Industrial clusters, entrepreneurial culture and the social environment: The effects on FDI distribution

International Business Review 18 (2009) 76–88

Contents lists available at ScienceDirect

International Business Review journal homepage: www.elsevier.com/locate/ibusrev

Industrial clusters, entrepreneurial culture and the social environment: The effects on FDI distribution Antonio Majocchi a,*, Manuela Presutti b,1 a b

Dipartimento Ricerche Aziendali, Facolta` di Economia, Universita` di Pavia, Via S. Felice, 7, I-27100 Pavia, Italy Dipartimento di Scienze Aziendali, Facolta` di Economia, Universita` di Bologna, Via Capo di Lucca, 32, 40123 Bologna, Italy

A R T I C L E I N F O

A B S T R A C T

Article history: Received 31 March 2007 Received in revised form 18 April 2008 Received in revised form 16 September 2008 Received in revised form 15 December 2008 Accepted 16 December 2008

Using balance sheet data from a sample of 3498 foreign firms in the manufacturing industry we analyse the distribution of foreign direct investments (FDI) in Italy at a very detailed geographical level, i.e. the provincial level, a region which comprises an urban area and the limited geographical area surrounding it. In this paper, we test the impact that agglomeration economies, entrepreneurial culture and social capital have on the distribution of foreign investments. While the findings regarding the social variables are mixed, the important role played by agglomeration economies is confirmed. Our analysis shows that investments by multinationals are attracted by those areas that combine industrial cluster characteristics with an agglomeration of foreign firms and that have a high level of entrepreneurial culture. The role that this last variable plays is fundamental and suggests the idea that multinational corporations (MNCs) invest in regions with entrepreneurial resources. ß 2008 Elsevier Ltd. All rights reserved.

Keywords: Agglomeration economies Entrepreneurship FDI Industrial districts

1. Introduction In recent years international business literature has frequently underlined the role that location has on firms’ competitiveness. Different authors, with different theoretical lens, have shown that multinational corporations (MNCs) are selecting the location of their foreign investments to tap knowledge linked to a specific local context (Almeida, 1996; Frost, Birkinshaw, & Ensign, 2002), to benefit from the opportunities of fast-growing markets (Brouthers et al., 1996) or to access valuable resources (Almeida & Kogut, 1997; Dunning, 1996; Frost, 2001). In this paper, we aim to make a contribution to the debate on the drivers of FDI geographical distribution shedding new light on previous findings. In particular, we studied the effects that industrial clusters and the agglomeration of foreign firms together with the social environment and entrepreneurial culture may exercise on the investment decision of multinational firms. First of all, we consider the effects of these variables at an individual level and then we study the effects of interaction among these variables. This approach allows us to validate the thesis that agglomeration economies play an important role in attracting foreign firms and we improve on this general conclusion by showing that not all industrial clusters attract multinational firms at the same level. Moreover, we illustrate the crucial role that entrepreneurial culture and, to a less extent, the social environment play in affecting foreign firm investment decisions. In fact Italy represents a significant setting to test for the effects of location variables on multinational firm attraction. Not only is the Italian economy characterized by a network of highly differentiated and competitive industrial clusters and a high level of territorial specialisation (Porter, 1998) but also the

* Corresponding author. Tel.: +39 0382 986467; fax: +39 0382 986228. E-mail addresses: [email protected] (A. Majocchi), [email protected] (M. Presutti). 1 Tel.: +39 051 2098062; fax: +39 051 2098074. 0969-5931/$ – see front matter ß 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.ibusrev.2008.12.001

A. Majocchi, M. Presutti / International Business Review 18 (2009) 76–88

77

country itself presents a great variety of institutional arrangements and marked differences among local entrepreneurial culture and social environment (Fernhaber, Gilbert, & McDougall, 2008; Narula & Zanfei, 2006). Moreover, unlike most studies, in this paper we focus on provincial characteristics as determinants of MNE location decisions in Italy. A province, according to Italian administrative law, covers a small area around a pivotal town and therefore allows us to examine the factors attracting foreign investments almost at town level. At the time of our analysis the number of provinces in Italy was 103. The average area of a province is 1120 squared miles (i.e. 1.6 times the size of the town of London) with an average number of urban areas of 78 and an average population of 562,000. While State or regions characteristics were commonly used as the determinants of foreign direct investment in the existing literature, the impact of more detailed factors at such a fine-grained level was examined only sporadically. In this way, we address both the limitations that Nachum (2000) attributed to previous research on firm location choice and agglomeration economies. The first limitation referred to the use of countries or large regions as a unit of analysis, a dimension that is clearly too large and ineffective for the purpose of measuring agglomeration economies. Using the small areas around pivotal town we adopt the correct spatial unit of analysis for the study of agglomeration economies. Secondly, we take the differences between the different industry sectors into account, controlling for possible industry effects. The paper is structured in the following way: the next two sections develop the theoretical framework and define the main research hypotheses; then we give a brief description of the data and of the methodology used and supply details of the empirical results, discussing our findings. The final section discusses the research and management implications of our main findings, highlighting the limitations of our analysis and possible future research development. 2. Theoretical background One of the main fields of research within international business literature is devoted to study the potential factors able to influence the geographical distribution of investments of multinational firms (Hood & Young, 1999; Malmberg & Maskell, 1999). This strong interest, manifest in the literature, can be explained by the dominant role of multinational firms in an increasingly globalised world, due to their capacity both to promote growth in their host country (Rugman, 2000; Van Den Bulcke & Verbeke, 2001) but also to the significance that the geographical distribution of the firm assets plays in shaping the firm competitiveness. From a theoretical and empirical point of view previous researchers have identified either firm-level characteristics or host country characteristics as the key determinants of the distribution of investment decisions. Firmspecific characteristics have only been analysed in a limited number of empirical works and range from firm size or technological capacity to previous international experience (Chung & Alca´cer, 2000; Pak & Park, 2005; Shaver & Flyer, 2000). The debate on host country characteristics has been much wider in scope and depth. In these studies, the concept of locational advantages, considered as significant place-specific factors able to influence the location decisions of some potential investors, has been well documented (Rugman, Verbeke, & Cruz, 1995). Really, while traditional studies have concentrated on factor endowments as the main locational advantages, a more recent approach tends to focus increasingly on ‘‘created assets’’ (Narula & Dunning, 2000), including knowledge-based assets, infrastructure and technology (Dunning, 1997; Rugman, 2000). This recent approach is also indicative of the influence of institutional development on the localization of inward foreign investment, since multinationals appear to react positively to any government policy that reinforces their own competences, knowledge and intangible resources (Bevan, Estrin, & Meyer, 2004). Finally, the new economic geography literature (Fujita et al., 1999) focuses on the influence of industry agglomeration and spatial clustering on the location decisions of multinationals, following the evidence that a significant concentration of related firms in a restricted place may strongly reinforce co-location by other firms (Maskell & Malmberg, 1999; Storper, 1997). Based on Dunning’s work (1997), all these factors can be classified into two broad main groups: economic and institutional factors. These two features are not to be considered as mutually exclusive but tend to reinforce each other. From a strictly economic point of view, the differences both in price of factors of production and in market size (Haddad & Harrison, 1993) have received consistent empirical support as drivers of MNC investment location. This means that the investments decisions of multinationals are closely tied to the comparative advantage of a country, which in turn affects the expected profitability of foreign investments (Casson, 1990). Taking this approach, vertical investments develop when foreign firms move their production process to countries in order to take advantage of less expensive factors of production, such as labour, mainly by delocalising the lowskilled production stages towards low-wage countries. Sun, Tong, and Yu (2002) in the Chinese context and Campos and Kinoshita (2003) in the countries of Central and Eastern Europe both empirically validate the conclusion that low labour costs attract foreign direct investors. By contrast, the FDI inflows to developed countries are mainly driven by marketseeking and strategic asset-seeking motives and they represent location of activities towards the foreign countries, which are potentially a large and significant market or a source of strategic resources (Narula & Dunning, 2000). Typically, marketseeking motives are the driving factor in the initial stages of the development of a new subsidiary, later followed by more complex goals. As Birkinshaw, Hood, and Young (2005, p. 228) clearly state: ‘‘The received wisdom today is that subsidiaries start out with market-seeking responsibilities . .and as subsidiaries develop resources and capabilities of their own they take on additional responsibilities’’. The responsibilities of these new subsidiaries range from the identification of new markets opportunities to the development of new ideas and products and of new resources to share within the firm. In order to take on this more developed role the subsidiaries need to look for valuable human resources and this explains the importance that the quality of the local labour force and its productivity plays in attracting multinational firms (Cantwell & Piscitello, 2002). Recent literature (Birkinshaw, 1997) has clearly demonstrated that entrepreneurial abilities are among the competences

78

A. Majocchi, M. Presutti / International Business Review 18 (2009) 76–88

requested by theses new subsidiaries. These skills are not only the result of resources developed internally but also of resources created through the subsidiary’s interaction with the external environment, a process which helps to shape its characteristics. Complementary to these economic factors other aspects have been identified in the institutional-culture literature. This approach emphasizes how the host institutional and cultural-environmental aspects are also likely to affect MNC operations. The importance of non-economic factors in MNC decision-making has been frequently highlighted in recent times in international business literature (Fujit, Krugman, & Venables, 1999; Grosse & Trevino, 2005; Rugman, 2000). This approach develops the fundamental contribution of North (1990) who underlined the role of institutions in promoting an efficient legal, political and administrative environment, reducing uncertainty and information costs, promoting the creation of a business environment and consequently attracting foreign firms (Mudambi & Navarra, 2002). Empirical research demonstrates the positive role of both country-level political and legal institutions in influencing MNC location decisions and also suggests how this creates new challenges for public policy. In particular, social and political stability may encourage MNCs to move in a particular context but also other non-economic factors have been considered such as the degree of corruption, law enforcement and administrative efficiency. From this perspective, legal and governmental arrangements, as well as informal institutions underpinning an economy, are able to influence corporate strategies (Aitken & Harrison, 1999), impacting both on the activity and on the performance of a foreign business (Storper, 1997). The rationale offered by the institutional culture perspective is that multinational firms prefer to invest in places where institutions minimize uncertainty increasing their chances of success (Flores & Aguilera, 2007). Strong support for the positive role of the institutional environment in the development of global firm-specific advantages is presented in recent literature (Bevan et al., 2004; Grosse & Trevino, 2005). Finally, at the crossroads of the two approaches economists and organisational theorists have emphasized the role of agglomeration economies. The new economic geography literature (Krugman, 1991) has suggested that investment location decisions by multinationals may be explained by agglomeration economies (Cantwell & Iammarino, 2000). Agglomeration economies emerge when many different economic units, with common characteristics, collect near each other due to the presence of factors like knowledge spillovers but also specialized labour markets, supplier networks and so on (Fujita et al., 1999; Maskell & Malmberg, 1999; Storper, 1997). The common characteristics of firms co-locating are of two different kinds. In some cases these firms are businesses operating in related industries and in the present paper we refer to this phenomenon as the ‘‘industrial cluster effect’’. In other cases the shared feature of the firms is their foreign origin and in this second case we simply refer to the phenomenon as ‘‘agglomeration economies of foreign firms’’. The main idea is that a cluster is a centre of accumulated competencies across a range of related industries and across various stages of production which can be very attractive to outside firms. These competencies are positive externalities that have positive effects on the productivity of those firms located in the area. These positive externalities can take different forms (McCann & Folta, 2008). Spatial concentration of similar firms favour industry-specific investments by workers and attract specialised suppliers such as research and training centres, distributors and consulting agencies. Moreover, geographical proximity between interrelated partners is generally assumed to reinforce knowledge acquisition and exploitation processes, since knowledge is partially tacit and localized (McEvily & Marcus, 2005) but also it is assumed to facilitate the consumers of an industry reducing their search costs and increasing the overall demand for the firms located in clusters. This process explains also the dynamic of clusters. Organisational theorists, for example, have argued that a local industrial structure with many firms competing in the same industry or collaborating across related industries tends to trigger processes which create dynamism and flexibility, but also new learning and innovation (Driffield & Munday, 2000). By this point of view, localized clusters of firms tend to have a self-reinforcing effect with an agglomeration of foreign firms and industrial activities that develops an environment beneficial to further agglomeration of firms (Gorg & Strobl, 2001) and of foreign investments (Mayer & Mucchielli, 1998). However, not all the clusters are the same. Beside geographical proximity, there are other aspects that characterise an industrial cluster. McCann, Arita, and Gordon (2002) identify three kinds of clusters according to the nature and the intensity of the relations between firms. The simplest form of cluster is characterised only by the geographical proximity of firms operating in related industries. In this kind of cluster there are no long-term relations between firms. A more complex kind of cluster emerges when these kind of long-term and stable relations are developed. Finally, when a common culture of mutual trust is developed within the cluster, the relations between firms are not only stable but also characterised by a low risk of opportunistic behaviours (Hendry & Brown, 2006). When such a local business environment emerges then transaction costs are low and the diffusion of reliable information is facilitated (Cantwell & Iammarino, 2000). Moreover, a local culture with specific norms, values and institutions makes it possible to transfer tacit forms of knowledge (Enright, 1998; Porter, 1998). These findings demonstrate that, when the effects of agglomeration are evaluated, also social variables should be included in order to distinguish different kinds of clusters. The benefits generated by industrial districts explain why firms, both local and foreign, are attracted by geographical agglomeration of firms so promoting further agglomeration. However, this process is stronger for foreign firms that tend to follow similar path when investing in a foreign country. Chang and Park (2005) show how the power of this mimetic behaviour and the search for legitimacy and for a lower level of risk explain the choice of firms investing in regional clusters. The concentration of foreign firms in a specific region indicates the existence of a favourable business environment and this channel promotes further aggregation of distant firms. Multinational corporations looking for the best location within a foreign country save on research costs and rely on the information given by other firms that have located their activities, over time, in the same place (Birkinshaw & Hood, 2000).

A. Majocchi, M. Presutti / International Business Review 18 (2009) 76–88

79

3. The conceptual framework and hypothesis development In this paper, we build on the analysis of economic and institutional variables in order to refine our understanding of the role of industrial clusters and of the agglomeration economies of foreign firms. As a first step, we concentrate on the effects that industrial concentration may exercise on the distribution of investments by multinationals. The main idea is that a cluster is a centre of accumulated competencies across a range of related industries which can be attractive to outside firms. There is a relatively large amount of literature that seeks to relate location specific factors to the determinants of inward investment across countries or industries, based on the importance of agglomeration effects (Head, Ries, & Swenson, 1995). In particular, studies about the spatial agglomeration phenomenon have sustained that the concentration of a significant number of firms inside a limited territorial context tends to have a self-reinforcing effect on foreign investment, since positive externalities tied to the economic agglomeration seem to be crucial for firm productivity (Ellison & Glaeser, 1997). For instance, Head et al. (1995) find industry-level agglomeration benefits play an important role in the location choice of Japanese manufacturing plants in the US. More recently, Cheng and Kwan (2000) in their study of the determinants of FDI in Chinese regions, also report the positive feedback effect of significant agglomerations of firms in a restricted area on the area’s capacity to attract foreign firms (Coughlin & Segev, 2000). Similar findings are illustrated in recent research carried out in the UK (De Propris, 2004; Devereux, Griffith, & Simpson, 2004; Driffield & Munday, 2000). Based on these suggestions, we formulate our first hypothesis of research: Hypothesis 1. Provinces with a higher level concentration of firms in similar industrial sectors attract a higher level of foreign investments. However, the definition of cluster cannot be limited only to the concentration of firms in related industries (Porter, 1998). Many studies (Bevan, Estrin, & Meyer, 2004; Driffield & Munday, 2000) have shown that the significant agglomeration of foreign firms inside a local context is relevant when investors have insufficient information about the host country (Coughlin & Segev, 2000; Tegarden, Hatfield, & Echols, 1999). The traditional advantages associated with the presence of industrial clusters for multinational firms can be reinforced by the presence of a significant concentration of foreign investments in a province. In fact, this element may be conceptualized as a signal of favourable investment environments, according to several empirical works (Ellison & Glaeser, 1997; Fujita et al., 1999; Meyer, 2001). The presence of a significant concentration of foreign firms in a province affects the perceptions of MNCs managers in a positive way with regards to uncertainty and to the degree of additional information necessary to know the new context (Narula & Zanfei, 2006). Thus, we may affirm the following hypothesis of research: Hypothesis 2. Provinces with a higher level concentration of foreign firms go on to attract a still higher level of foreign investment. The rationale behind the attractiveness of areas with industrial clusters and agglomeration economies lies in the fact that firm concentration lowers the cost of operating in a foreign context. However, since firms do not only look for lower costs but also for valuable resources we test to see if the presence of these resources can help better explain the factors influencing foreign firm decisions. The first variable we considered is the level of entrepreneurial culture of the province. This variable can be conceptualized as the new firm’s creation rate which is a significant indicator of a positive and active entrepreneurial spirit inside a province, which in our opinion should be studied independently from the existence of significant agglomeration economies or industrial clusters. Recent studies (Birkinshaw, 1997; Birkinshaw et al., 2005) have clearly shown how entrepreneurial culture is a crucial resource of the most innovative and successful MNCs that nurture entrepreneurial behaviour in their subsidiaries. Entrepreneurial capabilities are the result of both internal efforts and local environment conditions. Among these conditions the entrepreneurial spirit of a local context is a crucial variable in fostering subsidiary initiatives (Boojihawon, Dimitratos, & Young, 2007). A place where the rate of new firm creation is very high is strongly correlated with the presence of dynamic processes of knowledge creation, knowledge learning and innovation and is therefore very attractive to MNC firms. In fact, the localised nature of learning processes which sustain the rapid creation of new firms may change the geographical patterns of MNC investment decisions. This argument strengthens the link between knowledge creation and the geographical, social and institutional frameworks supporting the creation of new firms at the local level. Following Shane and Venkataraman (2000) we define entrepreneurship as the process through which new opportunities to create future goods and services are discovered, evaluated and exploited. The presence of significant entrepreneurial capabilities in a specific place could improve the development and the realization of new profitable business ideas, thus becoming an important factor in attraction of foreign firms. According to this idea, we may affirm our third hypothesis: Hypothesis 3. Provinces where entrepreneurship is higher in turn receive a higher level of foreign investment. Then, we consider the importance of social environment variables in attracting foreign firms. We sustain that while the literature on institutional-cultural effects is well developed, as illustrated above, it is still necessary to include a social perspective in the analysis, separately, from the legal, political and cultural dimensions. In particular, we focus on the quality of the social environment as MNCs decision makers are well aware of its pivotal influence, taking it into account when implementing various international strategies (Meyer, 2001; Nachum, 2000). Similar conclusions have been drawn in recent polls carried out annually by the consulting firm Ernst and Young (2007) on a representative panel of international decision

80

A. Majocchi, M. Presutti / International Business Review 18 (2009) 76–88

makers, regarding the perceived attractiveness of Europe as a location for foreign investments. The results of the poll show that managers strongly consider aspects such as quality of life, transparency of institutions and the legal environment when making their investment decisions. Since an objective measurement of these variables is difficult to define we use a negative indicator as the main index of the quality of a social environment, that is, the crime rate concerning both property and personal safety. These variables are strongly negatively correlated with the quality of local governance, seen as the capability of local institutions to favour local welfare and wealth and to improve local social capital. Moreover, crime rates are a much more objective measurement than the other variables often used such as the rate of corruption which is more business oriented but proves a much more difficult variable to measure since it depends heavily on manager perceptions. We keep the variables concerning the crime rate against property and the one against personal safety separated. This has been done because we expect that foreign firms would react differently to these phenomena. We assume that foreign firms in their investment decision will be more directly influenced by the level of crime against patrimony since this kind of crime is directly affecting the value of the investments. Moreover, the low level of correlation between the two variables (.49) seems to suggest that the two variables gauge different aspects of the same process. We suppose that independently from market, resource and technology factors, the quality of the local – provincial – social environment is highly significant in attracting multinationals and this leads us to posit our fourth hypothesis: Hypothesis 4. Provinces where the level of social environment quality is higher receive a higher level of foreign investments. Finally, we consider some interaction effects in order to catch possible different characteristics of industrial agglomerations. Gordon and McCann (2000) have underlined how industrial cluster analysis is mainly based on a very stylised and simplified notion of clusters that reduces the concept to a pure concentration of firms in the same location. Actually, the clusters can be very different in terms of the nature of the firms and in terms of the relationship between these firms. Therefore, using interaction variables, we try to gauge some distinctive features of industrial clusters. As Porter (1998) stated, clusters promote firms competitiveness by increasing firms productivity thanks to the presence of a pool of specialised labour, by promoting innovation thanks to the high level of competition, but also by stimulating the formation of new businesses which strengthens the cluster itself. Therefore, together with geographical proximity and industry concentration a peculiar characteristic of any industrial cluster should be a high rate of entrepreneurship represented by a high rate of new ventures creation. Therefore, we consider in our model an additional variable that is the result of the interaction between the industrial cluster and entrepreneurial capability. The second interaction variable included in the model gauges the interaction between industrial cluster and foreign agglomeration. Since foreign firms tend to have greater problems than local firms in collecting correct and reliable information about the main characteristics of different locations there is a natural tendency of foreign firms to locate in a particular area only after other foreign firms have already proved that that region is worth the investment. As Nachum (2000) shows, the agglomeration of foreign firms raises the probability that subsequent MNCs will invest that particular location. We finally posit our last set of hypotheses. Hypothesis 5a. Provinces with industrial clusters characterised by a higher level of entrepreneurial capability attract a higher level of foreign investment. Hypothesis 5b. Provinces with industrial clusters and a higher concentration of foreign firms attract a higher level of foreign investment.

4. Methods and data 4.1. Sample At the moment in Italy there is no public register of foreign firms available. Most of the previous studies (Basile, 2003; Mariotti & Piscitello, 1995) rely on data collected for research purposes or on small samples which are not fully representative of the large and differentiated network of foreign enterprises in the country. In order to bridge this gap the present research relies on data supplied by the database Aida published by the Bureau Van Dijck that gathers the balance sheets of Italian registered firms. The database Aida is the Italian sub-sample of the European database Amadeus and contains firms’ financial and commercial data for enterprises characterized by a turnover of at least one million euros, operating in Italy. This database is a reliable and detailed source of information and has been widely used in previous international business research (Brouthers, 2002). We extracted the total assets book value from the balance sheets of all the firms operating in the manufacturing sector in 2004 and with a majority of share capital in foreign hands, achieving a total of 3984 observations. In order to avoid any problems with the definition of majority control we selected only those firms where the percentage of foreign ownership was equal to or greater than 50%. This very tight definition of majority control allowed us to avoid any problems regarding the definition of ‘‘foreign firms’’. On the other hand this choice does not allow us to have a precise measure of the total value of foreign investment in the country, as none of the minority stakes will be considered in our analysis. As most of the previous studies on location dealing with large databases, the data available does not allow us to distinguish between greenfield investments and merger and acquisition so we could not separate those foreign firms which buy an existing asset from those that build up a new firm from start. However, since we use stock value, our dependent

A. Majocchi, M. Presutti / International Business Review 18 (2009) 76–88

81

variable is the result of an accumulated process of investment and divestment decisions and in this sense the target variable used in our analysis represents the results of accumulated decisions taken by firms over time and gives a picture of the distribution of foreign investments in the country. Besides, the variable gives a picture of the Italian manufacturing industry held by foreign investors that is coherent with the structures of the Italian industrial sector. Our sample of firms is largely made up of small enterprises with 2122 of the firms having less than 50 employees (61% of total firms), 962 (27.5%) firms having more than 50 but less than 250 employees and 414 large firms (11.5%) with more than 250 employees. Therefore, our sample mostly consists of SMEs (88.5%) a percentage not significantly different from the corresponding rate for the entire Italian manufacturing sector where SMEs are 97% of firm population. This feature of our sample can be seen as a particular strength of our analysis since most of the work so far has mainly considered large firms. Moreover, in industrial clusters a large share of firms is typically SMEs. So, from this point of view, our sample seems particularly well suited as a basis for the analysis of the effects of agglomeration economies. In order to cluster firms in few industrial sectors we categorised the firms according to the taxonomy set out by Peneder (1999). Using the NACE code we classified every firm in one of the 5 different categories used by Peneder (1999) and now widely used in other studies on European firms (i.e. Landesmann, 2003). The five sectors are: the technology-driven sector, the marketing-driven sector, the capital-intensive sector, the labour-intensive sector, and a final residual category of mainstream activities. In order to have a proxy of the investments by foreign firms in every Italian province for each of the five industrial sectors, we then summed up the total book value of the assets extracted from the balance sheet for the year 2004 for all the foreign firms located in every province. The main limitation of this choice is that this value does not represent the new investments realised in a year but the accumulated value, less depreciation and amortisation, through years. Of course more recent investments will weight more giving significance to the variable we use. Nonetheless, this cautionary note should be kept in mind when interpreting our results. An alternative measure could have been to measure the level of foreign investment by the number of foreign firms in every province (Basile, 2003). In this case, however we would not have weighted the investments by the capital invested and, given the great variety of firms in terms of dimension, this approach would have led us to consider investments, which were very different in terms of size, in the same way. Our choice allows us to proxy the level of total foreign investments that foreign firms have realised in the five industrial sectors in the 103 provinces and this is the strength of the analysis. 4.2. Methodology Following previous works (Sun et al., 2002) we use the transformed value of the investments realised in every Italian province, for the five sectors previously defined, as a dependent variable. Because the number of Italian provinces is 103 and the sectors are 5 the total number of observations is 515. Every observation measures the level of investments realised in the province j (with j going from 1 to 103) in the sector k (with k going from 1 to 5) in the year 2004. The level of the investment has been measured summing up all the values of the total assets held by single foreign firms located in the province J and active in the sector K. Since some provinces in some sectors did not receive any foreign investments the dependent variable sometimes assumed the value of 0. Therefore, we could not use the log transformation and we applied a simple squared root transformation. The resulting estimation model is the following: Y jk ¼ b jk X j þ e jk

ðJ ¼ 1; . . . ; j ¼ 103Þ

ðk ¼ 1; . . . ; k ¼ 5Þ

In our model the dependent variable is the root square of a stock value, i.e. the total assets invested by every single foreign firms in the province for the 5 industrial sectors, a value that has been used in other recent empirical works (Brenton, Mauro, & Lu¨cke, 1999). However, this choice generates the problem of the possible existence of simultaneity between the dependent variable and the location variables of our model. In order to alleviate this problem, we lagged the explanatory variables (Spanos, Zaralis, & Lioukas, 2004), which all refer to the year 2003 with the only exception being the data on population, which refers to the year of the national census, i.e. 2001. This strategy not only improves econometric methodology but also follows an economic rationale since firms make their investment and divestment decisions on the basis of past data. Using the procedures with the Intercooled-STATA 8 program we adopted a cross-sectional methodology with ordinary least square technique testing for standard assumptions underlying OLS. Following Aiken and West (1991) methodology we centered our variables. The descriptive statistics (Table 2) show that correlation between regressors is not high. In order to detect potential multicollinearity we examine also the variance inflation factors (VIFs) finding a maximum score of 2.61 and a mean VIF score of 1.84 well below the commonly used critical value of 10 (Cohen, Cohen, West, & Aiken, 2003). VIF scores are reported in Table 2. We then cross-check with the Breusch–Pagan test for heteroschedasticity. The p-value of the test was low suggesting the introduction of corrections for possible heteroschedasticity of the residuals. Following White (1980) we specify the Huber–White estimator of variance using OLS with robust standard errors in order to tackle issues concerning heterogeneity and lack of normality. Moreover, we perform a robust regression using iteratively re-weighted least squares. Comparing the results of robust estimation with the results of weighted estimation we find similar coefficients, standard errors and t and p values suggesting that the robust regression results are trustworthy. 4.3. Measures Industrial cluster effect and foreign firm agglomeration are the main focus of our analysis and have been measured with two different variables. A first variable has been developed in order to gauge the concentration of foreign firms in a specific

82

A. Majocchi, M. Presutti / International Business Review 18 (2009) 76–88

area. This variable (For_agglom) is the ratio of the total number of foreign firms in a sector over the total number of firms in the same sector for every province. A second variable (Ind_cluster) has been developed in order to measure the concentration of firms in the same sector regardless of their ownership origin. Following similar studies (Shaver & Flyer, 2000) this variable is the ratio of the total number of firms in the industry sector k in the province j over the total (national) number of firms in the industry sector k. This variable will be higher for those areas with a specific industrial specialisation, as is typical in Italian industrial clusters. Therefore, this variable measures the relative importance of firms in one industrial sector regardless of the geographical density of firms in the area. Entrepreneurship has been interpreted here as the dynamics of firms in terms of the net number of new ventures set up in a province over the total number of existing firms in the previous year. This variable measures the yearly percentage increase in the stock of firms in a province. The notion of entrepreneurship is surely a complex one and we are aware that this concept requires a multi-construct measure to be properly measured. Our variable catches only part of this complex concept. However, given data at hand, we could only develop the present measure of this seemingly important variable. The variable has been named Entrepren. Finally, two variables have been inserted in order to measure the level of social environment quality. In this case we measure crime levels in each province using the average number of crimes against people reported in a year (Crime_people) and the average number of crimes against property (Crime_prop). In order to gauge possible interaction effects two additional interaction variables have been inserted. One variable (Entrep_clust) is the result of the interaction between the rate of new ventures (Entrepren) and the industry cluster effects (Ind_cluster). The second interaction variable (Int_clust) gauges the combined effect of foreign firm agglomeration (For_agglom) and of industry cluster effects (Ind_cluster). Since the decision to invest by a foreign firm is the result of a complex process, we include a number of control variables in the analysis. To gauge the attraction exercised by large and densely populated areas, the so-called urbanisation economies, we add a variable made up by the total number of population in the province (Pop). The variable (Income) measures the disposable income per capita in every region and has been considered as a proxy of the purchasing power of the inhabitants. A specific variable (Transport) has been used to gauge the level of transportation infrastructure in the provinces, this variable has been constructed by adding the total kilometres of railway track and road present in every province, respectively, and then multiplying this value by the number of airport transportation facilities. Labour in terms of cost and of productivity is a factor that is typically considered in some depth in location studies. Concerning the cost of labour, Italian statistics do not allow precise distinctions to be made at a provincial level. Therefore, we were obliged to drop this variable. However, we feel that thanks to national collective bargaining this is not a real flaw as wage levels are roughly uniform at the national level within the country. The quality of labour is a very difficult variable to operationalise. The best measure is, without doubt, labour productivity but, unfortunately, again a precise measure of this variable is not available at the provincial level in Italy. Following previous studies, we use the proportion of the work force with secondary and tertiary education as a percentage of the total work force at the provincial level (Lab_quality) as the proxy variable. Similar problems arise relating to the definition of the quality of research and teaching centres in a province. In this case quality matters more than the number of institutions but, given the difficulties in measuring the quality of research, we use a simple measure counting the number of research institutions in a province (Instruction). For the quality of infrastructure an index number produced by a National Institute of Research (Istituto Tagliacarne) has been used. The index is a composite measurement and estimates the quality of environmental and energy structures (Environm). Finally, we include two additional control variables in the model: Fail, and Open. The variables measure respectively the failure rate of firms in a province and the ratio of the summation of export and import for every province over the total added value generated in the area. The source, the measurement, the expected effects and the descriptive statistics and correlation and VIF scores of the explanatory variables are reported in Tables 1 and 2. 5. Findings and discussion In Table 3 we report the main results of our cross-sectional analysis. We presented four different models. A general model with all the variables but the interaction terms, a more parsimonious model and two additional models with the interaction terms. Results produce some significant findings and demonstrate good explanatory power. Hypotheses 1 and 2 define the role of agglomeration economies with regard to both the role of industrial clusters and to the concentration of foreign firms. Both hypotheses are confirmed but the sign and the significance of the variable ind_cluster requires further comments. The ‘‘industrial cluster effect’’ is significant but only if the interaction terms are not inserted in the model. Italy is a country where the role of industrial clusters has always been considered relevant and our research confirms that this opinion is shared by foreign firms since they tend to concentrate their investments where a higher concentration of firms operating in the same industrial sector is found. However, our results suggest that this effect alone does not explain the distribution of investments in Italy. It is the contemporaneous presence of an industrial cluster and of an agglomeration of foreign firms that seems to be the catalyst for further foreign investments. In fact, the variable measuring foreign firm agglomeration is highly significant and fully confirms previous studies that found that foreign firms tend to concentrate their activity where other foreign firms are already located (Majocchi & Strange, 2007). In other words when firms invest in foreign countries they consider more important to exploit the positive externalities related to the agglomeration of foreign firms than to access the knowledge shared inside industrial clusters. Our results confirm that the gap in knowledge concerning the foreign location still remains more difficult to bridge, even in today’s increasingly globalised world, than the business knowledge which is usually offered by the location in an industrial cluster. These results

A. Majocchi, M. Presutti / International Business Review 18 (2009) 76–88

83

Table 1 Hypothesis, constructs, sources and descriptive statistics (variables not centered). Hps

Construct

Predicted effect

Variable name

H1

1. Industrial cluster

+

Ind_cluster

.027

.064

H2

+

For_agglom

.004

.008

H3a

2. Agglomeration foreign firms 3. Entrepreneurial culture

+

Entrepren

2.17

.81

H4

4. Social environment

-

Crime_people

3.023

1.344

H4

5. Social environment

-

Crime_prop

3.771

1.075

H5a

6. Ind cluster* Entrepreneurship 7. Ind cluster and agglomeration foreign firms 8. Control variable

+

Entrep_clust

.0001

+

Int_clust

Lab_quality

H5b

Mean

S.D.

Measurement

DATA source Infocamere (National CoC Association)

.0005

Firm in sector K in the prov. J on national firms in sector K (%) % of foreign firms on total firm New firms created in the year over total stock of firm existing the previous year (100) N. crime vs. people a year per 100 habitants N crime vs. property a year per 100 habitants Interaction 1  3

.0003

.0010

Interaction 1  2

.504

.048

% of labour force with secondary and tertiary education N of research and education institutions National index 1: 200

9. Control variable

Instruction

87.90

52.43

10. Control variable

Environm.

99.77

47.82

11. Control variable 12. Control variable

Pop Income

13 Control variable

Transport

14. Control variable

Open

15. Control variable

Fail

562,021 14,930

613,738 3,029

11,542

10,521

37.58 .175

24.01 .084

Population Disposable income per capita [Paved Roads (km) + railway (km)]  n of airports % of export + import over provincial GDP Yearly failure rate (%)

Istat and Aida (BvD) Tagliacarne Institute (Movimepresa)

National institute of Statistic (Istat) National institute of Statistic (Istat)

National institute of Statistic (Istat) Tagliacarne Institute Tagliacarne Institute (index) National census Tagliacarne Institute National institute of Statistic (Istat) National institute of Statistic (Istat) National institute of Statistic (Istat)

suggest that agglomeration economies are a feature that can attract foreign investments but at the same time they show that industrial clusters alone are not a sufficient condition to attract foreign investments unless some other foreign firm have already invested in the area. This view is reinforced by the results regarding our hypothesis 5b. This hypothesis, concerning the joint effects of industrial clusters and foreign firm agglomeration, is strongly confirmed by our statistical analysis. Firms look for industrial clusters but they show clear preferences for those areas presenting a higher concentration of foreign firms. In this sense our findings seem to suggest the existence of a ‘‘double’’ agglomeration effect that regard related industries and foreign firms. Of course, it could well be that some clusters are more attractive than other not because of the presence of foreign firms but because foreign investments have been attracted by other factors. We could only test some of these other possible factors. One is entrepreneurship. Our third hypothesis regards the role of entrepreneurship (Entrepren). We discuss this hypothesis together with hypothesis 5a which considers the joint effect of entrepreneurship and of the industrial cluster effect. The variable Entrepren has the expected sign and it is strongly significant. On the contrary, the interaction term Entrep_clust is not significant so Hypothesis 5a is not confirmed. The result indicates that foreign investments are attracted by areas where the degree of entrepreneurship is high but, at the same time, the combined effect of industrial cluster with a high entrepreneurship rate seems not to impact upon foreign firms’ behaviour. The rate of new ventures creation does not discriminate between different industrial clusters. However, our results show that this variable has an impact on FDI distribution. These findings demonstrate that, among the intangibles the entrepreneurial resources are important asset but this asset is not a crucial variable when they select an industrial cluster for their investments. The entrepreneurship variable seems an important variable even if it does not discriminate among different kind of industrial clusters. The finding is important by another point of view. Many studies have shown that firms look for geographic sources of innovation around the word (Frost, 2001) but, with few exceptions (Nachum, 2000), most of these studies concentrate on R&D capabilities. Our work suggests that the range of intangible capabilities that firms search is large and that, among these resources, entrepreneurial capabilities are an important factor driving firms’ location choices. Moreover, this result better specifies the general idea that firms are attracted by a well trained work force with higher productivity that enhances the competitive advantage of the firms that collocate in the area (Porter, 1998). We suggest that, among workforce abilities, managerial skills are also important and that entrepreneurship has to be considered as well, since foreign firms considered this variable among others when defining their localisation decisions. Actually, the concept that the role of external environment affects

84

Pop (1) Lab_quality (2) Income (3) Ind_cluster (4) For_agglom (5) Transport (6) Environm. (7) Fail (8) Instruction (9) Open (10) Entrepren (11) Crime_people (12) Crime_prop (13) Entrep_clust (14) Int_clust (15)

VIf score

(1)

2.32 1.61 2.42 1.85 1.72 1.41 2.61 1.61 1.81 1.87 1.35 1.81 1.72 1.26 2.24

1

Average VIF score = 1.84. * p-value at the 10% level. ** p-value at the 5% level. *** p-value at the 1% level.

.197* .035 .433 .020 .053 .211*** .471*** .464*** .070*** .039*** .401*** .468*** .168*** .183***

(2) 1 .319*** .075* .154*** .389*** .234*** .379*** .344*** .192*** .073* .238*** .256*** .034 .024

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

(12)

(13)

(14)

(15)

1 .204*** .162*** .082* .122*** .070 .260*** .226*** .024 .070

1 .332*** .326*** .566*** .219*** .400*** .220*** .018 .034

1 .369*** .271*** .148*** .311*** .318*** .029 .007

1 .034 .095** .358*** .091** .002 .051

1 .306*** .035 .131*** .017 .089*

1 .105** .088** .215*** .036

1 .464*** .062 .507

1 .073* .468***

1 .339***

1

1 .140*** .279*** .196*** .640*** .215*** .121*** .486*** 437*** .333*** .213*** .059 .027

1 .332 .027 .205*** .200*** .197*** .154*** .017 .191*** .216*** .305*** .507***

1 .110* .271*** .097** .150*** .213*** .128** .083* .005 .033 .468***

A. Majocchi, M. Presutti / International Business Review 18 (2009) 76–88

Table 2 Correlation.

A. Majocchi, M. Presutti / International Business Review 18 (2009) 76–88

85

Table 3 Regression results with robust standard errors for the whole sample. Model A For_agglom Ind_cluster Entrepren Crime_people Crime_prop Entrep_clust (Entrep* Ind_cluster) Int_clust (For_agglom*Ind_clust) Control variables Pop Income Transport Lab_quality Instruction Environm Fail Open Cons

***

Model B ***

Model C ***

Model D

78.13 (17.63) 892.15*** (304.05) 30 31** (11.9) .002 (.0122) .067* (.009) – –

78.63 (17.54) 895.44*** (298.36) 32.68*** (11.52) .002 (.0123) .0162* (.009) – –

153.31 (37.26)  14.12 (518.50) 30.88** (12.30) .0003 (.00116) .014 (.009) 473.23 (409.42) 421.36*** (154.86)

153.99*** (37.40) 19.12 (516.32) 31.92*** (11.65) .0008 (.0117) .013 (.009) 471.10 (412.59) 403.42*** (155.46)

.03*** (.00) .306*** (.005) .017 (.10) 138.10 (247.44) .315 (.269) .493* (.292) 168.51 (199.32) 2 89*** (.72) 211.97*** (10.99) F(15,499) = 16.5; p(F) = .0000; R2 = .5153; Adj R2 = .5286; RMSE = 249.52

.03*** (.00) .314*** (.005) – – .307 (.256) .537** (.304) – 2.83*** (.65) 541.9*** (89.22) F(10,504) = 17.23; p(F) = .0000; R2 = .5150; Adj R2 = .5244; RMSE = 249.06

.03*** (.00) .0284*** (.004) .037 (.098) 59.79 (230.66) .117 (.213) 536* (.283) 54.15 (165.82) 2.35*** (.61) 205.19*** (9.90) F(15,502) = 17.86; p(F) = .0000; R2 = .5694; Adj R2 = .5865; RMSE = 235.66

.02*** (.00) .0283*** (.004) – – .101 (.198) 550* (.289) – 2.35*** (.57) 205.19*** (9.87) F(12,502) = 20.49; p(F) = .0000; R2 = .5691; Adj R2 = .5827; RMSE = 235.03

Notes: (1) The sample consists of 515 observations. All the variables have been centered. (2) Standard errors are in brackets. * Coefficient is significant at the 10% level. ** Coefficient is significant at the 5% level. *** Coefficient is significant at the 1% level.

on subsidiary entrepreneurial capability was first introduced by Porter (1990) but his theory mainly concentrates on the effect of local competition. The idea that the external environment affects the degree of entrepreneurship in subsidiaries was subsequently developed by Birkinshaw (1997) and by Birkinshaw et al. (2005). Our research adds on these views suggesting that it is not only the competitive level but also the pervasiveness of entrepreneurship at the local level that is crucial. The results concerning the crime rate variables are mixed. Hypothesis 4 states that an area where the level of social environment quality is higher receives a higher level of foreign investments are only weakly confirmed. The variable that measures crimes against property has the expected sign and is significant at a 10% level of confidence, but only in the first two models, while the coefficients of crime rates concerning personal safety are not statistically significant. The role of political and social uncertainty in deterring investments has been shown in previous studies (Delios & Henisz, 2003). Our results partially confirm this hypothesis that foreign firms tend to avoid those areas where the crime rate is higher and the social environment is most degraded. The weak significance of these results can be probably explained by the fact that safety concerns are relevant but that at the same time these concerns are not decisive in the investing decisions taken at the subnational level. These results surely need further and deeper analysis. Finally, it must be noted that also some control variables are statistically significant. Most of these variables confirm the finding of previous analysis such as the highly significant variable concerning urbanisation, the one measuring income and the variable regarding openness to foreign trade. Surprisingly, the effect of the workforce education levels is not significant while the coefficient of environmental quality has a negative sign. Also these results surely deserve further investigation. 6. Conclusion This study sought to find evidence the importance of different factors which may exercise a significant influence in the distribution of investments by multinationals in Italy when choosing among different provinces. Our empirical findings show how many economic and also strategic place-specific factors are able to influence the decision of a multinational to invest in a province (or not). First of all, we confirm the idea that multinationals investments tend to concentrate where there are firms already operating in related industries (industrial clusters) and where other foreign firms have already developed their activities (foreign agglomeration economies). We refine these findings highlighting the role of entrepreneurship and of social environment quality. Our study makes several potential contributions, since we shed light on a phenomenon that is central to our understanding of the foreign direct investment process, and thus critical to scholarship in the domains of international business and strategic management. We suggest that these results could also provide useful insights regarding industrial cluster theories. At a general level, we verify that an analysis of the location strategies of foreign firms, which fails to simultaneously take into account institutional, economic, strategic and place-specific factors, is too limited. Our results are in line with many recent studies, which demonstrate that the agglomeration of firms has a self-reinforcing effect on foreign investment (Ellison & Glaeser, 1997), suggesting that investment decisions may well be influenced by the presence of agglomeration economies. This is confirmed in relation to both traditional industrial clusters and to local concentrations of

86

A. Majocchi, M. Presutti / International Business Review 18 (2009) 76–88

foreign firms. This last result provides important insights regarding a broad range of issues (Enright, 2005). From a research perspective, the analysis of the strategic behaviour of firms under agglomeration externalities is of crucial importance to effectively measure the effects of agglomeration on the investment decisions of multinationals. From a managerial perspective, our results suggest that, when agglomeration externalities exist, the strategic investment decisions of multinationals should take this agglomeration effect into account. However, from a firm’s point of view, even if agglomeration might facilitate hiring competitors’ employees and gaining access to technology spillovers, it also facilitates foreign competitors’ in hiring their workers and gaining access to their technology. In this sense, it is only when the relative costs and advantages of agglomerating are considered that an optimal location decision can be made by multinationals. From a policy perspective, the positive influence of agglomeration economies highlights the benefits of attracting large initial investments when developing concentrated production regions such as industrial parks (Rugman et al., 1995). Policies able to reinforce the agglomeration economies might become effective instruments to attract foreign investments. The presence of consolidated foreign firms in a province may act as a positive factor in building a locality’s reputation, reinforcing the attractiveness of that particular context. However, we further specify these results showing that along with agglomeration economies also the diffusion of entrepreneurial capabilities at the local level can be a driver of foreign investments. This conclusion seems to suggest that entrepreneurial capabilities are an intangible asset that is valuable for firms. This asset is a strategic resource when foreign firms invest within a country. Our view complements the much discussed idea of multinational corporations as a network of dispersed assets with a high innovative and entrepreneurial potential that can be leveraged in order to gain competitiveness (Rugman & Verbeke, 2001). Entrepreneurship is not only a virtue that needs to be internally nurtured but also depends on the resources that the subunit (i.e. the subsidiary) finds in a local environment (Ghoshal & Nohria, 1989). Our results show that the entrepreneurial attitude of the local market has an impact on MNC investment decisions thus confirming this view. Finally, in analysing the role of local institutions, we include a variable previously unstudied in this field of research, that is, the quality of social environment present in the territorial context, seen as a generic place-specific resource. It is expressed by crime rates in a province, allowing us to highlight the role of social institutions in generating socioeconomic well-being for communities and individuals which may, in turn, attract more multinationals. The mixed results produced by the analysis of this variable suggest that further research on this variable could better specify our findings. Given these strengths, the study suffers from several limitations. First, we are aware that the use of stock variable is not optimal and that a stream of flows values for different years would have greatly improved our methodology allowing a panel type approach to be applied and a more precise definition of the investment location decisions taken by MNCs. With no other measure available, we could not develop a better measurement of the distribution of foreign investments at a provincial level. Second, the level of foreign direct investment is determined by a number of factors larger than just location advantages (Dunning, 1993). These factors range from the characteristics of the investing firms to a series of specific macroeconomic and industry features. However, our aggregation procedure and the fine-grained level of geographical analysis do not allow us to have these data at a provincial level. Given this limits our results should be interpreted with caution. Similarly, our data are not able to explore the interesting effects of different firm strategies on location decisions. For example, we do not distinguish between different entry modes and, more specifically, between greenfield investments (when firms are free to choose their best location) and merger and acquisition when the choices of the acquiring firm are somehow constrained by the previous choices of the acquired target. Our database does not discriminate between different strategic approaches. It is logical to affirm that foreign firms looking for either markets or resources or strategic assets will be attracted by different factors and this point certainly requires further research. Third, as all the studies focused only on one country the problem of generalisation of results regarding other geographical contexts arises. Finally, while we reflect upon the role of agglomeration economies in attraction of foreign firms, this research suggests the importance of further in-depth analysis into the impact of industrial clusters on attracting multinationals investments. We consider the role that entrepreneurship and social environment quality have on the cluster’s ability to attract foreign investment but further differentiation in agglomeration of firms and more refined measurement of clustering could be developed. Moreover, the complex concept of social capital is captured by only two variables and we are aware that others more refined measures could be developed. Acknowledgements Support from MIUR, FIRB 2003, project DM2PMI is gratefully acknowledged. References Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park: Sage Publications. Aitken, B., & Harrison, A. (1999). Do domestic firms benefit from foreign investment? Evidence from Venezuela. American Economic Review, 89(3), 605–618. Almeida, P. (1996). Knowledge sourcing by foreign semiconductor industry. Strategic Management Journal, 17(Winter Special Issue), 155–165. Almeida, P., & Kogut, B. (1997). The exploration of technological diversity and the geographic localization of innovation. Small Business Economics, 9(1), 21–31. Basile, R. (2003). Acquisition versus greenfield investment: The location of foreign manufacturers in Italy. Regional Science and Urban Economics, 34, 3–25. Bevan, A., Estrin, S., & Meyer, K. (2004). Foreign investment location and institutional development in transition economies. International Business Review, 13(1), 43–64.

A. Majocchi, M. Presutti / International Business Review 18 (2009) 76–88

87

Birkinshaw, J. (1997). Entrepreneurship in multinational corporations: The characteristics of subsidiary initiatives. Strategic Management Journal, 18(3), 207–229. Birkinshaw, J., & Hood, N. (2000). Characteristics of foreign subsidiaries in industries clusters. Journal of International Business Studies, 31(1), 141–154. Birkinshaw, J., Hood, N., & Young, S. (2005). Subsidiary entrepreneurship, internal and external competitive forces and subsidiary performance. International Business Review, 14(2), 227–248. Boojihawon, D., Dimitratos, P., & Young, S. (2007). Influences and characteristics of multinational subsidiary entrepreneurial culture: The case of the advertising sector. International Business Review, 16, 549–572. Brouthers, L. E., Werner, S., & Wilkinson, T. J. (1996). The aggregate impact of firms’ FDI strategies on the trade balances of host countries. Journal of Business 1(3). Brenton, P., Di Mauro, F., & Lu¨cke, M. (1999). Economic integration and FDI: An empirical analysis of foreign investment in the EU and in Central and Eastern Europe. Empirica, 26, 95–121. Brouthers, K. D. (2002). Institutional, cultural and transaction cost influences on entry mode choice and performance. Journal of International Business Studies, 33, 203–221. Campos, N. & Kinoshita, Y. (2003). Why does FDI go where it goes? New evidence from the transitional economies. IMF Working Paper, WP/03/228. Cantwell, J. A. (1989). Technological innovation and multinational corporations. Oxford: Basil Blackwell. Cantwell, J. A., & Piscitello, L. (2002). The location of technological activities of MNCs in European regions: The role of spillovers and local competencies. Journal of International Management, 8, 69–96. Cantwell, J. A., & Iammarino, S. (2000). Multinational corporations and the location of technological innovation in the UK region. Regional Studies, 34(4), 317–333. Casson, M. (1990). Enterprise and competitiveness. New York: Oxford University Press. Chang, S.-J., & Park, S. (2005). Types of firms generating network externalities and MNCs co-location decisions. Strategic Management Journal, 26, 595–615. Cheng, L. K., & Kwan, Y. K. (2000). What are the determinants of the location of foreign direct investment? The Chinese experience. Journal of International Economics, 51, 379–400. Chung, W., & Alca´cer, J. (2000). Knowledge seeking and location choice of foreign direct investment in the United States. Management Science, 48(12), 1534–1554. Cohen, J., Cohen, P., West, S., & Aiken, L. (2003). Applied multiple regression/correlation analysis for the behavioral sciences (3rd edition). Mahwah, NJ: Lawrence Erlbaum Associates. Coughlin, C., & Segev, E. (2000). Foreign direct investment in China: A spatial econometric study. The World Economy, 23(1), 1–23. Delios, A., & Henisz, W. (2003). Political hazard, experience, and sequential entry strategies: The international expansion of Japanese firms, 1980–1998. Strategic Management Journal, 1153–1164. De Propris, L. (2004). Mapping local production systems in the UK: Methodology and application. Regional Studies, 197–212. Devereux, M. P., Griffith, R., & Simpson, H. (2004). The geographic distribution of production activity in the UK. Regional Science and Urban Economics, Elsevier, 34(5), 533–564. Driffield, N., & Munday, M. C. (2000). Industrial performance, agglomeration and foreign manufacturing investment in the UK. Journal of International Business Studies, 31(1), 21–37. Dunning, J. H. (1993). Multinational enterprises and the global economy. Reading: Addison Wesley. Dunning, J. H. (1996). The geographical sources of the competitiveness of firms: Some results of a new survey. Transnational Corporations, 5(3), 1–29. Dunning, J. H. (1997). Location and the multinational enterprise: A neglected factor? Journal of International Business Studies, 29, 45–66. Ellison, G., & Glaeser, E. (1997). Geographical concentration in US manufacturing industries: A dartboard approach. Journal of Political Economy, 105, 889–927. ¨ . So¨lvell, & P. Hagstro¨m (Eds.), The dynamic firm: The role of technology, strategy, Enright, M. (1998). Regional clusters and firm strategy. In A. D., Chandler, Jr., O organization, and regions. Oxford: Oxford University Press. Enright, M. (2005). Regional management centers in the Asia-Pacific. Management International Review, 45(Special issue 1), 57–80. Ernst & Young (2007). European attractiveness survey 2007. Ernst & Young. Retrieved 17.03.2008. Fernhaber, S., Gilbert, B. A., & McDougall, P. P. (2008). International entrepreneurship and geographic location: An empirical examination of new venture internationalization. Journal of International Business Studies, 39(2), 267–290. Flores, R., & Aguilera, R. V. (2007). Globalization and location choice: An analysis of U.S. multinational firms in 1980 and 2000. Journal of International Business Studies, 38(7), 1187–1210. Frost, T. S. (2001). The geographic sources of foreign subsidiaries’ innovations. Strategic Management Journal, 22(2), 101–123. Frost, T. S., Birkinshaw, J., & Ensign, P. C. (2002). Centers of excellence in multinational corporations. Strategic Management Journal, 23(11), 997–1018. Fujita, M., Krugman, P., & Venables, A. (1999). The spatial economy: Cities, regions and international trade. MIT Press. Ghoshal, S., & Nohria, N. (1989). Internal differentiation within multinational corporations. Strategic Management Journal, 10, 323–337. Gordon, I. R., & McCann, P. (2000). Industrial clusters, complexes, agglomeration and/or social networks? Urban Studies, 37, 513–532. Gorg, H., & Strobl, E. (2001). Multinational companies and productivity spillovers: A meta-analysis. Economic Journal, 111, 723–739. Grosse, R., & Trevino, L. J. (2005). New institutional economics and FDI location in Central and Eastern Europe. Management International Review, 45(2), 123–145. Haddad, M., & Harrison, A. (1993). Are there positive spillovers from direct foreign-investment. Evidence from panel-data for Morocco. Journal of Development Economics, 42(1), 51–74. Head, C. K., Ries, J. C., & Swenson, D. L. (1995). Agglomeration benefits and location choice: Evidence Japanese manufacturing investments in the United States. Journal of International. Economics, 38, 223–247. Hendry, C., & Brown, J. E. (2006). Organizational networking in UK biotechnology clusters. British Journal of Management, 17(1), 55–73. Hood, N., & Young, S. (Eds.). (1999). The globalization of multinational enterprise activity and economic development. London: Macmillan. Krugman, P. (1991). Increasing returns and economic geography. Journal of Political Economy, University of Chicago Press, 99(3), 483–499. Landesmann, M. A. (2003). Structural features of economic integration in an Enlarged Europe: Patterns of catching-up and industrial specialisation, European Economy, n. 181, January. Majocchi, A., & Strange, R. (2007). Economic liberalisation and the FDI location decision: Evidence from Italian firms in Eastern Europe. Journal of East-West Business, 13(2/3), 93–114. Malmberg, A., & Maskell, P. (1999). Localized learning and regional economic development. European Urban and Regional Studies, 6(1), 5–8. Mariotti, S., & Piscitello, L. (1995). Information costs and location of FDIs within the host country: Empirical evidence from Italy. Journal of International Business Studies, 26(4), 815–841. Maskell, P., & Malmberg, A. (1999). Localised learning and industrial competitiveness. Cambridge Journal of Economics, 23(2), 167–185. McCann, B. T., & Folta, T. B. (2008). Location matters: Where we have been and where might go in agglomeration research. Journal of Management, 34(3), 532–565. McCann, P. T., Arita, I., & Gordon, R. (2002). Industrial clusters, transactions costs and the institutional determinants of MNE location behaviour. International Business Review, 11, 647–663. McEvily, B., & Marcus, A. (2005). Embedded ties and the acquisition of competitive capabilities. Strategic Management Journal, 26, 1033–1055. Mayer, T., & Mucchielli, J. L. (1998). Strategic location behaviour: The case of Japanese investments in Europe. Journal of Transnational Management Development 3(3/4). Meyer, K. (2001). Institutions, transaction costs, and entry mode choice in Eastern Europe. Journal of International Business Studies, 32(2), 357–367. Mudambi, R., & Navarra, R. (2002). Institutions and international business: A theoretical overview. International Business Review 11(6) 635–464. Nachum, L. (2000). Economic geography and the Location of TNCs: Financial and professional service FDI to the USA. Journal of International Business Studies 31(3). Narula, R., & Zanfei, A. (2006). Globalization of innovation: Enterprises. In J. Fagerberg, D. Mowery, & R. R. Nelson (Eds.), The Oxford handbook of innovation. Oxford: Oxford University Press. Narula, R., & Dunning, J. H. (2000). Industrial development, globalisation and multinational enterprises: New realities for developing countries. Oxford Development Studies, 28(2), 141–167. North, D. C. (1990). Institution, institutional change and economic development. Cambridge: Cambridge University Press.

88

A. Majocchi, M. Presutti / International Business Review 18 (2009) 76–88

Pak, Y. S., & Park, Y. R. (2005). Characteristics of Japanese FDI in the East and the West: Understanding the strategic motives of Japanese investment. Journal of World Business, 40(3), 254–266. Peneder, M. (1999). Intangible investments and human resources. The new WIFO taxonomy of Manufacturing Industries, WIFO Working papers no. 114. WIFO: Wien. Porter, M. E. (1998). Cluster and competition. On competition. Boston: Harvard Business School Press. Porter, M. E. (1990). The competitive advantage of nations. London, Basingstoke: MacMillan. Rugman, A. M., Verbeke, A., & D’Cruz, J. (1995). Internalization and de-internalization: Will business networks replace multinationals? In G. Boyd (Ed.), Competitive and cooperative management: The challenge of structural interdependence (pp. 107–128). Aldershot. Rugman, A. M. (2000). The end of globalization. London: Random House. Rugman, A. M., & Verbeke, A. (2001). Subsidiary-specific advantages in multinational enterprises. Strategic Management Journal, 22(3), 237–250. Shane, S., & Venkataraman, S. (2000). The promise of entrepreneurship as a field of research. The Academy of Management Review, 25(1), 217–226. Shaver, J., & Flyer, F. (2000). Agglomeration economics, firm heterogeneity, and foreign direct investment in the United States. Strategic Management Journal, 21(12), 1175–1193. Spanos, Y. E., Zaralis, G., & Lioukas, S. (2004). Strategy and industry effects on profitability: Evidence from Greece. Strategic Management Journal, 25, 139–165. Storper, M. (1997). The regional world: Territorial development in the global economy. New York: Guilford Press. Sun, Q., Tong, W., & Yu, Q. (2002). Determinants of foreign direct investment across China. Journal of International Money and Finance, 21, 79–113. Tegarden, L. F., Hatfield, D. E., & Echols, A. E. (1999). Doomed from the start: What is the value of selecting a future dominant design? Strategic Management Journal, 20(6), 495–518. Van Den Bulcke, D., & Verbeke, A. (2001). Globalization and the small open economy. Aldershot: Elgar. White, H. (1980). A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. Econometrica, 48, 817–883. Antonio Majocchi. Is an associate professor of International Business at the Faculty of Economics, University of Pavia. His research and teaching interests focus on the internationalisation of SMEs and international marketing. Manuela Persutti. Is an assistant professor of Management at the Faculty of Economics, University of Bologna. Her main research interests focus on the dynamics of social capital diffusion and the internationalisation of start-ups.