University of Groningen
MSc. International Business &
Management
Uppsala University
MSc. Economics & Business
The Relationship between Internationalization and
Performance of SMEs in an Emerging Economy
August 2011
Author: Léone Slütter
Supervisor: Dr. M. P. van der Steen
2
The Relationship between Internationalization and
Performance of SMEs in an Emerging Economy
Léone Slütter
Abstract
The aim of this research is to examine the relationship between internationalization and performance of SMEs in an emerging economy. The study focuses on the internationalization strategies exporting and FDI, and firm performance in the form of profitability and growth. Using a multiple regression analysis for two samples of Croatian SMEs in the time period 2006-2009, evidence is found for a positive linear relationship between the level of export and profitability, as well as exporting and growth. There is also evidence for a positive linear relationship between FDI activities and growth, while the relation between FDI activities and profitability is U-shaped non-linear. These results suggest, SMEs in an emerging economy should explore or expand their internationalization activities and opportunities in the form of exporting or FDI to improve their profitability and growth rates.
Keywords
3 TABLE OF CONTENTS 1. INTRODUCTION 5 2. LITERATURE REVIEW 10 2.1. Introduction to Internationalization 10 2.2. Internationalization of SMEs 10 2.2.1. Stage Theory 11 2.2.2. Network Theory 13
2.2.3. International Entrepreneurship Theory 13
2.3. Forces of Internationalization of SMEs 14
2.3.1. Push Factors 14
2.3.2. Pull Factors 15
2.3.3. Interactive Factors 15
2.4. Internationalization and Performance 16
2.4.1. Main Arguments for a Positive Relationship 16
2.4.2. Main Arguments for a Negative Relationship 17
2.5. Forms of Internationalization 18
2.6. Hypotheses 20
2.6.1. Export and Performance 20
2.6.2. FDI and Performance 21
3. METHODOLOGY AND DATA 24
3.1. Research Methodology 24
3.2. Data Sources 26
3.3. Sample Selection and Design 26
3.4. Variables 27 3.4.1. Dependent Variables 27 3.4.2. Independent Variables 28 3.4.3. Control Variables 28 4. RESULTS 30 4.1. Descriptive Statistics 30
4
5. DISCUSSION 38
6. CONCLUSION 44
7. APPENDIX 46
5
1. INTRODUCTION
Over the last decades, international operations have become more and more important for organizations in virtually all industrialized nations. To maintain competitive advantages and sustain growth, expansion across borders is often a necessity (Hajela & Akbar, 2010). Even if a company itself does not expand its activities beyond borders of the home country, due to extremely integrated markets, global competition and the international environment, almost all organizations can be labeled as international. The international environment exerts a major influence on firms, for instance on their strategy, business models and operations (Majocchi & Zucchella, 2003). Companies tend to view internationalization as an instrument to increase their performance; this assumption has been fundamental to several internationalization theories (Oesterle et al., 2008). However, multiple studies and empirical evidence on the relationship between internationalization and performance of a firm, show different results.
The relationship between internationalization and firm performance has already been an important topic for research since the 1970s (Bausch & Krist, 2007). The first one to hypothesize a positive relationship between the degree of internationalization of a firm and its performance, as a result of emphasized benefits, was Vernon, in 1971. Several investigations followed; in contrast to the first investigations, during the 1980s and 1990s, researchers acknowledged that internationalization of firms could also provoke drawbacks to the before mentioned success, like risk or even failure (Bausch & Krist, 2007). Ever since, the discussion concerning the relationship between internationalization and firm performance continued. Several researchers hypothesized and proved positive linear relations (Siddharthan & Lall, 1982; Buhner, 1987; Kim et al., 1989; Baird et al., 1994; Delios & Beamish, 1999; Contractor et al., 2003; Elango, 2006), while others explored and found significant negative linear relationships (Brewer, 1981; Collins, 1990; Ramaswamy, 1992a) or non-linear relations (Buckley et al., 1977; Geringer et al., 1989; Haar, 1989). Besides these relationships, several studies proved U-shaped (Erramilli & De Souza, 1993; Lu & Beamish, 2001; Capar & Kotabe, 2003), inverted U-shaped (Daniels & Bracker, 1989; Hitt et al., 1997; Gomes & Ramaswamy, 1999; Elango, 2006) and S-shaped/cubic connections (Contractor et al., 2003; Lu & Beamish, 2006; Hsu, 2005). Obviously, the literature concerning the internationalization-performance relationship contains several conflicting results of investigations. As Bausch and Krist (2007) among others describe, the existing state of research has been mentioned as inconsistent (Harveston et al., 1999), contradictory (Tallman & Li, 1996; Geringer et al., 2000) and mixed (Doukas & Lang, 2003).
Most of the aforementioned studies focused on large
6
indicators, for instance the number of employees, profits, market position, annual turnover, number of outlets or total capital (Storey, 1994; Wolff & Pett, 2000; Stephanou & Rodriguez, 2008). According to the European Commission (2009) enterprises qualify as micro, small or medium-sized if they do not exceed 250 employees and if they have an annual turnover of maximum 50 million euro and/or an annual balance sheet total of maximum 43 million euro.
Previously, in consequence of financial and managerial shortcomings, SMEs were conceived as internationally low contributing companies (Golinelli, 1992; Oviatt & McDougall, 1994). However, as research on this topic by the Organization for Economic Cooperation and Development (OECD) (2005) shows, SMEs are developing faster internationally than before and are involved in more international operations (Chelliah et al., 2010). As a result of the rapid integration of economies over the world and declining government-imposed obstacles, the internationalization of SMEs can be expected to be continuously growing. Geographic expansion is one of the most crucial manners for a firm to ensure continuous growth (Lu & Beamish, 2001). Technological developments, declining trade barriers and ongoing globalization are enabling SMEs to be internationally more effective and efficient. As Lloyd-Reason (2003) points out, SMEs experience different competitive forces of domestic, as well as foreign rivals, which is a consequence of the opening up for foreign firms of home markets. SMEs are especially important for the social and economic welfare and growth of a nation, and play a critical role in handling the pressure from overseas companies entering the home market (Pollard, 2001). Therefore, it is important to broaden the area of research on the internationalization-performance relationship of SMEs. SMEs may not automatically be categorized as smaller versions of large corporations; they differ substantially on for instance the field of strategic and financial management, the importance of unique resources, ownership and their internationalization policy. Naturally, outcomes of investigations of large firms cannot simply be copied and used for smaller firms (Dana et al., 1999; Zucchella & Maccarini, 1999).
Previous investigations on the subject of SMEs mainly focused on prior developments and processes of internationalization, instead of a specific focus on the relationship between internationalization and performance (Preece et al., 1999; Wolff & Pett, 2000). Managers of SMEs are especially interested in the contribution of different forms of internationalization to the level of firm performance and the possible competitive advantages of expanding globally. SMEs have fewer resources and face greater risks than large firms when they expand internationally, therefore, research on this topic is crucial for these firms. Existing studies on this topic are often theoretical; there is a need for more empirical research, focusing on SMEs (Fillis, 2001; Lu & Beamish, 2001).
7
developing world‟; over 60% of all companies in emerging economies are SMEs (Tung & Aycan, 2008). As Kula and Tatoglu (2003) point out, SMEs provide significant employment and growth. Emerging economies are in a transitional phase between the developing and developed status. SMEs are responsible for more than half the employment and added value in most emerging nations and their presence correlates with several economic factors, including the growth of a nation‟s gross domestic product (UNCTAD, 1993; Newberry, 2006). The success of domestic internationalizing SMEs is crucial in developing economies, because it has a direct effect on the national economy, it can provide benefits reflected in growth, restructuring and enhanced transition. Besides, SMEs are a source of innovation and future successful large enterprises (Harvie & Lee, 2002).
SMEs in emerging economies need to keep up with their competitors regarding their internationalization process. Pressures of increasing foreign competition and liberalized global markets are pushing SMEs towards internationalization, because mainly internationalized companies will survive. On the one hand, developments in telecommunications and transport, Internet and international trade liberalization result in new business opportunities for SMEs, but on the other hand, they create fierce international competition. Increasingly, SMEs recognize that without globalization of activities, they will lose their position and fall behind in the world market. Recent studies (Mathews, 2006; Luo & Tung, 2007; Svetlicic, 2007; Aulakh, 2009) on the increased share and importance of SMEs show that these firms have to engage in globalization or internationalization to build competitive advantage. They mention that this is important for SMEs in all types of countries, but crucial for firms in less developed countries and European Union (EU) candidate states. Svetlicic et al. (2007) stress the internationalization of SMEs is especially important for countries at the doorstep of joining the EU, because they have to meet certain criteria before they may enter the EU. For example criteria concerning the national economy of the country, on which the internationalization of SMEs has a direct effect.
Earlier findings of investigations on firms in developed countries are often not completely applicable for firms in emerging markets. The environment of firms in emerging markets is characterized by greater governmental interference of market mechanisms, relatively poor governance and nascent institutions, which has an impact on the internationalization-performance relationship (Elango, 2006; Paittnak & Elango, 2007). Further, a great portion of the firms in emerging economies belongs to business groups; they hold political and market power and help affiliated firms with their competitive advantages. This can have a negative influence on the internationalization-performance relationship (Khanna & Rivkin, 2001; Paittnak & Elango, 2007).
8
al., 2008; Yang et al., 2009), India (Saini & Budhwar, 2008), Taiwan and Singapore (Sim & Pandian, 2003) and Vietnam (Thai & Chong, 2008). Because of the growing interest, globalization pressure, the crucial impact and the continually increasing presence of SMEs in emerging economies, it is important to focus empirically on these economies in this field of research (Xia et al., 2007; Chelliah et al., 2010). Besides, existing studies on this topic often use cross-section data, instead of pooling data across different years (e.g. Wolff & Pett, 2000; Fillis, 2001; Majocchi & Zucchella, 2003). In this manner, it is not possible to make causal interpretations or to detect a lag-structure. This study investigates causal connections between internationalization and future performance.
The relation between internationalization and performance of SMEs in emerging economies needs more empirical attention. Therefore, the purpose of this research is to provide a useful complement to existing studies by providing a better understanding of the relationship between internationalization and performance of SMEs in emerging markets. The emerging country this research will focus on is Croatia. A sample of Croatian SMEs is chosen because the economy in Croatia is defined by the International Monetary Fund (2011) as an emerging economy and currently the country is very close to joining the European Union (EU Observer, 2011). Further, there has not been much empirical evidence regarding SMEs in Croatia so far, while, as explained before, this is especially important for countries like Croatia.
There are several forms of internationalization. Multiple studies (e.g., Ramaswamy, 1992b; Majocchi & Zucchella, 2003; Lu & Beamish, 2006) mention exporting and FDI as two of the most prominent avenues of internationalization for SMEs. Hence, the focus will be on the impact of export and FDI on firm performance. Firm performance is a multi-dimensional concept; internationalization can have a different impact on different dimensions of firm performance, particularly for SMEs (Delios & Beamish, 2001; Lu & Beamish, 2006). The two most common goals of expansion to foreign markets are achieving profitability and increasing firm growth (Oviatt & McDougall, 1994). Therefore, the impact on firm performance will be divided into effects on profitability and growth of SMEs.
9
10
2. LITERATURE REVIEW
In this section relevant literature on internationalization of SMEs and the relationship between internationalization and performance will be provided. This section is divided into six parts. In the first part, the concept internationalization will be introduced. The most relevant and important theories of internationalization of SMEs shall be discussed in the second part, to provide a better understanding of internationalization and the different processes and theories on this topic. There will be elaborated on the Stage Theory, the Network Theory and the International Entrepreneurship Theory. In the third part, forces of internationalization will be discussed. SMEs experience pushing, pulling and interactive forces. The relation between internationalization and performance will be discussed in the fourth part; different research outcomes of positive as well as negative relations will be discussed (Oesterle et al., 2008). Main arguments for a positive relationship that will be described are: resource based arguments, organizational flexibility and arbitrage arguments, industrial economic effects, the portfolio theory and organizational learning effects. Arguments for a negative impact of internationalization on firm performance are: increasing costs of coordination and control, liability of foreignness and risks connected to foreign activities. Forms of internationalization will follow in the fifth part; the entry modes exporting and foreign direct investment will be discussed. Finally, in the sixth part the hypotheses of this research will be derived.
2.1. Introduction to Internationalization
There are varying definitions for the term internationalization; Welch and Luostarinen (1988) define it as ‘the process of increasing involvement in international operations’. It is also defined as „the process by which firms, both increase their awareness of the direct and indirect influence of international transactions on their future, establish and conduct transaction with other countries’, by Beamish (1999). In both definitions internationalization is mentioned as a process, instead of an event. Further, they describe outward movements of firms, which is also underlined in the definition of Johanson and Vahlne (1977): ‘Internationalization describes the outward movement in an individual firm’s or larger grouping’s international operations’. Lloyd-Reason (2003) adds to these definitions that internationalization is strategic, gradual and incremental. To internationalize, firms are investing abroad by for example selling their products or starting subsidiaries. Basically, all forms of international behavior of firms can be labeled as internationalization.
2.2. Internationalization of SMEs
11
1988, 1994). The Transaction Cost theory is based on the premise that business activities accomplished by external parties are costly and inefficient and therefore a firm should internalize all possible activities. According to Dunning, the decision to engage in foreign investments depends on the interplay of three variables: ownership-specific advantages, internalization advantages and locational advantages.
However, these traditional theories primarily explain behavior of multinationals; the specific characteristics of SMEs are not considered. Smaller companies often have limited access to resources and problems with resource commitments in conditions of environmental uncertainty, while they face great foreign competition from incumbent domestic firms, as well as large multinationals (Ruzzier & Konečnik, 2006). Given the differences between the types of firms, the transferability of findings is often difficult (Fillis, 2001). Furthermore, the mentioned traditional theories mainly focus on reasons behind international trade; the models in the next paragraphs also pay attention to intra-industry trade in a world with imperfect competition under economies of scale and heterogeneity of firm characteristics. These theories are dynamic, which means that they focus on the process of internationalization, instead of static theories that focus solely on the reason of internationalization. The next mentioned theories are dynamic theories that focus especially on SMEs; these internationalization theories suit better to the characteristics and competencies of smaller organizations. Over the past decades, the internationalization of SMEs has gained more and more attention. Several scholars (Laine & Kock, 2000; Axinn & Matthyssens, 2002; Mtigwe, 2006; Onkelinx & Sleuwaegen, 2008) distinguish three types of internationalization theories: Stage Theory, Network Theory and International Entrepreneurship Theory. In the next part these theories on internationalization will be described, to contribute to the understanding of the holistic internationalization process.
2.2.1. Stage Theory
12
I-models are process models on different stages of the internationalization process. In both models the central focus is the firm‟s involvement in the foreign market.
U-model
The U-model of Johanson and Wiedersheim-Paul (1975) and Johanson and Vahlne (1977) describes the internationalization process as ‘a gradual acquisition, integration and use of knowledge about foreign markets and operations, and a successively increasing commitment to foreign markets’. It shows the process as a practice of organizational learning, focusing on experience. The process evolves in interplay between on the one hand a growing commitment of resources to foreign markets and on the other hand the development of knowledge about these markets. There are four aspects for firms that must be faced when they decide to internationalize. The „stage aspects‟ market knowledge and market commitment, and the „change aspects‟ commitment decisions and current activities, which interact with each other. Market knowledge and market commitment affect both the commitment decisions and the way current decisions are performed. This, in turn, will change knowledge of foreign markets and operations, and commitment of resources in foreign market (Johansen & Vahlne, 1990; Andersen, 1993). Furthermore, Johanson and Wiedersheim-Paul (1975) distinguish between four different entering modes of a foreign market. The consecutive stages of entering in the model represent higher degrees of international involvement. The first stage shows basically no experience in the foreign market; there are no export activities performed by the firm. In the second stage, export occurs via independent representatives. The firm has an information channel to the foreign market and receives consistent superficial information about conditions of the market. An establishment of an overseas sales subsidiary is considered as the third stage. The succeeding business activities in the foreign market provide more differentiated and wide market experience. Finally, during the fourth stage, overseas production or manufacturing units are established. Each of these stages involves increased risk and commitment for the firm. The model explains how foreign market risks can be managed with the use of incremental change of commitment to foreign markets and tacit knowledge about those markets (Andersen, 1993; Mtigwe, 2006). However, according to Oviatt and McDougall (2005), the model focuses solely on traditional cross-border behavior, while it should also focus on accelerated internationalization or entrepreneurial behavior. Other critics dispute the incremental character of the model; they argue the possibility of skipping stages and achieve rapid, instead of gradual internationalization (Chetty & Campbell, 2003).
I-models
13
to start exporting, the number of stages and the description of the stages (Andersen, 1993; Onkelinx & Sleuwaegen, 2008). In these models, the decision to internationalize is considered as an innovation for a company; they focus on the learning sequence in connection with the adoption of an innovation. According to the best known model, the model of Cavusgil (1980), there are five stages to describe the internationalization process: the domestic marketing stage, the pre-export stage, the experimental involvement stage, the active involvement stage and the committed involvement stage. This model has been tested by Gankema et al. (2000), who agreed on the expansion of international involvement of SMEs over time and support the stage theory. They mention that the theory is correct in general; however, some SMEs skip stages or stop before reaching the final stage. To achieve their goals, as for example improved performance or growth, it is important for SMEs to follow an incremental stage process when they are internationalizing.
2.2.2. Network Theory
A challenging approach of the incremental Stage Theory is the Network Theory. According to Mtigwe (2006) the essential argument of this approach is that ‘modern high-technology firms do not exhibit the tortuous developmental process envisaged by the incremental theorists, but rather that the process is short-circuited owing to the experience and resources of network partners’. The theory is based on the assumption that companies share complementary competitive advantages with other companies to drive international growth. SMEs often have problems with specific amounts of knowledge and resources to overcome internationalization barriers. Companies with competitive benefits can leverage the specific advantages of their own company to give advantages to the company itself and the partnership as well (Onkelinx & Sleuwaegen, 2008). According to Johanson and Mattsson (1988), within this approach, internationalization is interpreted as a process with constantly establishing, developing, dissolving and maintaining relationships, with as purpose the achievement of the objectives of the company. Choices of foreign markets, entry modes and strategies of the firm are influenced by various network relations. It is hard for competitors to imitate these strategies, because it is difficult to copy and develop similar network relationships. Therefore, networks are a bridging mechanism to stimulate and support rapid and successful internationalization (Mtigwe, 2006).
2.2.3. International Entrepreneurship Theory
14
value creation process through which an individual engages in innovative, proactive, calculated risk-taking behavior designed to prosecute foreign business opportunities presented by multinational market successes and imperfections for financial and non-financial rewards’. Besides this definition, he distinguishes an important difference between the IET and the Network Theory, which are often erroneously used as synonyms. Contradicting to thoughts of the Network Theory, international entrepreneurship can internationalize in two manners, through formal networks, or without a network. It can exist outside formal networks, which occurs even far more often or at least industry-specific. Therefore, the two theories should be viewed as complements, rather than synonyms.
The discussed theories and models are not able to adequately explain the success of small and medium-sized firm internationalization by itself. However, different elements of the approaches can be combined by firms in their internationalization process. The internationalization behavior of SMEs can be labeled as a holistic process, combining different perceptions from multiple theories and models, like the Stage Theory, Network Theory and IET. The combination of these approaches in the internationalization process of SMEs can contribute to an improvement of the firm performance (Onkelinx & Sleuwaegen, 2008).
2.3. Forces of Internationalization of SMEs
There are several factors with an important influence on the internationalization of SMEs. Etemad (2004) categorized these factors into three groups, namely push factors, pull factors and interactive factors. The three forces are discussed in the following paragraphs.
2.3.1. Push Factors
15
strategic logic of internal operations, like additional resources of international partners (Etemad, 2004).
2.3.2. Pull Factors
Pull factors are the opposite of push factors. These factors are forces from the environment of the firm, thus external forces. Pull forces enhance competitiveness of the firm and attract it with certain incentives, thereby, pulling the firm into the internationalization process (Etemad, 2004). The forces are reactive, pushing the firm away from domestic markets and pulling it towards foreign markets. A key strategic pull driver for SME internationalization is forced reactive survival (Vasquez and Doloriert, 2011). Other important factors are liberalization of international markets and advancement in technologies of information, communication and transportation (Oviatt & McDougall, 1994; Coviello & McAuley, 1999). Foreign markets may provide access to advanced technologies, which are not obtainable in the home market. Firms can facilitate network operations and reduce costs (Etemad, 2004). Attraction of resources or partners can also be a pull factor, for example to increase the speed of internationalization or to shorten the internationalization process. A final pull factor is serving the international requirements of existing buyers and suppliers, to save existing relationships or in response to needs of previous domestic customers (Etemad, 2004; Hajela & Akbar, 2010).
2.3.3. Interactive Factors
16
be an interactive driving factor, like responding to internationalization of customers or taking advantage of international supply chains (Coviello & McAuley, 1999; Etemad, 2004).
2.4. Internationalization and Performance
As mentioned in the introduction, researchers disagree on the relationship between internationalization and firm performance. With the existence of these conflicting findings, several advantages and disadvantages of internationalization on the performance of SMEs have been raised. Oesterle et al. (2008) recapitulate the main reasons for a positive or negative relationship between internationalization and performance into several central arguments, which will be explained in the following paragraphs.
2.4.1. Main arguments for a Positive Relationship
Oesterle et al. (2008) mention five main arguments raised by scholars that discovered a positive relationship between internationalization and performance of SMEs. The categories of these arguments are: resource based arguments, organizational flexibility and arbitrage arguments, industrial economic effects, the portfolio theory and organizational learning effects.
The first arguments are resource-based and concern the distribution of a firm‟s resources to other countries than the firm‟s home country. Those resources are competitive advantages; they can be used internationally and are considered innovations in foreign markets. These competitive advantages generating resources are, for example, technological or firm-specific knowledge, trained and skilled employees, capabilities of the management team, organizational structure, capital and brands. It has been argued that the internationalization of a firm increases in proportion with the ratio of utilization of these resources (Morck & Yeung, 1991; Annavarjula & Beldona, 2000; Lu & Beamish, 2006). The use of these resources in other markets than the home market does not create extra costs; therefore, their utilization is related to increased firm performance (Qian, 1997; Lu & Beamish, 2006).
Second arguments are the organizational flexibility and arbitrage arguments. By increasing the number of countries in which they operate, firms increase their flexibility. Firms can realize higher profits with the use of arbitrage potentials, these are created by price differences between national markets, as a consequence of market imperfections and legal differences. Moreover, because of an international subsidiary network, companies can shift productions or redirect commodity flows in reaction to changing national market conditions (Kogut, 1985; Lu & Beamish, 2006; Bausch & Krist, 2007).
17
distribution of R&D expenses, but also costs of marketing and human resources. Fixed overhead costs of the firm can be dispersed over an increased allocation size when they are not directly connected to the production, because more markets are served simultaneously (Kobrin, 1991; Hitt et al., 1997; Bausch & Krist, 2007). Economies of scale and scope and experience are also often mentioned as positive arguments for the relationship (Markusen, 1984; Halkos & Tzeremes, 2009).
Fourth of the arguments are arguments concerning the portfolio theory. The portfolio theory, originally developed by Markowitz (1959), argues that risks can be reduced by spreading activities. Entrepreneurs want to maximize firm value and returns, while avoiding failure and high levels of risk. By internationalizing, firms are spreading activities over economically not integrated nations, with not-perfectly correlated markets, by which firms are reducing risks. Potential risks are for example changing political situations, bankruptcy threats or fluctuations in cash flow, total profit, demand or sales (Kim et al., 1989; Annavarjula & Beldona, 2000). Oesterle et al. (2008) explain that international SMEs are also used by individual investors to form internationally diversified portfolios, which is rewarded by the investors with an increased firm value.
Final arguments are based on organizational learning. Networks of subsidiaries in multiple countries are suppliers of diverse stimuli and new information to broaden the knowledge base of the SME and improve existing capabilities, thereby creating competitive advantages (Hitt et al., 1997; Lu & Beamish, 2006; Bausch & Krist, 2007). Ruigrok and Wagner (2003) explain that SMEs experience learning opportunities during the internationalization process, which provides them with cumulative knowledge, useful for further successful expansion and improvement of firm performance.
2.4.2. Main arguments for a Negative Relationship
Researchers who discovered a negative relationship raised arguments as well. According to Oesterle et al. (2008), the main three arguments in favor of a negative relationship are: increasing costs of coordination and control, liability of foreignness and risks connected to foreign activities.
18
geographical and cultural dispersion of multiple principals and agents within the SME. Exchange rate fluctuations and inflation will diminish benefits derived from portfolio diversification and political risks may enforce unanticipated change to foreign business environments, which will increase total costs (Ruigrok & Wagner, 2003).
Secondly, Oesterle et al. (2008) describe the argument of liability of foreignness, which refers to lack of local knowledge, experience and a reputation on the foreign market. These will be competitive disadvantages compared to the local companies. As a consequence of important differences between markets, the in the home market developed knowledge and capabilities are not suitable for the new market. Hence, it is necessary for the SME to acquire new knowledge and capabilities, which increases costs and takes time. However, this liability can be conquered over time when an SME builds a reputation on the new market and gains knowledge and experience of the country (Lu & Beamish, 2006; Barkema & Drogendijk, 2007; Bausch & Krist, 2007).
Thirdly, the risks as a consequence of foreign activities are stated. As mentioned before, risks can decrease because of the spreading of activities. However, there are reasons for a risk increase as well. Limited control, incomplete information and a self-fulfilling prophecy effect create greater risks for companies with overseas operations. Limited control and incomplete information are mainly consequences of geographical and cultural distance or agency problems. The self-fulfilling prophecy effect implies that SMEs with international investments tend to expect higher rates of return and therefore take higher risks, which leads to systematically increasing risk of foreign investments (Reeb et al., 1998; Oesterle et al., 2008).
2.5. Forms of Internationalization
19
Exporting is a broadly adopted strategy of SMEs to internationalize. Traditionally exporting has been seen as the first step to entering foreign markets, it is an easy and fast way with relatively low levels of risk and commitment (Kogut & Chang, 1996; Lu & Beamish, 2006). Export is a fast and effective access to new international activities, because firms have the opportunity to acquire meaningful international experience, with little necessary capital investment. Exporting is often an easy strategy to implement, because it is not as complex as several other internationalization strategies. Exporting firms are not dealing with complexities as establishing a foreign subsidiary or building new production facilities in foreign markets (Sullivan & Bauerschmidt, 1990; Zahra et al., 1997; Lu & Beamish, 2001). However, in certain circumstances, FDI becomes a more interesting alternative to internationalize for SMEs than exporting. According to definitions of the IMF (2001) and OECD (2008), a direct investment is „a category of cross-border investment made by a resident entity in one economy (the direct investor) with the objective of establishing a lasting interest in an enterprise (the direct investment enterprise) that is resident in an economy other than that of the direct investor’. The IMF (2001) expresses the owner of 10% or more of the capital of a firm as a direct investor. FDI may be more attractive, for instance, when a firm‟s assets are proprietary. Examples of proprietary assets of a firm are brand equity, trademarks or patents. In these cases, exporting can create risks in terms of opportunism, devaluation and asset appropriation. It can also be a better way to enter foreign markets with an FDI strategy when various barriers by host country governments heavily influence the achieved net benefits from the exporting strategy or when transportation costs are abnormally high (Rugman, 1982; Lu & Beamish, 2001). Both forms of internationalization are of great value if an SME decides to internationalize.
20
2.6. Hypotheses
In this section hypotheses will be derived. The first and second hypotheses concern export and performance while the third and fourth address FDI and performance.
2.6.1. Export and Performance
Exporting brings several economic advantages for SMEs, for instance economies of scale and scope (Markusen, 1984; Kogut, 1985; Halkos & Tzeremes, 2009). Costs of manufacturing in the host country are avoided by exporting and firms can gain advantages from their global sales volume. The flexibility of the firm increases, among others because the geographic scope can easily be changed (Hill, 2007). By exporting, SMEs are able to reach new and more profitable markets (Cavusgil & Zou, 2002), they can exploit a similar niche in different countries and thereby enhance their revenue as well as profit potential (Pangarkar, 2008). Furthermore, presence in multiple, divers markets can increase their market power (Ramaswamy, 1992a; Kim et al., 1993) and provide gains from revenue diversification (Ramaswamy, 1992b). As mentioned before, exporting is an easy and fast manner to enter a new and possibly more profitable market, without the need to establish a foreign subsidiary or production facility. Without significantly higher costs and with relatively little effort, firms can increase their returns and create higher profits. In addition, commitment and risk are relatively low with exporting as internationalization strategy, firms can quickly and easily withdraw from host country markets if this may be necessary as a result of political instability or fluctuating conditions on the market (Lu & Beamish, 2006).
These advantages are crucial for SMEs, because especially these firms are faced with resource constraints and cannot make excessive investments and resource commitments or be exposed to high risks. Because of the mentioned potential benefits and earlier summarized arguments, it is expected that the level of export of an SME is positively related to its profitability. The profitability of an SME is the future return as a result of the current level of internationalization. Therefore, the first hypothesis to be tested is the following:
Hypothesis 1: The level of export activities of an SME is positively related to its profitability.
Besides improvement of a firm‟s performance in terms of profitability, the most common objective of internationalization for SMEs is firm growth (Oviatt & McDougall, 1994; Jones, 1999; Lu & Beamish, 2006). Firm growth is the future growth of a firm as a result of current internationalization in the form of exporting; firm growth is a lagging factor indicating the performance of a firm.
21
economy of emerging countries, governments of these countries even initiated greater efforts in encouraging SMEs to involve in international businesses because growth is crucial in these countries. As Barringer and Greening (1998) explain, geographic expansion in the form of export is one of the most effective and important manners to accomplish growth of a firm. With their appearance in new markets, firms can increase their customer base and ensure larger production volumes and growth. Because of the different impact of internationalization strategies and the fact that growth is a crucial dimension, especially for emerging countries, it is important to test the relationship between the level of export and the dimensions growth and profitability separately.
Internationalization in the form of exporting can be regarded as an important growth strategy for SMEs (Vida et al., 2000). It can have a straightforward contribution to firm growth through sales increase. SMEs can expand their customer base and higher their sales volume by selling to clients in new geographic markets. Higher sales volume can lead to permanent higher production volume and expanded production capabilities, because firms need to meet the demand in new markets. By the broadening of markets and creation of room for expansion, exporting to foreign markets can be considered as a critical path for permanent firm growth. In this process, it is important to pursue developed theories and strategies to accomplish successful internationalization and subsequently lasting future sales growth. For instance, the Network Theory, with networks as bridging mechanisms to stimulate and support rapid and successful internationalization through exporting (Mtigwe, 2006) and the Stage Theory, to increase commitment and sales in foreign markets as an incremental and gradual process (Onkelinx & Sleuwaegen, 2008). By following proper strategies, obstacles and disadvantages like lack of knowledge and experience, limited control, cultural distance and incomplete information can be diminished or prevented, thereby making firm growth permanent and stable. Besides, the by exporting created larger production and sales volumes make it possible for SMEs to achieve economies of scale and increase efficiency of management and labor productivity (Kogut, 1985; Lu & Beamish, 2006). Hence, the second hypothesis states:
Hypothesis 2: The level of export activities of an SME is positively related to its growth.
2.6.2. FDI and Performance
22
learning effects. Exporting and FDI both enable a firm to develop useful knowledge about foreign markets through operations in the host country, while only FDI provides access to several specific location-based advantages (Kogut, 1985). In this manner, organizational learning can be promoted in multiple international markets and a firm‟s international competitiveness can be enhanced, thereby stimulating firm growth. These advantages are key benefits of FDI (Zahra et al., 2000). Besides, FDI creates the leverage of multiple location-based benefits, for example reduced prices of labor and opportunities to create knowledge or skills in technology and improve capabilities as competitive advantages (Rugman, 1982; Kogut, 1985; Shan & Song, 1997; Zahra et al., 2000; Lu & Beamish, 2001). FDI has a great effect when it comes to market expansion and higher sales, besides, future sustainable growth can be created through the development of knowledge and capabilities, which can feed back into greater competitiveness and realize sustainable growth of the firm (Lu & Beamish, 2006). Therefore, current FDI is crucial for future growth of SMEs in emerging countries. Because of these benefits of FDI and earlier mentioned arguments for the relationship, the third hypothesis is the following:
Hypothesis 3: The level of FDI of an SME is positively related to its growth.
23
future profitability (Tallman & Li, 1996; Hitt et al., 1997; Lu & Beamish, 2001). This relationship can be described as U-shaped non-linear (Hitt et al., 1997; Gomes & Ramaswamy, 1999). Prior research on this relationship at MNCs showed an S-shaped relation (Contractor et al., 2003; Vilas-Boas & Suárez-Gonzále, 2007). The first part of the S-form, where profitability is decreasing, is explained by liability of foreignness and organizational learning. The second decline in profitability can be explained by a phase where too many foreign investments are having a detrimental effect on performance of the MNC. SMEs are not in the position to have so many foreign subsidiaries, mainly because of resource constraints. This difference explains a U-shaped relationship for SMEs and an S-shaped relation for larger companies.
These arguments lead to the following hypothesis:
Hypothesis 4: The level of FDI of an SME is U-shaped non-linear related to its profitability.
24
3. METHODOLOGY AND DATA
The aim of this thesis is to investigate the relationship between internationalization and the performance of SMEs in Croatia, thereby providing a better understanding of this relationship for other emerging economies as well. The best method to examine this relation and answer the research questions is to use a multiple regression analysis. The dependent variables, profitability and growth, are modeled as functions of the independent variables, export and FDI activities. There will be elaborated on the methodology and data of this research in the following paragraphs. The research methodology will be explained in the first section, followed by parts on the data source, sample selection and design and variables.
3.1. Research Methodology
In previous research on the relationship between internationalization and performance, different techniques have been used to analyze the collected data, for instance, ANOVA tests or classical multiple regression analysis (Sullivan, 1994; Hitt et al., 1997; Hill et al., 2001; Ruigrok & Wagner, 2003). To answer my research question the technique of multiple regression analysis will be used. Aiken and West (1991) explain two disadvantages of using ANOVA tests: median splits of continuous variables throw away information, which reduces the power of the statistical test and existing significant results can be harder to detect. Regression analysis is used to investigate the statistical significance of an estimated relationship between variables, usually the quantitative causal effect of one variable upon another variable. A multiple regression analysis allows additional variables to enter the analysis separately; in this manner the separate effects of the different factors can be assessed. The impact of multiple simultaneous influences upon a single dependent variable, in this case profitability or growth, can be quantified. A multiple regression can even be essential when the focus is solely on the influence of one of the independent variables, in this investigation export or FDI (Sykes, 1993).
25
standard errors of the coefficients. Next, autocorrelation should be checked; autocorrelation means that the covariance between any pair of random errors is not equal to zero, a correlation of a signal with itself (Parr, 1999). Autocorrelation can be checked with the Durbin-Watson test (Field, 2009). Finally, variables need to be checked for multicollinearity, high correlation between variables. A correlation analysis will be used to test if there are significant relationships between all variables. Multicollinearity can be discovered by inspecting the correlation matrix (Field, 2009). The dependent and independent variables will be tested on correlation; to control for the variable industry, a distinction is made between manufacturing and nonmanufacturing companies. After these tests, the multiple regression analysis will be conducted. The results from the tests will be compared with results from previous investigations, for instance Lu and Beamish (2001; 2006) and Majocchi and Zucchella (2003).
There are different regression models used to investigate the four hypotheses. The following equations are used:
Model I: Profitability, exporting and control variables
Model II: Growth, exporting and control variables
Model III: Growth, FDI and control variables
Model IV: Profitability, FDI and control variables
26
Table 1. Definition of Variables
Variable Definition Abbreviation
Level of exporting activities Export ratio (Export revenue over operating revenue) EXP Level of FDI activities Number of foreign subsidiaries FDI Level of FDI activities (squared) Number of foreign subsidiaries (squared) FDI² Profitability Return on Assets (Net income over total assets) ROA
Growth Annual growth rate of sales Growth
Firm size (Ln) Number of employees Lnsize
Industry (dummy) Manufacturing (1) / non-manufacturing (0) dInd Leverage Debt ratio (Total debt over total assets) Lev
3.2. Data Sources
For this research, secondary data will be used. Time constraints make it too difficult to collect primary data and it is difficult to collect reliable primary data for private companies. Furthermore, all necessary data is available and easy to obtain from secondary sources. The variables necessary for this research will be derived from the Amadeus and ORBIS databases of Bureau van Dijk.
3.3. Sample Selection and Design
The sample in this research consists of SMEs in Croatia that have an internationalization strategy in the form of exporting or FDI activities and have the necessary data available. The SMEs are selected based on the aforementioned definition of the European Commission (2009), which states that an enterprise is small or medium sized if it does not exceed 250 employees and if they have an annual turnover of maximum 50 million euro and/or an annual balance sheet maximum of 43 million euro. Data needs to be available on the following variables: return on assets (ROA), annual growth in sales, export ratio, number of FDIs, number of employees, manufacturing/non-manufacturing firm and debt ratio.
27
There are two different databases used for the hypotheses. The first and second hypotheses concern the level of export, the third and fourth consider FDI. There are insignificant SMEs with data on the level of export as well as number of FDI activities. Therefore, it was necessary to create two separate databases. Dataset I contains data on export ratio, return on assets (ROA), annual growth in sales, number of employees, manufacturing/non-manufacturing firm and debt ratio. Tests for the first and second hypotheses will be conducted with this information. Dataset II it created with data on number of FDIs, return on assets (ROA), annual growth in sales, number of employees, manufacturing/non-manufacturing firm and debt ratio. With this data, tests for the third and fourth hypotheses will be conducted.
In this research, panel data will be used. Panel data is a combination of (simple) time-series data and cross-sectional data. Time series data solely measures changes through time and cross-sectional data are data on one or more variables collected at a single point in time (Brooks, 2002). Cross-sectional analysis measures the changes between individuals or firms. Panel data measures these changes through time. To predict or investigate the influence of one or more variables in one point of time on another variable in another point of time (for example, one year later), panel data is necessary. Panel data is very useful to study in depth complex economic and related issues, which could not be done with the use of time-series or cross-sectional data alone. Other mentioned benefits of panel data are the ability to distinguish inter-individual differences from intra-individual ones, the reduction of estimation bias and the reduction of multicollinearity problems (Hsiao, 1985).
Data is collected for the years 2006-2009. A larger time frame is not possible due to data constraints. For causal interpretations, the independent variables were lagged. According to Lu and Beamish (2006) it is arbitrary to assume a definitive time period for the independent variables to have effect on dependent variables, consequently, there has been experimented with 1-, 2-, and 3-year lag structures. However, the results from these lag structures were qualitatively the same. Therefore, the results will be reported for the year 2009, with a 1-year lag structure. The effect of the level of export and FDI in 2008 (t-1) upon the profitability and growth in 2009 (t) is measured. In this manner, the current internationalization of an SME is measured and analyzed in relation to future performance.
3.4. Variables
3.4.1. Dependent Variables
28
is the most popular method to measure performance, Bausch and Krist (2007) compared the indicators used in 95 studies and in 27 percent of the cases ROA was taken as variable. The numbers for this measure are relatively easy to collect and since it is widely used, it is easily comparable with results from other studies on this topic (Ruigrok & Wagner, 2003). Return on assets measures relative efficiency of firm effectiveness considering output, it measures the profitability of a company relative to the total amount of assets the owners have invested in the business; these assets include working capital as well as fixed assets (Fields, 2002). ROA is calculated by dividing the net income by total assets of the firm.
Likewise, there is a wide variety of used growth indicators, including, market share, enhanced venture capital funding, sales growth, turnover/sales, growth in assets, return on investment (ROI), employment, or a combination of these indicators (Hoy et al.,1992; Davidson et al., 2006; Lu & Beamish, 2006). Davidson et al. (2006) show that the indicator concerning turnover, sales is used in 30% of the investigated cases, hence, annual growth rate of sales will be used as indicator for growth. Similar to the previous reasoning, this indicator is easy to access and results are comparable to results of other studies. Annual growth in sales is computed involving the comparison of sales between two consecutive years.
3.4.2. Independent Variables
Independent variables to measure internationalization are exporting activities and FDI activities. The level of exporting activities will be measured through the ratio of exports, which is often used to measure the degree of internationalization (Ramaswamy et al., 1996). This ratio is calculated by dividing the export revenue by the operating revenue.
Lu & Beamish (2006) developed two manners to measure the magnitude of FDI activities. The first manner is the number of FDIs in which the parent firm has a 10 percent or greater equity share, the second one is the number of countries in which the company has FDIs. Several authors agree that these counts provide reasonable indicators for FDI activities (Ramaswamy, 1995; Delios & Beamish, 1999; Lu & Beamish, 2001). Because the number of FDIs in which the parent firm has a 10 percent or greater equity share is easy to obtain and compare, this variable will be used to measure the magnitude of FDI activities.
3.4.3. Control Variables
29
control variables have been added, namely, firm size, industry and financial leverage (Gomes & Ramaswamy, 1999; Majocchi & Zucchella, 2001). The indicators for these control variables are respectively, number of employees, manufacturing/non-manufacturing firm and debt ratio. These control variables are important because internationalization is not the only factor influencing profitability or growth.
Firm size can have different influences on firm performance. On the one hand, it is possible that small firms experience difficulties in the internationalization process, for instance because they lack the necessary broad resource base (Wolff & Pett, 2000). On the other hand, small firms may take advantage of a focused use of a narrow but critical set of skills. With this limited set of skills, a relatively narrow but international market segment can be pursued more quickly. Larger firms often experience increasing bureaucracy when they grow, and bureaucratic organizations are known to have a larger failing chance. However, larger firms have a greater resource base than small firms, which may have a positive effect on firm performance (Wolff & Pett, 2000). Obviously, the influence of the size of a firm can be interpreted differently. Therefore it is important to include the variable firm size, which controls for economies or diseconomies of scale.
Characteristics of industries can influence firm performance differently (Mitchell et al., 1993). For example, in some industries scale effects are not predominant, while they can be very important in others (Majocchi & Zucchella, 2001). Specific features of an industry will have an impact on the performance of firms operating in that industry in their own way. In this research, distinctions are made between manufacturing and non-manufacturing firms. This distinction is made because of differences in characteristics between these industries that are especially important for SMEs. Examples of these characteristics are the importance of access to specific firm resources or the availability to share expertise or knowledge (Fillis, 2001). The distinction between industries is made in the databases, by using the industry categorization NACE Rev. 2. With this categorization, companies are classified by their economic activities and can be divided into different main categories, for example manufacturing/non-manufacturing. For this variable, a dummy is included which states 0 for non-manufacturing companies and 1 for manufacturing companies.
30
4. RESULTS
4.1. Descriptive Statistics
Prior to the regression analysis, several aforementioned assumptions should be met. In order to understand and describe the most important features of the databases in quantitative terms, the descriptive statistics are presented and analyzed. In table 2 the variables of the database used for the first and second hypotheses and their abbreviation, number of observations, mean, standard deviation, minimum and maximum value are presented.
Table 2.
Descriptive Statistics Dataset I
Variable Abbr. N Mean SD Min. Max.
Level of Export EXP 1441 0.325 0.320 0.010 1.000 Return on Assets ROA 1441 0.113 0.131 -0.068 0.764 Sales Growth Growth 1441 -0.115 0.343 -0.799 5.465 Firm Size (Ln) Lnsize 1441 2.584 1.246 0.000 5.500 Industry (dummy) dInd 1441 0.320 0.465 0.000 1.000
Leverage Lev 1441 0.102 0.149 0.000 0.821
In table 3 the same statistics are presented for the data used for the hypotheses concerning FDI.
Table 3.
Descriptive Statistics Dataset II
Variable Abbr. N Mean SD Min. Max.
Level of FDI FDI 253 1.950 1.307 1.000 6.000
Level of FDI (squared) FDI² 253 5.498 7.514 1.000 36.000 Return on Assets ROA 253 0.013 0.102 -0.559 0.406 Sales Growth Growth 253 -0.147 0.273 -0.777 0.707 Firm Size (Ln) Lnsize 253 3.630 1.150 0.000 5.000 Industry (dummy) dInd 253 0.340 0.473 0.000 1.000
Leverage Lev 253 0.194 0.191 0.000 0.910
31
manufacturing industry and 168 in non-manufacturing industries. The databases were subject to an outlier and error test. Outliers bias the mean of a dataset and can inflate its standard deviation (Field, 2009). The most important consequence of outliers is the possible reduction of the power of statistical tests and increasing error variance (Zimmerman, 1994). There are two forms of outliers, data errors and legitimate outliers. Data errors should always be removed from the dataset, but this is not always the optimal option for legitimate outliers. Removing legitimate data points can lead to loss of valuable data and a not representative dataset. An outlier is generally considered as a data point that is far outside the norm for a variable or population (Rasmussen, 1988; Jarrell, 1994). If there are outliers detected with severe impact on the dataset there are several options of dealing with them. The data can be deleted from the sample, which only should be done when there is a plausible reason to believe that the case is not from the population that one intended to sample. Or the data can be transformed or replaced with another score when transformation fails (Field, 2009).
The data errors from both sets have been removed. Further, the variables have been standardized. The standardized residuals are distributed around a mean of zero, with a standard deviation of 1. All variables, except firm size, industry and FDI, are percentages. However, ROA and sales growth can have a negative value and a value above 100%. The negative values for ROA and sales growth are acceptable since it is possible for firms to have a negative growth ratio or return. Firm size, industry and leverage can have minima of zero. Firm size is a natural logarithm, thus, the minimum value of 1 employee is transformed into a value of zero. Industry is a dummy variable, meaning it has solely values of 0 and 1 and leverage is measured by long term debt ratio, firms without long term debt will give a long term debt ratio of zero.
To test whether the datasets are normally distributed, the Klomogorov-Smirnov and Shapiro-Wiki test are performed. The results of the normality tests are show in the appendix, in tables A and B. The tables include the test statistics, the degrees of freedom (should be equal to sample size) and the significance value of the test. A significant value (Sig. is less than 0.05) in the Klomogorov-Smirnov and Shapiro-Wiki test indicates a deviation from normality. From the tables it can be concluded that both tests are significant, indicating that both distributions of both datasets are not normal. However, in large sample these tests can be significant even when the scores are different from a normal distribution (Field, 2009). Additionally, Brooks (2008) argues that in studies with large sample sizes, violation of the normality assumption is virtually inconsequential because of the central limit theorem, which states that the sample mean converges to a normal distribution. Therefore, with the large sample sizes in this study, there is enough reason to assume that a possible deviation from normality is not biasing the results of the regression analysis.
32
Levene‟s test is significant (Sig. is less than 0.05), it can be concluded that the variances are significantly different and the assumption of homogeneity of variance is violated. If the results of the test are non-significant, the variances of the variables are roughly equal (Field, 2009). The results of Levene‟s test for both datasets are presented in tables C and D in the appendix. There are three significant variables in the first and one in the second dataset. However, as mentioned for the normality tests as well, Levene‟s test can be significant even when group variances are not very different when sample sizes are large (Field, 2009). For this reason, it can again be believed that these outcomes are not a bias for the results of the regression.
The Durbin-Watson test controls for autocorrelation. For any two observations the residual terms should be uncorrelated. A value of 2 as an outcome of the test means that the residuals are totally uncorrelated, a greater value indicates a negative correlation and a value below 2 a positive one. The results of the Durbin-Watson tests are shown in table E and F in the appendix. The values are very close to 2, except for the third hypothesis. The value for the third model is below 2, which indicates a positive relationship. This can be explained by the combination of the variables firm size and level of FDI in the model. Because of similarities between the variables, they are not totally independent. If the level of FDI rises, the number of employees is likely to increase as well. However, this value does not indicate a problem for the research because independence between the two is not necessary; they can have a combined influence on growth.
33
Table 4.
Test for Multicollinearity Dataset I Correlations
EXP ROA Growth Lev Lnsize dInd
EXP Pearson Cor. 1 ,099** ,060* ,024 -,044 ,049
Sig. ,000 ,023 ,359 ,093 ,065
ROA Pearson Cor. ,099** 1 ,201** -,227** -,109** -,161**
Sig. ,000 ,000 ,000 ,000 ,000
Growth Pearson Cor. ,060* ,201** 1 ,043 ,009 -,012
Sig. ,023 ,000 ,104 ,730 ,636
Lev Pearson Cor. ,024 -,227** ,043 1 ,147** ,134**
Sig. ,359 ,000 ,104 ,000 ,000
Lnsize Pearson Cor. -,044 -,109** ,009 ,147** 1 ,257**
Sig. ,093 ,000 ,730 ,000 ,000
dInd Pearson Cor. ,049 -,161** -,012 ,134** ,257** 1
Sig. ,065 ,000 ,636 ,000 ,000
**significant at 5%, *significant at 10%.
34
Table 5.
Test for Multicollinearity Dataset II Correlations
FDI ROA Growth Lev Lnsize dInd FDI²
FDI Pearson Cor. 1 -,040 ,134* ,001 ,223** ,246** ,973**
Sig. ,526 ,033 ,993 ,000 ,000 ,000
ROA Pearson Cor. -,040 1 ,218** -,159* -,024 ,001 -,016
Sig. ,526 ,000 ,011 ,702 ,988 ,805
Growth Pearson Cor. ,134* ,218** 1 -,009 ,195** ,031 ,158*
Sig. ,033 ,000 ,883 ,002 ,620 ,012
Lev Pearson Cor. ,001 -,159* -,009 1 ,050 ,066 ,003
Sig. ,993 ,011 ,883 ,430 ,299 ,964
Lnsize Pearson Cor. ,223** -,024 ,195** ,050 1 ,459** ,170**
Sig. ,000 ,702 ,002 ,430 ,000 ,007
dInd Pearson Cor. ,246** ,001 ,031 ,066 ,459** 1 ,161*
Sig. ,000 ,988 ,620 ,299 ,000 ,010
FDI² Pearson Cor. ,973** -,016 ,158* ,003 ,170** ,161* 1
Sig. ,000 ,805 ,012 ,964 ,007 ,010
**significant at 5%, *significant at 10%.
In the second dataset, all correlation coefficients are between -0.7 and 0.7 as well, except for FDI and FDI², they have a correlation coefficient of 0.973. Multicollinearity refers to a situation in which two or more explanatory variables in a multiple regression model are highly linearly related (Field, 2009). However, this correlation is logical, because the variable FDI² is created out of the variable FDI. It is not possible for them to not be related.
The next part shows the results of the regression analysis.
4.2. Multiple Regression Analysis
