1. Top of page
  2. Abstract
  3. Introduction
  4. Towards a theory of the M curve
  5. Research methodology
  6. Results
  7. Discussion and conclusions
  8. References
  9. Biographies

In this paper we address three challenges. First, we discuss how international new ventures (INVs) are probably not explained by the Uppsala model as there is no time for learning about foreign markets in newly born and small firms. Only in the longer term can INVs develop experiential learning to overcome the liability of foreignness as they expand abroad. Second, we advance theoretically on previous research demonstrating that the multinationality−performance relationship of INVs follows a traditional S-shaped relationship, but they first experience a ‘born global illusion’ which leads to a non-traditional M curve. Third, using a panel data analysis for the period 1994–2008 we find empirically that Spanish INVs follow an inverted U curve in the very short term, where no learning takes place, but that experience gained over time yields an M-curve relationship once learning takes place.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Towards a theory of the M curve
  5. Research methodology
  6. Results
  7. Discussion and conclusions
  8. References
  9. Biographies

Can the so-called international new ventures (INVs) actually become international from inception as argued by Oviatt and McDougall (1994)? Are INVs really born globals or do they expand within the home region of the triad, as do the large firms identified by Rugman and Verbeke (2004)? Do INVs expand internationally according to the traditional patterns observed in the multinationality and performance literature? To answer these questions we test empirically the actual internationalization process of Spanish INVs in the short term (no more than three years) and the longer term (over a 15-year period).

We embed our research in the mainstream literature on international business. This demonstrates that the process of internationalization (captured by the degree of internationalization (DOI)) represents a trade-off between the benefits and costs of expanding abroad. The benefits to a firm (large or small) are summarized by its firm-specific advantages (FSAs). The FSAs are unique capabilities possessed by the company (Rugman, 1981) such as R&D and marketing capabilities, experience and firm size (Rugman, Oh and Lim, 2011). The costs are the liability of foreignness (LOF), defined as the extra costs of doing business abroad derived from geographical and psychic distances (Zaheer, 1995). More specifically these costs include greater information costs across different countries, higher transportation cost, and increased coordination costs, among others. Firms can only succeed abroad if the FSAs outweigh the LOF. Obviously both FSAs and LOF need to be investigated in order to test the performance of INVs empirically.

The DOI thus represents the extent to which a firm expands beyond its domestic market into foreign countries (Hennart, 2011; Hult, 2011) and it is found in the literature under different names: multinationality; geographic diversification; international expansion; internationalization and globalization (Hitt et al., 2006). The relationship between the DOI and firm performance has not found any conclusive result over the last three decades. Initially, a linear form was tested with studies finding a positive relationship between the DOI and performance (Errunza and Senbet, 1981; Grant, 1987; Kim and Lyn, 1986); other studies find a negative linear relationship (Harveston, Kedia and Francis, 1999; Kumar, 1984; Siddharthan and Lall, 1982). Another stream of the literature has provided strong support for a curvilinear association. There are three main fits: a U-shaped relationship (Capar and Kotabe, 2003; Lu and Beamish, 2001; Ruigrok and Wagner, 2003); an inverted U-shaped relationship (Brock,Yaffe and Dembovsky, 2006; Gomes and Ramaswamy, 1999; Hitt, Hoskisson and Kim, 1997); and an S curve (Bae, Park and Wang 2008; Contractor, Kundu and Hsu, 2003; Lu and Beamish, 2004). There is also discussion of an inverted S curve (Chiang and Yu, 2005; Contractor, Kundu and Hsu, 2003; Ruigrok, Amann and Wagner, 2007), an M curve (Almodóvar, 2012; Lee, 2010) and an inverted M-curve relationship (Almodóvar, 2012).

As shown by the meta-analysis performed by Bausch and Krist (2007) and the review of the literature developed by Matysiak and Bausch (2012) among others, much of this extant multinationality−performance research has tended to focus on multinational enterprises (MNEs) and their foreign direct investments (FDI). However, we do not focus here on MNEs. Instead we target a very specific set of firms: INVs. These are small and medium sized enterprises (SMEs) with very little experience and very limited financial commitment abroad. Indeed they have exports but no FDI. We study how their multinationality−performance relationship evolves over time. In order to identify clearly the initial target of our paper, Figure 1 presents a 3 × 3 matrix where the horizontal axis represents size/experience and the vertical axis represents the mode/degree of financial commitment.


Figure 1. Matrix for the multinational activity

Download figure to PowerPoint

We have introduced this figure to clarify major confusion in the literature on international entrepreneurship. Our focus on INVs means that we study small and new firms which do international business mainly through exporting. Such INVs need to be contrasted with MNEs which are generally larger and older. MNEs also engage in deeper modes of foreign involvement such as joint ventures and wholly owned subsidiaries. Most of these MNEs are operating towards the top right cell of Figure 1. There are other combinations between the degree of foreign involvement and the size and experience of firms throughout the cells of Figure 1. Indeed MNEs may also engage in exporting. However, it is useful for us to restrict this paper on INVs by assuming that they alone engage in exporting.

Whilst it is not possible to develop a new synthesis of the literature by explaining all the nine cells of Figure 1, within the context of this paper it is essential that readers understand that our focus is upon the lower left cell which is bereft of MNEs and FDI. In the next section, we discuss literature which is relevant to this paper on INVs, and also offer some comments on literature about INVs which has not embedded itself fully in the international business literature. In general there are three major unresolved problems with the large current literature on INVs, all of which we address in the next section.

First, there is confusion about the nature and extent of internationalization. For INVs which are small, new, infant firms, it is highly unlikely that FDI is an option. Therefore it is unrealistic to apply MNE theory to such firms, although this has been attempted by Oviatt and McDougall (1994) and many subsequent authors. Instead, to test the internationalization of INVs we confine ourselves to export data.

Second, there is confusion about the age and experience of INVs. Obviously INVs of three years or younger have little experience of foreign markets. Therefore they are highly unlikely to have invested in market research about foreign markets in order to overcome the LOF. Probably their initial foreign involvement will be as a result of unsolicited orders and is unlikely to have been developed through careful strategic planning about the choice of location. Only in the industry-specific case of high-tech firms can it be assumed that such INVs are ‘born global’. Instead, the initial expansion abroad, especially under three years of age, will be in neighbouring markets in the home region of the triad.

Third, the opening of foreign markets to INVs is highly unlikely to be explained by any learning model, since learning takes time. By definition INVs are new infant, recently born, small firms in which learning has not taken place. Therefore, a knowledge based view of the firm should not be applied to INVs in a blanket manner. Instead, here we distinguish between very young ‘inexperienced’ INVs and somewhat older ‘experienced’ INVs as we model the stages of DOI, in the next section.

Overall, in this paper we are extremely careful to focus on the lower left hand cell of Figure 1 where each of these three problems in the literature is specifically addressed in our study. To advance on the current confused literature about INVs, here we propose and test that there must be an initial stage, prior to the traditional S curve, where INVs detect an opportunity abroad (an unsolicited order). Such opportunistic foreign sales will increase performance. In this initial phase, INVs appear to believe that they can cope with international markets without strong FSAs. This ‘born global illusion’ will subsequently lead to a fall in their performance (the first phase of the traditional S-curve relationship). Our new concept of the born global illusion adds an initial stage to the S curve, such that INVs shape an M-curve relationship between their DOI and performance. A similar M-curve relationship has been found by Lee (2010) when analysing INVs in Korea, but without the theoretical logic we develop in more detail in the next section.

We also contribute theoretically by addressing the issue of the final downturn in performance by appealing to the perspective of the regional nature of international business. This new thinking about how international activities are not globalized but regionalized (Rugman and Verbeke, 2004) has been confirmed, across various industries and time periods, by many papers (Beleska-Spasova and Glaister, 2009; Lee and Marvel, 2009; Rugman and Oh, 2010). Previously Rugman and Almodóvar (2011) and Almodóvar (2011, 2012) demonstrated that most international Spanish firms expand within their home region of the triad. It is difficult for them to transfer their business models to other regions out of the EU. We argue below that Spanish INVs face a similar inter-regional LOF.

We contribute empirically by using panel data for a 15-year period (1994–2008) which enables us to conduct a short-term and a longer-term analysis to capture the true relationship between the DOI and performance and to observe its evolution over time. The relevance of studying different time periods has been highlighted by several scholars such as McDougall and Oviatt (1996) and Thomas and Eden (2004) among others. Furthermore, we expand upon the empirical literature by examining the M-curve relationship, which we shall test using a quartic fit.

Finally, we join with other scholars who have noted that there is a need to study European firms to complement most of the research performed about INVs which has been focused on US companies (Stoian, Rialp and Rialp, 2011). Thus, our paper collects information on INVs from Spain which is one of the countries analysed by Guillén and García-Canal (2010) as the cradle of the so-called ‘new multinationals’. This new concept of MNEs has challenged scholars to reconsider some of the basic ideas of international competition because these MNEs do not follow traditional patterns. In the specific case of INVs, the Global Entrepreneurship Monitor (Bosma, Wennekers and Amorós, 2011) reports that Spain has lower than average levels of new entrepreneurs (5.8% of the active population) and lower than average levels in their international orientation (7% of new entrepreneurs).

We expand on these theoretical and empirical contributions in the next sections. First, we discuss the original and revisited Uppsala model in order to develop a theoretical background to support our four-phase model (M curve) and we present related hypotheses. We show that inexperienced INVs do not have time to learn about internationalization, so the Uppsala learning model does not apply and such INVs only operate in Phases One and Two, where we find an inverted U-shaped relationship. However, Uppsala is more relevant for experienced INVs, because they have time for learning and they are able to operate in Phases Three and Four of the M-curve model. Second, we explain the research methodology: the data set and the measurement of the variables. Third, we develop the statistical analyses for the short term (an inverted U-shaped curve) and for the longer term (an M curve) using panel data techniques. Fourth, we finish with a discussion of the results and conclusions.

Towards a theory of the M curve

  1. Top of page
  2. Abstract
  3. Introduction
  4. Towards a theory of the M curve
  5. Research methodology
  6. Results
  7. Discussion and conclusions
  8. References
  9. Biographies

The literature on INVs, starting with the innovative conceptual paper by Oviatt and McDougall (1994), demonstrates that INVs are not explained by Johanson and Wiedersheim-Paul (1975) and Johanson and Vahlne (1977) in their original Uppsala model. Although the Uppsala model has been revisited to update it with the business network approach (Johanson and Vahlne, 2009), it still does not explain the existence of small and medium INVs. Both models share the same core argument, firms become international gradually as they accumulate knowledge about foreign markets; the lack of such knowledge is an obstacle which will have to be overcome through a learning process over time. But, as INVs become international at inception, how and when do INVs learn? There is a need to examine, at least, two different time periods: (a) INVs in the short term (i.e. at their inception), labelled ‘inexperienced INVs’; and (b) INVs in the longer term (i.e. when they are older), labelled ‘experienced INVs’.

Original Uppsala model (1975/1977)

The original Uppsala model (Johanson and Vahlne, 1977; Johanson and Wiedersheim-Paul, 1975) is a dynamic model which argues that learning by experience enables companies to accomplish higher levels of overseas commitment. This model does not explain how the internationalization process starts; it only explains how the international commitment increases incrementally according to firms' experiential learning.

Inexperienced INVs do not have any market-specific experience because they are at their inception and there is no learning at this point (Oviatt and McDougall, 1994). The Uppsala model has recently tried to justify this reality by the previous international experience of the founder entrepreneur (Johanson and Vahlne, 2009); but a general model cannot expect that all the entrepreneurs have general and specific knowledge of both domestic and host markets. Thus, we argue that the learning by doing argument cannot be applied to inexperienced INVs; so this model cannot explain their existence and behaviour. Furthermore, Johanson and Vahlne (2003) state that performance is positively related to experiential learning (Barkema, Bell and Pennings, 1996); thus we may assume that inexperienced INVs will not be able to overcome the LOF and so they will face a decrease in performance.

Notwithstanding, in the longer term inexperienced INVs will learn from their international activity and they will eventually become experienced INVs. As time passes, they will be able to develop host market skills and accumulate specific market experience such that their inherent FSAs will overcome the LOF and will enhance performance. Consequently, increases in the international commitment may be expected. Therefore, we argue that experienced INVs learn from their international activities in line with the original Uppsala model.

A related aspect considered by the original Uppsala model is that companies tend to begin their international activity in countries characterized by a low psychic distance and progressively enter into more psychically dissimilar countries. This argument is consistent with the regional orientation framework developed by Rugman (2000, 2005) and it is applied here. Thus, inexperienced INVs are mainly expected to enter countries located in their home region, where the LOF is lower and easier to overcome; they are born regional, not born global. Alternatively, experienced INVs are expected to increase their presence first in the home region but also to start entering non-home region countries where the inter-regional LOF is higher and more difficult to overcome, which will impact negatively on performance.

Revisited Uppsala model (2009)

Because the international environment has changed since the original Uppsala model, the model has been revisited by introducing a business network view. The basic assumption is that all firms are connected in a network of relationships with customers and suppliers from which they accumulate knowledge, trust and, eventually, commitment (Johanson and Vahlne, 2009). Its main implication is that companies which are not in a relevant network are called ‘outsiders’. Finally, the psychic distance is related to the complexity of building new relationships abroad.

The rapid internationalization of INVs cannot be explained by this model because we cannot assume, as a rule of thumb, that all firms are embedded in a formal network before going abroad. In fact, Johanson and Vahlne (2003) states that ‘it takes time and resources to build relationships’ (p. 92) and small and medium inexperienced INVs have neither time nor resources to create a business network. The revisited Uppsala model tries to approach this issue with the prior-experience argument of the entrepreneur founder. In this case, the management team is expected not only to have the knowledge but also the access to the network. We state that this argument cannot be applied to small and medium inexperienced INVs. In the longer term, experienced INVs might be able to establish alliances and steady relationships with customers and/or suppliers. Whilst we consider it would be desirable for INVs to have network relationships in order to help reduce the LOF, in general INVs do not have the time to develop them.

To overcome the inability of both Uppsala models to justify the existence of INVs, we develop an M-curve model with four phases and we apply it to test their multinationality and performance.

The four phases of the M curve

The S-curve relationship combines a U-shaped and an inverted U-shaped relationship between DOI and performance. Initially, there is a U shape as the LOF causes performance to decrease but, as FSAs begin to assert themselves at higher levels of the DOI, then performance increases leading to a U-shaped relationship. Subsequently performance keeps increasing, driven by the FSAs, until an over-internationalization point develops as firms move beyond their home region of the triad from where performance decreases again (Contractor, 2007a; Contractor, Kundu and Hsu, 2003; Rugman and Verbeke, 2004). This leads to an inverted U curve.

We consider that this traditional S-curve logic (Contractor, Kundu and Hsu, 2003; Lu and Beamish, 2004) fails to explain how INVs can tolerate, at inception, the extra costs of doing business abroad and the investments needed to improve their FSAs. Thus, we add a new first phase where INVs detect a chance opportunity abroad (such as an unsolicited order) and, as a result, their performance increases. In this new Phase One, INVs appear to believe that they can cope with international markets without strong FSAs. This born global illusion will be the trigger of a decrease in their performance (first phase of the traditional S-curve behaviour and a new second phase of the M curve).

More specifically, the four-phase M-curve model is built upon the new idea that inexperienced INVs are less likely to possess strong FSAs in experiential knowledge and resources, necessary to overcome the LOF. However, as time goes by, inexperienced INVs will become experienced INVs by learning how to develop their FSAs in order to overcome the LOF, at the same time as they expand their international activity from the home region to other host regions. As we show below, in the four-phase model of the M curve, inexperienced INVs will find a chance opportunity which will generate an initial increase in their performance and this will generate a born global illusion; so this phase will be followed by a decline in performance derived from facing the LOF without strong FSAs. After learning from their international activity, experienced INVs will overcome the LOF and will have an increase in their performance but, finally, performance will decrease due to an over-internationalization (Contractor, 2007a; Contractor, Kundu and Hsu, 2003; Lu and Beamish, 2004). Here we explain this final downturn as being due to over-internationalization beyond the home region of the triad, using the regional logic of Rugman and Verbeke (2004). We use this thinking, along with a new concept in Phase One called the born global illusion, to expand the S-curve relationship into a four-phase M curve.

First phase: the born global illusion (positive slope)

Mainstream international business locational choice studies suggest that companies tend to start their internationalization process in neighbouring and culturally close countries (Davidson, 1983; Johanson and Vahlne, 1977). This ‘psychic distance’ principle is consistent with the home region orientation of firms where companies start their international career in their home region (Rugman, 2000, 2005; Rugman and Oh, 2007, 2010). According to this home region orientation, inexperienced Spanish INVs would start their internationalization in the EU. Previous work has found that, indeed, Spanish exporting firms average 91% of their sales in the EU, such that they obviously do not operate ‘globally’ (Almodóvar, 2011). In addition to this finding, here we argue that these Spanish INVs also suffer from a generic born global illusion. The definition of our born global illusion is the incorrect perception that a company with no formal international strategy or specific foreign market knowledge is able to succeed in international markets.

A born global illusion arises when INVs respond to opportunistic sales opportunities abroad, such as unsolicited orders. They rush to fill these orders without making any strategic assessment about either the LOF or the means by which they will sustain such export orders in the long term. However, INVs initially find that their performance improves as they begin to internationalize. In other words, initial international sales are related to an increase in performance in Phase One. There may be many other chance events which lead INVs to engage in opportunistic foreign sales, some of which have been examined in the export marketing literature (Katsikeas, 1996; Leonidou, 1995; Leonidou et al., 2007).

In this scenario, although performance may increase in the short run, this is an opportunistic orientation, not strategic behaviour based on a systematic and deliberate search but a chance event (Chandra, Styles and Wilkinson, 2009). Even when inexperienced INVs have no routines in place (Kuemmerle, 2002), a chance opportunity for sales in another country is alleged to have a positive impact on the firm performance.

  • H1: In an initial phase, the DOI of inexperienced INVs has a positive impact on performance.

This initial increase in their performance would be the trigger of a born global illusion so they will keep increasing their commitment abroad expecting higher outcomes. The opportunistic sales resulting from unsolicited orders are not the basis for a successful long-term international strategy. Thus, the born global illusion leads to a falling off in performance as the normal LOF arises. This leads to Phase Two.

Second phase: the liability of foreignness (negative slope)

Several classic studies on international business strategy and organizational structure suggest that a more strategic approach to international expansion is required. At low levels of foreign sales there are no significant changes in the infrastructure of the company (Bartlett and Ghoshal, 1989; Brock, Yaffe and Dembovsky, 2006; Egelhoff, 1982, 1988) because international activity is considered as an ‘appendage’ of their national activities (Daniels, Pitts and Tretter, 1984). However, the greater the DOI, the greater the need to implement new structures and routines in order to face the added costs of the LOF (Gomes and Ramaswamy, 1999; Hymer, 1960/1976; Zaheer, 1995). The LOF is due to the unfamiliarity with foreign language, culture and regulations. The LOF requires investments by the firm to learn and grow and this can be first achieved in the home region.

Small and medium inexperienced INVs are characterized by the paucity of FSAs such as (a) an inability to achieve scale economies to compensate for rising costs (Contractor, Kundu and Hsu, 2003) derived from the coordination complexity and information asymmetry of the new international structure (Lu and Beamish, 2004); (b) a lack of managerial skills (Qian and Wang, 1999) because these kinds of firms do not have a hierarchy of executives to filter complex decisions in the international markets (Buckley, 1997); and (c) inexperience and lack of foreign market knowledge (Lu and Beamish, 2004) to implement new routines and recombine them.

Thus, if the extra costs derived from the intra-regional LOF exceed the core FSAs of the firm then the potential benefits of internationalization are affected (Ruigrok, Amann and Wagner, 2007). This is because firms will need to invest in learning how to develop these FSAs and this will take time which will lead to a decline in performance.

  • H2: In a second phase, the DOI of inexperienced INVs has a negative impact on performance.

Third phase: international learning (positive slope)

In a third phase, the company will see its basic FSAs (scale economies; technology and marketing skills) overcome the intra-regional LOF. As an example of an important FSA, economies of scale may have a significant effect on firm performance (Contractor, 2007b; Contractor, Kundu and Hsu, 2003; Hitt et al., 2006). Hennart (2007) argues that many firms need to increase their DOI in order to achieve minimum efficient scale and, if there are no significant extra costs for selling products abroad, these economies will enhance firm profitability. In our case, INVs pursue exporting activities in the home region, so the assumption of no significant extra cost is suitable because all the production may be concentrated in Spain and there are few trade barriers within the EU (Rugman, 2005). Thus, economies of scale would have a positive impact on firm performance.

Verbeke (2009) explains that there are three levels of FSAs − stand-alone resources, routines, and recombination capabilities − and the main challenge for the success of an international company is developing FSAs to overcome the LOF. These FSAs come from linking their previous FSAs, such as technological or marketing knowledge, to location/country-specific advantages of the host market. The recombination of domestic and local knowledge is considered the highest level of FSA and it is critical for satisfying customer demand and succeeding abroad.

Some of this thinking may apply to exporting firms as they learn about host-country-specific advantages, even when there is no FDI involvement in host economies. Salomon (2002) found that firms learn from exporting activities because ‘exporters may receive valuable information about product preferences, competing products, and the local environment from local customers’ (p. 24). Thus, inexperienced INVs will become experienced INVs through this learning process which will strengthen their bundle of FSAs and will contribute positively to overcoming the intra-regional LOF. This situation will locate experienced INVs in a stronger position in international markets and will have a positive impact on performance.

  • H3: In a third phase, the DOI of experienced INVs has a positive impact on performance.

Fourth phase: over-internationalization (negative slope)

Sometimes companies associate a higher DOI with a higher level of performance without taking into account what the optimum level of internationalization is. Identifying the optimum level is crucial because, from this point on, the company will suffer from the loss of strategic control and higher information costs. Several authors have contrasted the negative impact of excessive internationalization on performance. For example, Merry (1995) explains that an extreme DOI pushes firms to a chaotic situation characterized by disorder, instability and ambiguity. Ruigrok, Amann and Wagner (2007) demonstrate that an extreme DOI leads to a ‘weak situation’ characterized by unstructured circumstances whose high complexity will lead executives to take ineffective procedures. Thus, over-internationalization will reduce company performance. These factors will cause a decrease in overall performance (Tallman and Li, 1996) due to the incremental costs and complexity that will be higher than the incremental benefits (Hitt, Hoskisson and Kim, 1997; Ruigrok, Amann and Wagner, 2007).

This thinking can now be adapted to incorporate recent work on the regional nature of firms (Rugman, 2005). Essentially the FSAs and business models are best exploited in the home region. Going to another region introduces major new costs. In short, when the firm expands beyond its home region to other regions in the triad, there is an inter-regional LOF such as communication, coordination, control and motivation (Gomes and Ramaswamy, 1999) which results in the decay of FSA transferability. Then such over-internationalization beyond the home region may reduce company performance. This point was recognized by Contractor (2007a) and it has been empirically supported by Lee (2010) in his test of the M-curve relationship between the multinationality and performance of Korean firms.

  • H4: In a fourth phase, the DOI of experienced INVs has a negative impact on performance.

We use this theoretical four-phase model to conduct tests of the performance of both inexperienced and experienced INVs. For the inexperienced INVs we use the theory behind Phases One and Two, where firms do not have time to learn yet and the Uppsala model does not apply. Thus, we expect to find an inverted U-shaped relationship between the DOI and performance. In contrast, for experienced firms, the Uppsala model does apply and learning can take place. Thus, we expect firms to operate in Phases Three and Four. Thus, INVs shape a more complex M-curve relationship between the DOI and performance over time. This relationship is illustrated in Figure 2.


Figure 2. Four-phase model: from the inverted U shape to the M-curve relationship between the degree of internationalization and performance (illustrative shape)

Download figure to PowerPoint

Research methodology

  1. Top of page
  2. Abstract
  3. Introduction
  4. Towards a theory of the M curve
  5. Research methodology
  6. Results
  7. Discussion and conclusions
  8. References
  9. Biographies

The sample of Spanish INVs

Data are from Survey on Business Strategies (SBS) which is a mature databank that gathers information about a representative panel of Spanish manufacturing firms with ten or more employees, and it is carried out by Foundation SEPI with the support of the Ministry of Industry of Spain. According to Podsakoff et al. (2003) and Chang, van Witteloostuijn and Eden (2010) common method variance (CMV) is often a problem with such survey data and they propose several remedies to avoid it. Yet, the SBS has multiple respondents; it assures anonymity; it asks mostly for non-perceptual measurements; and Foundation SEPI carries out different criteria for content validity (if there is a failure to comply with the consistency controls the company is required to submit documental justification). Furthermore, we use longitudinal analysis, and we include non-linear terms. Therefore the validity of our conclusions is not threatened by CMV.

We identify INVs and SMEs as follows. First, in order to identify INVs we choose the tightest and most followed thresholds to be in line with the literature: at least 25% of exports over total sales (Andersson and Wictor, 2003; Knight and Cavusgil, 1996, 2005; Kuivalainena, Sundqvista and Servais, 2007; Moen, 2002) in at least one of the first three years from inception (Autio, Sapienza and Almeida, 2000; Crick, 2009; Knight, 1997; Knight, Madsen and Servais, 2004; Madsen, Rasmussen and Servais, 2000; McDougall and Oviatt, 2000; McGaughey, 2007; Mort and Weerawardena, 2006; Zhou, Barnes and Lu, 2010). Second, we define SMEs as those firms with more than ten employees and fewer than 200 (Knight and Cavusgil, 2005; Reuber and Fischer, 1997). We are aware that the official cutoff point established by the European Commission is from ten to 250 employees in place of 200. We use 200 employees because the SBS uses this threshold when sampling the Spanish manufacturing sector. However, we have replicated the entire set of models with the 250 cutoff point and the results remain the same. The sample size is 110 firms that are small and medium INVs over the full period 1994–2008.

In order to offer accurate and robust results about INVs over time, we analyse two different time periods. In alternative (1), small and medium inexperienced INVs, we study Spanish INVs in the short term (i.e. at their inception) and when they become older than three years they are removed from the sample. This gives us around 17 companies per year on average over the 15-year panel data set. In alternative (2), small and medium INVs, we study Spanish INVs in the longer term (we allow the former inexperienced INVs to become experienced INVs over time). This gives us around 52 companies per year on average. We are aware of the low number of firms for alternative (1); in the section ‘Statistical analysis of the M curve’ we discuss which treatment is more appropriate for our data set.

Variables and measures

Dependent variable: performance

In order to measure the impact of the DOI on the overall performance of Spanish INVs, we follow the literature and use the return on sales (ROS) (Contractor, Kundu and Hsu, 2003; Geringer, Beamish and daCosta, 1989; Rugman and Oh, 2010).

Independent variables: degree of internationalization

We measure DOI through the ratio exports over total sales (Kundu and Katz, 2003; López, Kundu and Ciravegna, 2009; Moen, 2002). In order to capture the relationship between export intensity and firm's performance, we create three variables raising the original DOI to the power of two, three and four − DOI2/DOI3/DOI4.

We are aware that several measures of the DOI have been developed in the literature. Rugman and Oh (2011) review and discuss this and they conclude that scope metrics, such as the number of countries or the number of subsidiaries in each foreign country, ‘present misleading information about the breadth of foreign activity, because they assume countries to be of equal size’ (p. 203). The better metric is a scale one, namely foreign sales over total sales. Foreign sales include exports plus sales by overseas subsidiaries, and this metric has been used in previous papers in this journal (Balabanis, 2001; Gaur and Kumar, 2009). A similar good metric is foreign assets over total assets. Rugman, Yip and Jayaratne (2008) calculate return on foreign assets for large UK MNEs.

Here we use exports over total sales. Notwithstanding, this limitation does not appear to represent a significant bias because we analyse small and medium INVs which are expected to have low commitment modes abroad. Thus, because overseas subsidiaries' sales tend to zero, ‘foreign sales over total sales’ tends to be equal to ‘exports over total sales’.

Control variables

In order to avoid biased inferences and account for factors that may affect the results, we control for firm- and industry-specific variables that have been used in past literature.

We introduce the FSA R&D_intensity because it portrays the technological knowledge of the firm and, according to the resource-based view and the eclectic paradigm, firms with unique intangible capabilities might expect to enhance their performance when going international. This variable has been measured as R&D expenditure over total sales (Herrmann and Datta, 2005; Lee and Marvel, 2009; Lu and Beamish, 2004).

We measure the FSA marketing skills (Advert_intensity) through the proxy advertisement expenditures over total sales (Kotabe, Srinivasan and Aulakh, 2002; Lee and Marvel, 2009; Qian et al., 2010).

From the perspective of Uppsala stages theory, the internationalization process is affected by the firm's experience in overcoming the LOF. Furthermore, Bausch and Krist (2007) demonstrate that firm age affects the relationship between the DOI and performance. We measure the variable Age as the total number of full years between the firm's foundation and the year when the company is responding to the survey (Gaur and Kumar, 2009; Lee and Park, 2006; Qian et al., 2008).

Firm size has been used as a proxy of managerial and financial resources (Qian et al., 2008) and it has a direct impact on performance (Bausch and Krist, 2007). Among the different measurements to proxy the size of the company we employ the natural log of the number of employees LnSize (Acedo and Jones, 2007; Fang et al., 2007; Richter, 2007).

Fernandez and Nieto (2006) show how internationalization patterns are significantly affected by foreign ownership. Thus, the variable Foreign_ownership has been identified as the percentage of direct/indirect ownership of foreign capital on the Spanish firm.

We control for industry effects incorporating 19 dummy variables for the 20 manufacturing sectors identified in the SBS. Due to the large number of manufacturing sectors, we report only the existence or not of Industry_effects. This procedure has also been followed by Qian et al. (2010).

Statistical analysis of the M curve

While panel data sets have benefits over pooled data, with more variability and less collinearity among the regressors (Baltagi, 2001; Hsiao, 1986), our first step is testing what kind of treatment is better for our data: pooling data versus panel data. In our case we have to check the poolability tests in our two alternatives and both show that panel data are not poolable. Our second step is to apply the Hausman test, which compares the consistent fixed effects model with the efficient random effects model. In our case, with alternative (1) it is safe to use random effects. However, alternative (2) should be run using fixed effects.

We use a generalized least squares random effects model to prove whether the overall alternative (1) − inexperienced INVs − and its variables are significant. According to Greene (2002), when the individual effects are strictly uncorrelated with the regressors, then we can model the individual specific constant terms as randomly distributed across cross-sectional units. So, our formulation is

  • display math

where there are K regressors including a constant; the single constant term is the mean of the unobserved heterogeneity E[ziα]; the component ui = {ziα − E[ziα]} is the random heterogeneity specific to the ith observation and is constant through time, where zi is a set of observable, firm-specific variables. And, because we are working on an unbalanced panel, our full sample size is inline image instead of nT in order to allow unequal group sizes across years.

Alternative (2) follows a fixed effects regression where we assume that firm effects are correlated with xit and that differences across units can be captured in differences in the constant term. Its formulation is yit = xitβ + αi + εit where αi is ziα and represents all the observable effects; it indicates an estimable conditional mean and is treated as an unknown parameter to be estimated; vector zi contains all the observable effects and allows unbiased estimation of β (Greene, 2002).


  1. Top of page
  2. Abstract
  3. Introduction
  4. Towards a theory of the M curve
  5. Research methodology
  6. Results
  7. Discussion and conclusions
  8. References
  9. Biographies

Table 1 shows the means, standard deviations and correlations of dependent, independent and control variables for alternative (2).

Table 1. Correlation matrix and statistics of alternative (2)
 Standard deviation14.9727.074.728.812.141.0236.18

In Table 1 we do not have any case correlated over 0.5 which reveals that multicollinearity does not constitute a threat to coefficient estimations. We have created the correlation matrix for alternative (1); it presents similar results to Table 1 and, for that reason, we have not included it in the paper. Additionally, we have run different tests (such as causality and panel unit roots) and searched for the presence of influential outliers without detecting any bias.

We present our models in Table 2. Alternative (1) − inexperienced INVs − shows a generalized least squares random effects regression; and alternative (2) shows a fixed effects model. Both alternatives include three models in order to corroborate the consistency and robustness of the results depending on the variables included in the analysis. Model I introduces only the independent variables, model II reports results only for the control variables and model III presents the results for the full model. Since the results remain stable across models I, II and III in both alternatives (1) and (2), we will test our hypotheses using model III. Before analysing the coefficients of the models, we revise the Wald chi-squared test; the p-values of all the full models are lower than 0.05, so we can reject the null hypothesis; at least one coefficient is statistically different from zero, so we can proceed to study the coefficients.

Table 2. Results of random and fixed effect regression models for performance (ROS)
 Alternative (1): Inexperienced INVsAlternative (2): From inexperienced to experienced INVs
 Random effectsFixed effects
  1. Standard deviations are given in parentheses.

  2. +p < 0.1; *p < 0.05; **p < 0.01; ***p < 0.001.

DOI1.4528** 1.4224*0.6488* 0.5844+
 (0.5278) (0.5762)(0.3243) (0.3374)
DOI2−0.0549* −0.0548*−0.0325* −0.0354*
 (0.0239) (0.0261)(0.0146) (0.0151)
DOI30.0007+ 0.0007+0.0005* 0.0006*
 (0.0004) (0.0004)(0.0002) (0.0002)
DOI4−0.000003 −0.000003−0.000002+ −0.000003*
 (0.000002) (0.000002)(0.000001) (0.000001)
Age 2.8779**2.5416* −0.2649*−0.3084*
  (1.0852)(1.0529) (0.1299)(0.1320)
R&D_intensity 0.08870.0800 −0.0118−0.0177
  (0.0798)(0.0786) (0.0718)(0.0714)
Advert_intensity 0.07810.2212 −0.2060−0.1907
  (0.4794)(0.4885) (0.3348)(0.3328)
LnSize −1.3216−1.0381 5.8505***6.0812***
  (1.4524)(1.5108) (1.6672)(1.6565)
Foreign_ownership 0.0921**0.0726* 0.04440.0397
  (0.0330)(0.0347) (0.0320)(0.0319)
Industry_effects NoNo NoNo

In discussing the results for alternative (1) we observe that the behaviour of inexperienced INVs shapes a quadratic1 polynomial (only DOI and DOI2 have a p-value < 0.05). So, we get an inverted U shape where we first have a positive relationship (1.4224) followed by a negative one (−0.0548) with respect to the dependent variable ROS. A clearer presentation of the inverted U-curve behaviour is shown in Figure 3.


Figure 3. Performance and internationalization in the short term

Download figure to PowerPoint

Figure 3 depicts the predicted ROS by the DOI for inexperienced INVs. It presents two different intervals. First, when the DOI increases to 12.98% we observe a positive relationship getting a maximum of 9.23% on ROS; this behaviour is consistent with Hypothesis 1. After this turning point the relationship becomes negative on ROS giving support to Hypothesis 2.

Studying the control variables, we observe the significant effect that some of them exert on ROS. Age (p-value < 0.05) presents a coefficient of 2.5416 that is the average amount that ROS increases when the company becomes one year older, other variables being held constant. Foreign_ownership also appears to be relevant for ROS with an impact of 7.26% (p-value < 0.05). Finally, there are no significant industry effects.

Alternative (2) presents the same variables as alternative (1) and it includes both inexperienced and experienced INVs. Now we can observe the full evolution of INVs in the longer term. With respect to the DOI variables, we confirm Hypotheses 1, 2, 3 and 4 supporting the M-curve relationship between the DOI and ROS. The fixed effects model presents a weak significant value (p-value < 0.1) for the DOI and significant values (p < 0.05) for its quadratic, cubic and quartic transformations. When we fit a fourth degree polynomial, if the quartic term is significant, then we must retain those terms of smaller order than the quartic even if they were not significant. In our case, all of them are significant at different levels.

To further examine this new relationship in the literature, Figure 4 depicts an M-curve fit. In it we add a new descriptive table with the percentage of observations for each phase as well as the ratio of intra-regional sales (R/T, measured as domestic and rest of the region sales over total sales (Rugman and Verbeke, 2004)). In order to confirm that the R/T means are significantly different among phases, we have run ANOVA tests. Both the omnibus test and pairwise comparisons are significant (all p-values < 0.001), so we can reject the null hypothesis of equal population means.


Figure 4. Performance and internationalization of INVs for a 15-year period

R/T, intra-regional sales ratio.

***p < 0.001.

Download figure to PowerPoint

In Phase One, the first interval ranges from DOI values greater than zero to 11.1. This occurs across 13.9% of the observations. Phase One shows a strong home region orientation (R/T = 99.1%). The relationship between the DOI and performance is positive and it reaches a moderate maximum on ROS of 2.55%. Phase Two is within the range 11.1–46.32 with 45.43% of the sample and an R/T of 91.6%. But performance shows a marked decrease to a minimum of −2.57% on ROS. Phase Three has 36.29% of the observations and their R/T is 81.4% on average. It shows a sharp increase in performance, reaching figures of 9.83% on ROS with a maximum inflexion point of 95.38% of exports over sales. From this inflexion point the relationship becomes negative again as we enter Phase Four. Here we only have 4.38% of the observations and we detect a significantly lower commitment to the home region (R/T is 53.8% on average). In summary, Figure 4 demonstrates that the strong home region effect holds across the first three phases, but as sales outside the home region increase in Phase Four there is a new decrease in performance.

These results exemplify the support we obtain for Hypotheses 1, 2, 3 and 4. They also sustain our line of reasoning about the strong home region orientation of Spanish INVs and how, over time, they expand out of the EU. Hence, the strong home region orientation of INVs in Phase Two confirms the negative impact of the intra-regional LOF on performance. Finally, the lower home region orientation of INVs in Phase Four confirms our arguments about the negative effect of the inter-regional LOF on firm performance.

Among the control variables in alternative (2), older companies have a poorer performance than younger firms as evidenced by the negative coefficient of the variable Age (p < 0.05). However, larger firms have better performance than smaller ones in both models (the fixed effects model offers a coefficient of 6.0812 with a p < 0.001).

Regarding industry effects, as expected the fixed effects model shows no impact; this is due to the nature of the statistical technique that drops from the analysis those variables that show no change over time and sector variables remain stable through the years.

Discussion and conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Towards a theory of the M curve
  5. Research methodology
  6. Results
  7. Discussion and conclusions
  8. References
  9. Biographies

This paper contributes to the debate about the multinationality and performance relationship by focusing on a particular group of firms, fresh INVs, and their evolution over time. We find an inverted U shape in the short term (no more than three years) and an M-curve relationship in the longer term of 15 years. We advance theoretically and empirically upon previous research in three ways.

First, we find an initial phase characterized by a born global illusion where inexperienced INVs undergo a positive increase in performance that leads to an early international expansion with no sustainable FSAs to compete internationally.

Second, we find a decline in performance for inexperienced INVs in Phase Two. This is due to lack of experience and time for learning. Thus, we query the applicability of the original Uppsala model, and its learning−commitment assumption, for explaining the pace and path of INVs in the short term. In contrast, the Uppsala model does apply to experienced INVs, as we observe a positive impact on performance in Phase Three.

Third, we find that Spanish INVs are home region oriented during Phases One, Two and Three, but they reduce their home region orientation in Phase Four. This explains the final downturn in Phase Four of the M curve. Thus, when experienced INVs reduce their home region orientation (higher psychic distance) the LOF increases and it has a negative effect on performance.

Our research is relevant not only for scholars but also for entrepreneurs who are managing the international operations of new companies. As suggested by Lu and Beamish (2004), ‘managers need to take a long term view of internationalization’ (p. 607). Consequently, and focusing on the M-curve long-term view, entrepreneurs should always be aware of the risks of international operations. They should not be confused by the born global illusion derived from a chance opportunity in a foreign market, and they must take into account that any sustainable international strategy must rely on core FSAs. Then, they must be prepared for decreasing performance in a second phase, when the LOF arises. They should assume that learning costs will be experienced in this second phase. Finally, when achieving increasing performance in the third phase, due to their FSAs, they should pay close attention to the eventual downside of over-internationalization beyond their home region. Possibly, they should attempt to identify an optimum level of internationalization in order not to exceed the turning point in the fourth phase where costs derived from inter-regional expansion and rising complexity will exceed revenues.

Finally, the M curve could also be used as a managerial tool for identifying the position of the company in relation to its competitors. Here, the position of the competitors in the sector could be plotted using the coefficients presented in order to make a comparative analysis.

Our results also suggest other implications. Surprisingly, neither the R&D intensity nor the advertising intensity has any influence on firm performance for Spanish firms, despite being significant FSAs in the literature. During the three years analysed in alternative (1) it makes sense that none of them is significant because INVs do not have the capacity for their development. Yet, for alternative (2) over 15 years, the INVs enjoy learning (to reduce the LOF) leading to a positive link between FSAs and performance.

Other findings are that, while the firm is inexperienced, the level of foreign ownership is relevant for enhancing firm performance. This could be explained by the international experience or financial support that the foreign company could transfer to the firm in which they are investing. Second, the age of the company exerts a positive influence on performance during the first three years from inception but, as companies become older, the relationship becomes negative.

Although this study makes several contributions to the literature, it is not free from limitations which may provide fruitful avenues for future research. Our study may not be generalizable to other countries. The SBS guarantees the representativeness of the Spanish sample so the results may be extrapolated to the manufacturing sector, but any generalization apart from this country or sector should be undertaken with caution. Future studies may replicate this model in other contexts outside the EU, as well as for other industries such as services.

With regard to our metrics, there are several performance measurements: satisfaction, strategic, and financial metrics. Our study attempts to conform with the majority of the international business literature on this topic, so we use ROS. However, our M-curve fit could be replicated with different performance outcomes in order to offer new insights.

Finally, this research uses information gathered from a survey which allows us to generalize the results for the whole Spanish manufacturing sector. However, this line of research could be enriched by the use of qualitative methods (such as in-depth interviews) to understand the INV phenomenon in more detail.

  1. 1

    We have decided to exclude the cubic term from the analysis because its p-value is weakly significant and it could mislead the interpretation of the results.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Towards a theory of the M curve
  5. Research methodology
  6. Results
  7. Discussion and conclusions
  8. References
  9. Biographies
  • Acedo, F. J. and M. V. Jones (2007). ‘Speed of internationalization and entrepreneurial cognition: insights and a comparison between international new ventures, exporters and domestic firms’, Journal of World Business, 42, pp. 236252.
  • Almodóvar, P. (2011). ‘The home-region orientation of Spanish exporting firms’, Multinational Business Review, 19, pp. 213228.
  • Almodóvar, P. (2012). ‘The international performance of standardizing and customizing Spanish firms: the M curve relationships’, Multinational Business Review, 20, pp. 306330.
  • Andersson, S. and I. Wictor (2003). ‘Innovative internationalization in new firms: born-globals – the Swedish case’, Journal of International Entrepreneurship, 1, pp. 249276.
  • Autio, E., H. J. Sapienza and J. G. Almeida (2000). ‘Effects of Age at entry, knowledge intensity, and imitability on international growth’, Academy of Management Journal, 43, pp. 909924.
  • Bae, S. C., B. J. C. Park and X. Wang (2008). ‘Multinationality, R&D intensity, and firm performance: evidence from U.S. manufacturing firms’, Multinational Business Review, 16, pp. 5378.
  • Balabanis, G. I. (2001). ‘The relationship between diversification and performance in export intermediary firms’, British Journal of Management, 12, pp. 6784.
  • Baltagi, B. H. (2001). Econometric Analysis of Panel Data. New York: Wiley.
  • Barkema, H. G., J. H. J. Bell and J. M. Pennings (1996). ‘Foreign entry, cultural barriers, and learning’, Strategic Management Journal, 17, pp. 151166.
  • Bartlett, C. A. and S. Ghoshal (1989). Managing Across Borders: The Transnational Solution. Boston, MA: Harvard Business School Press.
  • Bausch, A. and M. Krist (2007). ‘The effect of context-related moderators on the internationalization−performance relationship: evidence from meta-analysis’, Management International Review, 47, pp. 319347.
  • Beleska-Spasova, E. and K. W. Glaister (2009). ‘The geography of British exports: country-level versus firm-level evidence’, European Management Journal, 27, pp. 295304.
  • Bosma, N., S. Wennekers and J. E. Amorós (2011). Global Entrepreneurship Monitor: 2011 Extended Report: Entrepreneurs and Entrepreneurial Employees Across the Globe. Babson Park, MA; London; Santiago de Chile; Malaysia: Babson College, London Business School, Universidad del Desarrollo, UniversitiTun Abdul Razak.
  • Brock, D. M., T. Yaffe and M. Dembovsky (2006). ‘International diversification and performance: a study of global law firms’, Journal of International Management, 12, pp. 473489.
  • Buckley, P. J. (1997). ‘International technology transfer by small and medium-sized enterprises’, Small Business Economics, 9, pp. 6778.
  • Capar, N. and M. Kotabe (2003). ‘The relationship between international diversification and performance in service firms’, Journal of International Business Studies, 34, pp. 345355.
  • Chandra, Y., C. Styles and I. Wilkinson (2009). ‘The recognition of first time international entrepreneurial opportunities: evidence from firms in knowledge-based industries’, International Marketing Review, 26, pp. 3061.
  • Chang, S. J., A. van Witteloostuijn and L. Eden (2010). ‘From the Editors: Common method variance in international business research’, Journal of International Business Studies, 41, pp. 178184.
  • Chiang, Y. C. and T. H. Yu (2005). ‘The relationship between multinationality and the performance of Taiwan firms’, Journal of American Academy of Business, 6, pp. 130134.
  • Contractor, F. J. (2007a). ‘The evolutionary or multi-stage theory of internationalization and its relationship to the regionalization of firms’. In A. M. Rugman (ed.), Regional Aspects of Multinationality and Performance, pp. 1129. Oxford: Emerald.
  • Contractor, F. J. (2007b). ‘Is international business good for companies? The evolutionary or multi-stage theory of internationalization vs. the transaction cost perspective’, Management International Review, 47, pp. 453475.
  • Contractor, F. J., S. K. Kundu and C. C. Hsu (2003). ‘A three-stage theory of international expansion: the link between multinationality and performance in the service sector’, Journal of International Business Studies, 34, pp. 518.
  • Crick, D. (2009). ‘The internationalisation of born global and international new venture SMEs’, International Marketing Review, 26, pp. 453476.
  • Daniels, J. D., R. A. Pitts and M. J. Tretter (1984). ‘Strategy and structure of United States multinationals: an exploratory study’, Academy of Management Journal, 27, pp. 292307.
  • Davidson, W. H. (1983). ‘Market similarity and market selection: implications for international marketing strategy’, Journal of Business Research, 11, pp. 439456.
  • Egelhoff, W. G. (1982). ‘Strategy and structure in multinational-corporations: an information-processing approach’, Administrative Science Quarterly, 27, pp. 435458.
  • Egelhoff, W. G. (1988). ‘Strategy and structure in multinational-corporations: a revision of the Stopford and Wells model’, Strategic Management Journal, 9, pp. 114.
  • Errunza, V. R. and L. W. Senbet (1981). ‘The effects of international operations on the market value of the firm: theory and evidence’, Journal of Finance, 36, pp. 401417.
  • Fang, Y. L., M. Wade, A. Delios and P. W. Beamish (2007). ‘International diversification, subsidiary performance, and the mobility of knowledge resources’, Strategic Management Journal, 28, pp. 10531064.
  • Fernandez, Z. and M. J. Nieto (2006). ‘Impact of ownership on the international involvement of SMEs’, Journal of International Business Studies, 37, pp. 340351.
  • Gaur, A. S. and V. Kumar (2009). ‘International diversification, business group affiliation and firm performance: empirical evidence from India’, British Journal of Management, 20, pp. 172186.
  • Geringer, J. M., P. W. Beamish and R. C. daCosta (1989). ‘Diversification strategy and internationalization: implications for MNE performance’, Strategic Management Journal, 10, pp. 109119.
  • Gomes, L. and K. Ramaswamy (1999). ‘An empirical examination of the form of the relationship between multinationality and performance’, Journal of International Business Studies, 30, pp. 173187.
  • Grant, R. M. (1987). ‘Multinationality and performance among British manufacturing companies’, Journal of International Business Studies, 18, pp. 7989.
  • Greene, W. H. (2002). Econometric Analysis. Upper Saddle River, NJ: Prentice Hall.
  • Guillén, M. and E. García-Canal (2010). The New Multinationals: Spanish Firms in a Global Context. New York: Cambridge University Press.
  • Harveston, P. D., B. L. Kedia and J. D. Francis (1999). ‘MNE's dependence on foreign operations and performance: a study of MNEs from the "triad" regions’, International Business Review, 3, pp. 293307.
  • Hennart, J. F. (2007). ‘The theoretical rationale for a multinationality−performance relationship’, Management International Review, 47, pp. 423452.
  • Hennart, J. F. (2011). ‘A theoretical assessment of the empirical literature on the impact of multinationality on performance’, Global Strategy Journal, 1, pp. 135151.
  • Herrmann, P. and D. K. Datta (2005). ‘Relationships between top management team characteristics and international diversification: an empirical investigation’, British Journal of Management, 16, pp. 6978.
  • Hitt, M. A., R. E. Hoskisson and H. Kim (1997). ‘International diversification: effects on innovation and firm performance in product-diversified firms’, Academy of Management Journal, 40, pp. 767798.
  • Hitt, M. A., L. Tihanyi, T. Miller and B. Connelly (2006). ‘International diversification: antecedents, outcomes, and moderators’, Journal of Management, 32, pp. 831867.
  • Hsiao, C. (1986). Analysis of Panel Data. Cambridge: Cambridge University Press.
  • Hult, G. T. M. (2011). ‘A strategic focus on multinationality and firm performance’, Global Strategy Journal, 1, pp. 171174.
  • Hymer, S. (1960/1976). The International Operations of National Firms. Cambridge, MA: MIT Press.
  • Johanson, J. and J. E. Vahlne (1977). ‘The internationalization process of the firm: a model of knowledge development and increasing market commitment’, Journal of International Business Studies, 8, pp. 2332.
  • Johanson, J. and J. E. Vahlne (2003). ‘Business relationship learning and commitment in the internationalization process’, Journal of International Entrepreneurship, 1, pp. 83101.
  • Johanson, J. and J. E. Vahlne (2009). ‘The Uppsala internationalization process model revisited: from liability of foreignness to liability of outsidership’, Journal of International Business Studies, 40, pp. 14111431.
  • Johanson, J. and F. Wiedersheim-Paul (1975). ‘Internationalization of firm: four Swedish cases’, Journal of Management Studies, 12, pp. 305322.
  • Katsikeas, C. S. (1996). ‘Ongoing export motivation: differences between regular and sporadic exporters’, International Marketing Review, 13, pp. 419.
  • Kim, W. S. and E. O. Lyn (1986). ‘Excess market value, the multinational-corporation, and Tobin Q-ratio’, Journal of International Business Studies, 17, pp. 119125.
  • Knight, G. A. (1997). Emerging Paradigm for International Marketing: The Born Global Firm. East Lansing, MI: Michigan State University.
  • Knight, G. A. and S. T. Cavusgil (1996). ‘The born global firm: a challenge to traditional internationalization theory’. In T. K. Madsen (ed.), Advances in International Marketing, pp. 1126. Amsterdam: JAI Press.
  • Knight, G. A. and S. T. Cavusgil (2005). ‘A taxonomy of born-global firms’, Management International Review, 45, pp. 1535.
  • Knight, G. A., T. K. Madsen and P. Servais (2004). ‘An inquiry into born-global firms in Europe and the USA’, International Marketing Review, 21, pp. 645665.
  • Kotabe, M., S. S. Srinivasan and P. S. Aulakh (2002). ‘Multinationality and firm performance: the moderating role of R&D and marketing capabilities’, Journal of International Business Studies, 33, pp. 7997.
  • Kuemmerle, W. (2002). ‘Home base and knowledge management in international ventures’, Journal of Business Venturing, 17, pp. 99122.
  • Kuivalainena, O., S. Sundqvista and P. Servais (2007). ‘Firms' degree of born-globalness, international entrepreneurial orientation and export performance’, Journal of World Business, 42, pp. 253267.
  • Kumar, M. S. (1984). Growth Acquisition and Investment: An Analysis of the Growth of Industrial Firms and their Overseas Activities. Cambridge: Cambridge University Press.
  • Kundu, S. K. and J. A. Katz (2003). ‘Born-international SMEs: BI-level impacts of resources and intentions’, Small Business Economics, 20, pp. 2547.
  • Lee, H. U. and J. H. Park (2006). ‘Top team diversity, internationalization and the mediating effect of international alliances’, British Journal of Management, 17, pp. 195213.
  • Lee, I. H. (2010). ‘The M curve: the performance of born-regional firms from Korea’, Multinational Business Review, 18, pp. 122.
  • Lee, I. H. and M. R. Marvel (2009). ‘The moderating effects of home region orientation on R&D investment and international SME performance: lessons from Korea’, European Management Journal, 27, pp. 316326.
  • Leonidou, L. C. (1995). ‘Export stimulation research: review, evaluation and integration’, International Business Review, 4, pp. 133156.
  • Leonidou, L. C., C. S. Katsikeas, D. Palihawadana and S. Spyropoulou (2007). ‘An analytical review of the factors stimulating smaller firms to export − implications for policy-makers’, International Marketing Review, 24, pp. 735770.
  • López, L. E., S. K. Kundu and L. Ciravegna (2009). ‘Born global or born regional? Evidence from an exploratory study in the Costa Rican software industry’, Journal of International Business Studies, 40, pp. 12281238.
  • Lu, J. W. and P. W. Beamish (2001). ‘The internationalization and performance of SMEs’, Strategic Management Journal, 22, pp. 565586.
  • Lu, J. W. and P. W. Beamish (2004). ‘International diversification and firm performance: the S-curve hypothesis’, Academy of Management Journal, 47, pp. 598609.
  • Madsen, T. K., E. S. Rasmussen and P. Servais (2000). ‘Differences and similarities between born globals and other types of exporters’, Advances in International Marketing, 10, pp. 247265.
  • Matysiak, L. and A. Bausch (2012). ‘Antecedents of MNE performance: blinded by the obvious in 35 years of literature’, Multinational Business Review, 20, pp. 178211.
  • McDougall, P. P. and B. M. Oviatt (1996). ‘New venture internationalization, strategic change, and performance: a follow-up study’, Journal of Business Venturing, 11, pp. 2340.
  • McDougall, P. P. and B. M. Oviatt (2000). ‘International entrepreneurship: the intersection of two research paths’, Academy of Management Journal, 43, pp. 902906.
  • McGaughey, S. L. (2007). ‘Hidden ties in international new venturing: the case of portfolio entrepreneurship’, Journal of World Business, 42, pp. 307321.
  • Merry, U. (1995). Coping With Uncertainty: Insights from the New Sciences of Chaos, Self-Organization and Complexity. Westport, CT: Praeger.
  • Moen, O. (2002). ‘The born globals: a new generation of small European exporters’, International Marketing Review, 19, pp. 156175.
  • Mort, G. S. and J. Weerawardena (2006). ‘Networking capability and international entrepreneurship − how networks function in Australian born global firms’, International Marketing Review, 23, pp. 549572.
  • Oviatt, B. M. and P. P. McDougall (1994). ‘Toward a theory of international new ventures’, Journal of International Business Studies, 25, pp. 4564.
  • Podsakoff, P. M., S. B. MacKenzie, J. Y. Lee and N. P. Podsakoff (2003). ‘Common method biases in behavioral research: a critical review of the literature and recommended remedies’, Journal of Applied Psychology, 88, pp. 879903.
  • Qian, G. M. and D. Wang (1999). ‘Factors that affect performance of U.S. small and medium sized technology-based enterprises: does multinationality matter?’, Journal of Business and Entrepreneurship, 11, pp. 119131.
  • Qian, G. M., L. Li, J. Li and Z. M. Qian (2008). ‘Regional diversification and firm performance’, Journal of International Business Studies, 39, pp. 197214.
  • Qian, G. M., T. A. Khoury, M. W. Peng and Z. M. Qian (2010). ‘The performance implications of intra- and inter-regional geographic diversification’, Strategic Management Journal, 31, pp. 10181030.
  • Reuber, A. R. and E. Fischer (1997). ‘The influence of the management team's international experience on the internationalization behaviors of SMEs’, Journal of International Business Studies, 28, pp. 807825.
  • Richter, N. (2007). ‘Intra-regional sales and the internationalization and performance relationship’. In A. M. Rugman (ed.), Regional Aspects of Multinationality and Performance, pp. 359381. Oxford: Emerald.
  • Rugman, A. M. (1981). Inside the Multinationals. New York: Columbia University Press.
  • Rugman, A. M. (2000). The End of Globalization: Why Global Strategy is a Myth and How to Profit from the Realities of Regional Markets. London: Random House.
  • Rugman, A. M. (2005). The Regional Multinationals. MNEs and Global Strategic Management. Cambridge: Cambridge University Press.
  • Rugman, A. M. and P. Almodóvar (2011). ‘The born global illusion and the regional nature of international business’. In R. Ramamurti and N. Hashai (eds), Research in Global Strategic Management: The Future of Foreign Direct Investment: Essays in Honour of Yair Aharoni, pp. 265283. Bingley: Emerald Group Publishing.
  • Rugman, A. M. and C. H. Oh (2007). ‘Multinationality and Regional Performance, 2001–2005’. In A. M. Rugman (ed.), Regional Aspects of Multinationality and Performance, pp. 3143. Oxford: Emerald.
  • Rugman, A. M. and C. H. Oh (2010). ‘Does the regional nature of multinationals affect the multinationality and performance relationship?’, International Business Review, 19, pp. 479488.
  • Rugman, A. M. and C. H. Oh (2011). ‘Methodological issues in the measurement of multinationality of US firms’, Multinational Business Review, 19, pp. 202212.
  • Rugman, A. M. and A. Verbeke (2004). ‘A perspective on regional and global strategies of multinational enterprises’, Journal of International Business Studies, 35, pp. 318.
  • Rugman, A. M., C. H. Oh and D. S. K. Lim (2011). ‘The regional and global competitiveness of multinational firms’, Journal of the Academy of Marketing Science, 40, pp. 218235.
  • Rugman, A. M., G. Yip and S. Jayaratne (2008). ‘A note on return on foreign assets and foreign presence for UK multinationals’, British Journal of Management, 19, pp. 162170.
  • Ruigrok, W. and H. Wagner (2003). ‘Internationalization and performance: an organizational learning perspective’, Management International Review, 43, pp. 6383.
  • Ruigrok, W., W. Amann and H. Wagner (2007). ‘The internationalization−performance relationship at Swiss firms: a test of the S-shape and extreme degrees of internationalization’, Management International Review, 47, pp. 349368.
  • Salomon, R. M. (2002). Spillovers to foreign market participants: assessing the impact of exporting and firm heterogeneity on innovative outcomes. Doctoral dissertation, New York University, Graduate School of Business Administration, NY.
  • Siddharthan, N. S. and S. Lall (1982). ‘Recent growth of the largest U.S. multinationals’, Oxford Bulletin of Economics and Statistics, 44, pp. 113.
  • Stoian, M. C., A. Rialp and J. Rialp (2011). ‘Export performance under the microscope: a glance through Spanish lenses’, International Business Review, 20, pp. 117135.
  • Tallman, S. and J. T. Li (1996). ‘Effects of international diversity and product diversity on the performance of multinational firms’, Academy of Management Journal, 39, pp. 179196.
  • Thomas, D. E. and L. Eden (2004). ‘What is the shape of the multinationality−performance relationship?’, Multinational Business Review, 12, pp. 89110.
  • Verbeke, A. (2009). International Business Strategy. Cambridge: Cambridge University Press.
  • Zaheer, S. (1995). ‘Overcoming the liability of foreignness’, Academy of Management Journal, 38, pp. 341363.
  • Zhou, L., B. R. Barnes and Y. Lu (2010). ‘Entrepreneurial proclivity and learning advantages of newness among international new ventures’, Journal of International Business Studies, 41, pp. 882905.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Towards a theory of the M curve
  5. Research methodology
  6. Results
  7. Discussion and conclusions
  8. References
  9. Biographies
  • Paloma Almodóvar is Senior Lecturer in the Business Organization Department at Complutense University of Madrid and Visiting Research Fellow at Henley Business School (University of Reading). She has received five national awards and several scholarships. Dr Almodóvar has been a Visiting Fellow at the Rotman School of Management (University of Toronto), CEIBS (Shanghai) and University of Auckland. Her research interest is international business: regional/global strategies and performance of international firms.

  • Alan M. Rugman is Professor of International Business and Head of the School of International Business and Strategy at Henley Business School (University of Reading). Previously he was at Indiana University, the University of Oxford and Canadian institutions (University of Toronto). He has published over 200 papers in major academic journals as well as over 50 books. He served as President of the Academy of International Business (AIB) from 2004 to 2006; he is the new dean of the Fellows of AIB.