The Export Propensity of MNC Subsidiaries
Evidence from CEE
By
LI HUANG
S1623362
University of Groningen
Faculty of Management and Organization
MSc International Business & Management
Landleven 5
9747 AD Groningen
tillyhuang@hotmail.com
Supervisor: Dr. Christoph Dörrenbächer
Dr. Henk Ritsema
The Export Propensity of MNC Subsidiaries Evidence from CEE
Abstract
This study examines the determinants of multinational company (MNC) subsidiary export propensity in manufacturing sector in five transition economies in Central and Eastern Europe (CEE) by analyzing the factors with regard to MNC strategy, host country environmental characteristics and subsidiary characteristics. A dataset with 401 foreign owned subsidiaries in Poland, Hungary, Slovakia, Slovenia and Estonia was used in this study. Independent sample T-test and multiple regression were performed to test the hypotheses. Empirical analysis finds that:(1)subsidiaries established through Greenfield investment have a higher export propensity than those through Merger&Acquisitions;(2) subsidiaries located in host countries with lower wage rate, higher labour productivity, better quality infrastructure, higher openness to trade and more efficient institution have a higher export propensity; (3) subsidiaries with lower autonomy, higher product development capability are more export oriented, and larger subsidiaries have higher export propensity, and no relationship is found between subsidiary age and subsidiary export propensity.
Keywords: FDI; MNC subsidiary export propensity; MNC strategy; host country environmental characteristics; subsidiary characteristics.
Acknowledgements:
Table of Contents
Chapter 1 Introduction ……….4
1.1 Problem Statement ………4
1.2 Main Research Question………5
1.3 Significance of the Research ………...………..6
1.4 Research Outline ………...7
Chapter 2 Theoretical Background and Hypotheses ………..8
2.1 Theoretical Background and Literature Review ………....8
2.2 Hypotheses ………...12
2.2.1. MNC Strategy ……….…….13
2.2.2. Host Country Environmental Characteristics ………..16
2.2.3. Subsidiary Characteristics ………21
2.3. Conceptual Model ……….………..25
Chapter 3 Research Methodology ………...26
3.1. Data Sources ...……….…...26
3.2. Operationalization of the variables ……….27
3.2.1 Independent Variables ………...…………27
3.2.2 Dependent Variable ………...30
3.3. Tests ………30
3.4. Sample Size ……….………...………...31
Chapter 4 Results ……….32
4.1. Descriptive of sample variables ………..32
4.2. Tests and Results ……….33
Chapter 5 Discussion ………..38
Chapter 6 Concluding Remarks, Policy implications and Limitation …………...42
6.1. Concluding remarks and Policy Implications ……….42
6.2. Limitations and Future Research ………44
Chapter 1. Introduction
“Exports have more value to people than is commonly appreciated… Exports enhance workers’ remuneration, benefits, skills, productivity and future prospects. Exports enhance corporate innovation, economic stability and ability to endure external shocks. Exports benefits workers and owners in small businesses, as well as large.” (Richardson and Karin,1995) 1
1.1. Problem statement
It is widely recognized that an increase in the exports of a country has a positive effect on the growth of a nation economy. Although foreign investors may mainly focus on local market of the host economy if a large local demand exists, the activities of multinational companies(MNCs) still have very important impacts on the growth of world exports (Kirkpatric andYamin,1981), since multinational company (MNC) subsidiaries are accounting for a substantial fraction of total manufactured exports of host economies and a large proportion of exports from MNC subsidiaries is intra-firm or inter-firm trade in character.
MNC subsidiaries are increasingly contributing to the growth performance of host countries, either directly through job creation, capital and tax revenues, or indirectly, via productivity, technology, management know-how spillovers to indigenous firms because of a more dynamic export strategy. Hence, the stimulation of MNC subsidiary exports is of great interest in public policy (Tavares and Young 2002). Policymakers in host countries need to know the factors influencing export orientation of MNC subsidiaries in order to take appropriate measures to encourage the subsidiaries to export. There is a large body of literature on the factors influencing domestic firms’ export performance, however, to my knowledge, few studies have been conducted on the determinants of export propensity in foreign subsidiaries. Egelhoff et al.(2000) point out that little work has been undertaken to explore the specific trade flows
produced by multinational affiliates (as cited by Tavares and Young, 2002). Therefore, the main objective of this thesis is to identify possible determinants of, and factors related to MNC subsidiary export propensity to add the limited existing literature on the determinants of export propensity of foreign subsidiaries in five transition economies in Central Eastern Europe(CEE), and to derive implications for host country policy.
1.2 Main Research Question
The reason why I focus on transition economies in CEE will be explained and main research question will be formulated below.
In view of the discussion above, the main research question is formulated as follows: What are the determinants of MNC subsidiary export propensity in the manufacturing sector in Central Eastern Europe? And what are the policy implications?
1.3. Significance of the Research
Insight into the factors that influence the export propensity of MNC subsidiaries can be useful for the decision making process of policymakers when they attempt to design appropriate measures to encourage foreign owned subsidiaries to export. And as pointed out before, only few studies on the determinants of MNC subsidiary export propensity have been conducted and this subject area remains relatively unexplored, although it is considered important. Besides, most research has been carried out in single countries and single industries (Dominguez &Sequeira, 1993;Hogenbirk & Kranenburg,2006). And as far as I know, no one has undertaken research on this subject using CEECs as a whole. This thesis aim to fill the gaps by using an existing “CEEC subsidiary database” 2 from multiple countries and multiple industries in the unique context of transition economies in CEE, namely,Poland, Hungary, Slovakia, Slovenia and Estonia. This study is thus expected to contribute to expending the limited existing literature on foreign owned subsidiary export propensity in the setting of transition economies in CEE. In addiction, institutional efficiency is used for the first time as an explanatory variable in the present study. The reason for studying the impact of institutional efficiency is that ‘corporate strategies in the transition economies can thus be explained only by incorporating the specific institutional context in the analysis.’ (Bevan et al., 2000,p.2), since individual CEE countries have achieved different degree of progress in transition which results in different institutional frameworks.
2
1.4. Research Outline
Chapter 2 Theoretical Background and Hypotheses
In order to investigate what factors influence MNCs subsidiary export propensity, we need to know why MNCs set up subsidiaries to export to the home country or a third country, since MNCs can serve overseas market by exporting from home country or building up plants in host countries through direct investment to serve host market. In other words, we need to understand why MNCs undertake export-oriented investment. And hypotheses will be formulated afterwards based on academic and empirical literature.
2.1. Theoretical Background and Literature Review
Theoretical paradigms from Vernon (1966), Frobel et al.(1980) and Dunning(1981,1988) provide explanation for MNC export-oriented investment. And as literature suggests that the roles of subsidiaries impact the propensity to export, I use the taxonomy from White and Poynter (1984) in this study, because it is the most commonly accepted and clearly defined. The above mentioned theories which will help us to explore the determinants of MNC subsidiary export propensity and previous research are presented in the following section.
happen in the maturity stage of product cycle with a greater importance of cost saving.
Frobel, Heinrichs and Kreye (1980) used their theory of new international division of labour to explain the international location of production process by MNCs. According to Frobel et al, technology innovations and advanced organizational methods enable certain production processes to be divided into different parts. Fragmented components of products requiring different levels of labour skills can be produced in different plants located in different host nations. MNCs can concentrate on higher value-added activities which need higher level skills in the home bases to minimize the utilization of skilled and expensive labour, and relocate the routine or standardized work which requires lower level skills in less developed countries to exploit the cost advantages of abundant unskilled or semiskilled but low-wage labours. In this way, MNCs can minimize the total unit cost of production of the final product. Meanwhile, the improved transportation and communication facilities make this kind of fragmented production more feasible. MNCs rationalize the production across borders on most economical bases to grow in terms of turnover and profits. Because the production process is fragmented, the production units in home and host countries are integrated vertically which results into intra-firm or inter-firm trade of intermediate products between the multinational parent and its subsidiaries, which in turn affects foreign owned subsidiary export propensity (as cited by Kumar,1994).
company; Resource Seeking(supply oriented):obtaining and ensuring raw material supplies for international production within MNC global networks and Efficiency Seeking (rationalized investment): making efficient use of resources by combining existing different factor endowments in host countries with mobile capital, technology. One more type is Strategy Asset Seeking (Dunnning, 1998): protecting, or enhancing ownership advantage by acquiring new assets, or by establishing partnership with a foreign firm. Most export oriented investments are efficiency seeking in essence and undertaken by MNCs relying on the presence of ownership-specific advantage, location specific advantage and internalisation. The ownership-specific advantage is necessary for export-oriented investment and includes organization and human skills to manage overseas production, a legally protected right or a commercial monopoly, economies of scope, and wide international distribution networks. Because of imperfect markets and uncertainty of markets, a firm should internalize its ownership-specific advantages instead of selling or leasing them to other firms. The higher degree of imperfect markets and uncertainty, the firm will exert a higher degree of control over its own activities and its overseas subsidiaries’ activities. Export-oriented investment can exploit internalisation advantage which decreases uncertainty of markets, threat of loss of trade and R&D secrets, maintains product quality and secures delivery time. In addiction, there should be favorable location conditions in host countries such as material or labour resources, infrastructure, political conditions, institution, outward-looking strategy undertaken by host government, since export-oriented investments put more attention on cost consideration and are highly integrated within international production networks. Therefore, the location specific advantage has an important impact on the attractiveness of host country for export-oriented foreign investment. MNCs’ decision regarding export-oriented investment will determine the exporting mandate designated to subsidiaries, as a result, the location factors attracting export-oriented investment will influence subsidiary export propensity.
repatriation restrictions have negative relationships with the location of export-oriented investment but political stability has a significantly positive impact on the location attractiveness of export-oriented investment. Kumar (1994) examines how the structural and policy factors influence the attractiveness of host country as an export-oriented production base by U.S. MNCs. His empirical results indicate that wage rates, infrastructure and open economy of the host country are important determinants of export-oriented production attractiveness of host country.
Tavares and Young (2002) use a survey-based dataset from four EU developed host countries, namely, Portugal, Spain, Ireland and UK to examine the influence of subsidiary roles and characteristics on subsidiary export propensity. Based on White and Poynter(1984), Tavares and Young categorized subsidiaries into three groups as Miniature Replica subsidiaries, Rationalized Manufacturer and Product Mandate subsidiaries. And their findings show that Miniature Replica subsidiaries have a negative relationship with export propensity, Rationalized Manufacturer and Product Mandate subsidiaries have a positive relationship with export propensity. Export oriented subsidiaries have a lower degree of autonomy. But their hypothesized relationships between subsidiary characteristics including age, size, specialized capability, entry mode and export propensity were rejected.
Hogenbirk and Kranenburg(2006) investigate the relationships between subsidiary roles, characteristics and export propensity of foreign owned subsidiaries in the Dutch electronics and electrical applications industry. And their empirical test provides some interesting results as follows. Subsidiaries in the Netherlands are more exported oriented since the size of host market is relatively small. Subsidiaries whose parents undertake multidomestic strategy mainly distribute and market their products in the host country. Their findings do not support the positive relationship between subsidiary size and export propensity. Different from Tavares and Young (2002), they find younger subsidiaries are more export oriented.
2.2. Hypotheses
location for export-oriented production in a significant manner. MNC strategy decision regarding export-oriented investment will determine the exporting mandate designated to subsidiaries, as a result, host country environmental characteristics will influence subsidiary export propensity. Besides, subsidiary strategic roles and other characteristics have important impacts on subsidiary export propensity. Hence, variables regarding MNC strategy, host country environmental characteristics and subsidiary characteristics and their relationships with subsidiary export propensity will be analyzed and hypotheses are generated in this coming section.
2.2.1. MNC Strategy
Firstly, since it is the headquarter of MNC that makes the decision to set up overseas subsidiaries to produce for exports, MNC strategy is one of the most important factors which determine subsidiary export propensity. Therefore, I explore the impact of a set of MNC strategy in this part.
2.2.1.1. International production organization: horizontal investment and vertical investment
takes place between parent companies and foreign subsidiaries, which should result in a relatively higher degree of subsidiary export propensity.
Building on the research by Caves (1971), Andersson and Fredriksson (1996) argue that the factors influencing subsidiary export propensity should correlate to the organization of international production. The variation in subsidiary export propensity depends on the extent to which firms adopt horizontal or vertical international production organization. They used data of Swedish owned subsidiaries in manufacturing sector to make the first empirical investigation for the trade pattern at subsidiary level. And their findings indicate that vertical integrated subsidiaries have a higher propensity to export. Meanwhile, they also find that subsidiary size positively influences foreign owned subsidiary export propensity and there is a negative relationship between the size of the host country market and subsidiary export propensity.
2.2.1.2 Multidomestic strategy and Global strategy
Bartlett and Ghoshal (1989) assert that worldwide companies pursue a variety of strategies to respond to changes in their global environment to facilitate their growth and international expansion. And they categorized the worldwide companies based on different strategy into four types of organizations used to manage international business: Multidomestic3, Global, International and Transnational Company which have specific characteristics of organizing international business activities. To study the export propensity of subsidiary, following Tavares & Young (2002) and Hogenbirk & Kranenburg(2006), I only focus on Multidomestic and Global strategy in this thesis, because these two strategies are the most widely accepted by international business researchers and clearly defined. Companies following multidomestic strategy highlight the importance of country-to-country differentiation. In order to successfully feed the different demands in the host market associated with different political, economic, social forces, local customers’ tastes, practices and preferences, multidomestic
3
companies decentralize their organizational assets and capabilities to grant foreign subsidiaries more autonomy to operate locally with a strong sensitivity and responsiveness to national differences. Subsidiaries adapt products to conform to the local tastes, practices and preferences. Therefore, we expect subsidiaries of multidomestic companies will mainly sell their outputs in the host economy to achieve a greater local market share. Companies following global strategy are pursuing global-scale efficiency oriented by exploiting potential global-scale economies in all international operation. ‘Building on cost advantages’, companies treat the worldwide market as ‘an integrated whole’ with the convergence of tastes and preferences, these companies will develop standardized products which are less sensitive to local requirement so that they can sell products worldwide easily from some concentrated location to get the most value from the least cost of their outputs. Hence, we can expect that subsidiaries of global companies will likely have a higher propensity to export than those of multidomestic companies.
2.2.1.3 Entry Mode
Multinational subsidiaries can be established through Greenfield investments or cross-border Mergers& Acquisitions (M&As). The impact of entry mode on subsidiary export propensity can be argued from two theoretical perspectives: linkages and coordination costs.
seeking a cheap production location for the labor-intensive manufacturing of products or fragmented components which are mainly produced by export oriented subsidiaries.
From a coordination costs perspective, mergers and acquisitions attach more transaction costs to integrate local firm into the parent production network because of different organizational structures, different corporate cultures and technology. And as discussed before, production cost considerations are more important for export-oriented foreign investment(Ekholm et al.,2003), hence, M&As is a less favorable entry mode for export-oriented investment.
In sum, both of the linkages perspective and coordination costs perspective would lead us to expect that
Hypothesis1: Subsidiaries established through Greenfield investments have a higher export propensity than those established through Mergers&Acquisitions.
2.2.2. Host country environmental characteristics
After MNCs make the decision to expand their international production networks, they will choose an optimal host country which provides favoured location advantages (Dunning, 1981) to locate export-oriented subsidiaries, which in turn impacts subsidiary export propensity. Therefore, after examining the determinants of subsidiary export propensity from the MNC strategy perspective, secondly, I will investigate the influence of host country environmental characteristics on subsidiary export propensity in the following part.
2.2.2.1 Labour of costs
Woodward and Rolfe(1993) argue that labour cost advantage is a primary motivation by MNCs to integrate less developed countries into their international production networks. The decision to establish assembly plants in less-developed countries is mainly influenced by a low labour costs in terms of wage per worker.
Bevan et al. (2000) argue that a low labour costs is an important location advantage of a potential host nation, especially for MNCs which are seeking optimal locations to produce standard or mature products for a wider global market. Given differences in labour costs, foreign investors exploit factor differences and establish export-oriented production in less developed countries with lower labour costs, particularly for labour intensive products.
Phisalaphong (2004) argues that as MNCs in industrialized countries innovate and steadily upgrade their domestic activities, they tend to more reply on capital and skilled manpower and remain heavily intensive in using unskilled or semi-skilled labour to take advantage of the economies of specialization (Dunning,1981). Facing intensive competition, firms have been under increasing pressure to relocate production activities to less developed countries that are still a step behind home country with a lower labour costs, which is somewhat similar to Frobel et al.(1980) theory of new international division of labour. Therefore, cheap labour costs have been a major location advantage to attract foreign investors to build up export-oriented manufacturing subsidiaries in less developed countries.
different locations to rationalize international production to seek efficiency with an attempt to maximize the profit. High labour productivity can decrease the unit cost of products, especially for high technology industries which require good quality labours with higher productivity.
Building on these arguments, I formulate hypotheses as follows:
Hypothesis2a: there is a negative relationship between wage rate and subsidiary export propensity.
Hypothesis2b: there is a positive relationship between labour productivity and subsidiary export propensity.
2.2.2.2. Institutional Efficiency of Host Country
Institutions have a significant effect on economic performance and enterprise strategy (Bevan et al.2000). Entering transition economies, foreign investors usually do not have sufficient information about local partners and ‘face unclear regulatory frameworks, inexperienced bureaucracies, underdeveloped court systems, weak protection of intellectual property and widespread corruption’ (Meyer, 2001). All of this increases their investment costs such as searching appropriate partners, negotiation, monitoring and enforcement costs associated with market transactions. Efficient institutions can provide the (formal and informal) rules of the game of a market economy which can reduce both transaction and information costs by decreasing uncertainty and building up a stable structure to facilitate transactions (Bevan et al.2000).
export-oriented foreign subsidiaries (Ekholm et al. 2003). The quality of laws or regulation may all affect the type of activity choices made by foreign entrants. Protection of both proprietary knowledge and property rights, for instance, the protection of intellectual property rights, influences foreign investors’ expectations of the stability of business environment in the host country within which firms operate, and therefore, it has an important impact on the choice only serving small local markets by setting up marketing satellites or developing large export-orientated manufacturing in host countries. Since manufacturing for exports requires more stable host environment (Holland and Pain,1998), therefore, it is expected that:
Hypothesis3: there is a positive relationship between institutional efficiency of host country and subsidiary export propensity.
2.2.2.3. Openness to Trade
Kumar (1994) suggests that outward-looking economies can be regarded to be an important policy factor influencing a country’s attractiveness for export-oriented foreign investment, and intra-industry trade shares tend to increase after the host government liberalizes trade. His findings from U.S. affiliates confirm that export orientation of subsidiaries is positive significantly related with their intra-firm imports and he asserts that countries setting up restriction for incoming from parents of MNCs are not probably to be selected as optimal locations for export-orientated production since these countries, to some extent, lose the location advantages.
obligations, will decrease the attractiveness for foreign investors to set up export oriented affiliates in the host nations.
Therefore, the expectation is that:
Hypothesis4: there is a positive relationship between openness to trade of host country and subsidiary export propensity.
2.2.2.4. Infrastructure Quality
As mentioned in the previous literature, advanced technology of transportation and communication make production frategemted feasible, MNCs can concentrate some particular value-added activities in a limited number of locations, and relocate other activities across borders (Dunning,1998). The quality of transportation and communication influences the cost of coordinating and supply of intermediate and end products, since multinational firms need to get access to the good quality infrastructure such as sea ports, roads and etc. Export-orientated investment puts more important concern about costs as it serves home country or a third country from the host country. Hence, it is more sensitive to the quality of infrastructure which is related to investment costs (Carr et al., 2002).
Kumar (2001) conducts an empirical analysis of the role of infrastructure in explaining the export-orientation of foreign affiliates by using data from 66 countries over 1982-94. And controlling other factors constant, his findings support that MNCs are more inclined to establish export-oriented subsidiaries in the host nations with better quality of infrastructure, especially when foreign subsidiaries are built up for serving the regional or global market, since MNCs want to achieve overall efficiency and competitiveness.
because good quality transportation infrastructure will make transporting goods produced and raw materials or components used for local assembly easier,which will save transportation cost and delivery time.
Hence, I suggest that:
Hypothesis5: there is a positive relationship between infrastructure quality and subsidiary export propensity.
.
2.2.2.5 Size of Host Country Market
Kravis and Lipsey(1982) argue that the role of local demand in host countries is expected to play a less important role when foreign owned subsidiaries focus on sales outside the host market. A large abroad market scope created by exporting can decrease unit costs of products produced when economies of scale exists in production. MNCs which want to gain sustainable competitive advantage have to capture a broader market to make effective use of their worldwide assets (Hogenbirk and Kranenburg 2006). Therefore, foreign owned subsidiaries in small countries are more export oriented because the small host demand is unable to generate sufficient market. In the case of Ireland, Barry and Bradley (1997) find that foreign owned subsidiaries of MNCs in manufacturing industry are almost wholly export-oriented because of the small size of local market of host economy. Empirical results provided by Andersson and Fredriksson (1996) confirm that there is a negative relationship between the host country market size and foreign subsidiary export propensity. Therefore, the hypothesis is formulated as follows:
Hypothesis6: there is a negative relationship between size of the host country market and subsidiary export propensity.
2.2.3. Subsidiary Characteristics
environmental characteristics perspective above, and finally, I focus on foreign subsidiary characteristics relevant to their export propensity based on existing theoretical and empirical literature.
2.2.3.1 Autonomy
As the academic literature on subsidiary development indicates ‘the subsidiary is semiautonomous entity capable of making its own decisions but constrained in its action by the demand of head office managers and by the opportunities in the local environment’ (Birkinshaw and Hood, 1998). The different level of autonomy vested in different subsidiaries reflects the different mandates designated to them by their parent companies (Majcen et al.2006).
Hypothesis7: there is a negative relationship between subsidiary autonomy and subsidiary export propensity.
2.2.3.2 Product Development Capability
Market integration influences foreign firms’ behavior(Phisalaphong,2004). International regional economic integration leads to a reduction in tariff barriers, intra-regional tariff, non-tariff barriers and common external barriers. This will increase the attractiveness of countries located in free trade area for export-oriented foreign investment. More and more foreign investors build up the plants to serve the local market and export for regional market or global market from limited locations to achieve economies of scale, which will intensify the competition between foreign subsidiaries. And product development capability is a critical determinant of which subsidiaries can survive (Birkinshaw and Hood 1998), so this kind of capability is very important to be utilized to gain competitive advantage in not only domestic but also export markets competition (Tavares and Young 2002).Hence,
Hypothesis8: there is a positive relationship between product development capability and subsidiary export propensity.
2.2.3.3. Subsidiary Size
(Pawlik, 2005) to achieve and sustain competitive advantage, and their empirical evidence suggests that there is a positive relationship between foreign affiliate size and export propensity.
Blomstrom (1990) compares the export propensity of Swedish and United States majority-owned affiliates in the same industries and countries and finds that United States subsidiaries are more export oriented because of the greater size which is along with greater access to distribution networks.
The work of Taggart (1996) on Scotland indicates that larger foreign subsidiaries tend to operate in a broader market, sell a higher proportion of products produced to other subsidiaries integrated within the respective MNC production network.
Therefore, the assumption is formulated as follows:
Hypothesis9: there is a positive relationship between subsidiary size and subsidiary export propensity.
2.2.3.4. Subsidiary Age
2006), in other words, younger subsidiaries have a higher degree of export propensity.
Therefore, I make the assumptions as follows:
Hypothesis10a: there is a positive relationship between subsidiary age and subsidiary export propensity.
Hypothesis10b: there is a negative relationship between subsidiary age and subsidiary export propensity.
2.3. Conceptual Model
In order to offer a clear picture of the hypotheses identified above, a conceptual model is formulated below based on the previous discussion.
Host Environmental Characteristics: Labour Costs
(wage rate and labour productivity) Institutional Efficiency of host country Openness to Trade
Infrastructure Quality Market size of host country MNC Strategy:
Vertical Integration & Horizontal Integration Strategy
Global & Multidomestic Strategy Entry Mode
Subsidiary Characteristics: Autonomy
Product development capability Subsidiary Size
Subsidiary Age
Chapter 3. Research Methodology
I use deductive research approach to conduct the present study. It can be seen from the thesis research outline that theory background and literature review are conducted first, and then hypotheses are formulated based on literature. After that, statistical tests are undertaken to examine the hypotheses, and empirical results are presented. In the coming section, the research methodology is elaborated in details to show how empirical findings come up step by step.
3.1 Data sources
One of the challenges we face for conducting research is data availability. Due to time constraints for this thesis, it is not realistic to obtain primary data by carrying out a survey which is commonly used in studies on MNC related subjects. Therefore, secondary data will be retrieved to generate a workable dataset for this thesis. Data sources are presented hereby.
Host country level data are derived from different international organization websites: WorldBank4, International Labour Organization(ILO) 5, European Bank for
Resconstruction and Development (EBRD) 6 and European Commission(EC) 7.
3.2. Operalizalisation of the variables
This study follows a quantitative research approach and presents a theoretical model in order to examine the relationship between MNC strategy, host country environmental characteristics, subsidiary characteristics, as independent variables and subsidiary export propensity as dependent variable. Due to data availability, only one MNC strategy named entry mode is measured. Host country environmental characteristics include labour costs, institutional efficiency of host country, openness to trade, infrastructure quality, size of the host country market. And subsidiary characteristics include subsidiary autonomy, product development capability, subsidiary size and subsidiary age. The operationalization explains how MNC strategy, host country environmental characteristics, subsidiary characteristics and subsidiary export propensity will be measured.
3.2.1 Independent Variables
3.2.1.1 MNC Strategy :
Entry Mode: In the existing “CEEC subsidiary database”, entry mode was not clearly defined, however, following Kokko&Kravtsova (2006), I assume if the year of the establishment is the same as the year when firms were registrated as foreign investment enterprises, this group of subsidiaries should be established through Greenfield Invest- -ments. Other subsidiaries were considered to be established trough Merger&Acquistio- ns(M&As).Therefore, I use a dummy variable to define entry mode: dummy equals to 1 for Greenfield entry mode, 0 for M&As entry mode.
3.2.1.2 Host Country Environmental Characteristics
Labour Costs
Wage Rate: This variable was measured in average monthly wages, original data are derived from ILO Yearlydata:http:www.laborsta.ilo.org. This website provides average monthly wages in local currency. In order to secure the data comparable, I converted different currency into US dollars via exchange rates.
Labour Productivity: Data are derived from Holland and Pain (1998), the authors used whole economy employment and output at constant 1995 prices converted using the 1995 Purchasing Power Parity(PPP) values to calculate the labour productivity.
Institutional Efficiency of Host Country: Following Holland and Pain (1998), institutional efficiency of host country was measured in the host country average score on the nine Transition Indicators for which the EBRD produce rankings, because the country-specific progress in transition can best capture the variation of institutional frameworks in CEE ( Bevan et al.,2000). EBRD use transition indicators to judge the country’s progress in transition in terms of small and large scale of privatization of enterprise ownership; governance and enterprise restructuring; price liberalization; trade and foreign exchange system; competition policy; banking reform and interest rate liberalization; securities markets and non-bank financial institutions. The scores of each indicator range from 1 to 4+, and higher scores indicate greater progress in transition.
Data is retrieved from http://www.ebrd.com/country/sector/econo/stats/tic.xls.
are drawn from the database of World Bank.
Infrastructure Quality: This variable was measured as the density of motorways in host countries. Data are retrieved from http://epp.eurostat.ec.europa.eu/.
Openness to Trade (OT): Following Kumar (1994), this variable was measured as the percentage of imports and exports in Gross Domestic Product (GDP) of host countries. Data are derived from the database of World Bank.
3.2.1.3. Subsidiary Characteristics
Autonomy: The information from the existing “CEEC subsidiary database” resulted from EU Fifth Framework Project allows us to specify subsidiary autonomy. In this EU funded survey, one of the questions was asked about the decision making process between the local affiliate and the parent company. The question asked was: which business functions are being undertaken (a) on your own only, (b) mainly on your own, (c) mainly by your foreign owner, or (d) by your foreign owner only. The value range is from 0 to 1 ( 0;0.33;0.67;1 according to a,b,c,d). So higher value, less autonomy. Therefore, following Männik et al. (2004), autonomy is measured by mean of business function autonomy.
Product Development Capability: In the same survey, foreign owned subsidiaries were asked who undertook the product development. Following Kokko and Kravtsova (2006), we can use the answers to define the subsidiaries’ product development capability based on the product development mainly undertaken by the subsidiaries or by the parent company. Therefore, product development capability was measured by innovative capability of product development. Lower score in the responsibility for product development capability indicates the higher initiative undertaken by the subsidiaries, the higher product development capability the subsidiaries have.
Subsidiary Age (SA)*: This variable was measured as Logarithm number of establishment years by using the formula: SA=lg (2007 – G) (G stands for the year when firms were registrated as foreign investment enterprises).
*The reason I measured subsidiary size and subsidiary age in Logarithm total number of employee and logarithm number of establishment years instead of total number of employee and number of establishment years respectively are explained below:
Linear regression tests assume that the scores of the continuous variables fall in a normally-distributed curve. When I run Histograms with SPSS to test the distribution of variables, I find that variables regarding subsidiary size and age do not follow the shape of the normal curve, so I transform these two variables into the form of Logarithm, then the scores are normally distributed. So I decide to use Logarithm of size and age in the regression test to avoid the violation of the normality assumption of linear regression to achieve reliable results. The transformed results are showed in Appendix1,2.
3.2.2. Dependent Variable
Export Propensity: This variable was measured as the percentage of exports in total sales of foreign owned subsidiaries.
3.3. Tests
In order to examine to what the extent by which MNC strategy, host country environmental characteristics and subsidiary characteristics influence subsidiary export propensity, after reviewing some statistics textbooks ( Pallant, 2001; Norusis, 2002 ), I make my decision to perform the following statistical tests to examine hypotheses by running SPSS 12.0.1 for Windows Programme.
Hypotheses 2ab,3,4,5,6; independent-samples T-test will be conducted. There are only five host countries investigated in this sample, the same values of environmental characteristics are repeated over different values of the dependent variable. I categorize different countries into two groups based on variable values. Therefore, the independent variables are categorical and export propensity as dependent variable is continuous. In this case, using independent samples T-test will find out whether there is a statistically significant difference in the mean scores for two groups to achieve the results of corresponding hypotheses.
Hypotheses 7,8,9,10ab; a multiple regression will be performed because I want to exam how well a set of firm-specific characteristics can influence subsidiary export propensity. Independent variables are continuous and dependent variable is continuous, which fulfills the requirement of a multiple regression.
3.4 Sample Size
In order to use statistics test, it is important to get a sample with significant observation units. There are different guidelines for the number of observation units needed to undertake statistical tests. According to Tabachnick and Fidell (1996), the number of observation units should satisfy the formula: N>50+8m (m stands for the number of independent variables) (as cited by Pallant, 2001).As statistical techniques are sensitive to outliers, extreme values are removed from the dataset and the observation units are reduced from 435 to 401. Since there are 11 independent variables in this study, the sample size fulfills the requirement for sufficient power of tests, so we can expect reliable empirical results.
Chapter 4. Results
4.1. Descriptive statistics of sample variables
Before starting hypothesis test, we need to review features of the data used in this study to get a general idea of: what host country environment foreign owned subsidiaries are located in, what strategy MNCs undertake to build up plant units and what kinds of subsidiaries are established in CEE countries, since I aim to explore how these factors influence subsidiary export propensity.
Table 1 summarizes the host country environmental characteristics in the dataset. Estonia, Slovakia and Slovenia are relatively smaller than Poland and Hungary which have larger local markets. The average wage rate is relatively lower in Hungary, Slovakia and Estonia compared with Poland and Slovenia. The labour productivity of Poland, Hungary and Estonia are lower than other countries in the sample. Slovenia has the highest density of motorway. Poland and Slovenia have a lower level of openness to trade. Poland and Hungary have made a more progress in transition than the rest of countries observed in this study.
Table 1 Descriptive statistics: host country characteristics
Country Population (million) Wage Rate (USD) Productivity Infrastructure (km/100 km²) OTrade % Institutional Efficiency Poland 38 713.99 0.89 0.1 29.0 38.77 Hungary 10 572.69 1.03 0.4 73.5 40.32 Slovakia 5 538.59 1.17 0.4 77.5 35.26 Slovenia 2 1029.67 1.88 1.4 57.5 34.90 Estonia 1 471.24 0.61 0.1 85.5 35.22 Average 11.2 665.24 1.12 0.48 64.6 36.89 Source:Worldbank(2001),ILO(2001),EBRD(1989-2001),EU(2003), Holland&Pain (1998) and author’s own calculation.
The average age of subsidiaries is 22, ranging from 8 to 143; the average amount of employees working in the foreign subsidiaries is 366, varying from 6 to 3500, the average score of autonomy enjoyed by the subsidiaries is 0.35, which suggests that parent companies grant a relatively higher autonomy to their subsidiaries in CEE countries because lower score indicates lower control by parent companies, therefore, higher autonomy is granted to the subsidiaries. Regarding product development capability, we can see that average 51% of product development initiative is undertaken by subsidiaries, and at the same time, subsidiaries still partly rely on the research and development from parent companies. The percentage of export share in total sales varies from 0 to 1.00, most subsidiaries export more than half of their output to serve foreign market. As I study subsidiary export propensity in the manufacturing sector, not only focus on exporting sector, I also include the subsidiaries which do not export into my dataset.41.4% of subsidiaries were established through Greenfield investment.
4.2. Tests and Results
In the coming section, hypotheses are tested and empirical results are presented against the statistical significance level which I chose p=0.05 for all tests in this study.
Firstly, I will perform independent samples T-test to test the hypotheses1 regarding MNC strategy, statistical results are indicated in Appendix 6.
Secondly, I will conduct independent samples T-test to test the hypotheses 2ab,3,4,5,6 regarding host country environmental characteristics. I categorize different countries into two groups based on the independent variable values: Group 1: those with actual values greater than the average values; Group 2: those with actual values lower than the average values. And then I perform the tests. The empirical results are shown in Appendix 7-12.
From Appendix 6, we can see that there is significant difference in mean scores for subsidiaries established through Greenfield investments (M=0.63,SD=0.37), and subsidiaries established through Merger&Acquisiton(M&As) [M=0.49,SD=0.36; t(283)=-3.264, p=0.001], p<0.05.Therefore, we can say that subsidiaries established through Greenfield investments have a higher export propensity than those established through M&As. Hypothesis 1 is supported.
From Appendix 7, we can see that there is significant difference in mean scores for higher wage rate (M=0.46,SD=0.38), and lower wage rate [M=0.58,SD=0.39; t(399)=-3.196, p=0.000], p<0.05. As lower wage has a higher mean score of export propensity, we can say that there is a negative relationship between wage rate and subsidiary export propensity, Hypothesis 2a is supported.
From Appendix 8, we can see that there is significant difference in mean scores for higher labour productivity (M=0.67, SD=0.37), and lower labour productivity [M=0.44,SD=0.37; t(273)=6.07, p=0.000], p<0.05 As higher labour productivity has a higher mean score of export propensity, we can say that there is a positive relationship between productivity and subsidiary export propensity, Hypothesis 2b is supported.
From Appendix 9, we can see that there is significant difference in mean scores for higher institutional efficiency (M=0.65,SD=0.37), and lower institutional efficiency [M=0.41,SD=0.36; t(399)=-6.441, p=0.000], p<0.05. As higher institutional efficiency has a higher mean score of export propensity, we can say that there is a positive correlation between institutional efficiency and subsidiary export propensity, Hypothesis 3 is supported.
From Appendix 11, we can see that there is significant difference in mean scores for higher motorway density (M=0.71, SD=0.31 and lower motorway density [M=0.48, SD=0.39; t(110.898)=-5.227, p=0.000], p<0.05. As higher motorway density has a higher mean score of export propensity, we can say that there is a positive relationship between infrastructure quality and subsidiary export propensity, Hypothesis 5 is supported.
From Appendix 12, we can see that there is significant difference in mean scores for larger population (M=0.33,SD=0.34) and smaller population [M=0.61, SD=0.37; t(298.39)=7.40, p=0.000], p<0.05. As smaller population has a higher mean score of export propensity, we can say that subsidiaries located in smaller host market has a higher export propensity. Hypothesis 6 is supported.
Finally, as mentioned before, I will run multiple regression to test the relationships between a set of subsidiary characteristics and export propensity which are indicated in hypotheses 7,8,9,10ab. Firstly, I undertake Person’s correlation to check if each of the subsidiary characteristics relates to subsidiary export propensity. And keep in mind that I chose the statistical significance level: p=0.05 for all the tests.
The Appendix 13 shows that there is correlation between subsidiary size and export propensity [r=0.150,n=396,P=0.003<0.05], there is no correlation between subsidiary age and export propensity [r=0.056,n=398,P=0.261>0.05] Hypotheses 10ab are rejected. There is significant relationship between autonomy and export propensity [r=0.437, n=401, P=0.000<0.05], and statistical correlation between product development capability and export propensity [r=0.183, n=388, P=0.000<0.05].
There are a few variables taken into account in the multiple regression model, so we need to check the multicollinearity which takes place when the independent variables are highly related. And multicollinearity may be a problem for analyzing the output of the multiple regression, if the independent variables are significantly related, which will reduce the efficiency of the multiple regression estimations. As tolerance indicates the strength of the linear relationships among the independent variables. We check the value of tolerance. According to Norusis(2002),when tolerance value is close to 1, it suggests that an independent variable has little of its variability explained by the other independent variables. When tolerance value is close to 0, it indicates both independent variables are highly correlated. Output presented in the table labeled Coefficients shows that tolerance value of each independent variable is more than 0.5 which is far from 0, so the results presented here do not suffer from multicollinearity for the independent variables used in this multiple regression model. Therefore, I do not violate the assumption of multiple regression because independent variables are respectable.
Before interpreting the relationships between independent variables and dependent variable, firstly, we investigate the robustness of the results by examining the value of R Square in the Model Summary box which indicates how much of the variance in the dependent variable is explained by the independent variables. We can see the value of R Square is .242, which means independent variables included in this model explains 24.2% of the variance in subsidiary export propensity. Secondly, we check whether the results of the model is statistically significant or not. From the ANOVA table which indicates the model as a whole is significant or not, we can see F(3, 379)=40.379, Sig.=0.000,(p<0.05), which indicates that the model as a whole is significant and the model goodness of fit is confirmed. Therefore, the independent variables included in multiple regression model have a respectable explanatory power for the dependent variable, although this model explains 24.2 percent of the variance in subsidiary export propensity. Statistical results are presented in Appendix 14
seen that subsidiary size has positive coefficient of 0.097 at the significance level of 0.034(p<0.05) which suggests that larger subsidiaries have a higher degree of export propensity. Hypothesis 9 is supported. And then we turn to the relationship between autonomy and export propensity and the correlation between product development capability and export propensity. Referring to the measurement of these two independent variables, the lower score, the higher autonomy and capability, so we should interpret them the other way round based on the measurement. Although the output table shows that product development capability has a negative coefficient of 0.178 at the significance level of 0.003(p<0.05), we should say that there is a significantly positive relationship between product development capability and export propensity, which means subsidiaries with a higher product development capability have a higher export propensity. Hypothesis 8 is supported. The same story applies to autonomy, because the output table shows that autonomy has a positive coefficient of 0.565 at the significance level of 0.000(p<0.05), we should say that there is a significantly negative relationship between autonomy and export propensity, which means subsidiaries with a lower degree of autonomy have a higher export propensity. Hypothesis 7 is supported.
The overall hypotheses test outcomes are summarized in table 2 as follows:
Table 2 Hypotheses and Outcomes
Hypotheses Expectated Relationship Results No.1 Entry Mode --> export propensity Greenfield: +; M&A: - supported No.2a Wage Rate --> export propensity - supported No.2b Labour Productivity --> export propensity + supported No.3Institution Efficiency --> export propensity + supported No.4 Openness to Trade --> export propensity + supported No.5 Infrastructure Quality --> export propensity + supported No.6 Size of Host Country Market --> export propensity - supported No.7 Autonomy --> export propensity - supported No.8 Product Development Capability --> export propensity + supported No.9 Subsidiary Size --> export propensity + supported No.10a Subsidiary Age --> export propensity + rejected No.10b Subsidiary Age --> export propensity - rejected
Chapter 5 Discussion
After conducting different statistical tests to investigate how the three level of factors influence subsidiary export propensity in five CEE transition economies, empirical results are presented in Chapter 4, hypotheses and outcomes of tests are summarized in table 2.And now I discuss the findings as follows.
The result with regard of entry mode variable indicates that subsidiaries established through Greenfield are more export oriented than those built up by Merger&Acquisition entry mode. As stated before, export oriented subsidiaries put more importance on cost saving. Greenfield ventures are set up with new facilities, and foreign owners can set up appropriate governance structure and company culture in line with the parent company’s overall strategy to decrease internal management and operation conflicts, which can decrease the coordination cost within the multinational international operation networks under common leadership. The result here conforms to the findings of Kalotay and Hunya (2000) who argued that privatization acquisitions in Hungary have lower propensity to export, smaller and less import intensive than Greenfield investments because acquisition investment attempts to make good use of the existing local linkage to serve local market.
With regard to labour costs variable, the hypotheses of the two perspectives, namely, wage rate and labour productivity related to export propensity are supported. The countries with lower wage rate or higher labour productivity are more attractive for export-oriented investment, which justify the arguments from Woodward and Rolfe(1993) and Cushman (1987) that lower wage rate and higher labour productivity can decrease the unit cost of products overall and they are important potential location advantages for export-oriented investment.
transition economies, however, after the host countries were categorized into two groups, the findings here do capture some difference regarding institutional efficiency impact on export propensity of subsidiaries. Poland and Hungary have made more progress in transition than the rest of countries observed in this study, so these two countries are more attractive for foreign investors to built up export oriented subsidiaries, which is in line with Bevan et al. (2000) who argue that with progress in transition, the host country has a stable structure and policy environment which can decrease the information, transaction costs and reduce uncertainty of investment environment. The countries offering this favorable business environment will be more attractive.
The result regarding openness to trade variable suggests that the host country with higher degree of openness to trade are more attractive for export-oriented investment as proposed by Hypothesis 5. The policy of host country influences the export behavior of subsidiaries. As we argued before, foreign owned affiliates are accounting for more and more substantial fraction of exports from host countries which are intra-firm or inter-firm trade in essence. MNCs relocate different fragmented production in the different locations across borders to minimize the cost overall. Under this situation, the intermediate or final products need to be moved within the global network under common leadership smoothly and with as least cost as possible. So host countries which undertake open outward-looking policy and put less requirements on local content and ownership or retractions on employment of foreign personnel, profit repatriation restriction will increase the attractiveness of location for export-oriented foreign investment, put it in other way, multinational parents will incline to designate the export mandate to local subsidiaries. Therefore, subsidiaries located in these countries will have a higher export propensity.
either the home country maket or a third country market is to reap the different factor endowments in the host economy through relocating different stage of production in the different locations to minimize the cost overall. In this case, the quality of transportation and communication influences the cost of coordinating the production and supply of intermediate and final products. Therefore, it is not surprising that foreign investors are more inclined to locate export oriented subsidiaries in the host country with better infrastructure.
The result regarding size of host country market variable meets the previous expectation, the host market size has a significantly negative relationship with subsidiary export propensity as proposed by Hypothesis 6. When foreign investors decide to enter certain countries, the size of market is put on the important position. In order to protect foreign market share, foreign investors build up plants in small countries. However, as the small size of local market of host economy can not generate sufficient demand, MNCs are under great pressure to explore new markets to make full use of assets including foreign ownership advantages and location advantages to achieve low unit cost overall and sustain competitive in global competition. The findings here justify that compared with Poland and Hungry with large local market, foreign owner affiliates in Estonia, Slovakia and Slovenia have a relatively higher export propensity because of small host markets.
The subsidiary size variable was significant and positive, as proposed by Hypothesis 9, which suggests that larger subsidiaries have higher export propensity. As I argue that internationalization requires sufficient various resources with regard to physical, financial perspectives and greater international marketing network. Meanwhile, larger subsidiaries are under greater pressure to make efficient use of large scales of existing resources to achieve economies of scale and scope to minimize the cost overall to capture efficiency and competitiveness. Hence, it is not surprising that larger subsidiaries have a higher export propensity.
proposed by Hypothesis 7. Based on literature concerning subsidiary autonomy, different types of subsidiaries are granted with different level of autonomy by parent company. Autonomy is determined by what mandate designated to subsidiaries or what roles subsidiaries play and how subsidiaries are integrated in the multinational company international production networks. Overseas subsidiaries either serving home country or a third country market are embedded in MNC global distribution network or production network. In order to secure the coordination of different subunits to achieve the common strategy executed by multinational parent, export oriented subsidiaries only can enjoy limited autonomy which satisfies the mandate they need to fulfill. The findings confirm the empirical result from Taggart and Hood (1999) who find less autonomy subsidiaries of German and Japanese manufacturing affiliates located in the British Isles have higher export propensity.
The product development capability variable displays the expected positive impact on subsidiary export propensity, Hypothesis 8 is supported. Regional economic integration results in a single market with less tariff or non-tariff barriers. Foreign owned affiliates located in the integrated regional area have access to the entire free trade area which provides a much larger market expending the limited host market. More and more foreign investors set up subsidiaries to compete for market share, which requires subsidiaries have sufficient capability to develop and produce new products to meet the demand of regional market.
Chapter 6 Concluding remarks, policy implications and limitations.
6.1. Concluding Remarks and Policy Implications
The present study has been undertaken to examine whether MNC strategy, host country environmental characteristics and subsidiary characteristics have significant influence on MNC subsidiary export propensity with an attempt to fill previous research gaps and provide some policy implication for host country policy makers. The determinants of foreign owned subsidiary export propensity are relatively unexplored, the present study uses a sample of five transition economies by conducting different statistical tests to investigate how the above three level of factors influence subsidiary export propensity to make a contribution to the expansion of the existing literature in the context of transition economies in CEE. Besides, institutional efficiency is used as an explanatory variable for the first time in this study and proved to be one of the important factors with potential impact on subsidiary export propensity.
After summarizing the related literature and previous research, the hypotheses with regards to the relationship between MNC strategy, host country environmental characteristics, and subsidiary characteristics and subsidiary export propensity are formulated. The independent sample T-test and multiple regression statistical tests are undertaken to achieve the empirical outcomes.
improve the labour productivity, host government need to invest considerable amount money to develop the education programmes and offer more opportunities for citizen to attain higher level of training. In order to attract export-oriented investment, host countries compete by offering insensitive, tax holiday, exchange rate depreciation which may decrease the welfare of host country (Oman ,2000), I argue here that host countries should focus on the development of public infrastructure in their countries, which both domestic ventures and foreign enterprises can benefit from and expedites the process of their development in the long term(Kumar, 2001). Coming to institutional efficiency, CEE countries make different progress in transition from central planning to market based economy, which results in different institutional frameworks. Host country should provide efficient institution system and stable politic environment which can reduce uncertainty of host country investment environment, therefore, increase foreign investors’ confidence to set up export-oriented subsidiaries integrated within the global networks. Subsidiaries located in small host countries are more export oriented because of small size of local market. Foreign owned subsidiaries account for more and more substantial fraction of exports from the host economies. In order to achieve export oriented foreign investment, policy makers should further reform outward-looking policy with certain reduction in requirements on local content and ownership or retractions on employment of foreign personnel, profit repatriation restriction, since these requirements reduce the attractiveness of host nations as favorable locations for export-oriented investment.
have a higher product development capability are more export oriented. I argue that, on one hand, since multinational companies have a higher control over export oriented subsidiaries, multinational parents are willing to let subsidiaries get access to the ownership advantage such as distribution channels, innovative technology which can contribute to the development of host economy with expending world market share and updating industry development. However, on the other hand, export oriented subsidiaries likely established by Greenfield investment have lower linkages with local firms, which may limit spillovers in terms of technical and organizational know-how towards indigenous firms from foreign investment, and these spillovers are important positive effects of FDI (Hastenberg,1999). Larger subsidiaries have a higher export propensity, which indicates that export-oriented foreign investment can create many job opportunities. Since foreign companies tend to provide in-house education and on-the-job training to employees, therefore, this may improve the skills and productivity of labour of host countries. And in turn host countries can attract more FDI because of this location advantage.
6.2. Limitations and Future Research
While this study has increased the insight on the determinants of subsidiary export propensity in CEE, due to time constraints in writing this thesis, there are some limitations which are presented as follows.
First of all, the time series dimension and the amount of sample countries in the dataset are limited. In order to generalize the findings of empirical study, a sufficiently large dataset needs to be used. The analysis would have more explanatory power if it is based on a larger sample and a longer time frame. Related to this is the fact that only five sample countries and a period of one year have been investigated in the present study. Further research is suggested to include more sample countries and a greater time series dimension, and thereby generate a more robust overview of the actual empirical results.
characteristics of different industries should be taken into account to see if there is any difference in empirical outcomes with respect to the industry specific impact on subsidiary export propensity.
Reference
Anderson, T., Fredriksson,T. (1996). International Organization of Production and Variation In Exports From Affiliates Journal of International Business Studies. 27(2), 249-263
Bartlett, C.A., Ghoshal, S. (1989). Managing Across Borders: The Transnational Solution Boston: Harvard Business School Press.
Blomstrom, M. (1990). Transnational Corporations and Manufacturing Exports from Developing Countries New York:UNCTC 35-36.
Bonaccorsi, A. (1992). On the Relationship Between Firm Size and Export Intensity Journal of International Business Studies, 23, 605-635.
Barry, F., Bradley, J. (1997) FDI and Trade: The Irish Host-Country Experience The Economic Journal, 107,1798-1811.
Birkinshaw, J., Hood, N. (1998). Multinational Subsidiary Evolution: Capability and Charter Change in Foreign-Owned Subsidiary Companies The Academy of Management Review, 23(4), 773-795.
Bevan A. A., Estrin, S. (2000). The Determinants of Foreign Direct Investment inTransition Economies William Davidson Institute Working Paper Number 342, 2000/10 1-57
Bevan, A., Estrin,S., Meyer, K. (2000). Institution Building and the Integration of Eastern Europe in International Production
