Assessing the performance consequences of multinationality is a key topic in the International Business (IB) literature (see Nguyen 2017 for a recent review). However, there are still 'more questions than answers' (Glaum and Oesterle 2007). Firms that venture beyond domestic borders can capitalize on their firm-specific assets (e.g., Buckley and Casson 1976; Caves 1971; Goerzen and Beamish 2003); access new resources (Benito 2015; Cuervo-Cazurra et al. 2015); achieve economies of scale and scope (e.g., Contractor et al. 2003; Hennart 2007; Miller et al. 2016; Tallman and Li 1996); and exploit arbitrage opportunities (Allen and Pantzalis 1996; Ghemawat 2007; Kogut 1985). Nonetheless, the profitability of multinational firms can be seriously compromised by the so-called liability of foreignness (LoF); that is, the additional costs in which multinationals incur because of the disadvantageous position that they hold in host markets compared to domestic incumbents (Hymer 1976; Zaheer 1995).
The literature on the LoF has acknowledged its multifaceted nature (Cuervo-Cazurra et al. 2007; Eden and Miller 2001). As such, some scholars have tried to refine our understanding of the concept by unpacking its triggers. For instance, Eden and Miller (2004) argue that the LoF arises from the unfamiliarity, relational and discriminatory hazards that multinationals are bound to face in foreign countries due to factors such as cultural and regulatory differences. Rugman and Verbeke (2007) go one step further and claim that costs of operating abroad are even higher when multinationals expand to another region, giving rise to the liability of interregional foreignness (LoIF). This additional obstacle may be the reason why some multinationals prefer to constrain their geographic reach to their home region at the expense of achieving global benefits, as the regionalization literature suggests (e.g., Oh and Li 2015; Rugman and Verbeke 2004). However, a clear-cut distinction among regions obviates the potential co-existence of similarities and differences within and across them (Dunning et al. 2007). Integrating the different conceptualizations of the LoF, Asmussen and Goerzen (2013) identified three different ramifications: cultural, institutional, and regional. They argued that the main costs that multinationals face when expanding abroad vary according to the dispersion of their locations across cultures, institutional settings, and regions. However, a comparative analysis of the impact of each of these dimensions of international dispersion on firm profitability has not been carried out so far.
We contribute to fill this gap not only by analyzing how each of these dimensions affects firm profitability, but also by examining the extent to which cross-industry differences moderate these relationships. Zhou and Guillen (2016) found that firms choose the type of foreign direct investments (FDI) that they undertake based on different country distance dimensions to minimize the impact of the LoF. This result suggests that some heterogeneity might exist in the relationship between the LoF and profitability at the firm or industry levels. Therefore, new research on the topic should determine the circumstances under which firms and/or industries may be less affected by the LoF. For this reason, we aim to answer the following research question: How does international dispersion affect the profitability of infrastructure firms (1) compared to that of their non-infrastructure counterparts? Although previous studies propose that infrastructure firms have an ability to navigate through different institutional contexts (Bonardi 2004; Guillen and Garcia-Canal 2010; Henisz 2003), the performance implications of their internationalization decisions have not been analyzed yet. We claim that these industries are less affected by the cultural, regulatory and regional dimensions of the LoF because for them (1) the importance of cultural fit in products is low; (2) their regulatory expertise facilitates navigating through different regulatory environments; and (3) there are limited aggregation opportunities at the regional level. We explain each of these points in more detail in subsequent sections.
Adopting an institution-based view angle (e.g., Peng 2002; Peng et al. 2008, 2009), we hypothesize that the profitability of multinationals decreases, on average, as their operations become more internationally dispersed. We also expect this relationship to be flatter for infrastructure multinationals. We test our hypotheses on a panel-data sample of Spanish listed firms (1986-2007). Our results reveal that higher degrees of international dispersion diminish profitability. By contrast, operating in infrastructure industries buffers the negative consequences of that higher dispersion, thus validating the arguments in favor of a flatter relationship between both variables. In fact, our findings show that infrastructure multinationals are even able to profit from venturing outside the boundaries of their home region. Therefore, it seems that these multinationals are better prepared to operate in different countries and regions.
We add to the existing literature in a variety of ways. First, we shed more light on the relationship between internationalization and performance by analyzing the multifaceted impact of the LoF on profitability. Second, we extend the institution-based view to make an initial attempt at empirically testing the factors that make some multinationals better able to mitigate the negative effects of managing their international footprint across borders. We do so by focusing on the different profitability consequences of international dispersion in infrastructure and non-infrastructure multinationals. This also responds to the call for a more context-based understanding of the outcomes of internationalization (e.g., Bausch and Krist 2007; Kim et al. 2015). Finally, our results add to the regionalization debate by confirming that we currently live in a semi-globalized world where regional borders still matter for most firms (Flores et al. 2013; Ghemawat 2003, 2005, 2007; Kim and Aguilera 2015; Rugman and Verbeke 2004, 2007).
Our findings carry important implications for managers, governments and policymakers. Despite the largely shared perception that multinationals should follow global strategies to pursue global opportunities (Verbeke and Asmussen 2016), managers should be aware that there is still a long way ahead for multinationals to fully profit from scattering their operations across different national and regional contexts even if it is true that some multinationals are in a better position to achieve it. Actually, multinationals are starting to display a higher caution in their foreign expansion. What The Economist calls "the retreat of the global company" (2) is just a reaction of many multinational companies to the high costs of operating at a global scale, and should challenge governments and policymakers to rethink their attitude towards FDI.
2 Conceptual Background
The literature on internationalization and performance--commonly measured as profitability--has drawn the attention of numerous scholars throughout the years. Early studies focused mostly on examining the relationship between degree of internationalization and profitability, albeit with inconclusive results (see Nguyen 2017 for a recent review). (3) Research on the topic has unearthed positive linear performance effects (e.g., Grant et al. 1988; Vernon 1971) as well as nonlinear patterns in the form of a U (e.g., Capar and Kotabe 2003; Lu and Beamish 2001), an inverted U (e.g., Geringer et al. 1989; Hitt et al. 1997), an S (e.g., Benito-Osorio et al. 2016; Contractor et al. 2003; Riahi-Belkaoui 1998) and, even, an M (e.g., Almodovar and Rugman 2014).
Some authors have explained these mixed findings based on the different definitions of internationalization used in the literature (e.g., Verbeke and Forootan 2012; Wiersema and Bowen 2011). Constructs such as foreign sales over total sales (FSTS), foreign assets over total assets (FATA), and number of operations do not account perfectly for the degree of dispersion of the multinationals' foreign footprint. As an example, two firms having the same amount of foreign operations could differ in the number and proximity of international markets entered.
To overcome this limitation, several studies have proxied the scope of internationalization either by a country count (e.g., Lu and Beamish 2001; Tallman and Li 1996) or an entropy index that considers both the number of national markets where multinationals operate and how important they are to them (e.g., Hitt et al. 1997). Nonetheless, these variables continue to neglect the inherent characteristics of each of the countries. To account for this fact, one should look at the international dispersion (4) rather than the degree of internationalization.
For this reason, current research is more concerned with the extent to which multinationals can successfully operate at a global scale. Challenging the existence of a flat world (Friedman, 2005), several studies have found that the LoIF tends to diminish profitability when expanding beyond the home region due to the increased complexity and costs (e.g., Mendoza et al. 2019; Oh and Contractor 2014; Rugman and Verbeke 2007). This is particularly salient in the regions often identified in the regionalization literature (e.g., Banalieva et al. 2012; Rugman and Verbeke 2004, 2008; Verbeke et al. 2016), which conform the so-called Triad; namely, European Union (EU). North American Free Trade Agreement (NAFTA), and Asia Pacific.
However, this research stream ignores that there can be varying degrees of intra and interregional heterogeneity (Galan and Gonzalez-Benito 2006), especially in institutional terms. For instance, although Spain and Germany are both members of the EU, they do not share the same...