Frank McDonald: Liverpool University Jens Gammelgaard: Copenhagen Business School Christoph Dörrenbächer, Berlin School of Law and Economics Heinz Tüselmann, Manchester Metropolitan University Andreas Stephan: Jönköping International Business School

Cities and the autonomy and network relationships of foreign subsidiaries: the case of Greater Copenhagen

1

Abstract This study considers the autonomy and network relationships in the subsidiaries of multinational corporations located in a competitive city region (Greater Copenhagen), compared to location elsewhere in Denmark. The study by focusing on city location expands and develops the literature that considers the importance of sub-national location. The paper enhances understanding on some of the major characteristics of the management systems of subsidiaries in different types of sub-national location. This increases knowledge on the complex ways by which spatial factors influences the strategies of firms, and thereby contributes to the growing literature on how not only national frontiers, but also sub-national frontiers, influence the strategies of multinational corporations. Using a combination of economic geography theories on competitive city locations and the resource based and network theories of the firm the paper examines the relationship between performance, and autonomy and network relationships of subsidiaries located in a competitive city region, compared to other locations in the same country. The theoretical underpinnings of the study lead to propositions that the autonomy and network relationship configuration in competitive city region differ from those of locations in other locations. The paper centres on the influence in a competitive city region location of characteristics in the autonomy and network relationship configuration compared to other locations in the same country. The propositions are tested by considering the effects on performance of autonomy and of inter and intraorganizational network relationships of subsidiaries in Greater Copenhagen compared to those located elsewhere in Denmark. The results reveal that subsidiaries located in Greater Copenhagen have a different autonomy and network configuration that relate positively to performance. The paper considers the implications of these differences for research on competitive cities and sub-national location. Keywords Autonomy, networks, competitive cities, sub-national location, subsidiaries

2

Introduction Research on the topic of the location strategies of multinational corporations (MNCs) is increasingly considering the issue of sub-national location (Dimitratos et al, 2009; Meyer et al, 2005; Nguyen et al, 2013; Santangelo, 2009). The motivations to move beyond only considering national frontiers arises from a renewed focus on the location component of the OLI paradigm (Cantwell, & Mudambi, 2011; Dunning, 2009; McCann, 2011), and the rise of resource and institutional based theories that highlights the complex interplay between spatial factors and the strategies of MNCs (Andersson et al, 2001; Peng et al, 2008 & 2009; Meyer, et al 2011). Research on supply and value chains and in the network relationships of firms (Dicken et al, 2001; Dicken and Malmberg, 2001; Hendersen et al, 2002; Yeung, 2005) influences the work of international business scholars seeking to extend location studies beyond only considering national frontiers. The work of economic geographers therefore exerts significant influence on the research approaches used by international business scholars interested in the impact of spatial factors on MNC strategies (Beugelsdijk, et al, 2010).

The impact of cities on the strategy of firms is a strong feature in research by economic geographers (Pain, 2012; Sassen, 2013). The importance of cities for economic and business developments is also prominent in government circles (OECD, 2006) and among commercial organizations (Economic Intelligence Unit, 2012). The development of digital and other radical technologies, combined with new economic, institutional and political environments have created a new global order in which certain types of cities provide wide ranging benefits not available in other areas (Storper, 2013). The benefits of locating in such cities arise from proximity to scarce and valuable resource leading to significant agglomeration benefits that 3

combine with network relationships competencies enabling firms and markets to link effectively across national and sub-national frontiers (Taylor, 2004). The concentration of agglomeration benefits and effective networking relationships in certain types of cities make them focal points in the global economy thereby exercising powerful influence on the location activities and strategies of firms (Beaverstock et al, 2000).

Economic geographers classify cities into different tiers that connect to the global operations of firms in a variety of ways thereby influencing the strategies of MNCs (Beaverstock et al, 1999; Taylor, 1997). Cities provide a wide and diverse range of resources and competencies that are necessary, and in some cases essential, for the ability of firms to fulfil their strategic objectives. The relationship between MNCs and cities is therefore a major driver of many of the most significant developments in the global economy (Sassen, 2013; Scott, 2012). There is however very little research by international business researchers on the role of cities in MNC strategies. The very limited research on location in cities by MNCs tends to centre on the attractiveness of ‘global cities’ as means to lessen the liability of foreignness that firms encounter when they cross national frontiers (Goerzen et al, 2013). The management systems used by MNCs to acquire and exploit the agglomeration and networking benefits available in ‘global cities’ to enable them to attain their objectives is considered, but is less central than the examination of city location to mitigate liability of foreignness.

The means by which the subsidiaries of MNCs embed in host locations and link with the other parts of their MNC to achieve their strategic objectives from their diverse geographical locations centres on the use by subsidiaries of inter and intra-organizational network relationships (Birkinshaw et al., 2005; Frost et al., 2002) and autonomy (Young and Tavares, 4

2004). Using the resource-based view (RBV) and network theory, studies find that complex interactions between autonomy and intra and inter-network relationships are necessary to extract the benefits that stem from the resource bundles and competencies available in host locations (Andersson et al, 2005; Gamelgaard et al, 2012). These studies however only consider autonomy and embeddedness at national level. There is a need therefore to more fully understand how MNCs organise their autonomy and network relationship configurations in the different geographical locations of their operations to achieve their strategic goals. This is part of a research agenda to discover how the complex global, national and subnational locations of MNCs are woven together to achieve the strategic objectives of MNCs (Beugelsdijk and Mudambi, 2013). Given the growing importance of cities in the global economy it is clear that it is necessary to have an increased understanding of how MNCs organize their strategies with regard to autonomy and network relationship configurations to effectively utilise the benefits available in such cities.

This paper increases our understanding on these matters by investigating autonomy and network relationships systems and performance in a competitive city region (Greater Copenhagen) compared to subsidiaries located elsewhere in Denmark.

MNCs and location Economic geographers often regard MNCs as being primarily a type of multi-establishment firm that has establishments in different countries. The cross frontier aspect of MNCs is normally not central to analysis of location strategies as compared the drivers that lead to firms becoming geographically dispersed normally across a number of countries and regions in the global economy (Scott, 1986). Economic geographers often focus on relational 5

configurations between firms and territory (sub-national locations) and cities are normally a prominent feature in this type of analysis (Dicken and Malmberg, 2001; Taylor, 2004). These relational approaches focus on the local business environments in which firms are embedded and the linkages of these local environments to other locations in the global economy (Yeung, 2005). International business scholars tend to take a somewhat different view of MNCs. They regard national frontiers has being of profound importance and they have developed a host of theories to examine the implications for location of national frontiers (Dunning, 2009). The focus in international business research is normally on the drivers of the internationalization process of firms that leads them to engage in extensive cross frontier operations. At the core of much of contemporary international business research is the examination of the major challenges of managing effectively across national frontiers. This often resolves around problems associated with the liability of foreignness that MNCs face in their foreign operations (Zaheer, 1995). These liabilities are considered to be rooted in differences in national institutional systems between home and host locations (Peng et al, 2008 & 2009). The growing importance of geographical factors in international business research focuses attention on the importance of spatial factors connected to agglomeration benefits and networking relationship advantages available to firms in various volumes and qualities in different host locations, including sub-national locations (Beugelsdijk et al, 2010). This has led to calls to consider national frontiers as a type of ‘spatial discontinuity’ (Beugelsdijk and Mudambi, 2013: 411). In this context, MNCs face sharp and discontinuous national frontiers, but are then faced with more smooth and continuous internal frontiers within national frontiers that delineate sub-national areas in which firms seek the locations that best facilitate the achieving of the strategic objectives of firms (Beugelsdijk, 2007). The management 6

systems associated with the various and diverse locations of MNCs must be developed and sustained in a complex spatial environment in which both national and sub-national frontiers are crossed. As many economic geographers regard certain types of cities as being at the core of proximity benefits and networking competencies (Beaverstock et al, 2000; Pain 2012; Taylor, 2004) examination of MNC strategies at city level provides a more nuanced view of the affects of spatial factors on the strategies of MNCs. This type of analysis of the effects of location on the strategies of MNCs is called for by international business researchers (Beugelsdijk and Mudambi, 2013).

The arguments forwarded for why certain cities have competitive advantages over other areas centre on wide, deep, and scare resources and competencies leading to agglomeration benefits, and networking relationships characterised by ‘buzz’ and other proximity related benefits (Scott, 2012; Stroper, 2013). These characteristics suggests the firms located in cities need to embed effectively in host locations and to relate effectively to other parts of the MNC to take full advantage of the range of competitive advantages available in such locations. A key question is whether MNC’s located in competitive city locations have different characteristics in their use of autonomy and network relationship configurations from those situated in other types of locations. In other words, does the configuration of autonomy and network relationships of subsidiaries differ in competitive city locations compared to other areas?

There is extensive research by economic geographers on MNC location strategies in competitive regions, including competitive city regions (Diez and Bergerô, 2005; Musil, 2009; Yeung, 2000; Zhao and Zhang, 2007). Much of this literature centres on the ability of 7

firms to exploit proximity effects by using effectively their network relationship systems (Eraydin and Fingleton, 2006; Jones, 2007; Wu and Radbone, 2005). These studies focus on inter-organizational network relationships, but place less emphasis on intra-organizational relationships with other parts of the MNC, and rarely consider the affect of the autonomy of subsidiaries. The work considers comparative regional differences (Burger et al, 2013) but not in terms of autonomy and network relationship systems. There is therefore no evidence from this literature on whether autonomy and network relationship configurations in subsidiaries vary in different types of sub-national location, or if different configurations of autonomy and network relationships in competitive city regions are associated with performance.

International business research on subsidiary location factors often centres on the OwnershipLocation-Internalization (OLI) paradigm (Dunning, 2009; McCann and Mudambi, 2004). The work of Hymer (1971) explicitly or implicitly also underpins many of the studies on MNC location. In Hymer’s model MNCs concentrate key activities in locations with large pools of skilled labour and good international communications, with more routine operations based in locations with less deep and rich pools of skilled labour and less extensive communications. These locations are identified by Hymer as major global cities for the former, and lesser cities in the latter case (Driffield and Love, 2005). The cities aspects of Hymer’s approach to MNC location has however faded from most studies. The most extensively studied aspect of subnational location using the OLI type approach is the influence on foreign direct investment (FDI) of industrial clusters (Birkinshaw and Hood, 2000; Chang and Park, 2005; Driffield, 2006; McCann and Mudambi, 2005; Shaver and Flyer, 2000). This literature focuses on institutional distance and liability of foreignness, the spillovers of knowledge to domestic 8

firms, and the affects of firm heterogeneity issues such as the strategic orientation of MNCs, technological transfers and R&D intensity. These studies do not explicitly focus on management structures such as autonomy and network systems that are required to mobilise and effectively utilise resources in different locations. Literature based on the RBV and network theory does highlight the role of autonomy and networks relationship configurations for the performance of subsidiaries (Andersson et al., 2001 & 2007; Gammelgaard et al, 2013; Schmid and Schurig, 2003) but does not consider the influence of sub-national location.

An international business literature on location is emerging that relates to economic geography research on competitive cities. This literature centres on examination of key factors influencing FDI to locate in competitive city regions. The key reason for the attractiveness of competitive city regions is seen as the ability to limit problems of liability of foreignness (Zaheer, 1995) by networking with the concentrations of other prominent MNCs and the sophisticated institutional systems in such cities. These network relationships help MNCs to adjust to additional costs of foreign locations (Nachum and Wymbs, 2007). The network relationships with other MNCs and also with institutional systems result in the development of interconnectiveness that stimulates diverse and high quality provision of services to MNCs in these regions. The provision of high quality services for MNCs in competitive city regions consequently lowers the liabilities of foreignness and permits MNCs to exploit effectively the unique or rare resource bundles found in such city regions (Goerzen, et al., 2013). There is evidence that this type of behaviour by MNCs has become more important in the latter half of the twentieth century and has continued into the twenty first century (McCann and Acs, 2011). This literature has however not considered the association 9

between autonomy and network relationship configurations and performance in competitive city regions compared to other locations.

Competitive City Regions

The topic of regional competitiveness stimulates much debate about how certain regions acquire distinctive competitive advantages and about the characteristics of such advantages (Boschma, 2004; Kitson et al., 2004; Harrison, 2007; Parr, 2008; Ward and Jonas, 2004). Competitive city regions may contain several clusters in different industries in the same city region making them in effect a cluster of industrial clusters (OECD, 2006). Evidence suggests that competitive city regions are more likely to have large pools of highly skilled labour (Eriksson and Lindgren, 2009; Hospers, 2003) and substantial pools of labour with creative skills and competencies (Florida, 2005; Sassen, 2013; Scott, 2006). Concentrations of high knowledge-based assets combined with effective networking and close proximity to commercial organizations, governmental and non-governmental agencies facilitate help to harness the potential to benefit from potential competitive advantages available in these cities (Glaser et al., 2001; Jones, 2007; Storper and Venables, 2004; Stroper, 2013).

Economic geographers identify competitive city regions in terms of a hierarchy in which differing volumes and qualities of resources and competencies are evident (Beverstock et al, 1999 & 2000; Taylor, 1997). There are three main views on the relationships between MNCs and competitive cities. The Global Commodity Chains (GCC) position centres on global supply chain system approaches (Gereffi and Korzeniewicz, 1994). In this view, certain cities have well developed and highly valued clusters of industries that grant firms located in such 10

cities high valued operations not easily available in other regions. The advantages arising from these clusters of industries allow MNCs located in these cities to play dominant HUB role in global supply chains. Locating in these types of cities is considered essential for the effective operation of global supply chains. The World City Network (WCN) view focuses on the centrality of key ‘world cities’ for development of the global economy. These ‘world cities’ provided specialised and effective economic and business infrastructures that enable MNCs to direct and control international flows of goods, services, labour and especially knowledge from these locations. Location in these ‘world cities’ enables MNCs to acquire and develop competitive advantages (mostly centred on knowledge-based assets) that enable them to be major and innovative operators in the global economy (Sassen, 2002 & 2013). The major difference between WCN and GCC is that the former highlights the importance of effective networks in certain cities that can utilise the high volume and quality of knowledgebased assets that are available in major world cities, and also transmit these knowledge-based assets across the global economy. The GCC view is more focus on agglomeration benefits arising from clusters of industries that make such cities HUBS for directing and controlling global supply chains. The Global Production Network (GPN) stance focuses on embeddeness in city locations founded on the relationships that firms have with their territorial location (Dicken et al, 2001; Henderson et al, 2002; Hess and Yeung, 2006). The embeddeness of firms in their territory (often cities) is founded on the network relationships of firms with other firms, governmental and other organizations in cities. The GPN view does not require a city to be a ‘world city’ or to have substantial concentrations of MNCs that are headquarters, or that have a dominant HUB role in global supply chains. In the Global Production Network approach, a city need not be a world city (or hub city) neither must it have a wide ranging stock of knowledge-based assets. Competitive city regions in the GPN view have network 11

relationships and stocks of scare and desirable assets that can make them useful as major HUBS in global supply chains, or centres for R&D and innovation for wide ranging operations in the global economy. Alternatively, some competitive cities in the GPN view provide important contributions (but not necessarily core contributions) for global operations, or they are good locations to direct and control regional (eg European), or national strategic objectives.

In this view, competitive cities regions may host MNC subsidiaries whose

primary role is not as headquarters, or to play a prominent part in global supply chains, or to be global centres of R&D and innovation. Subsidiaries may be located in competitive cities to take advantage of unique or rare resource bundles that enable subsidiaries to gain competitive advantages in what may be important but specialised and limited strategic objectives. The GPN approach therefore, permits examination of relative competitiveness in cities, as it does not depend on cities having a dominant global position, as either a headquarters or a pivotal part of a global supply chain (Mans, 2013).

The benefits of location in competitive city regions largely connect to proximity to assets that have high knowledge content, especially in the approaches of WCN and GPN. Such knowledge-based assets often require sophisticated and close contact between agents to acquire the benefits of such assets (Teece, 1986 and 1998). In other words, the comparative advantage of competitive city regions largely stems from relatively large stocks of high quality knowledge-based assets and the ability to use network relationships to effectively use these assets to generate and sustain dynamic capabilities (Teece et al, 1997). This suggests that to use scare and desirable resources and networking competencies effectively in competitive city regions is likely to require subsidiary organization systems that have different characteristics from those in other regions in the same national location. The 12

configuration of autonomy and network relationships systems are at the heart of the management systems that enable firms to attain and exploit knowledge-based assets (Andersson et al., 2007; Gammelgaard et al, 2013; Teece, 2000; Venaik, et al., 2005). In competitive city locations therefore it is likely that subsidiaries have different configurations of autonomy and network relationships than those that prevail in other host locations.

Copenhagen Greater Copenhagen (the City of Copenhagen and the North Eastern part of Zealand) is a city region with a large pool of high knowledge-intensity firms (Winther and Hansen, 2006). Greater Copenhagen was a centre of manufacturing activities in the early post-war years, but it evolved into a services and knowledge-based region (Maskell, 1986). The evolution of Copenhagen as a knowledge-based city region followed from such developments as the emergence of the electro-medical instruments cluster arose from cooperation between businesses and the State University Hospital in Copenhagen (Lotz, 1993). The knowledge base of the city grew with the integration of research skills in Danish universities leading to the creation of the Medicon Valley cluster, which is the third most successful biotechnology cluster in Europe (Drejer et al., 1999; Steinfield and Scupola, 2008). Services and knowledge-based industries cluster more strongly in Copenhagen compared to the rest of Denmark as is reflected in employment patterns, which is also evident in foreign owned companies located in Denmark (Nielson et al., 2009). Although there are pockets of knowledge-based industries and firms outside of Copenhagen, the largest concentration of knowledge-based industries is in Copenhagen (Drejer el al., 1999). Labour productivity is considerably higher in the Copenhagen labour market areas compared to the rest of Denmark (Timmermans and Boschma, 2013). 13

Greater Copenhagen has approximately 20% of the population of Denmark, but has clearly larger concentrations of knowledge-based industries than is suggested by the proportion of population and higher productivity. The picture of Copenhagen indicates that it is, on average, a competitive city region compared to the rest of Denmark. The high density of population in Greater Copenhagen relative to the rest of Demark and the concentration of firms, governmental and non-governmental agencies appear to give this region a resource bundle that confers the status of a competitive city region as compared with the rest of Demark.

Copenhagen is not classified by the many indices of world cities as being a top ranked city. It is usually located in the middle of the ranking of world cities (for example AT Kearney, 2014). However, Copenhagen has a leading role in knowledge based industries that give t a competitive edge in Denmark and possibly, together with Stockholm, in the Nordic countries. In terms of the GPN approach, Copenhagen is likely to confer competitive advantages in Denmark and possible the Nordic countries, and may possess niche advantages in knowledge based industries in the global economy. Given the position of Copenhagen it is likely that MNCs will use different organizational systems in that city compared to the rest of Denmark in order to exploit the competitive advantages of the city. This suggests that subsidiaries located there will have different configurations of autonomy and network relationships than in subsidiaries located elsewhere in Denmark. Network and Autonomy Relationships Most of the views on the nature of the competitive advantages of cities have at their core wide and rich pools of scarce and valuable resources and good networking competencies 14

(Dicken et al, 2001; Scott, 2001; Storper, 2013). This implies that MNCs located in such cities will require management systems that will enable them to embed into host locations, this requires extensive and effective inter-organizational networks. To make best use of these resource and networking competencies in host locations also requires embedding into the rest of the MNC to enable access to the firm’s resources to make best use of assets available in host locations, and also to supply the products and knowledge to global markets from host locations (Birkinshaw et al 2005; Frost, et al, 2002). This requires subsidiaries located in competitive cites to have good intra-organizational networks. Granting subsidiaries autonomy to enable local decision-making to make best use of inter and intra-organizational networks is also often required to obtain the competitive benefits from host locations (Andersson et al, 2005; Young and Tavares, 2004). A complex configuration of autonomy and inter and intraorganizational networks is therefore seen as being important to leverage the competitive advantages of host locations (Gamelgaard et al, 2012).

Given the competitive benefits to

MNCs of concentrating location of higher valued activities in competitive cities that provide deep and rich pools of resources and effective means communication channels (Hymer, 1971) it is likely that subsidiaries in such cities will have a configuration of autonomy and inter and intra-organizational networks that differs from the configurations in subsidiaries not located in such cities.

Linking to inter-organizational networks, such as customers, suppliers, competitors, and supporting government and non-government agencies provides improved access to the scare resource bundles available in host locations thereby enhancing the development of the firm specific advantages of subsidiaries. Subsidiaries also need to connect via intra-organizational relationships to other parts of their MNCs to facilitate the transfer of assets and competencies 15

to enable effective exploitation of the desirable assets and network relationships in host locations (Andersson et al., 2005; Frost et al., 2002; Holm, et al., 2003). Competitive city regions tend to have a richer menu of knowledge-based assets than other regions (Martin et al., 2011). Extracting and utilizing such knowledge is likely to require subsidiaries to develop extensive inter and intra-organizational network relationships systems (Andersson et al., 2007). It is therefore likely that a more pronounced relationship exists between network relationships and performance in competitive city locations compared to other regions. This leads to the first general proposition -

P1 Network relationships display a more pronounced relationship to performance in competitive city locations compared to other locations.

Autonomy helps subsidiaries utilize effectively the competitive advantages in host locations by reducing the time and costs expended in negotiating with headquarters for permissions (Andersson et al., 2005; Birkinshaw et al. 2005; Taggart, 1998, Young and Tavares, 2004). The reduction in time negotiating to get permissions from headquarters and the ability to alter quickly strategies and operations that can follow from autonomy can enhance performance. Studies find that marketing innovations and subsidiary growth connect to granting autonomy (Fenton-OCreevy et al, 2008; Johnson and Medcof, 2007; Vernaik et al., 2005).

The

concentration of knowledge-based assets and dense network connections in competitive city regions suggest that subsidiaries in these locations will benefit from autonomy to permit them to take quickly and at low cost take advantage of the ‘buzz’ in such cities (Storper and Venables, 2004). There is therefore a likelihood of a more pronounced relationship between

16

autonomy and performance in competitive city regions compared to those not located in such regions. The second general proposition is therefore -

P2 Autonomy has a more pronounced relationship to performance in competitive city locations compared to other locations.

The ability to exploit effectively competitive advantages in host locations depends on the configuration of autonomy and both intra- and inter-organizational network relationships. The interaction between autonomy and network relationships is likely to exercise a positive influence on performance as it enables subsidiaries to develop appropriate management systems to make best use of resources and competencies available in host location and in the other locations of the MNC (Birkinshaw et al., 2005). The ability to have effective interaction between autonomy and network relationships systems in competitive city regions appears to influence MNCs to locate in such regions (Goerzen et al., 2013). Evidence exists, at a national level, that interaction between autonomy and networks positively affects performance. Indeed these interactions seem to have a more powerful effect than the direct effects of these factors on performance (Gammelgaard et al., 2013). It seems that subsidiaries need to combine autonomy and inter- and intra-network relationships to create managements system that can embed into host locations and link effectively to the rest of their MNCs to enable them to fulfil their strategic objectives. One of the main means therefore by which MNCs exploit effectively the competitive advantages available in host locations is to create appropriate configurations of autonomy and networks. In competitive city locations the concentration of large volumes and high quality of knowledge-based assets, combined with good networking relationships, lead to attractive stocks of possible competitive advantages. 17

For subsidiaries to gain and exploit these potential competitive advantages requires not only location in such competitive cities, but the development of an appropriate configuration of autonomy and network relationships. In other locations with lower stocks of potential competitive advantages, the rewards (and therefore motivation) to develop sophisticated configurations will be lower especially as the transaction and coordination costs of these complex management systems are likely to be large. The final general proposition is therefore P3 Interactions between autonomy and networks in competitive city locations affect performance in a more pronounced way compared to other locations, and are likely to have a more significant effect on performance than the individual parts of the autonomy and network configuration.

The pathways in the configuration of autonomy and networking relationships linking to performance is shown in Figure 1. --------------------Figure 1 about here ---------------------

Data Gathering and Analysis To test the propositions that subsidiaries in Copenhagen have different configurations of autonomy and network relationships, a survey of foreign-owned subsidiaries located in the Denmark was undertaken provided the data. The design, administration, and procedures of the mail survey followed the main techniques recommended by Dillman (1991). These were 18

supplemented with specific steps informed by Harzing (2000) and Harzing and Noorderhaven (2006), which were designed to increase response rates.

The target audience for the survey was the managing directors of the subsidiaries. The sampling frame was constructed from information provided by the Experian database and yielded 2,996 firms covering all foreign owned subsidiaries in Denmark.

The survey

achieved a response rate of 15%. Tests for representativeness on industry characteristics indicated no significant differences; therefore it is unlikely that the survey has a problem of non-coverage error (Dillman, 1991). Non-response bias was tested using wave analysis, based on the observation that late respondents to mail surveys tend to be more similar to nonrespondents (Fowler, 1993). The comparison of early and late respondents on the variables of broad industry, age, entry mode, and nationality of managing director, did not reveal any significant differences in response.

The questionnaire collected data on the salient characteristics of the foreign-owned subsidiaries. Over half of all of the subsidiaries were German, Dutch or Swedish owned. The majority of subsidiaries employed less than 100 employees, and most managers were hostcountry nationals. Most subsidiaries had been under foreign ownership for more than 10 years. The majority of subsidiaries operate in high/medium-high and knowledge-intensive industries.

A partial least square (PLS) approach to structural equation modeling was used to test the proposition. A PLS modeling approach to subsidiary research has been adopted, among others, by Vernaik et al. (2005) and Fey, Morgulis-Yakushev, Park, and Björkman (2009). 19

This technique is preferable to the Lisrel and AMOS techniques when given a smaller sample. The PLS model operates with two sets of linear equations: an inner model that specifies relationships between latent variables, and an outer model analyzing relationships between the latent variables and associated manifest variables. This permits the simultaneous analysis of the path coefficients between latent variables, and the path coefficients between these variables and their constructs (Fey at al., 2009). This allows for an assessment of the reliability and validity of the measurement model, as well as an assessment of the structural model (Hulland, 1999). Furthermore, as Vernaik et al. (2005) argue because models and measures in international business are still in the initial stages of development, the regressionbased approach is more appropriate than covariance-based methods like LISREL. Finally, the PLS method is effective in guarding against inadequacies, such as skewed distributions of manifest variables, multi-collinearity within blocks of manifest variables and between latent variables, and omissions of data (Cassel, Hackl, & Westlund, 1999). Through the use of Smart PLS 2.0, t-statistics emerge through bootstrapping procedures. This technique makes the results more reliable, as it uses repeated random samples (Vernaik et al., 2005). Furthermore, total effects are calculated. This is advantageous, as it includes both direct and indirect effects (Sönke, 2010).

The model has four main constructs: ‘subsidiary autonomy’, ‘inter-organizational network relationships’, intra-organizational network relationships’, and ‘subsidiary performance’. Data for these constructs were gathered in relation to both the current period and five years ago. Previous work has shown that a five-year time span provides more accurate information than longer periods (Peng & York, 2001). A five-year period establishes a mathematically

20

explicit relationship between observed scores (or manifest variables) and latent variables (the change variable) (Borsboom, Mellenbergh, & van Heerden, 2003).

To capture rich data, the constructs used multiple questionnaire items that used five-point Likert scales. Respondents were asked to report the number and frequency of network relationships for a variety of their inter-organizational relations for the situation five years ago and for the present time. The latent variable then becomes an amalgamation of the changes in the number and frequency of the various inter-organizational network relationships. In PLS, each variable is assigned a weight – a coefficient – that reflects the importance of the manifest variable for the latent variable. The t-tests for the outer relations (manifest variables) indicate whether those coefficients (weights) are significant. The coefficients for the manifest variables are determined and the R-square for the inner relation maximizes the structural model. All constructs are based on self-reported information and therefore subsidiary performance uses subjective measures. This method is widely used in the literature and, in general, there is evidence of general reliability (Venkatraman & Ramanujam, 1986). There are however problems of measuring the performance of MNC operations related to the multi-faceted nature of performance (Miller, Lee, Chang, & Le Betron-Miller, 2009). There are also problems of collecting accurate, valid performance measures using questionnaires (Luo, 2007). Management decisions however are not guided solely by objective performance indicators, but are likely to depend to a large extent on the perceptions of managers, which although influenced by objectives measures, also contains elements of believe and opinions (Thompson, 2003). In addition, many objective indicators at the subsidiary level, especially 21

financial indicators, are suspect because of the reporting arrangements of MNCs (Guest, Michie, Conway, & Sheehan, 2003). The use of internal transfer prices and related tax issues also affects the validity of subsidiary turnover as a good measure of size (Demirbag et al., 2007). Given these reservations and following similar studies objective measures of performance were not used. Various studies employing subjective measures of performance ask respondents to assess performance in relation to their competitors (Ellis, 2007). This facilitates the comparison of establishments across size categories and industries. Subsidiary performance uses a five-item measurement encompassing items frequently and reliably used in other studies (Birkinshaw et al., 2005):

•

‘Sales growth by volume’,

•

‘Sales growth by value’,

•

‘Productivity’,

•

‘Customer satisfaction’, and

•

‘Market share’.

Respondents assessed each of these performance items relative to their market competitors on a scale of one (a lot better) to five (a lot worse). The constructs intra- and inter-organizational networks followed Holm and Pedersen (2000). These items measure the number and frequency of a subsidiary’s relationships with a range of partners. Intra-organizational partners included:

•

‘Buyers’, 22

•

‘Suppliers’,

•

‘R&D and innovation centers’, and

•

‘Other units within the MNC organization’.

Inter-organizational partners included:

•

‘Customers’,

•

‘Suppliers’, and

•

‘Competitors’.

•

Governmental institutions

•

Universities, Science centers

Both inter- and intra-organizational relationships were measured as the number of relationships on a scale ranging from one (none) to five (many), and as the frequency of contact with networks on a scale of one (low) to five (high). Measurement of autonomy followed Young and Tavares’s (2004) using strategic decision making (policy decisions) and operational decision making (tactical decisions). The measurement of strategic and operational decision-making authority uses approaches and measurement scales adapted from Birkinshaw and Hood (2000), and Taggart and Hood (1999). The items related to strategic decision making authority were: •

‘Market areas supplied’,

•

‘Product range’,

•

‘R&D and new product development’,

•

‘Production of goods or services’, 23

•

‘Financial control’, and

•

‘Human resource management’.

Areas of operational decision making were: •

‘Marketing activities’,

•

‘R&D and new product-development activities’,

•

‘Activities related to producing goods or services’,

•

‘Financial management practices’, and

•

‘Human resource management practices’.

For the strategic and operational decision making items, respondents were asked to assess the extent of their decision-making autonomy on a scale from one (exclusively by headquarters) to five (exclusively by the subsidiary). The control variables included in the model were: (a) home country, (b) size, (c) type of industry, and (d) entry mode (greenfield, acquisition). These types of control variables have been used in other PLS tests (Fey et al., 2009). Turnover is not used as a control for size because the sample includes service industries (including financial services firms) where size is measured by the value of assets. The study therefore used number of employees as the best control for size. Results Table 1 provides the composite variables reliabilities, Cronbach’s alpha values, and the R². These values indicate that the composite variables are robust. The results of the PLS test shown that there is only significant direct effects (path coefficients) on performance in Copenhagen from intra-organizational networks. There are no significant direct effects on 24

performance outside of Copenhagen (see Table2). This provides support, P1 but only for intra-organizational networks. There is no support for P2 as autonomy has no direct effect on performance either within or outside of Copenhagen.

Consideration of total effects (indirect effects to performance) however reveals that the configuration of autonomy and networks within Copenhagen displays strong connections between the different components of the configuration. Outside of Copenhagen these connections are considerably looser. All combinations of the variables are significant within Copenhagen, but only three total effects are significant outside of Copenhagen (see Table 2). Autonomy in combination with inter-organizational networks significantly affects performance in Copenhagen, and there are strong links between autonomy and both inter- and intra-organizational networks. Outside of Copenhagen fewer of the total effects are significant and in all cases the means of the total effect are smaller than those for within Copenhagen. The picture that emerges for Copenhagen is that the interactions within the configurations of autonomy and networks are more important for performance than is the case for other locations in Denmark. This provides support for P3.      ---------------------

Table 1 and Table 2 about here Discussion These findings add a sub-national (in particular a city) dimension to the relationships between performance and autonomy and network configurations found at national level (Andersson et al., 2005; Gammelgarrd et al., 2013; Lou, 2003). The results also add to the literature on MNC strategy in competitive city regions (Goerzen et al., 2013; McCann and Acs, 2011; 25

Zhao and Zhang, 2007) by providing evidence on the importance of autonomy and network relationships configurations and the link to performance in such locations. The findings provide evidence that the strategy of MNCs, in terms of autonomy and inter and intraorganizational configurations, is affected by location in a competitive city compared to other locations within the same host country. This contributes to the development of theory of MNCs occupying a complex spatial environment that involves consideration of both national and sub-national locations (Beugelsdijk and Mudambi, 2013).

The evidence about sub-national location in this study is not on the decision on where to locate within a particular country, or on how decisions are made between different possible sub-national locations with countries. The focus of these results is on the effects on management systems connected to the autonomy and network configuration of subsidiaries. The results highlight that interaction between these factors helps subsidiaries to benefit from the large reservoirs of high quality knowledge-based assets and effective networking infrastructures available in competitive city regions. This contributes to a better understanding of how MNCs take advantage of the greater ‘face to face’ and ‘economic buzz’ potential and agglomeration benefits available in competitive city regions (Asheim et al, 2007; Stroper, 2013; Storper and Venables, 2004;). These results therefore develop the literature that regards the proximity benefits available in competitive city regions as involving complex interactions and relationships between agents (Boschma, 2004 and 2005, Dicken and Malmberg, 2001; Eraydin and Fingleton, 2006; Yeung, 2005). The development to these theories is to focus attention on the importance of the autonomy and inter-and intraorganizational networks configurations of the subsidiaries of MNCs located in competitive cities. It seems that management systems in subsidiaries in competitive cities are more 26

complex than those not in such locations, involving embedding in both host locations and with the wider MNC that are woven together (involving autonomy at subsidiary level) to take advantage of the competitive benefits available in some cities. This adds a further dimension to Hymer’s view of the unequal development affect of the strategy of MNCs.

The finding of differences in subsidiaries in Copenhagen compared to locations elsewhere in Denmark highlights the importance of considering the importance of the hierarchies of cities for MNC strategies (Beaverstock et al, 2000; Taylor, 1997 & 2004). Copenhagen is not a significant ‘world city’ and therefore not the focus of the WCN view (Sassen, 2013). It is also not an important HUB in global supply chains in a variety of industries as in the GCC view (Gereffi and Korzeniewicz, 1994). It does however appear to fit with the GPN approach (Henderson et al, 2002) in that it provides competitive advantages, many not of major global significance, but more likely of national, and regional importance, and possibly in niche areas of innovation and R&D that are important in the global economy. Taking a WCN or GCC approach may well shut out a host of competitive city locations that has important implications for subsidiary strategies of MNCs. It seems that, at least in terms of the complexity of management systems, a process of unequal development is evident in Denmark, although Copenhagen is not a WCC or a significant GCC city. It is possible that MNC strategy has important implications for unequal development arising from the characteristics of cities that may stretch beyond a few major world cities.

This connects to the importance of examining the strategies and operations of MNCs as occupying a complex set of spatial environments that is likely to have important implications for how they locate their activities (Beugelsdijk and Mudambi, 2013). Te results of this study 27

indicate that the examination of these activities should include consideration of the autonomy and inter- and intra-organizational configuration. Most of the existing research by international business scholars currently centres on location decisions and neglects management systems implications. Economic geography scholars focus on networks relationships, but mainly on inter-organizational networks, with little attention to intraorganizational networks and autonomy, and with no clear emphasis on the interaction between these major components of the management systems of MNCs. The results of this study suggest that the configuration of autonomy and networks in the management systems of MNCs is an important factor in how location in competitive cities affects the strategies and activities of MNCs, and thereby makes an important contribution to the evolution of the global economy.

Conclusion Location in competitive cities by MNCs is clearly an important factor in the evolution of the global economy. The increased focus by international business scholars on sub-national location, including location in cities, is a welcome development to enable a better understanding of the influence of the complex spatial environment that MNC occupy for their strategies and management systems. The bulk of the research thus far has centred on strategy for location issues with the major management focus being on sub-national location to reduce liability of foreignness. The work of economic geographers on competitive cities has begun to influence the research agenda of international business scholars leading to more attention being paid to the importance of competitive cities in international business research. The influence of the autonomy and network configuration in the management systems of subsidiaries has however not played a major role in research on competitive cities, either by 28

economic geographers or international business scholars. The findings of this study suggest that this issue should be an important part of research on MNC location by both groups of scholars.

The differences in these configurations in Copenhagen compared to other locations in Denmark, suggest that management systems have subtle but important differences in competitive city regions from location elsewhere in the same country. The scare and desirable resource bundles and networking relationship benefits available in competitive cities leads to a more complex and interacting autonomy and network configuration than in other locations, leading to more sophisticated management systems in competitive cities. Subsidiaries located in such cities seem to be able to use interactions between autonomy and networks, including networks in the host location and with other parts of the MNC, to achieve performance benefits not available in other locations. Subsidiaries in competitive cities appear to have management systems that lead to more effective autonomy and network configurations that those located in other areas. This implies that not only are more high valued added and knowledge intensive activities centred in subsidiaries located in competitive cities, but they also have more complex autonomy and network configurations requiring higher quality resources to manage effectively these subsidiaries. It seems that competitive cities are on a win- in position whereby success in creating and sustaining attractive resource bundles and networking competencies motivates not only MNC location but also the development of sophisticated management systems that are likely to further the add to the resource bundles and networking competencies in such cities. In this type of a global economy the prospects for cities that have low global competiveness and in areas of low urbanisation may face a bleak future. 29

30

References Andersson U, Björkman I, Forsgren M, 2005, “Managing subsidiary knowledge creation: the effect of control mechanisms on subsidiary local embeddedness” International Business Review 14 521-538 Andersson,U., Forsgren, M., and Holm, U., 2001, Subsidiary Embeddedness and Competence Development in MNCs A Multi-Level Analysis, Organization Studies, 22, 1013-1034 Andersson U, Forsgren M, Holm U, 2002, “The strategic impact of external networks: subsidiary performance and competence development in the multinational corporation” Strategic Management Journal 23 979-996 Andersson,U., Forsgren, M., and Holm, U., 2007, “Balancing subsidiary influence in the federative MNC: a business network view”, Journal of International Business Studies 38, 802–818. Almeida, P. and Phene, A, 2004, “Subsidiaries and Knowledge Creation; The Influence of the MNC and Host country on Innovation, Strategic Management Journal, 25 847-864 Armstrong J, Overton T, 1977, “Estimating non-response bias in mail surveys”, Journal of Marketing Research 14 396-402 Asheim, B., Coenen, L., and Vang. J., 2007, "Face-to-face, buzz, and knowledge bases: sociospatial implications for learning, innovation, and innovation policy" Environment and Planning C: Government and Policy 25, 655 – 670 AT Kerney, 2014, Global Cities, Present and Future, http://www.atkearney.co.uk/documents/10192/4461492/Global+Cities+Present+and+FutureGCI+2014.pdf Baltagi B, 2003 Econometric Analysis of Panel Data 3rd Ed. (John Wiley, London) Bathelt H, Gräf A, 2008, “Internal and external dynamics of the Munich film and TV industry cluster, and limitations to future growth” Environment and Planning A 40 19441965 Beaverstock, J. V., Smith, R.G., & Taylor, P.J. 1999. A roster of world cities. Cities, 16(6), 445-458. Beaverstock, J. V., Smith R. G. and Taylor, P. J. 2000. “World-city network: a new metageography?” Annals of the Association of American Geographers 90: 123–134. Beugelsdijk, S. and Mudambi, R. 2013. MNCs as border-crossing multi-location enterprises: The role of discontinuities in geographic space, Journal of International Business Studies, 44(5): 413-416. Beugelsdijk, S., Mudambi, R., & McCann, P. 2010. Place, space and organization: Economic geography and the multinational enterprise. Journal of Economic Geography, 10(4): 485– 493. Birkinshaw J, Hood N, 2000, “Characteristics of foreign subsidiaries in industry clusters, Journal of International Business Studies” 31 141-154 Birkinshaw J, Hood N, Young S, 2005, “Subsidiary entrepreneurship, internal and external competitive forces, and subsidiary performance” International Business Review 14 227-248 Boschma R, 2004, “Competitiveness of regions from an evolutionary perspective” Regional Studies, 38 993–1006 Boschma R, 2005, “Proximity and Innovation: A Critical Assessment” Regional Studies 39 61–74 Burger, M.J., va der Knapp, B., Wall, R.S., 2013, “Revealed competition for greenfield investments between European region”, 13, Journal of Economic Geography, 619-648 31

Cantwell J, Santangelo G, 2002, “The new geography of corporate research in Information and Communications Technology (ICT)” Journal of Evolutionary Economics 12 163-197 Cantwell, J. A., & Mudambi, R. 2011. Physical attraction and the geography of knowledge sourcing in multinational enterprises. Global Strategy Journal, 1(3–4): 206–232. Chang S, Park S, 2005, “Types of firms generating network externalities and MNCs' colocation decisions” Strategy Management Journal 26 595-615 Chang S, van Witteloostuijn A, Eden, L, 2010, “From the Editors: Common Method Variance in International Business Research” Journal of International Business Studies 41(2) 178 – 184 Demirbag M, Tatoglu E, Glaiser K, 2007, “Factors influencing the perceptive performance: The case of Western FDI in an emerging market” International Business Review 16 310-36 Diez J, Bergerô M, 2005, “The role of multinational corporations in metropolitan innovation systems: empirical evidence from Europe and Southeast Asia” Environment and Planning A 371813-1835 Dicken, P., Kelly, P. F., Olds, K., & Yeung, H. 2001. Chains and networks, territories and scales: Towards a relational framework for analyzing the global economy. Global Networks, 1(2): 89–112. Dicken, P., & Malmberg, A. 2001. Firms in territories: A relational perspective. Economic Geography, 77(4): 345–363. Dimitratos, P., Liouka, I., and Young, S. 2009. Regional location of multinational corporation subsidiaries and economic development contribution: Evidence from the UK, Journal of World Business, 44(2): 180-191. Drejer I, Kristensen F, Laursen K, 1999, “Cluster studies as a basis for industrial policy: the case of Denmark” Industry and Innovation 6 171-190 Driffield N, 2006, “On the search for spillovers from foreign direct investment (FDI) with spatial dependency” Regional Studies 40 107-119 Driffiled, N., and Love. J.H., 2005. Intra-industry trade foreign direct investment, uneven development and Globalisation: The legacy of Stephen Hymer, Contribution to Political Economy, 24 55-78 Dunning J, 2009, “Location and the multinational enterprise: John Dunning’s thoughts on receiving the Journal of International Business Studies 2008 Decade Award” Journal of International Business Studies 40 20-34 Economist Intelligence Unit 2012. Hot spots – benchmarking global city competiveness. London, New York, Hong Kong & Geneva: The Economist. Eraydin A, Fingleton B, 2006, “Network relations and local economic development: some causes of differentiated network structures and intensities among Turkish industrial firms” Environment and Planning A 38 1171-1186 Eriksson R, Lindgren U, 2009, “Localized mobility clusters: impacts of labour market externalities on firm performance” Journal of Economic Geography 9 33-54 Florida R, 2005, Cities and the Creative Class, (Routledge, New York) Foss, N, 1997, Resources, Firms and Strategies: A Reader in the Resource-Based Perspective, (Oxford University Press, Oxford). Frost T, Birkinshaw J, Ensign P, 2002, “Centers of excellence in multinational corporations” Strategic Management Journal 23 997-1018 Gamelgaard, J. and Dörrenbächer, C, 2006, “Subsidiary Role Development: The Effect of Micro-Political Headquarters-Subsidiary Negotiations on the Product, Market, and ValueAdded Scope of Foreign Owned Subsidiaries”, Journal of International Management, 12 266-22012 32

Gamelgaard, J., McDonald, F., Tüselmann, H.J., Dörrenbächer, C. and Stephan, A (2012) “The impact of changes in subsidiary autonomy and network relationships on performance”, International Business Review, 20(5), 1158-1172 Glaister K, Buckley P, 1999, “Performance relationships in UK international alliances” Management International Review 39 123-47 Goerzen, A., Asmussen, C.G., Nielsen, B.B., 2013, “Global cities and multinational enterprise location strategy”, Journal of International Business Studies, 44, 427–450 Harrison J, 2007 “From competitive regions to competitive city-regions: a new orthodoxy, but some old mistakes” Journal of Economic Geography 7 311-33 Henderson, J., Dicken, P., Hess, M., Coe, N., & Yeung, H. 2002. Global production networks and the analysis of economic development. Review of International Political Economy, 9(3): 436–464. Holm U, Pedersen T, 2000, The Emergence and Impact of MNC Centres of Excellence – A Subsidiary Perspective (MacMillan Press, Basingstoke) Holm U, Malmberg A, Sölvell Ö, 2003, “Subsidiary impact on host-country economies – the case of foreign-owned subsidiaries attracting investment into Sweden” Journal of Economic Geography 3 389-408 Hospers G, 2003, “Creative cities-breeding places in the knowledge economy” Knowledge, Technology and Policy 16 143-162 Hymer, S. 1971, The multinational corporation and the law of uneven development, in Economics and World Order, ed by Bhagwati, J.W. (London: Macmillan) Jones A, 2007, “More than ‘managing across borders?’ the complex role of face-to-face interaction in globalizing law firms” Journal of Economic Geography 7 223-246 Kitson M, Martin R, Tyler P, 2004, “Regional competitiveness: an elusive yet key concept” Regional Studies 38 991–999 Lotz P, 1993, “Demand as a driving force in medical innovation” International Journal of Technology Assessment in Health Care 9 174-188 Luo Y, 2003, “Market-seeking MNEs in an emerging market: how parent-subsidiary links shape overseas success” Journal of International Business Studies, 34 290-309 Malecki, E. J., 2007, "Cities and regions competing in the global economy: knowledge and local development policies" Environment and Planning C: Government and Policy 25, 638 – 654 Mans, U., 2013, “Revisting city connectivity”, Journal of Economic Geography, Advanced published January 30, 2013.   Maskell P, 1986, Industriens Flugt fra Storbyen: Årsager og Konsekvenser. Copenhagen: Handelshøjskolens Forlag McCann, P. 2011. International business and economic geography: Knowledge, time and transaction costs. Journal of Economic Geography, 11(2): 309–317. McCann, P., and Acs, Z.J., 2011, “Globalization: Countries, Cities and Multinationals”, Regional Studies, 45, 17-32 McCann P, Mudambi R, 2004, “The location and behavior of the multinational enterprise: some analytical issues” Growth and Change 35 491–524 McCann P, Mudambi R, 2005, “Analytical differences in the economics of geography: the case of the multinational firm” Environment and Planning A 37, 1857- 1876 Mu S, Gnyawali D, Hatfield D, 2007, “Foreign subsidiaries’ learning from local environments: an empirical test” Management International Review 47 79-102 Mudambi R, 2008, “Location, control and innovation in knowledge-intensive industries” Journal of Economic Geography 8 699-726 33

Mudambi, R. and Navarra, P, 2004, “Is knowledge power? Knowledge flows, subsidiary power and rent-seeking within MNCs”, Journal of International Business Studies,35, 385406. Musil R, 2009, “Global capital control and city hierarchies: an attempt to reposition Vienna in a world city network” Cities 26 255-265 Meyer, K.E., and Nguyen, H.V. 2005. Foreign Investment Strategies and Sub-national Institutions in Emerging Markets: Evidence from Vietnam, Journal of Management Studies, 42(1): 63-93. Meyer, K.E., Mudambi, R. and Narula, R. 2011. Multinational enterprises and local contexts: the opportunities and challenges of multiple-embeddedness. Journal  of  Management  Studies,   48(2):235-­‐252.    

Nachum, L., and Wymbs, C., 2007, “The Location and Performance of Foreign Affiliates in Global Cities”, in Alan M. Rugman (ed.) Regional Aspects of Multinationality and Performance (Research in Global Strategic Management, Volume 13), Emerald Group Publishing Limited, pp.221-259 Nielsen P, Hansen C, Nielsen M, 2009, Udenlandske Firmaer i Danmark (Statistics Denmark, Copenhagen)Oerlemans L, Meeus M, 2005, “Do Organizational and Spatial Proximity Impact on Firm Performance?” Regional Studies 39 89–104 Nguyen, T. V., Le, N.T.B., and Bryant, S.E. 2013. Sub-national institutions, firm strategies, and firm performance: A multilevel study of private manufacturing firms in Vietnam, Journal of World Business, 48(10): 68-76. OECD 2006, Competitive Cities in the Global Economy (OECD, Paris) Pain, K. 2012. Spatial transformation of cities: Global city-region? Mega-city region? In B. Derudder, M. Hoyler, P.J. Taylor & F. Witlox (Eds), International Handbook of Globalization and World Cities: 83-93. Cheltenham; Northampton, MA: Edward Elgar. Parr, J, (2008), “Cities and regions: problems and potentials Environment and Planning A 40 3009-3026 Peng, M.W., Wang, D.Y.L., Jiang, Y. (2008) An institution-based view of international business strategy: a focus on emerging economies, Journal of International Business Studies, 39(5): 920-936. Peng M.W., Sun, S/L., Pinkham, B., and Chen, H. (2009) The institution-based view as a third leg for a strategy tripod, Academy of Management Perspective, 23(4): 63-81. Peng M, York A, 2001, “Behind intermediary performance in export trade: Transactions, agents, and resources” Journal of International Business Studies 32 327–46 Podsakoff P, Organ D, 1986, “Reports in organizational research: problems and prospects” Journal of Management 12 531-45 Sassen S, 1994, Cities in a World Economy (Pine Forge, Thousand Oaks, CA) Sassen, S. 2013. Cities in a world economy, 4th ed. Thousand Oaks, CA: Pine Forge/Sage. Santangelo, G. D. 2009. MNCs and linkages creation: Evidence from a peripheral area. Journal of World Business, 44(2):192-205. Schmid, S., and Schurig, A., 2003, “The development of critical capabilities in foreign subsidiaries: disentangling the role of the subsidiary’s business network”, International Business Review, 12, 755-782 Scott A,J. 2001, Global City-regions: Trends, Theory & Policy (Oxford University Press, Oxford) Scott, A.J. 2006, Creative Cities: Conceptual Issues and Policy Questions, Journal of Urban Affairs, 28(10), 1-17 34

Scott, A.J. 2012. A World in Emergence: Cities and Regions in the 21st Century, Cheltenham; Northampton, MA: Edward Elgar. Scott A, Storper M, 2003, ``Regions, globalization, development'' Regional Studies 37 579594 Shaver J, Flyer F, 2000, “Agglomeration economics, firm heterogeneity, and foreign direct investment in the United States” Strategic Management Journal 21 1175-1193 Simmie J, Sennett J, Wood P, Hart D, 2002, “Innovation in Europe; A Tale of Networks, Knowledge and Trade in Five Cities” Regional Studies 36 47-64 Steinfield C, Scupola A, 2008, “Understanding the role of ICT networks in a biotechnology cluster: an exploratory study of Medicon Valley” The Information Society 24 319-333 Storper, M. 2013 Keys to the City: How Economics, Institutions, Social Interaction, and Politics Shape Development, New Jersey: Princeton University Press. Storper M, Venables A, 2004, “Buzz: face-to-face contact and the urban economy” Journal of Economic Geography 4 351-376 Taggart J, 1998, “Strategy shifts in MNC subsidiaries” Strategic Management Journal 19 663-681 Taylor, P. J. 1997, Hierarchical tendencies amongst world cities: a global research proposal” Cities 14: 323–332 Taylor, P. J. 2004 World City Network: A Global Urban Analysis London: Routledge. Teece, D, 1986, “Profiting from Technological Innovation”, Research Policy, 15 285-305 Teece, D, 1998, “Capturing value from knowledge assets: the new economy, markets for know-how and intangible assets”, California Management Review, 40 55-79. Teece, D, 2000, “Strategies for Managing Knowledge Assets: the Role of Firm Structure and Industrial Context”, Long Range Planning, 33 35-54 Teece, D.J., Pisano, G., and Shuen, A. 1997, Dynamic Capabilities and Strategic Management, Strategic Management Journal, 18(7), 509-533 Timmermanns, B., and Boschma, R., 2013, “The effect of intra- and inter-regional labour mobility on plant perforamcne in Denmark: the significance of related labour inflows”, Journal of Economic Geography, Advance Access, January 24, 2013 Thompson E, 2003, “A grounded approach to identifying national competitive advantage: a preliminary exploration” Environment and Planning A 35 631 – 657 Turok I, 2004, “Cities, regions and competitiveness” Regional Studies 38 1069-1083 Venkatraman N, Ramanujam V, 1986, "Measurement of Business Performance in Strategy Research: A Comparison of Approaches" Academy of Management Review 11(4) 801 – 814 Venaik, S,. Midgley, D.F,. and Devinney, T.M., 2005, “Dual paths to performance: the impact of global pressures on MNC subsidiary conduct and performance”, Journal of International Business Studies, 36, 655–675 Ward K, Jonas A, 2004, “Competitive city-regionalism as a politics of space: a critical reinterpretation of the new regionalism” Environment and Planning A 36 2119- 2139 Winther L, Hansen H, 2006, “The economic geographies of the outer city: industrial dynamics and imaginary spaces of location in Copenhagen” European Planning Studies 14 1387-1406 Wu, J and Radbone, I (2005) Global integration and the intra-urban determinants of foreign direct investment in Shanghai. Cities 22 (4), 275-286. Yeung H, 2000, “Embedding foreign affiliates in transnational business networks: the case of Hong Kong firms in Southeast Asia” Environment and Planning A 32 201- 222 Yeung, H. 2005. Rethinking relational economic geography. Transactions of the Institute of British Geographers, 30(1): 37–51. 35

Young S, Tavares A, 2004, “Centralization and Autonomy: Back to the Future” International Business Review 13 215-237 Zaheer, S., 1995, “Overcoming the liability of foreignness”, Academy of Management Journal, 38, 341-363 Zhao S, Zhang L, 2007, “Foreign Direct Investment and the Formation of City Regions in China” Regional Studies 41 979-994

Figure 1: Conceptual Model

Increases in number of relationships within corporation

Increases in frequency of relationships within corporation

Intra-org

Increases in strategic decision-making rights

Networks

Performan

Autonomy

ce

Increases in operational decision-making rights

Improves in performance (compared to market competitors)

Inter-org Networks

Increases in number of relationships outside corporation

Increases in frequency of relationships outside corporation

36

Table 1: Composite Reliabilities, Cronbach’s Alphas, and R-squareds

Within Copenhagen/outside Copenhagen* Composite

Cronbach’s

R-squareds

Reliability

Alphas

Autonomy

0.94/0.94*

0.93/0.93*

-

Inter-organizational networks

0.89/0.84*

0.85/0.80*

0.18/0.00*

Intra-organizational networks

0.88/0.84*

0.85/0.81*

0.56/0.41*

Performance

0.86/0.89*

0.79/0.84*

0.37/0.20*

Note: Composite should be above 0.70 for each construct (Fornell & Larcker, 1981). Cronbach’s alpha values should be above 0.70 (Hulland, 1989). When using the PLS technique, one variable is ‘locked’ and R² are reported in relation to this variable.

37

Table 2: PLS Test Within Copenhagen/ Outside Copenhagen

Inter-organizational

Path Coefficient Mean

Total Effect Mean

t-statistics Total Effects

- 0.00/0.13

0.29/0.25

2.40***/

network relationships AND

1.26

performance Intra-organizational

0.52/0.24

0.52/0.24

network relationships AND

3.39***/ 1.13

performance

Inter-organizational

0.56/0.51

0.56/0.51

network relationships AND

6.16***/ 3.16***

intra-organizational network relationships Autonomy

AND

0.15/0.23

0.44/0.33

performance

Autonomy

3.25***/ 1.89*

AND

organizational

inter-

0.42/0.08

0.42/0.08

network

2.85***/ 0.12

relationships

Autonomy

AND

organizational

intra-

0.30/0.41

network

0.54/0.44

3.61***/ 2.39***

relationships

t-values in parentheses are without controls. *p