Introduction

Globalisation has been a key topic in recent decades with many scholars, politicians and economists predicting that the world would become increasingly integrated on a global basis (cf. Friedman 2005, Ghemawat 2005, Economist 1997). Nonetheless, a renewed interest in the phenomenon of regionalisation and increasing scepticism about globalisation can be observed in the international business (IB) literature (cf. Rugman and Hodgetts 2001, Dunning et al. 2007). In fact, the world is not as global as previously believed. For example, many multinational corporations (MNCs) operate regionally rather than globally. Prominent studies by Alan Rugman and his peers (cf. Rugman and Hodgetts 2001, Dunning et al. 2007, Flores and Aguilera 2007, Rugman and Verbeke 2004) largely build on the argument that firms’ operations are not scalable beyond regional boundaries. These studies show that MNCs manage their businesses regionally. For example, more than 70 per cent of large European MNCs’ sales and assets are concentrated in Europe (Oh and Rugman 2012). Other scholars, who postulate that MNCs seek a form of ‘semi-globalization’ (Ghemawat 2003), propose an intermediate version of MNCs’ global integration in which global and national strategic advantages are balanced (Yip 1989). Consequently, the call for more research into the more common type of MNCs – home-region-oriented MNCs – has become louder (Sammartino and Osegowitsch 2013).

At the same time, the topic of organisational structures and management has become popular as a potential means for firms to cope with the complexities of global business. Organisational mid-layers, such as regional headquarters (RHQ) (Lasserre 1996, Nell et al. 2011) or regional management centres (Enright 2005), have been proposed as a better way to manage units in host regions. Regional management allows for the development of regional strategies (Schütte 1997), helps in the sensing of local business opportunities and innovation (Hoenen et al. 2014) and reduces management complexity by dividing authority between MNCs’ headquarters and their subsidiaries (Paik and Sohn 2004).

These two literature streams address the issue of why regionalisation occurs. The first stream argues that MNCs are region-bound due to the limited scalability of their operations, while the second regards regional management as a way to better manage a global organisation. While the two streams have been kept separate thus far, we draw on corporate strategy literature to argue that they are linked. We show that the tendency to regionalise reflects the interplay between the boundaries of a firm’s scope and its capabilities to manage it. As Europe – especially the European Union (EU) – represents one of the most advanced forms of integration of regional organisations (Johnson and Turner 2006, p. 60), we use it as an exemplary region in this chapter.

A corporate strategy perspective on regionalisation

Organisations constantly face corporate-level issues, which can be grouped into the categories of scope and value creation (Johnson et al. 2008). On the one hand, firms need to decide which products to sell and where to sell them. On the other hand, they need to determine how to create maximum value, which parenting role to assume and how to manage their business portfolios. Issues of scope pertain to a firm’s products and its international diversity. Issues of value creation are concerned with the role of headquarters and the management of dispersed business operations, which are essentially questions of organisational structure. Consequently, a regional focus among MNCs can be motivated by two factors – scope and structure. The former refers to what and where to sell, while the latter pertains to value creation through organisational structure (Figure 3.1).

Figure 3.1   Framework corporate-level issues related to regionalisation

Figure 3.1 summarises the two distinct but complementary streams of literature that have evolved around regionalisation. The first is concerned with MNCs operating regionally and the second focuses on MNCs organising regionally. As such, the first stream touches on the question of scope and analyses MNCs’ tendencies to regionally limit their operations. The second stream centres on the question of structure and focuses on organisational mid-layers, such as RHQs, as additional regionally focused management and control bodies.

The magnitude of regionalisation

Scholars have long shown an interest in the magnitude of globalisation and regionalisation, and the scope of MNCs’ operations. Ohmae (1987, p. 17) postulated that global competition occurred in a tetrahedral world in which 85 to 90 per cent of all value-added, high-technology products were manufactured and consumed in only three regions – the United States, Europe and Japan. Research by Michalak and Gibb (1997, p. 266) supported these results and showed that intra-regional trade accounted for 38 per cent of merchandise imports and exports, while inter-regional trade represented only 10 per cent. The study by Rugman and Hodgetts (2001) of trade data from 1997 also provided strong evidence for triad-shaped world trade. Rugman (2005, p. 11) further revealed that 84.2 per cent of firms in his sample made more than half of their sales in their home region of the triad (Europe, North America, Asia). As such, Rugman (2005) was the first to provide evidence of MNCs’ home-region orientation and to illustrate the limits of the scalability of a firm’s operations and, thereby, its scope. Only 6.6 percent of the companies in his sample generated less than 50 per cent of sales in their home region and at least 20 per cent in two regions (including the home region), which made them bi-regional (Rugman 2005, pp. 12–13). The author characterised only nine of these firms as truly global, thereby refuting the prominent assumption that MNCs generally operate on a global basis. In general, firms prefer to internationalise within their home regions before moving into new territories (UNCTAD 2007). Firms’ assets also show patterns of regionalisation as around 80 per cent of MNCs’ assets are deployed in their home region (Rugman and Verbeke 2008, Rugman and Oh 2012), confirming that most MNCs are not global but regional in scope. Large European MNCs are mostly home-regional, as less than 30 per cent of their sales and assets are outside Europe (Oh and Rugman 2012, p. 495). Syed and Colleen (2011) highlighted an increasing trend towards regionalisation and reported a significant increase in intra-regional trade within Europe. Intra-regional exports in the EU increased by 163 per cent from 1,910 billion euros in 2002 to 3,115 billion euros in 2016 (Eurostat 2017).

Banalieva and Dhanaraj (2013) went in a different direction. These authors analysed which MNCs were the most prone to being home-region oriented, and found that European MNCs were more home-region oriented than MNCs from the USA or Japan. The authors suggested that this finding was the result of comparably low institutional diversity in Europe, which can be attributed to the presence of the EU – the most advanced level of regional integration in the world (Blevins et al. 2016).

All of these studies aim to improve our understanding of MNC scope, which is a key issue in strategy research (Johnson et al. 2008). Whether there is a relationship between multinationality and performance has been intensively discussed (Cardinal et al. 2011), but the results are inconclusive (Verbeke and Forootan 2012). A variety of multinationality-performance relationships, including positive, negative, curvilinear, S-shaped, M-shaped and W-shaped, have been proposed (Powell 2014). Despite their divergence, all of these studies view multinationality as a continuum ranging from no multinationality to global coverage, thereby ignoring limits to multinationality arising from regional confines. The scope literature on regionalisation opposes this view, as it regards multinationality as a set of stages in which each stage relates to a certain region. Essentially, a firm’s spread is argued to be region-bound and its scope is limited by regional boundaries.

What defines a region?

A region is a collective term referring to neighbouring countries or markets. As such, a regional orientation reflects a firm’s deliberate choice to enter close, attractive markets in order to strengthen its competitive position (Sammartino and Osegowitsch 2013). Regional clusters draw homogeneous countries closer together (Schlie and Yip 2000). The meaning of the term ‘closer’ depends on the chosen distance dimension(s): geographical, administrative, economic and cultural (Ghemawat 2001). A region combines markets with low distance among them (Sammartino and Osegowitsch 2013) based on one or more distance dimensions.

Aguilera et al. (2007, pp. 8–9) argue that ‘physical immediacy is a precondition for a sense of unity or shared properties’. The most prominent geography-based regional classification pattern is the triad-concept introduced by Ohmae (1985), which was initially composed of the United States, Europe and Japan, and later extended to a ‘broad triad’ including North America, Europe and Asia Pacific (Rugman and Oh 2012). Time zones are another aspect of geographical distance often used by firms to efficiently organise their operations (e.g. facilitate correspondence (Nan et al. 2009) or allow for closer monitoring (Elango 2004)). Regional classification patterns that minimise administrative and economic distance are often manifested in regional trade agreements (RTAs) (e.g. EU, NAFTA, ASEAN) (Fratianni and Oh 2009) that minimise institutional diversity. Notably, MNCs strive for maximum uniformity in their institutional environments (Banalieva and Dhanaraj 2013) in order to facilitate administrative work, increase efficiency, support regional scalability and simplify organisational control. Along these lines, RTAs lead to the formation of regional economic areas that bring about economic proximity, tax and familiarity benefits; reduce tariff and non-tariff barriers; and facilitate governmental and business coordination (Hejazi 2007). This regional focus, or regionalism (Suder 2015), often gives rise to a trade bias for RTA-based regions (Fratianni and Oh 2009). It is important to remark, however, that RTAs do not necessarily require geographic proximity (Suder 2015), although that is often the case. Rather, an alignment in institutional matters is the key focus of RTAs, which allows them to span large geographical distances and facilitate trade among geographically distant countries (Suder 2015). The EU, which is the most far-reaching attempt at regional integration among independent nations in modern times, includes economic, social and political elements (Kolk et al. 2014). The EU not only encompasses a common European market, including uniform tariffs and trade regulations, but also a common governmental body (the European Commission), a political agenda and a common currency (at least in part) (Blevins et al. 2016), which further reduce perceived distance (Verbeke and Asmussen 2016).

Cultural clusters that minimise (perceived) cultural distances are another way to define a region (Banalieva and Dhanaraj 2013). They often share the same or similar languages as language is an antecedent of national culture (Ronen and Shenkar 2013). Dunning et al. (2007) confirmed the existence of culturally induced regions in a study on foreign direct investment.

There is little agreement on what determines a region or how it can best be operationalised (Dunning et al. 2007). We understand ‘region’ as a spatial concept that combines proximate markets in terms of geographical, administrative, economic or cultural distance, where geography is the most prominent determinant (Banalieva and Dhanaraj 2013). When they decide on their region(s) of operation, firms choose their spatial scope and a corresponding set of boundary conditions (Verbeke and Asmussen 2016). Therefore, the perceived compound distance across distance dimensions within the region is lower than the distance to any area outside the region’s border. In other words, a region’s ‘outsiders’ are characterised by a greater distance relative to a region’s ‘insiders’ (Verbeke and Asmussen 2016). On the basis of salience theory, which suggests that the choice of a subject or referent partly depends on the relative salience of the objects (Tversky 1977), we challenge this view. Salience theory predicts firms to choose their regions of operations based on a few firm-specific requirements that are salient compared to others. This argument is supported by distinctiveness theory, which asserts that individuals focus on aspects that are more peculiar than others (McGuire and Padawer-Singer 1976). The degree of peculiarity then influences the relative importance of certain features (Mehra et al. 1998).

Observing firms with operations in Europe, this idea is confirmed. European-based MNCs are seen as the strongest inter-organisational regional networks (Cantwell and Janne 1999). They prefer to internationalise their R&D activities in this region due to its attractiveness in terms of knowledge base, industry-specific and cluster-based spillovers, and technological specialisation (Cantwell and Piscitello 2002). Cantwell and Piscitello (2002) assert an increased regional focus based on technological advancement. Along these lines, due to a favourable regulatory framework within the EU, automobile supply chains are clustered intra-regionally (Rugman and Oh 2012). Service activities are also more local than global (Rugman and Oh 2012) and are likely to be clustered regionally. Within Europe, service firms tend to group their operations according to language. For example, Bain and Company, in parts, internally combines the markets of Austria, Germany and Switzerland, which all share German as the dominant language. In such situations, a common language appears as the most salient factor for setting operational boundaries.

In this respect, changes in the regulatory environment and language barriers appear to be natural fault lines that induce the splitting of groups (Lau and Murnighan 2005). In other words, one or more distinguishing factors align in their uniqueness and form distinct schisms (Li and Hambrick 2005) that then determine the boundaries of a subgroup. Therefore, while the European continent itself offers a natural regional definition, firms select certain factors that they find most salient given their needs and then determine their operational space accordingly. Thus, regional fault lines depend on the characteristics of an individual firm and its operational requirements (e.g. technological know-how, regulatory environment, common language). Rather than being determined by firms, a region can be seen as an induced ‘space’ limited by schisms in the proximity of firm-specific, salient, regional denominators. These denominators form the same regional outer boundary. Therefore, regions do not always form based on a single factor but can be the result of the interplay among similar specificities and a number of factors.

Why does regionalisation occur?

Multinational corporations are encouraged to ‘design strategies and adopt structures that focus on markets close to their countries-of-origin’ (Goerzen and Asmussen 2007, p. 66). Although MNCs often do so, they may also operate in host regions. In the European region, it is important to distinguish between European home-region MNCs and European host-region MNCs. Given that firms face corporate-level issues that are concerned with either the firm’s scope (i.e. strategy) or structure (Figure 3.1), we aim to explain why many firms confine their operations to Europe or smaller regions therein.

Structure

The second stream explaining the phenomenon of regionalisation bases its arguments on aspects of organisational structure. Structure-related corporate-level issues include decisions regarding the role of headquarters and how to manage the portfolio of subsidiaries in order to add maximum value to subsidiaries (Figure 3.1). The parenting literature suggests that headquarters’ main purpose is to add value to the MNC’s portfolio of subsidiaries (Nell and Ambos 2013) – in other words, to settle on a particular parenting strategy (Goold 1996). Essential to the right parenting strategy is the fit between the parent’s characteristics and the subsidiary’s parenting needs, as well as the parent’s familiarity with the subsidiaries (Goold et al. 1994). A suitable organisational structure is therefore crucial. In this respect, hierarchical structures, especially RHQs, are undergoing a revival (Nell et al. 2011) (e.g. the number of European RHQs increased by 76 per cent from 2000 to 2010 (Ambos and Schlegelmilch 2010)).

RHQs are relatively autonomous units (Enright 2005) responsible for managing a firm’s operations in a host region (Yeung et al. 2001). As such, RHQs are a phenomenon limited to MNCs’ host regions. As intermediaries between headquarters and subsidiaries, RHQs face the constant challenge of responding to both parties. Two main control relationships result – one between headquarters and RHQs and one between RHQs and the regional subsidiaries (see Figures 3.2 and 3.3). Thus, an RHQ limits a headquarters’ sphere of control by taking on the task of subsidiary oversight.

Figure 3.2   Organisation of MNC without regional structures/RHQs

Figure 3.3   Organisation of MNC with regional structures/RHQs (adapted from Johnson et al. (2008))

RHQs are believed to have intra-regional and inter-regional effects. As Yeung et al. (2001) observe, RHQs have a mandate to exercise control over subsidiaries that require local management despite assumed regional market homogeneity. As MNCs need to simultaneously be locally responsive and globally integrated in order to benefit from global efficiencies (Doz and Prahalad 1984), management from a distance may be inappropriate. An RHQ is a control unit that manages the subsidiaries in its realm (Yeung et al. 2001), balances the conflicting demands of globalisation and localisation as an intermediary unit (Paik and Sohn 2004) and makes simultaneous exploitation possible (Schlie and Yip 2000). RHQs serve as ‘two-way conduits of influence’ (Kriger and Rich 1987, p. 45) or ‘strategic mid-way houses’ (Yeung et al. 2001, p. 165) with the purpose of implementing a firm’s global strategies on a regional level. Regionalisation can thus be seen as ‘a half-way house that meets these balancing and often conflicting pressures’ (Khan 2010, p. 28).

In Europe, RHQs were introduced by many MNCs to manage operations and carry out headquarters’ functions in Central and Eastern Europe (CEE) after the fall of the Iron Curtain (Schuh 2013). The main benefits of setting up an RHQ in CEE were regional knowledge bundling, adaptation of global strategies to a regional level, headquarters’ complexity reduction and synergy creation (Schuh 2013), as well as RHQs’ positive nurturing effect for core competencies (Filippaios and Rama 2008). RHQs enable MNCs to leverage synergies across countries in a host region (Werth 2006). Moreover, they serve as a coordinating body for host-region operations that bundles supporting services for subsidiaries, thereby reducing costs (Mori 2002). They also support knowledge transfer between headquarters and subsidiaries (Asakawa and Lehrer 2003). Overall, RHQs are an organisational solution for maximising the value added to subsidiaries. While headquarters are restricted in their parenting resources, RHQs can execute the headquarters’ role on a regional level. As such, they can play an important part in coordinating GVCs with internationally dispersed, but regionally focused value-chain activities that are inherent to a single MNC.

Ghemawat (2005) critically reflects on the role of RHQs and challenges the view that RHQs are the enabling unit in regional strategies. While RHQs can help in the realisation of regional strategies, it is far more essential for MNCs to determine how they plan to add value to their subsidiaries in a region. Regional organisational structures may be useful, but a fit between the parent’s intention to add value and the roles and capabilities of the RHQ is essential.

In sum, this literature stream views regionalisation as motivated by the desire to add maximum value to subsidiaries by means of regional organisational structures. RHQs provide MNCs with a means to limit headquarters’ span of control and manage a region from a distance (Yeung et al. 2001). Furthermore, RHQs are host-region bound – they are established because they better understand the local environment (Lasserre 1996) and, therefore, allow for regional synergy realisation. However, as Ghemawat (2005) stresses, RHQs should not necessarily be regarded as the best method for implementing regional strategies. Rather, they are one of several alternatives.

Discussion – a holistic view on regionalisation

MNC regions are defined in terms of regional denominators, such that markets within a region are perceived as closer than markets outside that region. The firm-specific saliency of certain regional denominators is likely to induce fault lines that determine regional boundaries. The formation of regions is explained in two streams of the IB literature centring on arguments of scope and structure. We posit that an integration of not only these two streams, but also other views on what defines a region would help answer the question of why regionalisation occurs, and lead to a holistic understanding of the phenomenon of regionalisation and its contingencies. We have identified two areas in which we see considerable potential benefits from such an integration.

First, a firm’s scope and structure are interrelated. Whether firms regionalise and to which degree is dependent on two main corporate-level issues (see Figure 3.1) – decisions regarding scope (i.e. which products to sell where) and decisions regarding structure (i.e. which parenting role to choose and how to manage the portfolio of subsidiaries). Given the principle of ‘structure follows strategy’ (Chandler 1962, p. 314), the chosen organisational structure (i.e. headquarters’ decision on its parenting role/how to manage the portfolio of subsidiaries) must fit the chosen strategy as reflected in the choice of firm scope. Therefore, both perspectives must be taken into account. A firm’s regional choice reflects a compromise between the maximum feasible scope from a strategic perspective and the maximum complexity that an organisation can handle.

Second, the phenomenon of regionalisation is evident in both home and host regions. Even though the majority of MNCs are predominantly oriented towards their home regions, bi-regionally oriented firms do exist (Rugman 2005). In this respect, an analysis of the phenomenon from both the home- and host-region perspectives is beneficial. Arguments that certain firm-specific advantages do not qualify for redeployment beyond regional confines may also apply to advantages associated with host regions. As in a home region, host-region countries can be assumed to be homogeneous (e.g. the European market as a host region consists of proximate countries on a variety of distance dimensions, such as regulatory environment or geography). When explaining home-region orientation, ignoring host-country scope-related issues in the debate about RHQs or arguments about organisational structure and control can lead to certain biases. Clearly, both home- and host-region orientations exist. They should therefore be analysed jointly rather than in isolation.

Conclusion

In drawing from corporate strategy, this chapter offers a better understanding of the phenomenon of regionalisation by discussing and challenging popular perspectives on the topic found in the IB literature. While regionalisation seems to be well explained on an abstract level, these explanations leave significant room for clarification. We are forced to acknowledge that the extant research does not offer a clear answer as to why regionalisation occurs. Therefore, we call for more research on this topic.