. As many authors have done it before, the size is represented by the number of employees in the subunit (Egelhoff and Gates, 1984; Chang and Taylor, 1999; Varblane, Mannik and Hannula, 2005; Kokko and Kravtsova, 2006). Here I speak of the absolute size of the subsidiary instead of the relative size (relative size compared to the MNC as a whole), since information on the overall number of employees in the MNC were neither included nor disclosed in the questionnaire. Therefore, question number 2 (Q2) – What is the total number of employees employed in your com-pany? – was used for testing the second hypothesis. Since there was a great disparity in the size of the companies as stated in their answers, no normal distribution was ex-pected. In that case two dummy variables were introduced; one for the medium sized companies (100-500 employees) and another one for the large companies (more than 500 employees).

subsidiary on one side, and other subsidiaries of the same MNC, the parent company and actors from the subsidiary’s external business network, on the other side (Chang and Taylor, 1999; Egelhoff and Gates, 1984; Cray, 1984). They proposed that these in-teractions can be estimated through number of purchases and sales among the above mentioned parties. In order to test the third hypothesis two questions were taken out of the questionnaire, number 10 (Q10) – Please indicate the structure and value of your sales (in%) according to origin of the buyers - and question number 11 (Q11) - Please indicate the structure and value of your supplies (in%) according to origin of the buyers. The value of importance was estimated as taking over the percentage of sales/supplies to/from subsidiary’s foreign owner and other domestic subsidiaries of the foreign owner as they were answered in questionnaire. This would give the frequency of interactions within the MNC which a condition for level of importance.

auton-omy. More autonomous would also be subsidiaries whose personnel are represented by experienced and powerful labor unions. Additional determinants of the institutional in-fluence were given by Dikova and Witteloostuijn (2005). Among other factors they speak of are the level of corruption in the country and the efficiency of the legal system. High corruption means institutional instability therefore increasing the level of control in the subunits, while strong legal institutions enhance the level of autonomy. These four determinants accompanied by the country’s discrimination against foreign owners were taken into consideration for defying the institutional influence in the subsidiary’s home country as part of this research. Although country’s discrimination against foreign owners does not really express the institutional power it represents the attitude of the home institutions towards foreign companies in their country, thus having indisputable influence over the control level in the subsidiaries.

Table 2: Level of institutional influence by indicator per country

Educational

System Legal System Corruption Labor Unions

Discrimina-tion against foreign own-ers lo w me diu m hi gh lo w me-diu m hi gh lo w me-diu m hi gh lo w me-diu m hi gh lo w me-diu m hi gh Estonia 9 9 9 9 9 Hungary 9 9 9 9 9 Slovenia 9 9 9 9 9 Slovakia 9 9 9 9 9 Poland 9 9 9 9 9

Author’s calculation (source-www.state.gov)

In the next step each of the indicators was valued by their ratings with HIGH getting value 3, MEDIUM getting value 2 and LOW getting value 1, except for the corruption level where HIGH got value 1, LOW got value 3 and MEDIUM remained unchanged (according to Dikova and Witteloostuijn (2005) high corruption contributes to low insti-tution efficiency leading to low level of autonomy in the subsidiaries). At the end aver-age of these values per country was calculated and three groups of countries with differ-ent level of institutional influence were formed. Slovenia as country with the strongest institutions and highest institutional influence among those in the sample (average value of 2.8), Estonia and Hungary coming second with mid-level institutional influence (av-erage value of 2.2 each) and at the end of the tail were Poland and Slovakia with the weakest institutions in the region (average value of 1.6 each).

Dependent Variables

in which they enumerate number of control indicators sorted into three groups: market-ing decisions, manufacturmarket-ing decisions and financial decisions. This framework was later used in the works of Chung, Gibbons and Schoch (2000), Chang and Taylor (1999) and can be found in the article of Bowman, Duncan and Weir (2000), although the later do not referred to Egelhoff and Gates. Since the various types of control are interrelated and can be used combined instead of individually and since in this thesis I do not con-sider types of control but rather the degrees of control, a single level of control had to be establish. From the questionnaire two questions were extracted. Those were number 7 (Q7) – Please tick which business functions are being undertaken (a) on your own (b) mainly on your own (c) mainly by your foreign owner (d) by your foreign owner only - and number 14 (Q14) – Who has undertaken initiatives for changes in the following areas: in organization of business functions, in number of lines of businesses and in sales and exports - the same four possibilities to answer Q7 were given for Q14 as well. Five out of thirteen business functions were chosen from Q7 ac-companied with all three initiative areas from Q14 to form panel of control indicators in accordance with the Egelhoff and Gates’ framework. These control indicators are as fol-lows: Marketing decisions - marketing, market research and determining the product price; Manufacturing decisions - organization of business functions, number of lines of businesses, sales and exports, operational management and strategic management of planning; Financial decisions - accounting and finance of operations, investment fi-nance. The level of control was estimated as an average of the percentage by which cer-tain functions and initiatives were undertaken by ether the parent or the subsidiary. The control level is estimated as follows, between 0%-33% means only the subsidiary or mainly the subsidiary enjoys the decision-making power, 33%-67% is mainly the sub-sidiary or mainly the parent, and 67%-100% is mainly the parent or only the parent.

3.3. Results and Analyses

sub-sidiaries among the observed countries. And lastly, cross-country regressions were run by picking each of the country as a base scenario. Than, variables from every single country were individually compared to the same variables from the rest of the countries, in that case allowing us to determine which of the factors creates the biggest control gap. Quick introduction of the results says that the most consistent variables are the im-portance of the subsidiaries and the size of companies. They found support in both indi-vidual- and regional-base analyses. The institutional influence also proved to be a rele-vant factor of the control differences. Most disputed variable was the subsidiaries’ age (as part of the MNC). It was rejected in the regional analyses and in the individual analyses as well. Number of business lines when considered regionally was important control factor, whereas individually it was insignificant. Detailed summary of the results and the tables is offered in the remaining of this chapter.

Table 3: Regression on the individual variables in the whole region

Variable Coefficient Std. Error t-Statistic Prob.

CONTROL 0.468262 0.097401 4.807579 0.0000 LINES_5_50 -0.034030 0.023855 -1.426536 0.1551 LINES_50 -0.070629 0.034398 -2.053267 0.0412 MEDIUM_COMP 0.038751 0.024804 1.562283 0.1197 LARGE_COMP 0.080159 0.029192 2.745897 0.0065 AGE -0.005386 0.003770 -1.428770 0.1545 OLD_COMP 0.037389 0.039326 0.950742 0.3428 IMPORTPERC_OWNER 0.001299 0.000334 3.884232 0.0001 IMPORTPERC_SUBS 0.001380 0.000914 1.510302 0.1324 SALESPERC_OWNER 0.002030 0.000287 7.076179 0.0000 SALESPERC_SUB 0.002286 0.000820 2.786321 0.0058 INSTIT_INFLUENCE -0.073627 0.040246 -1.829402 0.0687

(at the level of significance α=0.05), leading to conclusion that the difference in the control level cannot be explained by the different age.

The size of the subsidiary shows significance only for the large companies (variable LARGE_COMP) giving support to the first assumption in the hypothesis that the larger subsidiaries would be more controlled. However, even though the assumption also holds for the medium sized subsidiaries (variable MEDIUM_COMP) as well, its importance is insignificant therefore we cannot claim full support to this hypothesis.

Importance of the subunit for the level of control was proved to be both significant and in accordance with the assumption in the hypothesis. All four variables (IMPORT-PERC_OWNER, IMPORTPERC_SUBS, SALESPERC_OWNER and SALES-PERC_SUB) signify that when the subsidiary is more important to its parent, expressed through imports from the HQ and from other subsidiaries and also through sales to the HQ and other subsidiaries, it will be more controlled than when it is less important. All the variables except one (IMPORTPERC_SUBS) were significant and gave support to the hypothesis.

Subsidiaries that had more than 50 business lines (variable LINES_50) were signifi-cantly less controlled than those with fewer number of business lines. Subsidiaries with 5-50 business lines (variable LINES_5_50) also had more autonomy than those with lower number of business lines but in this case the difference was insignificant. These results give support to the suggestion in the H4b stating that more business lines mean reduced control in the subunit.

Table 4: General control level among the countries

Variable Coefficient Std. Error t-Statistic Prob.

CONTROL (SLK) 0.447090 0.022947 19.48335 0.0000

HUN -0.078446 0.032998 -2.377306 0.0180

SLO -0.075295 0.034125 -2.206420 0.0280

POL -0.031124 0.028379 -1.096720 0.2735

EST -0.098724 0.032998 -2.991845 0.0030

Table 4 gives outline on the differences in control level over the subsidiaries taking into consideration all the factors together. Slovakia was chosen as the base country in this case and was compared to the remaining countries. Interpretation of the result is that the subunits from all the other countries have greater degree of autonomy then those from Slovakia with Poland coming next followed by Slovenia, Hungary and finally, Estonia as a country with most autonomous subsidiaries. The control discrepancies were signifi-cant between Slovakia and the rest of the countries apart from Poland. This alludes that the variances in control level do not originate only from the individual control factors but they show some general patterns as well.

In addition to the previous regressions five more series of regressions were run with each of the countries having a role of base scenario. Influence on the control level of each of the independent variables from the base country was estimated against the influ-ence of the same variable from the rest of countries. This was conducted in order to show the individual relevance of these factors when measured on a country- and not on a regional- level. Eventually 50 more regressions were made and they are all presented in the Appendix A.

between the control levels in the large subsidiaries with Estonia’s large subunits repre-senting the highest amount of autonomy before Poland, Hungary, Slovenia and Slova-kia. The same could not be said for the small subsidiaries since the relations between the countries were insignificant. Importance of the subunit could also not explain the differences in control between subunits coming from various countries. The number of business lines is taught to play no part, too.

INSERT TABLES 5-14 FROM APPENDIX A

Slovakia in the base scenario exampled that difference in age and the number of busi-ness lines are not able to give explanation of the control gap. Importance of the subsidi-ary complements this group with all but one relation being insignificant. Size difference was only significant when compared to Estonia. Slovakian subsidiaries were more con-trolled when large, whereas Estonian subsidiary had less autonomy when small.

INSERT TABLES 15-24 FROM APPENDIX A

Hungary was the third country in the base scenario. Age, again was found irrelevant fac-tor in explaining the control variances on the subsidiary level in almost all cases. Slove-nia was the only country where older subsidiaries had significantly higher autonomy than their counterparts in Hungary. Between the Hungary and the rest of the countries no relevant relationship was found regarding the factor size. Importance of the subunits in Poland gives the subsidiaries more decision making freedom than to those in Hun-gary but no link was detected with the rest of the countries. Number of business lines was unimportant in this situation.

INSERT TABLES 25-34 FROM APPENDIX A

INSERT TABLES 35-44 FROM APPENDIX A

The last base scenario is represented by Slovenia. In this case all the indicators were found insignificant excluding the size of the subsidiaries but only when related to the Estonia. Estonia’s large subunits are less controlled whereas the small once had more restrictions in comparison with the subunits from Slovenia.

INSERT TABLES 45-54 FROM APPENDIX A

To summarize, according to the results, it can be said that when applied to a broader as-pect (the region) hypotheses H3, H4b and H5 were supported, H2a was partly accepted and H1, H2b and H4a were rejected. It was also shown how control levels resemble various patterns in each of the countries without being particularly linked to any of the control factors individually. Finally, we saw that some of the factors when examined separately are not appropriate for clarifying the control differences. The most irrelevant were the age and number of business lines of the subunit. Size and importance of the subsidiaries occasionally found some support in the examples.

3.4. Discussion

The results from the previous chapter gave, to a certain extend, expected answers com-pared to what the theory preceding this research has found in the same field. Here I spe-cifically mean of the factors like the importance of the subsidiary and the level of

insti-tutional influence whose significance in determining the level of control was confirmed

Comfortably said the most surprising case of all was age of the subunits as part of the MNC whose significance in relation to the control level was largely denied in this re-search. Literature states many reasons for delegating more autonomy to the subsidiaries that have been working in the MNC for longer period. Whether it was the trust factor (Egelhof and Gates, 1984), the creation of its own networks during the time (Tavares and Young, 2004) or the evolutionary process of mandate gaining (Birkinshaw, 1996),

none of them eventually mattered when it comes to CEE subsidiaries.When looking for

reasonable explanations answers could be found in the claims that in the CEE countries before and during the transition period companies that turned later to MNC subsidiaries were involved in activities drastically different from the once they faced after entering the world market. Accordingly, it is difficult to believe that the resource base they had previously created was strong enough to earn them more autonomy within the MNC (Drogendijk, 2005). Since their market orientation in pre-subsidiary stage was basically local and their marketing and financial skills fairly poor it makes these subunits

consid-erably dependent on their parents especially in these areas (Uhlenbruck, 2004).

Al-though the relations between the “old” subsidiaries among the countries were insignifi-cant they still showed tendencies in the control level. Hungarian and Slovenian subsidi-aries enjoyed the most freedom while the Polish and Slovakian the least. This could be the case because the average age of the subunits was more or less the same in all coun-tries except in Hungary that started the market openness process some time before eve-ryone else, thus, had the “oldest” subsidiaries in the area (Varblane, Mannik and Han-nula, 2005). In Poland and Slovakia on the other hand enterprises had received substan-tial state’s help and protection, so half way through the transition period when they be-came foreign property, they had not really created solid resource base to yield more

ne-gotiating force against the parent’s directives (Bohle and Greskovits, 20071). Despite

the irrelevance in this research age of the subsidiaries should not be dismissed as an im-portant factor since it has found strong support in other earlier findings. The only thing that could be altered in the future is considering the absolute age of the unit (since its foundation) and not only its subsidiary status. This might explain the differences in the creation of resource base, networks establishment and performances, hence can be re-lated to the level of control in the subunits.

Speaking of the number of business lines, when going through the tables one could note that the results comply with the assumption about the greater number of business lines

determinant was measured between the countries it showed no relevance. Summarizing, the number of business lines could be an important factor in determining the control level in the subsidiaries (Varblane, Mannik and Hannula, 2005; Cray, 1984) however, aiming to discover the real meaning of this factor, we have to look beyond its scope and

turn to its structure i.e. business functions. The structure of business functions would

show that there are certain types of functions that are traditionally more decentralized than the others. For instance the financial control and global marketing are probably the most controlled functions whereas the local marketing and the employment manage-ment are usually left out to the subsidiary’s responsibility (Varblane, Mannik and

Han-nula, 2005; Bowman, Duncan and Weir, 2000; Egelhoff and Gates, 1984). Drawing a

parallel line with the CEEC several things come to mind. For a long period of time CEE companies had been oriented towards the local market thus possessing extensive knowl-edge about the market and its preferences. That is why most probably they could have more freedom in respect to the local marketing and maybe employment policies. On contrary to this, the financial decisions and the general marketing decisions would re-main in the hands of the HQ, since the CEE subsidiaries as part of the previous regimes and even after worked below their efficiency levels with no global market interests, sheltered by their government’s “saving hand”. It is also not excluded that the HQ would run the employment policies in their subunits as well, since some practical ex-amples show infiltration of expatriate managers from the HQ to their subsidiaries (O’Donnell, 2000).

As the hypothesis for higher degree of control in larger subsidiaries was supported and it complements to the findings of Chang and Taylor (1999) and Dhanaraj and Beamish (2004) still does not really solve the controversy surrounding the size of the

subsidiar-ies.In the general estimation the size was found significant and positively correlated to

indi-vidually created resource base, consequently attracting more surveillance from the HQ. In the cross-country regressions Hungarian and Slovenian subunits were leaders in the decision-making power they possessed (the results were insignificant). It once again gives prove of these countries advanced participation in the multinational activities and the influence of the FDI which in these two states arrived earlier than in the rest of CEEC (Kokko and Kravtsova, 2006).

Importance of the subsidiary and the institutional influence were two indicators that had prominent role in determining the control levels in the subsidiaries. For better compre-hension of the former factor it could be useful to take an insight view in a topic closely related to the control issues and that is the subsidiary’s role. It was deliberately avoided until this point in order not to be confused with the main research problem in the thesis, which is the control level in the subsidiaries. Nevertheless, short reference to the sub-sidiaries’ roles could bring us closer to understanding the importance of the subsidiary as a control factor. Subunits within the MNC are given various roles. These roles to large extend define the intensity of the interactions (subsidiary’s importance) between the HQ and the subunits. Moreover, they have crucial meaning for the degree of control in these units (Kokko and Kravtsova, 2006; Birkinshaw and Morrison, 1995).

relative to the subsidiaries from the other CEE countries, hence will have more omy. Polish subsidiaries were among those who also enjoyed higher levels of auton-omy. Explanation for that might be sighted in the findings of Bohle and Greskovits

(20071), who claim that Poland belongs to the group of countries that successfully

ca-nalized the inward FDI in the targeted technology- and skill- intensive industries. This leads to conclusion that Polish subsidiaries would be engaged in more R&D activities, would have higher innovative capacities and therefore will be more independent (Kokko and Kravtsova, 2006). Considering the subsidiaries’ importance Hungarian and Slova-kian subunits were least autonomous.

The institutions in the CEE region are considered to be relatively new although some of them were inherited from the previous regimes most of them emerged during the transi-tion period. Their introductransi-tion and development was carefully observed and guided by the transnational institutions like the EU and WTO. This led to believe that the institu-tional influence could not be as noticeable as in the advanced western economies or it will not largely vary within the region. On contrary to this, this research exhibited facts that there are differences in the institutional influences among the CEEC and that these differences are significant. Slovenia followed by Estonia and Hungary were the coun-tries with the most influential institutions, whereas Poland and Slovakia had the weakest

ones. Bohle and Greskovits (20071) again give rational explanations to this situation. In

their view Slovenia was the only CEE country that managed in the process of transition to preserve and upgrade the strong institutional base while opening the market slowly and cautiously for the foreign investors. Estonia was one of the countries that based its growth strategy on macroeconomic stability and attraction of the massive number of foreign enterprises. The market was highly liberalized but the institutions were doing good job of maintaining the economic and social fluctuations stable. The remaining countries were characterized with well managed industrialization process but fail to provide macroeconomic security. Their institutions were overpowered by the large for-eign companies and some of them were even intentionally discriminated in order not to obstruct the transition goals. Hungary was the only one that stood out of this group be-cause the early transformation made the home institutions slightly more prepared in the process of market liberalization.

that respect. This general pattern should not be taken for granted without considering several more factors like for example geographical proximity, cultural differences, in-dustry type and etc (Strijker, 2006; Dikova and Witteloostuijn, 2005). After all the lev-els of control in the subsidiaries are not given for good and can be easily changed by the slightest change in any of the above mentioned indicators.

4. Conclusions and Limitations

In the final part of the thesis the aim would be to summarize the answers provided for the questions asked at the beginning of this research as well as to point out several limi-tations and suggestions for future research. The feedback provided by the parallel calcu-lation of the subsidiaries characteristics’ and environment’s influence over the level of control in the subunits confirmed that both factors indeed have substantial role in de-termining this level. What it turned out not be a clear cut is whether the combination of both really enhances subsidiary’s position within the MNC, since some of the units with influential characteristics had little support from their home institutions and vice versa. Even so, the importance of the institutional environment proved to be much more straightforward than the one of the subsidiary’s individual characteristics. Further on, the findings in this research suggest that among the CEE countries the levels of control over the subsidiaries differ one from another however, reasons for that should not be only sought in the subsidiary aspect (characteristics/environment). For better and more assuring conclusions on the control level in the subsidiaries, extended number of factors would have to be considered. Here I think of the HQ’s aspect as well, including the HQ’s home environment and characteristics, company’s overall strategy, together with factors like the type of industry, physical and cultural distance between the parent and the subunit and etc. At the end, as stated before, control in the subsidiaries could not be fixed to a certain level as it is frequently changing due the many influences it is exposed to.

It could be interesting to see how the autonomy of the CEE subsidiaries evolved espe-cially after their entry in the European Union in year 2004.

5. References

Bandelj, N. (2003) ‘Varieties of capitalism in Central and Eastern Europe’ For the

presentation at the Society of Comparative Research, Princeton University

Bartlett, C. and Ghoshal, S. (1989) ‘Managing across Borders: The transnational So-lution’ Boston, MA: Harvard Business School Press

Bjorkman, A. (2007) ‘Towards explaining the use of control mechanisms in foreign subsidiaries MNCs’ Master Thesis, Swedish School of Economics and Business

Ad-ministration, Helsinki, Finland

Birkinshaw, J. and Morrison, A. J. (1995) ‘Configurations of strategy and structure in subsidiaries of multinational corporations’ Journal of International Business

Stud-ies, 1995, Vol. 26 Issue 4, p. 729-753

Birkinshaw, J. (1996) ‘How Multinational Subsidiary Mandates Are Gained and Lost’ Journal of International Business Studies, Vol. 27, No. 3. pp .467-495

Birkinshaw, J. and Hood, N. (1998) ‘Multinational Subsidiary Evolution: Capability and Charter Change in Foreign-Owned Subsidiary Companies’ The Academy of

Management Review, Vol. 23, No. 4, pp. 773-795

Birkinshaw, J. (2001) ‘Strategy and Management in MNE Subsidiaries’ Oxford

Handbook of International Business, Chapter 14, pp. 380-402

Bohle, D. and Greskovits, B. (20071) ‘Capitalist diversity in Eastern Europe’

Eco-nomic Sociology-the European electronic newsletter, Volume 8, Number 2

Bohle, D. and Greskovits, B. (20072) ’Neoliberalism, Embedded Neoliberalism, and

Bowman, S., Duncan, J. and Weir, C. (2000) ‘Decision-making autonomy in multina-tional corporation subsidiaries operating in Scotland’ European Business Review,

Vol. 13, No. 3, pp. 129-136

Chang, E. and Taylor, M.S. (1999) ‘Control in Multinational Corporations (MNCs): The Case of Korean Manufacturing Subsidiaries’ Journal of Management Vol. 25,

No .4, 541-565

Chung, L.H., Gibbons, P.T. and Schoch H.P. (2000) ‘The influence of the subsidiary context and head office strategic management style on control of MNCs: the ex-perience in Australia’ Accounting, Auditing and Accountability Journal, Vol. 13, No.

5, pp. 647-666

Cray, D. (1984) ‘Control and coordination in multinational corporations’ Journal of

International Business Studies, Fall, 85-98

Dikova, D and Van Witteloostuijn, A (2005) ‘Managerial satisfaction with subsidiar-ies performance: The influence of the parent MNE’s specificitsubsidiar-ies and subsidiary’s environment’ Desislava Dikova’s Doctoral Dissertation: Studies on Foreign Direct

Investments in Central and Easter Europe: Establishment Modes, Performance and Strategies of Western European Enterprises, University of Groningen

Dhanaraj, C. and Beamish, P.W. (2004) ‘Institutional Environment and Survival of Overseas Subsidiaries, Journal of International Business Studies, June

Doz, Y. L. and Prahalad, C. K. (1981) ‘Headquarters Influence and Strategic Con-trol in MNC’ Sloan Management Review Vol. 23, No. 1, p.15

Drogendijk, R. (2005) ‘The development of network relations of MNC subsidiaries: how internal MNC and external (local) relations evolve’ Tilburg University, No. 2005-128, ISSN 0924-7815

Dyer, J. H. and Singh, H. (1998) ‘The Relational View: Cooperative Strategy and Sources of Interorganizational Competitive Advantage’ The Academy of

Egelhoff, W. G. (1984) ‘Patterns of Control in U. S., UK, and European Multina-tional Corporations’ Journal of InternaMultina-tional Business Studies, Vol. 15, No. 2, Special

Issue on Strategic Planning, Autonomy and Control Processes in Multinational Corpo-rations, pp. 73-83

Ferner, A. (2000) ‘The underpinnings of bureaucratic control systems: HRM in European Multinationals’ Journal of Management Studies 37:4

Ferner, A., Lmond, P., Ian, C., Colling, T., Edwards, T., Holden, L. and Muller-Camen, M. (2004) ‘The dynamics of central control and subsidiary autonomy in the man-agement of human resources: Case study evidence from US MNCs in UK’

Organi-zation Studies Vol. 25, No .3, pp. 363-391

Fong, E. A. and Tosi, H.L. (2007) ‘Effort, Performance and Conscientiousness: An Agency Theory Perspective’ Journal of Management, 33; 161

Forsgren, M., Holm, U. and Johanson, J. (2005) ‘Managing the Embedded Multina-tional, A business network view’ Department of business Studies, Uppsala University,

Sweden, Edward Elgar Publishing, Inc.

Gates, S. and Egelhoff, W.G. (1986) ‘Centralization in Headquarters-Subsidiary Re-lationships’ Journal of International Business Studies, Vol. 17, No. 2, pp. 71-92

Geppert, M. and Williams, K. (2006) ‘Global, national and local practices in multi-national corporations: towards a sociopolitical framework’ Intermulti-national Journal of

Human Resource Management 17:1, 49-69

Ghauri, P. N. (1992) ‘New structures in multinationals based in small countries: A network approach’ European Management Journal 10, pp. 357-364