Master Thesis
Major disruptive events and its consequences for SMEs:
An empirical analysis of the internationalization and performance of SMEs
during highly uncertain times.
By
Eyzo J.B. Bakker
18 January 2021
University of Groningen
Faculty of Economics and Business
MSc International Business and Management
Abstract
Recent decades have been marked by a number of disruptive events (e.g., the 2007-2009 global financial crisis and the current COVID pandemic), which have had a strong impact on small and medium-sized enterprises (SMEs) in particular. As such, it becomes pertinent to understand the strategic behavior and performance of these firms. The current article therefore investigates the effect of a major disruptive event on the internationalization strategies of SMEs, and how these internationalization strategies in turn influence the SME’s financial performance. We hypothesize that a major disruptive event significantly decreases the number of international SMEs, and that the degree of internationalization will also be significantly lower. Furthermore, we pose that (1) de-internationalization leads to lower financial performance, (2) internationalization leads to higher performance, and (3) international SMEs outperform domestic SMEs during a major disruptive event. Using a sample of 1,512 Turkish SMEs from the Orbis database for the time period 2016-2019, with the 2018 Turkish currency and debt crisis as the major disruptive event, the empirical analyses offer partial support for the proposed model. Contrary to our expectations, SMEs actually increase their degree of internationalization, showing their immense resilience and entrepreneurial spirit in the face of strong environmental uncertainty. However, there was no significant change in the number of international SMEs across the years. Furthermore, although the analysis provides evidence for a direct and positive effect of the degree of internationalization on firm financial performance, the effect of de-internationalization on firm performance was insignificant. Finally, the results show that international SMEs outperform their domestic counterparts. The study contributes to the existing body of IB and entrepreneurship literature in several ways and provides managers with useful recommendations for the organization of their business during a major disruptive event.
Table of Contents
1. INTRODUCTION ... 5
2. LITERATURE REVIEW ... 9
2.1 The Internationalization Process... 9
2.2 SME Internationalization and Performance ... 12
2.3 Major Disruptive Event and SMEs ... 14
3. HYPOTHESES DEVELOPMENT ... 16
3.1 Disruption as a Threat ... 16
3.2 Disruption as an Opportunity ... 17
3.3 Internationalization and Financial Performance ... 19
4. METHODOLOGY ... 23
4.1 Sample and Data Collection ... 23
4.2 Variables ... 24
4.2.1. Dependent variable ... 24
4.2.2. Independent variables ... 24
4.2.3. Control variables ... 25
4.3. Statistical Analysis and Procedure ... 27
5. RESULTS ... 29
5.1 Descriptive Statistics and Correlation Matrix ... 29
5.2 McNemar’s Test, Wilcoxon Signed-Rank Test, and Welch’s T-test Results ... 31
5.3 Panel Regression Results ... 34
5.5 Robustness Check ... 36
6. DISCUSSION ... 38
6.1 Empirical Evidence and Theoretical Implications ... 38
6.2 Managerial Implications ... 40
6.3 Limitations and Future Research ... 41
7. CONCLUSION ... 43
REFERENCES ... 44
APPENDIX ... 55
Appendix A: Additional Descriptive Statistics, Spearman Correlations, and VIF Values... 55
Appendix B: Frequency Tables of the Number of International Turkish SMEs. ... 57
Appendix C: Frequency Table of the Distribution of Industry Sectors Among Turkish SMEs. ... 59
Appendix D: McNemar’s Distributions ... 60
Acknowledgements
1. INTRODUCTION
Globalization has shaped our world into a more competitive place than ever before (Dabic et al., 2019). Due to rapid technological developments, the ease of doing business, and the globalizing landscape (Caputo, Pellegrini, Dabic, & Dana, 2016), internationalization has become a strategic priority. Consequently, internationalization strategy is one of the most important strategic decisions a manager can make (Brammer, Pavelin, & Porter, 2006). Internationalization is the process in which firms expand to foreign markets and locations (Hitt, Ireland, & Hoskisson, 2007), and has been one of the core issues in international business (IB) literature (Pangarkar, 2008; Sullivan, 1994).
The impact of major disruptive events on firm’s decision making and performance has been the subject of studies for more than a decade. Overall, research has found a negative impact of major disruptions on firm investment (Bernanke, 1983) and performance (Claessens, Djankov, & Xu, 2000; Notta & Vlachvei, 2014). In these past few decades, the vast majority of this research has focused on the effects of such events on large enterprises’ strategic decisions and performance (Athukorala, 2003; Campello, Graham, & Harvey, 2010; Claessens et al., 2000; Johnson, Boone, Breach, & Friedman, 2000). However, by predominantly focusing on the effects of a major disruption on the performance and decisions of large enterprises, the current literature may not have provided us with a true picture. According to surveys from the World Bank, SMEs are actually more heavily affected compared to their larger counterparts. SMEs appear to have a relatively stronger decline in sales, more delayed payments, and less capacity utilization (World Bank Enterprise Surveys, 2020). The evidence that SMEs are affected differently by a disruptive event (compared to their larger counterparts) is not surprising, given their distinct characteristics. They differ fundamentally in terms of organizational structures, processes, resources, and management systems (Carrier, 1994; Lu & Beamish, 2001; Smith, Gannon, Grimm, & Mitchell, 1988). Consequentially, SMEs cannot be viewed as smaller versions of larger businesses (Shuman & Seeger, 1986), which in turn invalidates the concept of generalizing results from studies conducted for large enterprises and applying them to SMEs.
2008). Although the antecedents and outcomes of an SME’s internationalization process are quite well understood under normal/stable circumstances, relatively little is known about this process during a major disruptive event. Hence, in an attempt to fill this gap, the current research aims to shed light on the following questions:
(1) Do major disruptive events increase or decrease the SME’s degree of internationalization?
(2) Does an increase (decrease) in the degree of internationalization lead to an increase (decrease) in the SME’s financial performance?
(3) Do international SMEs outperform domestic SMEs during a major disruptive event?
Based on arguments from the investment and IB literature, we propose that during a major disruption (1) relatively more SMEs stay domestic, and that (2) relatively more international SMEs de-internationalize (i.e., downsize or cease foreign operations) as compared to stable periods. Furthermore, we propose that the latter is likely to negatively impact the SME’s performance as the loss of foreign sales is unlikely to be compensated by the SME’s current domestic sales (Onkelinx, Manolova, & Edelman, 2016a). Finally, we propose that internationalization leads to higher financial performance, as it enables SMEs to achieve benefits from an increased customer base (Lu & Beamish, 2001).
Findings based on the empirical context above reveal that contrary to our predictions, Turkish SMEs show an increase in the degree of internationalization during (2018) and one year after (2019) the currency and debt crisis. Additionally, we find no evidence for a negative relationship between de-internationalization and firm performance. However, the results do provide support for a positive and direct relationship between the degree of internationalization and performance. Finally, international SMEs substantially outperform their purely domestic counterparts.
2. LITERATURE REVIEW 2.1 The Internationalization Process
Internationalization has been defined by numerous scholars in several ways. However, most researchers reach consensus that internationalization is the process in which firms expand to foreign markets and locations (Hitt et al., 2007). As such, in the majority of these definitions, internationalization is viewed as a process in which firms increase their foreign activities. However, once a firm has started the process of internationalization, there is no guarantee that the firm will continue its international operations in the future (Welch & Luostarinen, 1988). As not all firms that internationalize meet their objectives (Onkelinx et al., 2016a), some of them decide to exit foreign markets, or de-internationalize (Benito & Welch, 1997). De-internationalization is the process of reducing a firm’s engagement in foreign activities (Benito & Welch, 1997). Most frequently, firms de-internationalize partially. However, in extreme cases, a firm may decide to withdraw from foreign markets completely. As such, de-internationalization can be partial or complete (Turcan, 2013). Since firms can both increase and decrease their international commitment, we follow Calof and Beamish (1995: 116) who define internationalization as “the process of adapting a firm’s operations (strategy, structure, and resource, etc.) to environments”.
Furthermore, numerous theories have been developed to explain the process of firm internationalization. Although there is a vast body of literature that examines how firms internationalize (e.g., Johanson & Vahlne, 1977; Knight & Cavusgil, 2004; Oviatt & McDougall, 1994), the variety of theories can be divided into two overarching perspectives: incremental and radical perspectives. Whereas the incremental perspectives argue that firms gradually increase their commitment to foreign markets by learning from prior experience (Andersson, Gabrielsson, & Wictor, 2004; Johanson & Vahlne, 1977), the radical perspectives argue that some firms, right from their conception, expand to foreign markets in an effort to obtain superior international business performance (Knight & Cavusgil, 1996). However, not all firms internationalize through an incremental process, and not all firms internationalize radically. Indeed, the paths taken by firms are rarely smooth and often irregular (Welch & Luostarinen, 1988). Consequently, explaining the internationalization process of firms in general cannot be limited to one theoretical perspective as it would provide an inaccurate picture of their internationalization process in reality (Wright, Westhead, & Ucbasaran, 2007). Therefore, the current paper includes the implications of both perspectives. In doing so, we follow Spence & Crick (2006) who propose a holistic adoption of the different approaches towards firm internationalization.
Second, international expansion provides firms with access to better and more (in some cases country-specific) resources (Hennart, 2007). For example, international firms can have access to better technology, cheaper labor, and other (country-specific) resources (Contractor et al., 2003). Third, international firms have an additional option of shifting production and sourcing locations to markets with more favorable conditions, which minimizes the effects of changes in a country’s environment on their operations (Kim et al., 1993). Finally, being present in multiple countries provides firms with learning opportunities (Hennart, 2007). The cultural differences between the firm’s home and foreign country can lead to a more diverse set of ideas (Hitt et al., 1997), and increased decision-making quality (Stahl, Miska, Lee, & De Luque, 2017). Furthermore, they can acquire valuable knowledge and essential skills from working with leading producers and demanding customers overseas (Hitt, Bierman, Uhlenbruck, & Shimizu, 2006). This, in turn, enables them to develop more diverse capabilities than their purely domestic competitors (Hitt et al., 1997; Kim et al., 1993).
Conversely, not all firms that internationalize meet their objectives (Onkelinx et al., 2016a). As such, we see many firms leaving foreign markets (Bonaccorsi, 1992; Ozkan, 2020), with Calof and Beamish (1995) reporting that in 23% of the cases, firms withdrew partially or completely from foreign operations. However, scholars have devoted far less attention to the study of firms withdrawing from foreign markets compared to firms entering foreign markets (Turcan, 2013). As such, there is considerably less evidence for the (negative) relationship between de-internationalization and firm performance. Although they are few in number, there are some studies that investigate this relationship. These studies predominantly point towards a negative impact of de-internationalization on firm performance (Girma, Greenaway, & Kneller, 2003; Gnizy & Shoham, 2018; Onkelinx et al., 2016a). De-internationalization may lead to reduced economies of scale (Ruigrok & Wagner, 2003), lower productivity levels (Girma et al., 2003), and lower employment levels (Onkelinx et al., 2016a). These negative consequences of de-internationalization decrease the firm’s profit levels (Siegfried & Evans, 1994; Wagner, 2008), and may ultimately lead to firm failure (Onkelinx et al., 2016a). Therefore, withdrawal from foreign markets is likely to negatively impact the firm’s financial performance.
2.2 SME Internationalization and Performance
While scholars reach consensus that internationalization is beneficial for many firms, the relationship is less clear-cut for SMEs, primarily due to concerns about their ability to compete internationally and their (predominantly internal) resource constraints (Pangarkar, 2008). Prior literature has shown that SMEs in particular face numerous constraints as they internationalize (Morais & Fereirra, 2020). SMEs often end up experiencing a greater shock when entering a foreign market as compared to larger firms (Carr, Haggard, Hmieleski, & Zahra, 2010), because SMEs are (1) smaller and have considerably less resources (Aldrich & Auster, 1986), (2) have smaller market shares and as a result less brand awareness (Freeman, Carroll, & Hannan, 1983), (3) have little knowledge about how to do business in international markets (Zaheer, 1995), and (4) have more difficulties in claiming key roles in international networks (Johanson & Vahlne, 2009). Furthermore, due to their lack of (financial) resources, establishing international involvement might be difficult for SMEs, as creating and maintaining cross-border operations requires substantial investments (Hsu et al., 2013; Melitz, 2003; Onkelinx, Manolova, & Edelman, 2016b). Besides transportation costs and currency risks, international firms also have to perform market research, adjust products and services to foreign customers’ demands, and set up foreign distribution networks (Onkelinx et al., 2016b). Therefore, if firms desire to internationalize, they need substantial financial resources to overcome the high costs associated with these barriers (Melitz, 2003; Melitz & Ottaviano, 2008).
therefore help SMEs to grow, and can be especially relevant for those SMEs operating in small domestic markets (Barringer & Greening, 1998).
Firm profitability is another outcome of internationalization, and has been widely studied by international business scholars. Most research in this area has found a positive influence of internationalization on SME profitability (Chelliah, Sulaiman, & Yussof, 2010; Kuivalainen & Sundqvist, 2007; Pangarkar, 2008). There are several arguments that explain why SMEs can experience an increase in profitability when they internationalize. First, the larger sales volume associated with internationalization enables SMEs to achieve economies of scale and scope (Martineau & Pastoriza, 2016). This leads to significant cost reductions in their operations and hence positively influence the SME’s subsequent performance in terms of profitability (Lu & Beamish, 2006). Additionally, having a presence in international markets enhances the firm’s market power (Kim et al., 1993). Furthermore, by operating in foreign countries, international SMEs have relatively more and greater learning opportunities, enabling them to develop more diverse capabilities compared to their domestic counterparts (Zahra, Ucbasaran, & Newey, 2009). The additional knowledge and experience accumulated by international SMEs may lead to competitive advantages regarding adaptability to changing consumer preferences (Harrison, Dalkiran, & Elsey, 2000) and the ability to develop and produce high-quality products (George et al., 2005). Although these advantages are not specific to SMEs, they are highly relevant for these firms, as the performance boost (due to the international expansion) is likely to be disproportionally greater for SMEs compared to their larger counterparts (Loth & Parks, 2002).
economies of scale and lower productivity levels, which is likely to decrease the firm’s profitability (Siegfried & Evans, 1994; Wagner, 2008). Finally, it can lower the SME’s adaptability to changes in consumer preferences (Harrison et al., 2000), which has the potential to decrease the SME’s profitability in the long run. Thus, de-internationalizing SMEs generally tend to experience a decrease in their financial performance.
2.3 Major Disruptive Event and SMEs
The majority of IB research, for SMEs and their larger counterparts alike, solely investigate the consequences of increasing foreign commitment (e.g., Contractor et al., 2003; Pangarkar, 2008; Ruigrok & Wagner, 2003). In doing so, they do not consider a crucial note about the internationalization process: “there is no inevitability about its continuance” (Welch & Luostarinen, 1988: 37). Indeed, several studies report that a significant number of international SMEs in reality withdraw from foreign markets (e.g., Calof & Beamish, 1995; Ozkan, 2020). Although there is empirical evidence for the de-internationalization of SMEs (and firms in general), this research area remains relatively unexplored. This is rather surprising given (1) the significant drop-out rate of international firms (Calof & Beamish, 1995), which appears to be particularly high for SMEs (Bonaccorsi, 1992), and (2) the growing body of literature concerning the relationship between internationalization and firm performance – which has been a key part of IB research for more than a decade (Matlay et al., 2006; Schwens et al., 2018).
and the firms operating in them (Belke, Dubova, & Osowski, 2018; Smallbone et al., 2012). Recent examples include the 2014 Russian financial crisis, the 2015 Chinese stock market crash, and the Turkish currency and debt crisis in 2018. Such disruptions compromise both domestic and international trade, and they are often characterized by a strong increase in economic uncertainty (Bloom, 2009). Hence, a major disruption has the potential to severely impact the internationalization strategies of SMEs.
The current research aims to address the aforementioned gaps by investigating the following set of research questions:
(1) Do major disruptive events increase or decrease the SME’s degree of internationalization?
(2) Does an increase (decrease) in the degree of internationalization lead to an increase (decrease) in the SME’s financial performance?
3. HYPOTHESES DEVELOPMENT
3.1 Disruption as a Threat
Although internationalization can provide firms with several benefits, there are also substantial costs associated with creating and maintaining cross-border operations (Hsu et al., 2013; Onkelinx et al., 2016b). Consequently, internationalization requires substantial investment (Melitz, 2003). However, acquiring the (internal or external) financial resources to make such investments when there are exogenous shocks might be troublesome, especially for SMEs. First, obtaining a bank loan or finding investors (i.e., external financial resources) might be challenging, as disruptive events tend to decrease corporate investment (Gaiotti, 2013; Gulen & Ion, 2016) and shrink bank loans to business (Francis, Hasan, & Zhu, 2014). The lack of external financial resources may be especially problematic for SMEs, as they typically depend heavily on external capital; not only for development and investment, but also for maintaining operations – even more so than their larger counterparts (Abor & Quartey, 2010; Levy et al., 1999). Due to the lack of external capital, most SMEs are unlikely to internationalize during a major disruptive event.
Secondly, most SMEs do not possess the internal resources for investment and growth, given their liability of smallness (Nakos & Brouthers, 2002). Therefore, investing the SME’s little internal funds in the pursuit of international expansion might not be deemed appropriate by the owner/manager, mainly because these investments tend to be specific and difficult to redeploy (Maekelburger, Schwens, & Kabst, 2012). Since financial resources are scarce during a major disruption, preserving internal funds becomes even more vital for the survival of the SME. Therefore, the optimal decision for firms facing high levels of uncertainty might be to cut their investments and increase their cash holdings in preparation for the period in which conditions (to invest) become more favorable (Bernanke, 1983; Bloom, Bond, & Van Reenen, 2007; McDonald & Siegel, 1986). Because internationalization requires substantial investments (Hsu et al., 2013), SMEs are likely to preserve their internal funds in this regard.
international environment and perceives it as relatively stable, the firm may decide to increase its foreign commitment (Figueira-de-Lemos & Hadjikhani, 2014). On the contrary, if the firm views the environment as unstable, it is likely to respond to this uncertainty by decreasing foreign commitment (Gnizy & Shoham, 2018). As such, environmental changes can severely impact the internationalization strategies of firms.
Indeed, research has shown that SMEs in particular are strongly impacted by changes in their environment (Kuivalainen et al., 2012). A major disruption might cause the SME’s owner/manager to change the internationalization strategy as a consequence of an increase in the perceived uncertainty (Liesch et al., 2011). Research has found that in response to a highly uncertain environment, firms generally decide to divest (Bernanke, 1983; Bloom et al., 2007; Boddewyn, 1979). As operating across borders requires substantial investments (Onkelinx et al., 2016b), and major disruption are characterized by high levels of uncertainty (Mody, Ohnsorge, & Sandri, 2012), SMEs are likely to decrease their international involvement, and might even withdraw from foreign operations completely. Thus, SMEs are likely to de-internationalize when they experience (or perceive) their environment as highly uncertain.
In conclusion, due to high levels of uncertainty and the decreasing availability of both internal and external financial resources during a major disruptive event, we expect (1) SMEs that were planning to internationalize to stay domestic and (2) the internationalized SMEs to divest in their foreign operations, and thus de-internationalize.
Hypothesis 1a. The proportion of internationalized SMEs is likely to fall post disruptive event.
Hypothesis 1b. Internationalized SMEs are likely to de-internationalize (or downscale their exports) post disruption.
3.2 Disruption as an Opportunity
income deferrals, tax deferrals, and wage subsidies (OECD, 2020). Furthermore, European governments appear to be stimulating and supporting internationalization among SMEs in particular. Indeed, the European Commission has launched two projects in 2016 with the intention to help SMEs internationalize and provide them with concrete assistance in this process (European Commission, 2020). Because SMEs receive several forms of government support during a major disruption, including information on how to internationalize successfully, they could actually decide to increase their international involvement during a major disruptive event.
Furthermore, SMEs are embedded in a network of actors, including customers, suppliers, competitors and public and private support organizations (Coviello & Munro, 1995). As globalization has led to a removal of (trade) barriers (Hitt, Ireland, Camp, & Sexton, 2001), several SMEs are now joining international networks, which are providing them with an increased array of openings and opportunities (Dana, 2001). During periods of disequilibrium and economic uncertainty, several large firms and public sector organizations withdraw from markets (Acs & Storey, 2004; Grilli, 2011). This provides an opportunity for the best entrepreneurs to recognize the opening and, in turn, exploit this market gap (Cowling et al., 2015). Because SMEs are inherently entrepreneurial and flexible (Smallbone et al., 2012), its managers are most likely to recognize and seize this opportunity (Cowling et al., 2015; Grilli, 2011). Since international SMEs are operating in an international network, they are (compared to domestic SMEs) more likely to recognize the market exit of large firms as an opportunity (Matlay et al., 2006), and could increase their international involvement accordingly.
they are able to provide a better service, particularly to their larger counterparts (Pangarkar, 2008). As such, SMEs might internationalize, even in the face of a major disruption.
Thus, during a major disruption, we see several firms exiting markets, which in turn creates opportunities for firms to internationalize. Due to their entrepreneurial nature, SMEs are most likely to recognize an opportunity when it presents itself. As such, SMEs are most likely to seize the opportunity to internationalize. Especially in combination with the additional governmental support SMEs receive during external shocks and the pivotal role they play in (international) supply chains, some SMEs could take advantage of the situation and, as such, may very well decide to internationalize. Hence, we argue the following:
Hypothesis 2a. The proportion of internationalized SMEs is likely to rise post disruptive event.
Hypothesis 2b. Internationalized SMEs are likely to further internationalize (or increase their exports) post disruption.
3.3 Internationalization and Financial Performance
So far, it has been explained how SMEs are likely to respond to a major disruptive event. We predict that domestic SMEs withhold from internationalizing, and that internationalized SMEs de-internationalize during an external shock. Alternatively, we propose that domestic SMEs actually internationalize, and that international SMEs internationalize even further during a major disruption. The current part builds on the aforementioned and will, consequently, investigate how different internationalization strategies – i.e., internationalizing and de-internationalizing – impact the SME’s financial performance. In doing so, we will primarily focus on the financial performance of (de-)internationalizing SMEs.
recover the loss of foreign sales in the short run (Onkelinx et al., 2016a), these firms tend to have lower productivity and employment levels (Girma et al., 2003), which in turn are likely to decrease the firm’s profitability (Siegfried & Evans, 1994; Wagner, 2008). Furthermore, due to higher margins, foreign sales often yield better returns than domestic sales (Manova & Zhang, 2012). Losing these sales will therefore result in a decrease in the firm’s financial performance (Onkelinx et al., 2016a). This is especially likely during a major disruptive event as local demand and the resulting sales tend to decline when economic uncertainty is high (Bloom, 2014). Other firm-level consequences from exiting a foreign market include (1) an increase in the market share of existing competitors, (2) an opportunity for new firms to enter that market, and (3) a potential loss of the firm’s reputation (Karakaya, 2000). These consequences lower the competitive position of the focal firm (Li, Zhou, & Shao, 2009), and could therefore negatively impact the firm’s subsequent performance. Because internationalizing entails several negative implications, de-internationalizing SMEs are likely to experience a decline in their financial performance.
Finally, having an additional customer base (i.e., foreign consumers) can be very relevant as it enables firms to sell products to a larger audience (Lu & Beamish, 2001). Especially during a major disruptive event, having a larger customer base can be very useful because economic uncertainty tends to decrease consumption levels (Giavazzi & McMahon, 2012). As such, firms that de-internationalize are likely to report lower performance since they will not have access to these extra consumers.
In conclusion, we expect SMEs that de-internationalize (i.e., who downscale their international operations or exit from foreign markets completely) to experience a decline in performance, leading us to hypothesize the following:
Hypothesis 3a. De-internationalization will lead to lower SME financial performance.
Because of their entrepreneurial nature (Smallbone et al., 2012), some SMEs might recognize this opportunity for international expansion, and internationalize (Cowling et al., 2015), providing them with a number of potential benefits. First, international firms have a broader customer base, which enables them to achieve larger production volumes and grow (Lu & Beamish, 2001). Secondly, international expansion provides firms with access to better and more (in some cases country-specific) resources (Hennart, 2007). Third, international firms have an additional option of shifting production and sourcing locations to markets with more favorable conditions, which minimizes the effects of changes in a country’s environment on their operations (Kim et al., 1993). Finally, being present in multiple countries provides firms with learning opportunities (Hennart, 2007; Hitt et al., 2006). Although these advantages are not specific to SMEs, and indeed apply to firms of all sizes, they are highly relevant for these firms. The performance boost (due to the international expansion) is likely to be disproportionally greater for SMEs compared to their larger counterparts (Loth & Parks, 2002), which is why the aforementioned advantages may be particularly relevant for SMEs. There are however, a few additional advantages associated with internationalizing which apply primarily to SMEs. First, by internationalizing, SMEs might be able to provide their larger multinational clients (i.e., multinational enterprises) with better service (Pangarkar, 2008) and so these SMEs can gain additional business from these clients and increase their financial performance consequentially (Hennart, 2020). Furthermore, due to their size and flexibility, SMEs are likely to be more efficient and more adaptable in redeploying their resources than their larger counterparts. Whereas the structures of large enterprises are often characterized by strong control systems, cumbersome planning, and inertia (Carrier, 1994), the structures of SMEs are generally viewed as flexible, adaptable, and simple (Levy & Powell, 1998). As such, research has found that SMEs are able to respond and adapt relatively quickly to external shocks (Grilli, 2011; Smallbone et al., 2012). Because of their flexibility, SMEs might be best suited to respond to the internationalization opportunities, created by the major disruption, in an effective manner, and as such are likely to reap relatively more benefits when they internationalize during such an external shock.
demand and general consumer spending (Giavazzi & McMahon, 2012), these internationalization benefits may be of particular relevance for the SME’s performance, and ultimately, its survival. Hence, we hypothesize the following:
Hypothesis 3b. A higher degree of internationalization will lead to higher SME financial performance.
Hypothesis 3c. International SMEs will have higher performance than purely domestic SMEs.
In conclusion, major disruptions appear to have a significant impact on firms, and SMEs in particular. However, how this shapes their strategic decisions and their subsequent financial performance remains rather unclear, and underlines the need for empirical analyses. Therefore, the current research aims to investigate the conceptual model below (see Figure 1.).
4. METHODOLOGY 4.1 Sample and Data Collection
To test whether a major disruptive event influences an SME’s internationalization strategy and whether, in turn, a different internationalization strategy subsequently affects the SME’s financial performance, this study draws on SME data from the Turkish currency and debt crisis in 2018. According to the European Commission (2003), SMEs are firms which have at least two out of the following three characteristics: they have less than 250 employees, have a yearly turnover equal to or less than €50 million, and have a balance sheet total equal to or less than €43 million. Only firms with at least two of these characteristics were included. More specifically, firm-level data was collected from Turkish SMEs for the period 2016-19. This sample covers data from two years before the major disruptive event (2016), one year before the major disruptive event (2017), during the major disruptive event (2018), and one year after the major disruptive event (2019). We chose to study a sample of Turkish SMEs, which were operating during their country’s 2018 currency and debt crisis, for several reasons. First and foremost, this crisis represented a significant interference of the regular flow of goods and services within, from and to Turkey (Arbaa & Varon, 2019), which categorizes it as a major disruptive event. Furthermore, Turkish SMEs account for 73.9% of overall employment levels in Turkey and have a value-added share of 53.9% of all firms (European Commission, 2019). This shows that these firms are essential for job creation, growth, productivity, and competitiveness and as such are tremendously important for the Turkish economy (OECD, 2016), which makes studying Turkish SMEs particularly relevant and highly representative.
firms. However, after close inspection of the entire dataset, we discovered that the majority of our firms (20621 to be exact) had missing data and were therefore excluded from further analyses. Furthermore, there were also several subsidiaries – which were part of a larger multinational corporation – included in the original sample. Since the current study’s sole focus is on SMEs, these firms, although relatively few in number (87), were also excluded from our final sample. Consequently, our final sample includes data on 1512 Turkish SMEs.
4.2 Variables
4.2.1. Dependent variable
To measure SME financial performance, we used an accounting-based measure of firm performance: return on assets (ROA). We adopted this measure for two reasons. First, because firm internationalization is associated with firms trying to achieve economies of scope and scale, looking at the firm’s ROA gives a good indication of how well they are able to realize these efforts (Kim, Hwang, & Burgers, 1989). Second, by using the same performance measure as previous work (e.g., Hsu et al., 2013; Javalgi & Todd, 2011; Lu & Beamish, 2001; Majocchi & Zucchella, 2003; Sardo, Serrasquiero, & Alves, 2018), this study’s results can be directly compared with prior literature.
4.2.2. Independent variables
will focus on the ratio of export revenues/total revenues as a proxy for the degree of internationalization of Turkish SMEs.
Additionally, we measured the de-internationalization of Turkish SMEs in 2018 and in 2019. In line with Onkelinx et al. (2016a), we created a dummy variable as a proxy for de-internationalization. The variable was computed by subtracting the average of the degree of internationalization in 2016 and 2017 from the degree of internationalization in 2018 (i.e., De-internationalization 2018 = DOI 2018 – ((DOI 2016 + DOI 2017)/2)). The same method was used for 2019 (i.e., De-internationalization 2018 = DOI 2019 – ((DOI 2016 + DOI 2017)/2)). The firms with a negative value were assigned a ‘1’, the firms with a neutral or positive value were assigned a ‘0’. As this study focuses on the de-internationalization of firms in 2018 and in 2019, only dummy variables were created for 2018 and 2019.
4.2.3. Control variables
We included several control variables. The first control variable is firm size. This control variable was included for two reasons. First, research has shown that larger firms tend to achieve higher financial performance than small firms (Doğan, 2013). Second, recent data from the World Bank has confirmed that smaller firms experience more difficulties during a major disruptive event as compared to their larger counterparts (World Bank Enterprise Surveys, 2020). Consequently, larger firms generally perform better during such an external event. To account for this possible bias in our sample, we controlled for firm size. Firm size will be measured through SME turnover transformed by the natural logarithm, following prior research (Hsu et al., 2013; Lu & Beamish, 2004).
organizations and bodies’. As the current study’s sample includes Turkish (i.e. European) firms only, it is a valid method to classify the SMEs in our sample. To control for firm industry, 16 dummy variables were created, taking the value of ‘1’ if the SME operated in that industry and ‘0’ if otherwise. The current set-up implies that the ‘Other service activities’ industry sector is the reference group.
Another control variable included in this study is prior financial performance. This variable was included because a firm’s previous performance can act as an anchor for a firm’s future performance (Riahi‐Belkaoui, 2003). Specifically, the Turkish SME’s performance in 2017 (measured in terms of the firm’s ROA) was used as control variable for its subsequent financial performance in 2018 (also measured in terms of the firm’s ROA), which is in line with the methodology used by Baer and Frese (2003).
Furthermore, in line with previous research investigating the relationship between internationalization and performance, we controlled for firm age (Onkelinx et al., 2016b). Older firms may be less flexible to adapt to strong environmental changes and may not recognize opportunities when they are presented (Love, Roper, & Zhou, 2016). Furthermore, Cowling, Liu, and Zhang (2018) have shown that younger firms are more resilient to a major disruption, and recover more quickly compared to their older counterparts. As such, older firms might act suboptimal during a major disruptive event which could lead to lower financial performance. Therefore, we also controlled for firm age. Firm age was measured in the number of years since the business was established (Javalgi & Todd, 2011). For each year, the year of incorporation was subtracted from the year of reference.
Next, we included prior exports as a control variable for financial performance. Internationalization is a process that takes time and resources (Lu & Beamish, 2004). As such, performance benefits of internationalization can be lagged (Lu & Beamish, 2001). In other words, internationalization strategies of an SME in time ‘t’ could influence its performance in t+1. As such, prior DOI is included as a control variable for ROA.
each year in our sample. Currency exchange rates were collected through the European Central Bank (2020) database. The exchange rate for 2016, 2017, 2018, and 2019 are 3.433, 4.121, 5.707, and 6.357 respectively, in which the euro is taken as the reference currency for the Turkish lira.
4.3. Statistical Analysis and Procedure
To test the hypotheses in the current research, several analytical tests were conducted. To analyze whether the proportion of internationalized SMEs decreases (H1a) or increases (H2a) during a major disruptive event, we used McNemar’s one degree of freedom chi-square test. This test is useful when testing for the equality of proportions in pairs of matched, binary data (Eliasziw & Donner, 1991). To that end, we first had to create a dummy variable for internationality, as we are interested in testing whether there is a statistically significant difference between the number of international SMEs one year before the major disruption (2017), during the major disruption (2018), and one year after the major disruption (2019). The dummy variable takes a value of ‘0’ when the firm has a score of 0.00 on the degree of internationalization (and is thus ‘not international’) and a ‘1’ when the firm has a positive value on the degree of internationalization (i.e., > 0.00 and is thus ‘international’). As the data regarding this hypothesis is binary and paired, McNemar’s test is appropriate to use for these hypotheses. The test will show whether there is a statistically significant difference in the number of international SMEs between two years.
Furthermore, to test hypothesis 1b (internationalized SMEs are likely to de-internationalize post disruption) and 2b (SMEs are likely to internationalize post disruption), we tested whether the mean of the two paired groups in our sample is statistically significantly different from zero. Because the variable of interest (degree of internationalization) is not normally distributed, we used a non-parametric test: the Wilcoxon signed-rank test. This test is a useful alternative when the difference between two sample means cannot be assumed to be normally distributed (McDonald, 2014).
To determine whether a fixed or random effects regression was appropriate, we performed the Hausman test. This test was performed for the relationship between de-internationalization and performance (hypothesis 3a) and between internationalization and performance (hypothesis 3b). The test proved significant for the former (p = 0.035) and insignificant for the latter (p = 0.975) at the 0.05 level. As such, we performed a fixed and random effects regression analysis for hypothesis 3a and 3b respectively. Additionally, we tested for heteroscedasticity in the data. To that end, we performed a modified Wald test for the fixed effects regression analysis and a robust test for the random effects regression analysis. Wald’s test proved significant (p = 0.000) at the 0.001 level, pointing towards heteroscedasticity in the data. Furthermore, the robust test for the random effects regression also pointed to heteroscedasticity in the data. As such, we used heteroscedasticity robust standard errors for both analyses.
Furthermore, to increase the validity of our results, we performed several additional regression analyses which can be found in Appendix E. For the robustness test of hypothesis 3a, we performed an analysis of covariance (ANCOVA). Furthermore, to test for the robustness of hypothesis 3b we performed a multiple linear regression analysis. Both models are types of regression models, and therefore need to fulfill several key assumptions (Oakes & Feldman, 2001). As the data shows (1) heteroscedasticity issues and (2) a non-normal distribution, the key assumptions are not satisfied. Therefore, robust analyses were performed for both additional regression models.
5. RESULTS 5.1 Descriptive Statistics and Correlation Matrix
First, we will elaborate on the characteristics of the firms in our sample. Table 1a provides the descriptive statistics for the Turkish SMEs in our sample in 2016, Table 1b for 2017, Table 1c for 2018, and Table 1d for 2019. Additional information about the firms in our sample, the number of international firms in 2016, 2017, 2018, and 2019, and the distribution of firms across the NACE industries can be found in Appendix A, B, and C respectively. The total sample includes data from 1512 Turkish SMEs, from all years mentioned above. Out of the 1512 SMEs in our sample, 539 had an international presence in 2016, 563 in 2017, 575 in 2018, and 582 in 2019. Of the 563 SMEs with an international presence in 2017, 299 internationalized in 2018, and 308 de-internationalized in 2019. Finally, the SMEs in our sample predominantly operated in the ‘wholesale and retail trade; repair of motor vehicles and motorcycles’ and the ‘manufacturing’ sectors. These two industries together accounted for 84.59% of the total number of firms in our sample (see Appendix C).
Additionally, the tables below (Table 1a, 1b, 1c, and 1d) include information on the correlation and variance inflation factors (VIFs) of the variables in our study. Because some variables were non-normally distributed, we used the Spearman Rank correlation analysis. Furthermore, industry dummies were excluded from the tables for the sake of clarity and brevity. As can be seen in the tables below, ROA is positively correlated with DOI in 2017, 2018, and 2019. All correlations were below the commonly used cut-off threshold of 0.7 (Dikova, Sahib, & Van Witteloostuijn, 2010). Furthermore, to test for multicollinearity among the regressor variables, we computed additional diagnostics: the VIF values. The VIF values are reported for each year. Since the VIF values are below the widely used threshold of 10, there appears to be no issue of multicollinearity in the dataset (Olibe, Michello, & Thorne, 2008).
Table 1a. Descriptive Statistics and Correlations for 2016.
Table 1b. Descriptive Statistics and Correlations for 2017.
Variables Min Max Mean Std. Dev. 1. 2. 3. 4. VIF
1. ROA -64.56 75.95 4.33 7.98 - -
2. DOI 0.00 100.00 9.69 23.33 .152* - 1.85
3. Firm size 1.11 11.41 8.18 1.45 .062* .313* - 2.80
4. Firm age (years) 1.00 68.00 15.91 9.91 -.027 .067* .250* - 1.06 Note: N=1,512; * p < 0.05
Table 1c. Descriptive Statistics and Correlations for 2018.
Variables Min Max Mean Std. Dev. 1. 2. 3. 4. VIF
1. ROA -89.02 77.65 3.27 10.33 - -
2. DOI 0.00 100.00 9.98 23.45 .162* - 1.90
3. Firm size 1.17 11.94 8.14 1.43 .043 .354* - 2.47
4. Firm age (years) 2.00 69.00 16.91 9.91 -.040 .056* .227* - 1.05 Note: N=1,512; * p < 0.05
Table 1d. Descriptive Statistics and Correlations for 2019.
Variables Min Max Mean Std. Dev. 1. 2. 3. 4. VIF
1. ROA -53.19 63.48 3.88 8.66 - -
2. DOI 0.00 100.00 10.55 24.04 .195* - 1.92
3. Firm size 2.08 12.16 8.16 1.46 .149* .376* - 2.30
4. Firm age (years) 3.00 70.00 17.91 9.91 .001 .067* .203* - 1.04 Note: N=1,512; * p < 0.05
Variables Min Max Mean Std. Dev. 1. 2. 3. 4. VIF
1. ROA -93.56 61.24 3.46 9.81 - -
2. DOI 0.00 100.00 8.92 22.29 .048 - 2.01
3. Firm size 0.815 10.80 8.08 1.50 -.005 .306* - 2.70
5.2 McNemar’s Test, Wilcoxon Signed-Rank Test, and Welch’s T-test Results
To test whether there was a significant difference in the number of international firms before the major disruptive event (i.e., between 2017 and 2016) during the major disruptive event (i.e., between 2018 and 2017) and one year after the major disruptive event (i.e., between 2019 and 2018 and between 2019 and 2017), we performed McNemar’s one degree of freedom chi-square test. The results of this test are shown in Table 2. A more detailed version of the most relevant test results can be found in Appendix D. As the results show, we find no support for hypothesis 1a or 2a. Although there is a positive and significant difference in the number of international SMEs
between 2017 and 2016 (2 = 4.57, d.f. = 1, p = 0.033), the analyses display that the difference in
the number of international firms between 2018 and 2017 is positive (2 = 1.11, d.f. = 1) but
insignificant (p = 0.293), as are the differences between 2019 and 2018 (2 = 0.37, d.f. = 1, p =
0.541) and between 2019 and 2017 (2 = 2.09, d.f. = 1, p = 0.149). This implies that during (2018)
and shortly after (2019) the major disruptive event in Turkey (in 2018), the number of international Turkish SMEs did neither significantly increase nor decrease.
Furthermore, to test hypothesis 1b and 2b, we performed the Wilcoxon signed-rank test. Because the distribution of the difference in the degree of internationalization between the years in our sample is not normal, the Wilcoxon singed-rank test is the most appropriate. Results from this test are reported in Figure 2. Contrary to our main prediction (hypothesis 1b), but in line with the alternative prediction (hypothesis 2b), the Wilcoxon test shows that SMEs in 2018 (i.e., during the major disruptive event) were significantly more internationalized at the 0.001 level. This was reflected in a higher degree of internationalization (mean DOI 2018 = 9.98, mean DOI 2017 = 9.69; p = 0.000). Additionally, the same SMEs were even more internationalized in the next year, 2019 (i.e., shortly after the major disruptive event). This difference was significant at 0.05 level, reflected in a higher degree of internationalization (mean DOI 2019 = 10.55, mean DOI 2018 = 9.98, p = 0.035). Furthermore, the empirical analyses show that the average DOI has only increased over the years: from an average DOI of 8.92 in 2016 to an average DOI of 10.55 in 2019. As such, we find convincing support for hypothesis 2b.
Table 2. McNemar’s Test. 2016 2017 2018 2019 2016 – 2017 4.57* – 2018 8.20** 1.11 – 2019 10.22** 2.09 0.37 –
Note: N=1,512; * p < 0.05, ** p < 0.01; Chi-square statistics are reported.
Figure 2. Wilcoxon Signed-Rank Test
Note: Wilcoxon signed-rank test for internationalization is significant at 0.01, 0.001 and 0.05 level respectively.
Figure 3. Welch’s T-test
5.3 Panel Regression Results
The results of the panel regression are reported in Table 3 and Table 4. The upper parameter shows the coefficient including the significance level, and the lower parameter shows the standard error. As can be seen in the tables below, several control variables were excluded. Due to the nature of the data, prior performance and prior internationalization (both measured in t-1) could not be included. Additionally, industry dummies could not be included because of the nature of the analysis. As panel models are designed to study the causes of change within a particular variable (firm performance in this case), including a time-invariant/constant variable (e.g., industry sector) is not possible as it is constant for each observation (i.e., for each firm in every year). It therefore results in collinearity, which is why industry dummies were omitted. Furthermore, due to heteroscedasticity in the data, robust analyses were performed for both the random and fixed effects regressions.
Table 3. Panel Regression for ROA with Fixed Effects
Variables Model 1 Model 2
Independent variable De-internationalization 0.353 (0.746) Control variables Firm size 2.936*** (0.379) 2.119* (0.066) Firm age 1.679*** (0.412) 0.582* (0.234) Exchange rate -1.654*** (0.388) - Industry NO NO Constant F-value R-squared Observations -39.616*** (5.649) 28.48*** 0.023 1,512 -23.905** (7.377) 6.30*** 0.038 1,512
Note: Independent and control variables are from 2018. Industry sector is omitted due to collinearity. Robust results are reported.
*** p < 0.001, ** p < 0.01, * p < 0.05, + p < 0.1
Table 4. Panel Regression for ROA with Random Effects
Variables Model 1 Model 2
Independent variable Degree of internationalization (DOI) 0.036** (0.011) Control variables Firm size 0.914*** (0.140) 0.790*** (0.144) Firm age -0.040* (0.019) -0.038* (0.019) Exchange rate -0.016 (0.083) -0.032 (0.083) Industry NO NO Constant F-value R-squared Observations -2.976* (1.239) 44.27*** 0.023 1,512 -2.265* (1.265) 53.41*** 0.032 1,512
Note: Independent and control variables are from 2018. Industry sector is omitted due to collinearity. Robust results are reported.
5.5 Robustness Check
Several robustness checks were performed for the linear models reported in the tables above. The first robustness checks are already included in the subsection ‘Statistical Analysis and Procedure’ in order to establish the correct methodology. Results of the additional tests can be found in Appendix E. Table E1 and Table E2 show the regression results of the variables in our study for 2018 (t) and 2019 (t+1). Due to heteroscedasticity, robust analyses were performed for both years. Table E1 and E2 show the results for both the regression and ANCOVA analysis in 2018 and 2019 respectively. All industries were included in Model 1 to Model 4. Exchange rate, however, was omitted due to collinearity. Model 1 only includes the regression results of the dependent variable and the control variables. Model 2 includes both the de-internationalization dummy and the control variables. As such, the ANCOVA methodology is applied in Model 2. Model 3 includes the regression results for the degree of internationalization as well as the control variables. Additionally, Model 4 shows the regression results after including both the de-internationalization dummy and the degree of de-internationalization. Finally, Model 5 and Model 6 are structured similarly to Model 2 and Model 3 respectively, but do not include industry 3 and industry 7 (which accounted for 84.59% of our data). Accordingly, we could check whether our results in the robustness test were driven by a particular industry sector. The tables show that de-internationalization is a negative but insignificant predictor of firm performance in 2018 (Table E1 Model 2; = -0.067, p = 0.921) and in 2019 (Table E2 Model 2; = 0.367, p = 0.449). Thus, we find no empirical support for hypothesis 3a in 2018 nor in 2019, in line with our main analysis. Furthermore, the tables show that the degree of internationalization is a positive but insignificant predictor of firm performance in 2018 (Table E1 Model 3; = 0.035, p = 0.264) as well as in 2019 (Table E2 Model 3; = -0.003, p = 0.880). As such, we do not find empirical support for hypothesis 3b in 2018 nor in 2019 in this robustness test. Furthermore, the results from Model 5 and Model 6 in both Table E1 and Table E2 show that these results, which remained insignificant, are not driven by a particular industry in our sample.
short-term impact of such an event on an increased sample of firms. Results are reported in Table E3 (ANCOVA) and Table E4 (regression) in Appendix E. Contrary to the results in Table 3 and Table 4, Table E3 does provide evidence for the negative impact of de-internationalization on firm performance (Table E3 Model 2; = -0.587, p = 0.029) at the 0.05 level. Additionally, Table E4 provides evidence for the positive impact of a firm’s internationalization on its subsequent financial performance (Table E4 Model 2; = 0.049, p = 0.001) at the 0.01 level.
6. DISCUSSION
The current research investigates the internationalization and performance of SMEs during a major disruptive event. In order to do so, this study draws on a sample of 1,512 Turkish SMEs in the years 2016 (two years before the crisis), 2017, (one year before the Turkish currency and debt crisis), 2018 (during the Turkish currency and debt crisis) and 2019 (one year after the crisis). The results of this study broaden the general IB literature by focusing on both decreasing and increasing forms of internationalization, and adds valuable insights on this particular process during times when SMEs experience severe external uncertainty. The interpretation of the results and its implications for theory, the managerial implications, the limitations, and the directions for future research will be discussed below.
6.1 Empirical Evidence and Theoretical Implications
The first set of competing hypotheses predicted that a major disruptive event would lead to a decrease (H1a) (or alternatively, increase: H2a) in the number of international SMEs. However, empirical evidence from this study shows that there is no statistically significant difference in the number of international firms in 2018 nor in 2019 (compared to 2017). Furthermore, the competing hypotheses 1b and 2b predicted that a major disruptive event would lead to a de-internationalization of SMEs (1b) and to an internationalization of SMEs (2b). Contrary to our main prediction (1b), but in line with the alternative hypothesis (2b) we find that firms actually internationalized. The results of hypothesis 2b provide evidence for the resilience of SMEs. This finding is echoed by Ter Wengel and Rodriguez (2006), who show that the SME sector can actually grow during a crisis. Especially during times characterized by strong environmental uncertainty, SMEs in particular appear to recognize, identify, and exploit business opportunities when they present themselves. This underlines the general entrepreneurial nature of SMEs (Smallbone et al., 2012), and their incredible importance for a country’s development (Matlay et al., 2006) and employment levels (OECD, 2016). As such, SMEs, particularly during and shortly after a major disruptive event, function as the backbone of our economy (Cowling et al., 2015; Grilli, 2011).
de-internationalization on firm performance (Onkelinx et al., 2016a), we do not find convincing evidence for this relationship during a major disruptive event. As such, it is premature to conclude that de-internationalization leads to a decrease in the SME’s financial performance during a major disruption. There are several theoretical explanations as to why a de-internationalization does not necessarily lead to a decrease in financial performance. SMEs, and especially international SMEs, need to adapt to their constantly changing environment (Kuivalainen et al., 2012). This is a phenomenon called uncertainty acclimatization (Liesch et al., 2011). As a major disruptive event is likely to increase the uncertainty perceived by the SME’s manager(s), a logical response would be to internationalize. Although there is no empirical evidence for a significant increase in internationalizing SMEs during a major disruption, there were indeed some SMEs that de-internationalized. These firms may have done so to decrease the uncertainty in their environment. If such an adaptation is done effectively and with caution, de-internationalization does not necessarily have to lead to lower performance as it can enable the firm to restructure its current (domestic) operations (Gnizy & Shoham, 2018) as well as free up funds for its key activities (Campello et al., 2010).
such, the SMEs that recognize and exploit this opportunity, and thus internationalize, appear to achieve higher financial performance.
Additionally, this study provides evidence that international SMEs outperform their domestic counterparts (hypothesis 3c). These results hold true both during (2018) and one year after (2019) the major disruptive event. Again, these results point towards the resilience and entrepreneurial nature of SMEs (Smallbone et al., 2012), and to the strong positive performance implications of internationalizing (Pangarkar, 2008), especially during uncertain times. Furthermore, the empirical evidence supports the notion that SMEs are highly flexible (Levy & Powell, 1998), and that these firms in particular are able to recover relatively quickly after a major disruptive event (Cowling et al., 2018).
In sum, the empirical results discussed above provide several contributions to the IB literature. First, the current study adds valuable insights to a research area that is central to the IB literature: between the degree of internationalization and firm performance (Pangarkar, 2008). Furthermore, it adds new information to this research field by not only focusing on increasing forms of internationalization (i.e., internationalizing) but also on decreasing forms of internationalization (de-internationalizing). Third, by investigating the (de-)internationalization – performance relationship during a major disruptive event, the current study provides valuable insights about the (de-)internationalization strategies of SMEs during highly uncertain times, which is a research area that remains relatively understudied (Cowling et al., 2015; Saradakis, 2012).
6.2 Managerial Implications
Furthermore, although the results indicate that de-internationalization does not necessarily lead to a decrease in performance in the short run, its impact on long-term performance is less clear. Conversely, Onkelinx et al. (2016a) show that de-internationalization negatively impacts the SME’s financial performance, both short- and long-term, during stable environmental circumstances. As such, we would advise managers of SMEs to stay international, as long as the cost of maintaining the international operations can be recovered from the foreign sales.
Naturally, there are some caveats that need to be taken into consideration when looking at the aforementioned arguments. Internationalization is a costly process (Lu & Beamish, 2004), which takes substantial time, effort, and coordination to embark upon and maintain (Hsu et al., 2013; Onkelinx et al., 2016b). Following this, it becomes apparent that internationalization in itself is not a guarantee for ‘success’ or financial performance. However, if managed effectively, this process can help SMEs to achieve significantly higher performance, enabling them to thrive during times when other firms struggle to survive.
6.3 Limitations and Future Research
Although the current research has many strengths, including the diversity of the analytical techniques employed and the variety of industries in our sample, it is not without limitations. One limitation is that the present article focusses on a major disruptive event, but mainly in a local setting. The major disruptive event under investigation, namely the Turkish 2018 currency and debt crisis, severely affected Turkey. Although the crisis also influenced banks in other parts of Europe, its general impact on other countries has remained relatively small (Arbaa & Varon, 2019). As such, drawing conclusions for global disruptive events based on the present article should be done with due care, as global and local disruptive events are not one and the same. Therefore, future research could add valuable insights to the IB literature by focusing on the effects of a major disruptive event on SME (de-)internationalization in a global setting (such as the current pandemic), and look at the performance implications of both increasing and decreasing international commitment.
the use of the current measurement for the degree of internationalization means that other modes of entry, such as foreign direct investment (FDI) (Lu & Beamish, 2001), were not included. Although most SMEs, especially in their early stages, engage in exporting rather than other modes of entry, such as FDI or joint ventures (JVs) (European Commission, 2015; Love et al., 2016), studying the same model but with a different measurement of the degree of internationalization would serve as an extremely valuable complement to this study. Specifically, including multiple measures of internationalization in the same study would increase the explanatory power of the results, as internationalization is a multi-dimensional construct (Sullivan, 1994).
7. CONCLUSION
REFERENCES
Abor, J., & Quartey, P. 2010. Issues in SME development in Ghana and South Africa. International Research Journal of Finance and Economics, 39(6): 215-228. Acs, Z., Storey, D. 2004. Introduction: entrepreneurship and economic development. Regional
Studies, 38(8): 871-877.
Aldrich, H. E., Auster, E. R. 1986. Even dwarfs started small: Liabilities of age and size and their strategic implications. Research in Organizational Behavior, 8: 165-198.
Andersson, S., Gabrielsson, J., & Wictor, I. 2004. International activities in small firms: examining factors influencing the internationalization and export growth of small firms. Canadian
Journal of Administrative Sciences, 21(1): 22-34.
Athukorala, P. C. 2003. Foreign direct investment in crisis and recovery: Lessons from the 1997– 1998 Asian crisis. Australian Economic History Review, 43(2): 197-213.
Arbaa, O., & Varon, E. 2019. Turkish currency crisis - spillover effects on European banks. Borsa
Istanbul Review, 19(4): 372-378.
Baer, M., & Frese, M. 2003. Innovation is not enough: Climates for initiative and psychological safety, process innovations, and firm performance. Journal of Organizational Behavior:
The International Journal of Industrial, Occupational and Organizational Psychology and Behavior, 24(1): 45-68.
Barringer, B. R., & Greening, D. W. 1998. Small business growth through geographic expansion: A comparative case study. Journal of Business Venturing, 13(6): 467-492.
Bausch, A., & Krist, M. 2007. The effect of context-related moderators on the internationalization-performance relationship: Evidence from meta-analysis. Management International
Review, 47(3): 319-347.
Belke, A., Dubova, I., & Osowski, T. 2018. Policy uncertainty and international financial markets: the case of Brexit. Applied Economics, 50(34-35): 3752-3770.
Benito, G. R. G., & Welch, L. S. 1997. De-internationalization. MIR: Management International
Review, 37: 7-25.
Bernanke, B. S. 1983. Irreversibility, uncertainty, and cyclical investment. The Quarterly Journal
of Economics, 98(1): 85-106.
Berry, H. 2013. When do firms divest foreign operations? Organization Science, 24(1): 246-261. Blackhurst, J., Dunn, K. S., & Craighead, C. W. 2011. An empirically derived framework of
Bloom, N. 2009. The impact of uncertainty shocks. Econometrica, 77(3): 623-685.
Bloom, N. 2014. Fluctuations in uncertainty. Journal of Economic Perspectives, 28(2): 153-176. Bloom, N., Bond, S., & Van Reenen, J. 2007. Uncertainty and investment dynamics. The Review
of Economic Studies, 74(2): 391-415.
Boddewyn, J. J. 1979. Foreign divestment: magnitude and factors. Journal of International
Business Studies, 10(1): 21-26.
Bonaccorsi, A. 1992. On the relationship between firm size and export intensity. Journal of
International Business Studies, 23(4): 605-635.
Brammer, S. J., Pavelin, S., & Porter, L. A. 2006. Corporate social performance and geographical diversification. Journal of Business Research, 59(9): 1025-1034.
Bugert, N., & Lasch, R. 2018. Supply chain disruption models: A critical review. Logistics
Research, 11(5): 1-35.
Bureau van Dijk. 2020. Orbis: Overview.
https://www.bvdinfo.com/en-gb/our-products/data/international/orbis. Accessed on November 20, 2020.
Calof, J. L., & Beamish, P. W. 1995. Adapting to foreign markets: Explaining internationalization. International Business Review, 4(2): 115-131.
Campello, M., Graham, J. R., & Harvey, C. R. 2010. The real effects of financial constraints: Evidence from a financial crisis. Journal of Financial Economics, 97(3): 470-487. Caputo, A., Pellegrini, M. M., Dabic, M., & Dana, L. P. 2016. Internationalization of firms from
Central and Eastern Europe. European Business Review, 28(6): 630-651.
Carr, J. C., Haggard, K. S., Hmieleski, K. M., & Zahra, S. A. 2010. A study of the moderating effects of firm age at internationalization on firm survival and short‐term growth. Strategic
Entrepreneurship Journal, 4(2): 183-192.
Carrier, C. 1994. Intrapreneurship in large firms and SMEs: a comparative study. International
Small Business Journal, 12(3): 54-61.
Chelliah, S., Sulaiman, M., & Yusoff, Y. M. 2010. Internationalization and performance: Small and medium enterprises (SMEs) in Malaysia. International Journal of Business and
Management, 5(6): 27-37.
