The Consequences of Trade Deliberalization in Economic Unions:
The effect of trade policy uncertainty on investment patterns
from remaining member-states to a withdrawing state
Msc Business Administration – International Management
Master’s Thesis
Student: Manon Macharis
Student ID: 11779691
Supervisor: Dr. N. Pisani
Second Reader: Dr. V. Scalera
Date of Submission: June 22
nd, 2018
Statement of Originality
This document is written by student Manon Macharis who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not the contents.
Abstract
The consequences of trade deliberalization are widely unknown. Previous literature has extensively covered how trade liberalization, in the form of economic unions, influences firm behavior. However, little research has been conducted on how firms react to trade deliberalization in an economic union. In light of this, the purpose of this study is to understand the influence of trade policy uncertainty caused by a country’s withdrawal of an economic union on the investment patterns of firms from remaining member-states. For this, the study analyzes whether trade policy uncertainty has a negative effect both on subsidiary location and equity commitment. Moreover, it examines the moderating effects of firm experience on the underlying relationship between trade policy uncertainty and a remaining member-state’s firm internationalization patterns to the withdrawing state. The study researches the occurrences in the empirical setting of Brexit. Using cross-sectional data on French multinationals’ investment patterns between 2014 and 2016, the empirical results show support for the hypothesized negative effect of trade policy uncertainty on equity commitment in subsidiaries located in the withdrawing state. The findings, however, neither provide evidence for the effect of trade policy uncertainty on subsidiary location, nor for the moderating effect of firm experience. Based on the findings of this research, important implications for managerial decisions as well as limitations and ideas for further research are discussed.
Keywords: trade deliberalization; trade policy uncertainty; economic union; subsidiary location; equity commitment; firm experience.
Table of Contents
List of figures ... 5 List of tables ... 5 I. Introduction ... 6 II. Literature review ... 9 I. Terminology: Economic unions and MNE internationalization ... 9 a. Free trade and economic unions ... 9 b. MNE internationalization ... 11 II. Institutional environment ... 13 a. Effect of free trade agreements on international business ... 13 b. Effect of withdrawal on international business: trade policy uncertainty ... 16 III. Research Gap ... 17 III. Theoretical framework ... 19 I. Effect of trade policy uncertainty on subsidiary location ... 19 II. Effect of trade policy uncertainty on equity commitment ... 20 III. Firm experience as a moderating effect ... 22 IV. Methods ... 23 I. Data sample and methodology ... 24 II. Empirical setting ... 24 III. Measurement ... 26 a. Independent variable ... 26 b. Dependent variables ... 26 c. Moderator variable ... 27 d. Control variables ... 28 IV. Data analysis ... 30 V. Discussion ... 37 I. Academic relevance ... 38 II. Practical relevance ... 41 III. Limitations and further research ... 42 VI. Conclusion ... 43 VII. Acknowledgement ... 46 VIII. References ... 47List of figures
Figure 1: Conceptual model ... 23
List of tables
Table 1: Operationalization of the study’s variables ... 29Table 2: Change in subsidiary location and equity commitment ... 30
Table 3: Variable means, standard deviations, and correlations ... 35
I. Introduction
Despite the ever-evolving traits of globalization, the economic unions remain an important concept in today’s economy. Kohl and Brouwer (2014) find that proximity and economic
integration agreements are key concepts to explain the tenacity of regional trade blocs in a
globalizing world economy. Since their origin, economic unions have facilitated the spread of peace and democracy amongst their member-states. The reasons for which countries enter regional trade agreements are multiple, and their benefits have been widely studied. Regional trade agreements create traditional trade gains, strengthen domestic policy reforms, increase multilateral bargaining power, guarantee access to a larger market, and create a strategic linkage to surge security and multilateral interplay (Whalley, 1998). Most importantly, foreign direct investment (FDI) proliferates between member states of an economic union, as these unions reduce trade and investment costs (Dhingra et al., 2016). As such, countries have widely benefitted from setting up economic and political unions and from complying with free trade agreements. In addition to the above-mentioned economic gains, another benefit is the reduction of risk and uncertainty from belonging to the same trade agreement.
Nevertheless, countries have decided to withdraw from economic unions in the past years. This has caused an upsurge of risk attached to investments in that country because of the uncertain consequences of trade deliberalization (Sampson, 2017). In the post-war era, leaving an economic or political union simply implied reversing the trade effects of joining it. However, today the effect of leaving an economic union is far more complex than it was in the post-war era (Hine, 2017). The dynamics of adjustment to trade deliberalization are complex and not yet studied in the academic literature. By looking at FDI determinants, the influence of uncertainty on international management, and firm characteristics, the effect of trade policy uncertainty on investment patterns in economic unions can be better understood.
Many factors influence FDI flows in the international business. Dunning’s (1998) eclectic paradigm highlights the importance of ownership, internationalization and location scope as dimensions of FDI. Research has abundantly covered the first two dimensions, however the issues related to location dimensions have only recently become a topic of research (Blevins et al., 2016). Moreover, it comes without much surprise that institutional characteristics influence FDI decisions (Blevins, Moschieri, Pinkham & Ragozzino, 2016; Cantwell, Dunning & Lundan, 2010; Chi, 2015). Since the work of North (1990), FDI research has extensively focused on institutional influences. One manifestation of this is that FDI is extremely vulnerable to uncertainty (North, 1990). Furthermore, much research has investigated the effect of risk and uncertainty on international management (e.g. Miller, 1992). Risk refers to the unpredictability in corporate outcome variables, and uncertainty refers to the unpredictability of environmental organizational variables that impact corporate performance. Thus, as uncertainty decreases the likelihood of corporate performance, it increases risk (Miller, 1992). Uncertainty and risk are therefore important determinants for the entry mode choice of multinational enterprises (MNEs) (Müllner, 2016). Studies have frequently focused on researching how managerial and organizational traits affect FDI behavior in the face of risk (Buckley, Chen, Clegg & Voss, 2016). However, political uncertainty and risk have often been researched with regards to the effect on MNEs and their investments in developing countries (e.g. Banalieva & Sarathy, 2010).
Hence, the dynamics of adjustment to trade deliberalization are widely understudied in the sphere of international business (Sampson, 2017). More specifically, there is a need to shed light on the effects of trade policy uncertainty caused by a country’s withdrawal from an economic union on international ventures. How the remaining countries of an economic union adjust their investment patterns to a withdrawing country following the country’s decision of withdrawal thus needs further assessment.
The European Union (EU) has created a strong trading area of 28 member states acting on an international single market. The union has positively impacted the economy of its member-states, as is shown by the boost in FDI within the economic area. Data on bilateral FDI flows between 1985 and 2013 show that there is a statistically significant positive effect of being in the EU on inward FDI (Dhingra et al., 2016). The UK has been a major recipient of FDI, with an estimated stock value of £1 trillion, of which half derives from other EU member states (Dhingra et al., 2016). Following the June referendum, the Sterling depreciated enormously, which has contributed to a rise in inflation from 0.5 per cent in June 2016 to 2.6 per cent a year later (Sampson, 2017). Hence, the statistics show that merely the announcement of the referendum has harmed the UK economy. The economic decline succeeding the Brexit announcement demonstrates the impact of trade policy uncertainty following the announcement of withdrawal. Brexit thus serves as a perfect case study to analyze the adjustments to trade deliberalization of remaining member-states. This trend towards increased protectionism has sent shockwaves through Europe (Hobolt, 2016). It is therefore important to study this occurrence in order to understand the effect of trade policy uncertainty in economic unions.
The following question will guide the present research: What is the effect of trade
policy uncertainty caused by raising trade barriers in economic unions? And more specifically, how do remaining member states’ investment patterns with the departing state change after the announcement of withdrawal? This paper hypothesizes that trade policy
uncertainty between a country that is part of an economic union (A) and a country that is withdrawing from it (B) is negatively related to the likelihood of a firm based in country A entering into country B. Moreover, it hypothesizes that trade policy uncertainty between these two countries is negatively related to the level of equity commitment of a firm based in
country A investing into country B. Finally, it hypothesizes that firm experience positively moderates both these previous relations.
These hypotheses are empirically tested through a cross-sectional data analysis of France’s 100 biggest MNEs’ investment patterns in the U.K. between 2014 and 2016. The results show statistical evidence for the negative relationship between trade policy uncertainty and equity commitment. Moreover, the results provide insights on the moderating effect of firm experience, and the influence of structures such as firm size, industry type, firm revenue and the public firm. Therefore, this study contributes to the academic literature by shedding light on the relationship between trade policy uncertainty and investment patterns within economic unions.
The paper is structured as follows. The first section reviews the relevant literature on economic unions, withdrawal from economic unions and the effect of trade policy uncertainty on MNEs in economic unions. Subsequently, the hypotheses are built and presented. Thereafter, the methods section specifies the variables used in the research as well as the methodology and statistical findings. The following section discusses the statistical results, considers the limitations of the study, and offers advice for further research. The final section of the paper draws conclusions.
II. Literature review
I. Terminology: Economic unions and MNE internationalization a. Free trade and economic unions
The extension of the establishment of free markets to an international scope translates into
free trade. The objective of free trade is to achieve maximum productivity for citizens of the
trade restrictions such as tariffs, quotas, subsidies and regulations of price and quantity of traded products (Nollen & Quinn, 1994). Moreover, removing trade barriers enables a global political economy in which states cooperate, consequently increasing the economic efficiency and the peaceful environment of the involved states. Thus, free trade not only generates economies of scale and extends markets abroad; it also procures national defense and political authority (Ma & Lu, 2011).
There are multiple levels of integration in formal regional trade agreements (RTAs). These range from establishing a preference in tariffs to full-scale economic integration (Frankel et al., 1997). Deep integration influences laws and institutions that could have been preserved as domestic privileges even with a high level of trade integration (Frankel et al., 1997). There are two types of deep integration: the common market and the economic union. A common market entails, in addition to the free exchange of goods and services, the free movement of factors of production (labor and capital). This applies both to portfolio investments as to FDI. An economic union goes beyond the free movement of goods, services, and factors, by involving the harmonization of national economic policies (Frankel et al., 1997). These policies include taxes and a common currency.
An economic unions’ purpose is mainly to “correct inefficiencies created by national policies with international spillovers” (Etro, 2002, p. 187). A union of two countries is possible when the instated policies are strategically complementary and when it reduces the scope for free riding of the outsiders (Etro, 2002). Countries have widely benefitted from setting up economic and political unions and from complying with free trade agreements. An example of this is how the EU has boosted foreign direct investment (FDI) within the economic area. After having recognized the importance of non-tariff barriers, the EU implemented the Single Market Program in the late 1980s. This led to the elimination of internal barriers of cross-border trade and investment, consequently increasing FDI in Europe
(Pain & Young, 2004). Data on bilateral FDI flows between 1985 and 2013 shows that there is always a statistically significant positive effect of being in the EU on inward FDI (Dhingra et al., 2016). Dhingra and his colleagues (2016) find these estimates to be between 14 per cent and 28 per cent, which is consistent with previous research that found this to be between 25 per cent and 30 per cent (Campos & Coricelli, 2015).
b. MNE internationalization
MNEs are companies that own and control activities in two or more countries (Buckley & Casson, 2009). Thus, MNEs use FDI to create income-generating assets abroad. Internationalization of a firm in the form of FDI is one of the most prominent features of globalization. One type of FDI is termed mergers and acquisitions (M&As). This entails the recombination of two or more companies in order to attain a strategic financial objective. Another source of FDI is Greenfield investments, which involves the construction of new facilities to create a new company from the ground up (Chala & Lee, 2015). MNEs need to consider multiple factors when strategizing an internationalization process.
There are various institutional factors that affect FDI attractiveness in a country (North, 1990). Firstly, political stability, democracy and rule of law attract FDI, while corruption, tax rates and cultural distance discourage it. Secondly, environmental effects such as development levels, region of destination and competitive environment moderate the above relationships (Bailey, 2017). Thirdly, research conducted by Chakrabarti (2001) shows a positive relationship between market size (as measured by per-capita GDP) of a host country and FDI. The institutional environment of countries is thus amongst the most important determinants of FDI, and institutional quality therefore has a significant positive effect on FDI levels (North, 1991).
Internationalization knowledge is another necessary element for MNEs to successfully enter and develop competitive strategies abroad (Riviere, Suder & Bass, 2018).
This kind of firm-specific knowledge is accumulated from all foreign-operated markets, and by engaging in foreign activities. On the one hand, it shapes the internationalization strategy of the firm as knowledge is integrated. On the other hand, it alters the content and process of the firm’s internationalization strategy to enhance long-term competitiveness. Thus, the content and process of a firm’s internationalization strategy is partly formed by the advancement and improvement of the international knowledge base (Riviere, Suder & Bass, 2018). This idea of internationalization knowledge stems from Johanson and Vahlne’s (1977) “Internationalization Process model”. This model, also known as the Uppsala model, is a powerful tool to interpret the progression over stages of a multinational’s international growth and the underlying processes that build the firm’s capabilities. The model perceives the development of the MNE as a process of gradual acquisition, where the MNE uses assimilated knowledge of foreign markets in order to successively increase its commitment in the foreign market (Santangelo & Meyer, 2017). Yet, this model is static in theory, and does not account for the non-linear and discontinuous dynamics of the internationalization process over time, such as the path-braking commitments that increase a multinational’s exposure to risk.
Santangelo and Meyer (2017) provide a model for the internationalization of a multinational as an evolutionary process. In order to do this, they add an explicit time dimension to the internationalization process, which allows for creating alternative paths the process can take. Moreover, they argue that interpreting the model in an evolutionary perspective offers opportunities to explain non-linearity and discontinuities such as commitment decisions or the scale and novelty of commitments. The authors propose that path-breaking resource commitments increase the exposure to threats and opportunities, and consequently the possible outcomes. This implies that path-breaking commitments are associated with both outstanding results and higher rates of failure, of both the subsidiary and
the parent company (Santangelo & Meyer, 2017). This shows the importance of taking into account multiple factors such as institutional influences and trade policy when strategizing internationalization.
II. Institutional environment
a. Effect of free trade agreements on international business
The most important RTAs are signed between members of the chunk of developed countries. Following this observation, Baltagi, Egger and Pfaffermayr (2008) researched the effect of regional trade agreements on FDI. Their findings suggest that preferential trade agreements affect both the trade and FDI of the involved countries and their host countries (Baltagi, Egger and Pfaffermayr, 2008). The rationale behind the redirection of FDI to the liberalizing area is that export platforms created in multinational networks contribute to the tariff reduction in the involved economies. Consequently, this makes it cheaper for multinationals to deliver goods to consumers inside the liberalizing area, from the export platform situated inside the area. As such, bilateral FDI increases between member-states of the liberalizing area, though FDI decreases in countries situated outside of the regional trade agreement (Baltagi, Egger and Pfaffermayr, 2008). These findings are in line with the findings of Yeyati, Stein and Daude (2003) on the effect of trade agreements on FDI stocks. The research shows that common membership of two countries in a regional trade agreement increases FDI flows by approximately 27 per cent (Yeyati, Stein & Pfaffmayr, 2008).
Banalieva and Sarathy (2010) have analyzed the effects of variation in three key components of regional integration on the degree of global orientation of emerging market MNEs. These components are RTA diversity, RTA potential market size and RTA experience, and the results suggest that if these characteristics are unfavorable in their home market, the emerging market MNEs will internationalize outside of their home region and consequently
adopt a higher degree of global orientation (Banalieva & Sarathy, 2010). Even though this research focused on emerging countries, some of its findings are useful for general internationalization studies. One relevant finding is that MNEs situated in regions with high levels of economic liberalization are more home-region oriented than MNEs of other regions. Moreover, the institutional environment and institutional variation play an important role on the firm’s internationalization strategy (Wolf et al., 2008). Finally, the MNE’s experience with the RTA in question is important in the process of internationalization, as greater experience suggests greater organizational learning within the RTA (Banalieva & Sarathy, 2010).
Blevins and his colleagues (2016) have analyzed how regional integration affects MNE decisions. Based on the facts that governance and investment location influence the success of MNEs, they have researched MNE’s entry mode decisions during a period of institutional change. The empirical setting of this research is grounded in Europe from 1990 to 2012, where they analyzed all full acquisition deals between European firms that were not from the same country of origin. Their strongest finding is that economic geography is a direct determinant of an MNE’s entry mode decision. Thus, the level of integration of an economic region shapes the entry mode decisions of a firm. Their results indicate a positive relationship between regional integration and the preference of acquisitions over strategic alliances. Moreover, they show that this choice is equally more likely to occur when the target firm is situated within the same economic region than when it is located outside of it. However, the research shows that there are time-dependent characteristics of governance choice by the firm. The more the economic region integrates, the weaker the previous relationships become. This finding is explained by two factors. First, the temporal evolution of the economic region, which decreases transaction costs of firms investing within the region, and directly affects their governance choices. Second, countries’ membership to an
economic region provides direct and indirect effects to its MNEs, allowing these to focus on strategic issues instead of external factors (Blevins et al., 2016).
Moschieri and colleagues (2014) have explored how the integration within the European Union has created a more fluid environment for firms seeking international expansion through M&As. Their results indicate that uncertainty avoidance of the country of the investor and the political risk of the host country matter enormously for the M&A decisions made by the investing MNE early on in the life of the economic union. However, these effects do not play a role on the M&A decision once steps towards further integration (such as the adoption of a common currency) have been taken (Moschieri, Ragozzino & Campa, 2014).
Research is equally keen to find out whether RTAs encourage FDI in the form of Greenfield investments. A Greenfield investment produces inflows of managerial skills, know-how and the establishment of new jobs (Chala & Lee, 2015). This is, therefore, an important factor to consider when companies need to make investment decisions, especially in RTAs. Chala and Lee (2015) have investigated how Greenfield investments respond differently to RTAs’ common membership for high income and low-income host countries. They find that common membership to an RTA discourages Greenfield investments in OECD-high income country pairs, while it promotes Greenfield investments in OECD-low income country pairs (Chala & Lee, 2015). Their findings suggest that differences in the economic growth stage of country pairs should be properly considered when investigating the effect of RTA common membership on bilateral FDI flows (Chala & Lee, 2015).
However, in the face of risk, the MNEs need to adapt their international strategies. The “real options perspective” on internationalization holds that small investments with little risk allow firms to postpone large strategic investments until environmental contingencies at the host-country level are solved or until the firm has gained internationalization knowledge.
This model suggests that if a firm’s investment can be either delayed or not easily reversed, the firm may choose a ‘‘wait-and-see’’ approach to equity commitment (Wooster, Blanco & Sawyer, 2016). As such, the real options approach to equity commitment complements the leading theories of MNE strategies and provides a natural framework to explore the effect of uncertainty on further investment choices (Wooster, Blanco & Sawyer, 2016).
b. Effect of withdrawal on international business: trade policy uncertainty
Despite the prevalent clauses authorizing withdrawal from treaties, scholars and theorists have not studied their impact on regional integration (Helfer, 2005). Denunciation or withdrawal clauses permit member states to withdraw from a treaty if they wish to do so (Helfer, 2005). These are a necessary component of trade agreements, as treaties and their contexts are not static, and shifts in the political background or domestic preferences destabilize a treaty's objectives. However, due to the lack of precedents in withdrawal of a country from a trade agreement, the future of trade policies bears a new kind of risk for international management. In the literature, risk refers to the unpredictability in corporate outcome variables, and uncertainty is used in strategic management and organization theory as referring to unpredictability of environmental organizational variables that impact corporate performance. Thus, as uncertainty decreases the likelihood of corporate performance, it also increases risk (Miller, 1992).
The role of uncertainty and risk is an important determinant of entry mode choice, though theories disagree on its effect. Müllner (2016) summarizes the two main stances. On the one hand, research taking the transaction cost-view associates high uncertainty with ownership-based entry modes that permit high organizational control. On the other hand, research taking the stance of the real options theory expects firms to avoid ownership in risky environments in order to increase their flexibility. Müllner builds on work by Liesch and his colleagues to integrate the distinction between risk and uncertainty into entry mode
frameworks. Liesch et al. (2011) argue that risk and uncertainty coevolve, but differ in their effects on internationalization. They claim that risk, as opposed to uncertainty, is in part manageable by MNEs entering a foreign country. Recognizing the existence of manageable sources of risk helps to create a realistic model of entry mode choice. As such, Müllner’s model identifies that MNEs are proficient in diversifying, transferring or mitigating identifiable sources of risk through various risk management instruments, such as capital structure or contractual and network-based risk management. Moreover, the model shows that when uncertainty is composed in large parts by manageable sources of risk instead of true uncertainty, risk management permits MNEs to enter foreign markets without the cost accompanying hierarchical means of internationalization (Müllner, 2016).
Henisz (2000) studies the institutional environment for multinational investment. The author provides the following definition for political hazards: “the feasibility of policy change by the host country government which either directly (seizure of assets) or indirectly (adverse change in taxes, regulations or other agreements) diminishes the MNE’s expected return on assets” (Henisz, 2000, p. 334-335). Thus, it could be derived that the withdrawal of an economic union provides a political hazard to any remaining member state’s MNEs. The effect of political hazards on the MNE’s entry mode choice varies across the firms based on the extent to which they face expropriation hazards from their potential joint-venture partners in the host country (Henisz, 2000). The results show that as political hazards increase, the MNE faces an increasing threat of opportunistic expropriation by the government. However, partnering with a host-country firm that has a comparative advantage in interactions with the host-country government can safeguard against this hazard (Henisz, 2000).
III. Research Gap
The effects of economic unions and trade agreements have been widely discussed over the past decades. Yet, trade deliberalization in economic unions remains a subject for research.
Since the work of North (1990), FDI research has extensively focused on institutional influences (e.g. Blevins, Moschieri, Pinkham & Ragozzino, 2016; Cantwell, Dunning & Lundan, 2010; Chi, 2015). These studies, however, do not find consistent findings. Hence, the effects need to be studied under individual conditions instead of generalized conditions. Moreover, research has investigated the effect of risk and uncertainty on international management (e.g. Miller, 1992; Müllner, 2016). However, these studies have too often taken a linear perspective on internationalization instead of an evolutionary one that takes into account changing patterns of political influences and trade policies (Santangelo & Meyer, 2017). As a result, these studies cannot account for the sudden change in trade policies that cause risk and uncertainty in a host country.
Although research has built on the link between risk and FDI in international business, research is lacking in the consequences of trade policy uncertainty stemming from trade deliberalization in economic unions. Most studies of how firms respond to host country risk have focused on the early stages of economic integration and organizational capabilities. Moreover, the majority of studies concerned with trade policy influences on MNEs have researched the effects in developing markets, which behave in different ways than the western developed economy.
The implications of exiting an economic union in the 21st century need further exploration, as the dynamics of adjustment to trade deliberalization remain widely understudied (Sampson, 2017). There is a disruption in post-war economic integration, which questions the ramifications of protectionist measures. The uncertainty and risk following these protectionist measures in a certain country are unprecedented in economic unions. The OFDI patterns following a country’s announcement to exit a trade union have received little attention thus far. It remains to be studied how the remaining countries of an economic union adjust their patterns of investment following a country’s referendum for departure. This begs
the question: What is the effect of trade policy uncertainty caused by raising trade barriers in
economic unions? And more specifically, how do remaining member states’ investment patterns with the departing state change after the announcement of withdrawal?
III. Theoretical framework
I. Effect of trade policy uncertainty on subsidiary location
With the creation of region-wide trade rules and standards, countries were able to operate in a fair and competitive environment. The implementation of trade agreements was accompanied by a drastic decrease in trade policy uncertainty, which consequently stimulated imports and exports within acceptable product-level markets. Hence, the reduction of trade policy uncertainty caused by being a member of an economic union positively affects the export market, consequently making the country more productive and profitable (Feng & Swenson, 2017). Conversely, the opposite can be true as well. The decision of a country to leave a trade agreement causes unprecedented trade policy uncertainty within that economic region, which in turn would affect how firms of remaining member-states continue their investment patterns to the withdrawing state.
Political instability makes a country less attractive for MNEs to invest in, as it creates an unpredictable environment that might disrupt economic activity (Bailey, 2018). However, while these theoretical foundations insinuate that political stability would be negatively related to FDI, the results in the literature provide mixed results. After having conducted a meta-analysis, Bailey (2018) finds that undesirable institutional factors do indeed have a negative and strongly significant effect on FDI. Thus, host countries whose institutions are more negative than the home country, deter FDI (Bailey, 2018). However, the strength of the results depends on country-level determinants, and the study shows stronger results for the relationship between political stability and FDI in developed host countries than in the rest of
the sample. This means that MNEs from developed countries will seek to invest in host countries with high political stability.
Moreover, the literature shows that as a result form economic unions, bilateral FDI increases between member-states of the liberalizing area, and FDI decreases in countries situated outside of the regional trade agreement (Baltagi, Egger and Pfaffermayr, 2008; Yeyati, Stein & Daude, 2003). Therefore, it is most likely that firms from remaining member-states of the economic union will avoid future investments in the departing country, as its unprecedented environment causes uncertainty that firms want to prevent. This means that the countries will prefer to invest in another member-state where this uncertainty is not present. Hence the following hypothesis:
Hypothesis 1: Trade policy uncertainty between a country that is part of an economic union (A) and a country that is withdrawing from it (B) is negatively related to the likelihood of a firm based in country A entering into country B.
II. Effect of trade policy uncertainty on equity commitment
Another major variable to consider more deeply is how trade policy uncertainty impacts the country’s equity commitment to the departing country. Managers are required to make decisions about further investment in foreign subsidiaries, while the outcomes of these decisions are difficult to forecast. Political risk affects the value of an MNE through changes in future cash flows and investors' required return. Thus, by looking at political risk as the
risk of adverse consequences arising from political events, we understand that it is the
unfavorable effects of political risk that divert firm value. This demonstrates how important it is for managers to take political risk into account (Butler & Joaquin, 1998).
Managers’ subjective perception of political risk plays an important role in how a firm manages risk through strategic decision-making (Giambona, Graham & Harvey, 2017). Nearly 50 per cent of firms avoid FDI due to political risk. Moreover, firms with risk-averse executives are more likely to avoid investment in politically risky countries. This relationship increases when agency problems are more likely to be severe (Giambona, Graham & Harvey, 2017). Thus, there is an important link between executive risk aversion and FDI commitment. Because the withdrawal of a country from an economic union is unprecedented, executive managers will interpret risk differently based on individual opportunities and threats. Therefore, executive managers’ carefulness, especially alongside the presence of agency problems with local subsidiaries in the withdrawing host country, negatively affects the equity commitment in a withdrawing country.
Fisch’ (2011) real options model justifies the MNE’s decision of enlarging a new foreign subsidiary by subsequent investments after an initial investment was made based on opportunities and threats. From this model, it is possible to study the influence of environmental uncertainty on decisions to enlarge foreign subsidiaries by extra investments. Their results suggest that the interaction of economic volatility and irreversibility has a negative effect on subsequent investment (Fisch, 2011). Following the real options model, the uncertainty around the future investment environment of a withdrawing state would provide a scenario in which firms of remaining member-states will not extend investments in the withdrawing state, or wait for a reduction in trade policy uncertainty. This insinuates that there would be a decline in equity commitment in a country after it has withdrawn from an economic union in which the investing country is still part of. Taking the two theoretical findings together, the following hypothesis holds:
Hypothesis 2: Trade policy uncertainty between a country that is part of an economic union (A) and a country that is withdrawing from it (B) is negatively related to the level of equity commitment of a firm based in country A investing into country B.
III. Firm experience as a moderating effect
As the Uppsala model demonstrates, learning and knowledge are crucial elements in the internationalization process of an MNE. Experiential learning and the portfolio of subsidiaries of an MNE are two sources of knowledge that shape the way in which firms internationalize (Hutzschenreuter & Matt, 2017). Firms learn from their previous practice, transform routine into knowledge, and thus develop critical capabilities. Therefore experience is valuable in dealing with political risk arising from unfair behavior of political authorities (Jiménez et al., 2018). Firm experience is seen in the literature as a component of organizational learning, a source of competitive advantage, a foundation for superior performance, and it has a significant impact on the internationalization strategy of an MNE (Jiménez et al., 2018). Firms not only learn from the intensity of experience dealing with policy risk, but equally from the exposure to a more diverse range of political environments (Jiménez et al., 2018). Hence, the more experience the firm has gathered, the better it can adapt to a multitude of different situations.
Firm experience is expected to positively moderate the two hypotheses. This means that the higher the firm experience of the MNE, the less these MNEs will negatively react to the withdrawal of a country from an economic union. Because firm experience increases as the firm becomes older, it is safe to say that the older the MNE, the more it will have gathered experiential learning over a multitude of locations. Consequently, firms that are older will be able to adapt their strategy through experiential learning, and are less affected by the subsidiary location’s external pressures. Hence the following two hypotheses:
Hypothesis 3: Firm experience positively moderates the relationship hypothesized in hypothesis 1.
Hypothesis 4: Firm experience positively moderates the relationship hypothesized in hypothesis 2.
The four proposed hypotheses are represented in Figure 1.
Figure 1: Conceptual model
IV. Methods
In this section, the relationship between trade policy uncertainty, subsidiary location, equity commitment and firm experience is analyzed. The hypotheses created in the previous section will be tested using a hierarchical multiple regression analysis. First the sample and subsequent data set are reviewed. Then the different variables and measures are discussed.
I. Data sample and methodology
This research adopts a quantitative research methodology. It focuses on gathering a representative quantitative data sample and subsequently on analyzing it, using statistical processes to generalize findings across companies. The research uses data from before and after the withdrawal of the UK from the EU to observe the effects of trade policy uncertainty, as caused by the withdrawal of a country from an economic union, on the investment choice of MNEs situated in the remaining member-states. As such, the years 2014 and 2016 are chosen as reference years. Given that the data is measured at two points in time, the study takes the form of a cross-sectional analysis.
For the measurement to be valid and reliable, the study needs a large sample. The overall sample of MNEs used in this study consists of the 100 biggest French firms, which are retrieved from the Forbes Global Fortune 500 list published in 2017. The French setting is interesting, as French firms play an important role in international FDI outflows (Chen & Moore, 2010). Data on the companies is gathered from the ORBIS database. Using this database is the most practical way to collect secondary data efficiently, as it provides a large collection of firm data. Moreover, the information from the ORBIS database is supplemented with firm-specific data found in the firms’ annual reports obtained on their websites. In order to analyze the desired effect, the first step is to remove all companies without subsidiaries from the sample. Then, the companies that did not have subsidiaries in the UK are equally ignored, and those that had British subsidiaries in 2016 but not in 2014 are added to the sample. This results in a final sample of approximately 40000 subsidiaries.
II. Empirical setting
The reasoning behind the formation of the EU was to generate a trade agreement that underpins strategic alliances and hence creates security arrangements among the member states (Whalley, 1998). As a result of European integration, the EU’s institutions and policies
play an important role in the determination of international trade and investment patterns by its member states. Consequently, the Union’s policies and institutions bear a significant role on the location of economic activity (Pain & Young, 2004). The UK’s withdrawal from the EU will therefore bear significant effects on the remaining member states’ outward FDI decisions to the withdrawing country.
Scholars claim that the political and economic consequences of Brexit will likely be significant and prolonged, both for the UK and the EU as a whole. On the one hand, Brexit will make the UK poorer than it would have been if it remained in the EU, due to the increase of trade barriers that are accompanied by it (Sampson, 2017). The UK will suffer from a fall in FDI, a decline in migration, and the negative byproducts of constructing trade barriers (Sampson, 2017). The living standards in the UK will be negatively affected due to a decline in inward FDI, and as a result, the UK’s economic growth will decline (Pain & Young, 2017). Recent research also suggests that the short-term impact of Brexit on goods exports may cause the UK’s exports to the European Union to decrease by 2 per cent (Kee & Nicita, 2017).
On the other hand, research proves that the remaining European countries will suffer economically from Brexit, but will yield lower losses than the UK (Sampson, 2017). There is wide uncertainty over what the exact costs of Brexit will entail in the long run (Sampson, 2017). It is predicted that Brexit will likely cause a reduction of 22 per cent of FDI inflows to the UK (Dhingra et al., 2016). Moreover, research shows that economic growth in the UK has suffered due to the uncertainty about the referendum’s future (Kierzenkowski et al., 2016).
Analyzing how the remaining member states of the EU adjust their OFDI based on the decision of the UK to exit the union will provide deeper insight into this matter. The UK’s recent announcement to exit the EU demonstrates the need to understand the effects of trade policy uncertainty following from trade deliberalization. The lack of precedents has
caused a wave of uncertainty, as well as the possibility that other states will follow in the UK’s footsteps. Thus, researching the effects of trade policy uncertainty on investment patterns of remaining member-states’ MNEs in the empirical setting of Brexit can provide valuable insights for the field of international business.
III. Measurement a. Independent variable
The independent variable in this research is the trade policy uncertainty that is caused by a country’s withdrawal from an economic union. Much research has been conducted on the effect of risk and uncertainty on international management (e.g. Miller, 1992). However, the effect of uncertainty caused by withdrawal of a country from an economic union on investment patterns needs further investigation. The independent variable is represented in this research by the consequences of the UK’s withdrawal from the EU. Hence, the fiscal year 2014 represents the time without manifestation of trade policy uncertainty, and the fiscal year 2016 represents trade policy-uncertainty due to the UK’s withdrawal from the EU. This is measured with a dummy variable, zero being pre-Brexit and one being post-Brexit. As Brexit was voted in July 2016, the results published at the beginning of 2017 from the fiscal year 2016 are expected to illustrate the reaction to fear of the uncertain future of the EU and the risky investment environment in the UK.
b. Dependent variables
The dependent variables in this research are subsidiary location and equity commitment. The research aims to find out how parent companies react to risk caused by uncertainty by adjusting their internationalization strategy subsequent a risk-producing external action. Thus, the study aims to find out the relationship between trade policy uncertainty and both subsidiary location and equity commitment.
First, subsidiary location is measured by a dummy variable, where one represents the location of the subsidiary being inside the U.K. and zero represents elsewhere. Research shows that countries being part of an economic union favor other member-states over non-member states to invest in (e.g. Baltagi, Egger & Pfaffermayr 2008). Therefore it is interesting to see whether the French MNEs continue to increase the number of subsidiaries in the UK after the announcement of Brexit. Secondly, equity commitment is measured through the company’s stake in the affiliates, on a range varying from 1 to 100 per cent. For this, the firm’s direct and total stakes in the subsidiaries are assembled into one new variable, as otherwise there is too much missing data when only choosing one of the two. Previous studies show that uncertainty and risk affect managerial decisions, and that economic volatility and irreversibility have a negative effect on subsequent investment (Giambona, Graham & Harvey, 2017; Fisch, 2011). Thus, equity commitment is used to measure the variation of invested resources by French MNEs in their English subsidiaries before and after the presence of uncertainty.
c. Moderator variable
The moderation model tests whether the prediction of a dependent variable from an independent variable differs across levels of a third variable. A moderator variable thus affects the strength of the relation between a predictor variable and an outcome variable (Fairchild & MacKinnon, 2009). Firm experience is the moderator in this research, as it is expected to influence the relationships in the two main hypotheses. Firm experience is measured through the age of the firm, which is defined by the number of years since the foundation of the company until 2016. Older firms are anticipated to have more international knowledge and experience that positively impact internationalization (Singla & George, 2013). Moreover, the more years of experience the MNE has gathered, the more resilient the
firm becomes towards external pressures. To measure this, the year the MNE was founded is subtracted from 2016, which creates a new variable.
d. Control variables
Different control variables at the firm level are included in this research in order to ensure these are not influencing the results of the regression analyses. Firstly, firm size is accounted for, and measured through the number of employees the MNE retains in 2016. Firm size is commonly used as a control variable in international business research (e.g. Banalieva & Dhanaraj, 2013). A firm that employs more employees is predicted to have a higher foreign sales intensity (George et al., 2005). Moreover, larger firms have more experience and possess a larger pool of resources, and therefore have the ability to establish economies of scale and scope across operations, and consequently become more resilient to external influences. Finally, Kogut and Singh (1988) claim that firm size positively impacts FDI decisions of MNEs.
Secondly, the type of industry is controlled for, since internationalization motives and characteristics vary across industries (George et al., 2005). Different types of industries can create different reasons for expanding abroad, and external pressures can differently affect industries. Bailey (2018) finds that the type of industry influences the relationship between institutional factors and FDI. For example, manufacturing industries will be more sensitive to institutional factors because of the high investment mobility in the competitive environment (Bailey, 2018). As such, all MNEs are divided according to being manufacturing companies or other industries. This means that all banks, financial services, insurance companies and other services are coded in a dummy variable in opposition to all manufacturing industries.
Thirdly, firm revenue is controlled for in the analysis. MNEs that have more revenue can more easily react to external pressures and deploy the necessary actions. Chen and Moore (2010) find that a firm’s choice of host countries differs significantly with regard to their total
factor productivity, as more productive firms are significantly more likely to invest in countries with a small market potential, high entry barriers and large fixed costs of investment. Firm revenue is in this research measured through the total turnover of the MNE at the end of 2016.
Finally, publicly traded companies are controlled for in this research. Publicly traded companies generate extra funds by selling shares on the stock market. Hence, this provides a distinctive revenue from their sales and activities that private companies do not have. In contrast, privately held firms are often undercapitalized and the allocation and exploitation of resources is likely to dominate managerial decisions (George, 2005). Thus, the importance allocated to continuing resource allocations in a certain location can be influenced by the firm being privately or publicly held. This variable is computed in the form of a dummy variable that represents one if the firm is public and zero if the firm is private.
An overview of all variables and their operationalization is presented in the following table.
Table 1: Operationalization of the study’s variables
Variables Operationalization of Variables
Dependent Variable
Subsidiary Location A dummy variable (1 = inside UK; 0 = elsewhere) Equity Commitment Firm's total and direct stakes in subsidiary
Independent Variable
Trade Policy Uncertainty A dummy variable representing Brexit (1 = post-Brexit; 0 = pre-Brexit) Moderating Variable
Firm Experience Years of existence since foundation year of firm (until 2017) Control Variables
Firm Size Total number of employees
Industry A dummy variable (1 = manufacturing; 0 = others) Firm Revenue Firm total revenue in US billion dollars
IV. Data analysis
This study measures the impact of trade policy uncertainty, as caused by a country’s withdrawal from an economic union, on the remaining member-states’ equity commitment and subsidiary location in that withdrawing state. The data is cleaned as to identify and remove all outliers, and a list wise deletion is applied to deal with missing values in the dataset. Before investigating the data, Table 2 provides raw insights on the change in subsidiary locations and equity commitment over the analyzed timeframe without controlling for anything. This helps to provide an overview on the variation in the analyzed variables over the reference years. From the table, it can be deduced that the number of subsidiaries, both within and outside the UK, increase drastically from 2014 to 2016. Moreover, the average equity commitment in subsidiary locations increases by almost 3 per cent.
Table 2: Change in subsidiary location and equity commitment
Subsidiary Location Average Equity Commitment (%)
Non-UK UK Non-UK UK Total
2014 17249 1189 52.63 39.14 51.75
2016 19754 1215 55.16 44.91 54.56
In order to further understand the variables in the research, a table with the descriptive statistics is created (Table 3). This table includes the dependent variables, moderator variable, control variables as well as the predictor variable of the study. Taking both reference years together, the average subsidiary location is .06, and the average equity commitment in the subsidiaries is 53.24 per cent. This means that most observations are recorded outside of the UK, and that the level of equity commitment is average. Moreover, the descriptive statistics show that the mean of pre-Brexit and post-Brexit observations is of .53. Hence the sample is well distributed across both reference years. Finally, the average firm size is of 85849 employees, the average turnover is of 532,306.671 US million dollars, and the average firm
experience in the sample is of 75.83 years. Consequently, the data sample comprises large and very large firms, which all have accumulated a great amount of experience.
Since the study uses a hierarchical multiple regression analysis, all variables are tested for multicollinearity, so that the empirical results are not misinterpreted or overestimated (Field, 2013). Multicollinearity poses problems when correlation among the predictor variables is too high (> 0.8). As shown in Table 3, none of the variables express a high correlation. To further control for multicollinearity, it is convenient to study the variance inflation factors (VIF) and tolerance statistics of the variables (Field, 2013). The results of the collinearity statistics show that the VIFs are for the predictor variable are always below 10, which means there is no cause for concern. However, the average of the VIFs is above 1, which means the regression may be biased (Field, 2013). The values of the tolerance statistics are always well above 0.2, which again points to the fact that there is no cause for concern of multicollinearity in this research (Field, 2013). Hence, multicollinearity is does not seem problematic in this analysis and the variables can be studied through a hierarchical multiple regression.
The hypotheses of this research are tested using Ordinary Least Squares (OLS) regression models. The relationships between the independent variable, trade policy uncertainty, and the two dependent variables (subsidiary location and equity commitment) are studied separately. The results are summarized in Table 4, including 3 models for the variable subsidiary location and 3 models for the variable equity commitment. The table specifies for each individual model the variables’ standardized beta coefficients, the standard errors, the level of significance, and general information regarding the model fit.
To study the effect of trade policy uncertainty on the dependent variables, a hierarchical multiple regression is completed. As both dependent variables are tested separately, the first hierarchical regression is executed for the dependent variable subsidiary
location. The first step of the regression analysis is to examine the ability of the control
variables to predict subsidiary location (model 1). This model shows that the control variables in the research account for only 1.3 per cent of the variance in subsidiary location as the dependent variable. Moreover, the results show that the model is statistically significant as the F-value is 131.490 with a probability less than .001. When looking further into the effect of the control variables, all variables except firm size illustrate a significantly positive effect on subsidiary location. Industry effects (ß = .051; t = 8.051; p = .000), firm revenue (ß = .069; t = 10.366; p = .000) and public firms (ß = .050; t = 7.975; p = .000) thus cause an increase in subsidiary location.
Then, the independent variable is added in the model (model 2). Adding the independent variable to the model after having controlled for all the control variables leads to a variance of 1.3 per cent that is explained by the predictor variables, which is constant to the previous model (R Square change = 0; F = 105.806; p = .000). This shows that the control variables do not have any effect on the research. Moreover, the model shows that the relationship between subsidiary location and trade policy uncertainty is negative but not statistically significant (ß = -.009; t = -1.744; p = .081). Therefore, the results do not show support for the first hypothesis, which is consequently rejected. However, the moderation effect is still analyzed, as this could explain the lack of results in the previous step.
To analyze the moderation effect of firm experience on the relationship in hypothesis one, a third step is first added to the hierarchical multiple regression. The interaction term of the independent variable and moderator variable is computed, and added as a new variable in SPSS. The results of this are added into the model (model 3), which explains 1.4 per cent of the variance in the dependent variable and is statistically significant (F = 80.71; p = .000). The results of the model show that firm experience is negatively related to risk (ß = -.032; t = -4.277; p = .000). The results for the interaction term show a positive relationship, which is
not significant (ß = .001; t = .092; p = .927). Hence, the moderation itself negatively affects the dependent variable, but in combination with the independent variable the impact on the subsidiary location is positive. However, the interaction term is not statistically significant, hence the moderation hypothesis is rejected.
In order to test the effect of the independent variable on equity commitment, the same efforts as for the first dependent variable are applied. Model 4 shows the effect of the control variables in the hierarchical multiple regression. The model is significant (F = 7950.586; p = .000), and it can be deducted that the control variables explain 45 per cent of the variance of equity commitment in the subsidiary. In contrast to the results of the subsidiary location, all control variables illustrate a statistically significant effect on equity commitment. Firm size (ß = -.105; t = -24.843; p = .000), industry effects (ß = -.488; t = -102.192; p = .000), firm revenue (ß = -.325; t = -65.234; p = .000) show these have an important negative impact on equity commitment in this research as the results are statistically significant. Public firm (ß = .087; t = 18.530; p = .000) is significantly positively related to equity commitment.
Next, the independent variable is included in the model (model 5). The results demonstrate that subsidiary location provides only little additional explanatory power by increasing the R Square by 0.01 per cent (R Square change = .001; F = 6365.702; p = .000). The model indicates that the relationship between trade policy uncertainty and equity commitment is negative and statistically significant (ß = -.015; t = -3.851; p = .000). Therefore, the results show support for the second hypothesis, and the second hypothesis is thus accepted.
To analyze the moderation effect of firm experience on the relationship in hypothesis two, the interaction term of the independent variable and moderator variable is again used in the model. The moderation effect of firm experience is investigated using a multiple regression that includes the corresponding moderating variable and interaction term in
addition to the independent variable and control variables. The model explains an additional 0.09 per cent increase, leading to 46 per cent of the variance in the dependent variable, and is statistically significant (F = 4725.033; p = .000). The results of this regression analysis show that firm experience is positively related to trade policy uncertainty and is statistically significant (ß = .109, t = 19.348, p = .000). The results for the interaction term show a negative relationship that is not significant (ß = -.009, t = -1.220, p = .000). Hence the fourth hypothesis is rejected.
Table 3: Variable means, standard deviations, and correlations
Variables Mean Std. Dev. 1 2 3 4 5 6 7 8
1. Trade Policy Uncertainty .53 .499 1
2. UK Subsidiary Location .06 .239 -.014** 1
3. Total Equity Commitment 53.24 4.248.536 .033** -.068** 1
4. Firm Experience 75.83 53.316 .020** -.050** .246** 1
5. Industry Effects Dummy .51 .500 -.053** .052** -.583** -.122** 1
6. Firm Size 85849.00 59.965.805 -.004 .024** -.083** -.294** -.219** 1
7. Total Turnover (US millions) 532.306.671 6.048.365.614 -.056** .105** -.456** -.188** .271** .283** 1
8. Public Firm Dummy .76 .429 .034** .061** .107** -.001 -.319** .107** .380** 1
Table 4: Summarized results of the hierarchical multiple regression
Dependent variable: Subsidiary Location Dependent variable: Equity Commitment
Variables Model 1 Model 2 Model 3 Model 4 Model 5 Model 6
Hypothesis 1 Hypothesis 3 Hypothesis 2 Hypothesis 4
Control Variables
Firm Size .010 (.000) .010 (.000) .000 (.000) -.105*** (.000) -.105*** (.000) -.071*** (.000)
Industry Dummy .051*** (.003) .051*** (.003) .045*** (.003) -.488*** (.406) -.488*** (.406) -.469*** (.408)
Firm Revenue .069*** (.000) .069*** (.000) .067*** (.000) -.325*** (.000) -.326*** (.000) -.321*** (.000)
Public Firm Dummy .050*** (.003) .050*** (.003) .050*** (.003) .087*** (.466) .088*** (.467) .088*** (.463) Independent Variable
Trade Policy Uncertainty -.009 (.002) -.043*** (.004) -.015*** (.321) -.009 (.554)
Moderating Variable
Firm Experience -.032*** (.000) .109*** (.005)
Firm Experience x Trade Policy Uncertainty .001 (.000) -.009 (.006)
Model Fit N 39407 39407 39407 38788 38788 38788 R2 .013 .013 .014 .451 .451 .460 Adj. R2 .013 .013 .014 .450 .451 .460 F-Value 131.49 105.806 80.71 7950.586 6365.702 4725.033 P-Value .000 .000 .000 .000 .000 .000
V. Discussion
The present study is guided by the following question “What is the effect of trade
policy uncertainty caused by raising trade barriers in economic unions? And more specifically, how do remaining member states’ investment patterns with the departing state change after the announcement of withdrawal?” This study explores how the
emergence of trade policy uncertainty, as caused by the withdrawal of a country from an economic union, influences the subsidiary location and equity commitment in that country. In addition, the moderating effect of firm experience is studied. Based on previous literature, it was expected that subsidiary location and equity commitment would be negatively impacted by trade policy uncertainty, and that firm experience would moderate these relationships. The findings of this study disprove most predictions.
The results of the hierarchical multiple regression provide interesting findings. In none of the models testing the moderation, the interaction term is significant. However, there are changes in the significance of the independent variables when introducing the moderator variable in the model. In model 2, the independent variable (trade policy uncertainty) is not significant, but when the moderator variable (firm experience) is added in model 3, then the independent variable becomes significant. This could mean that trade policy uncertainty is related to firm experience that is added in the models. Because of this relation between the independent variable and the moderator variable, the independent variable already reflects the effect of the moderator variable in model 2, as the moderator variable is not yet in the model at that point. In the second half of the analysis the opposite happens. In model 5, the independent variable is significant when the moderator variable is not yet in the
model. Because the two are related, the independent variable already absorbs the effect of the moderator variable. So the independent variable is actually significant because it is related to the moderator variable. When the moderator variable is added to the model, it can exert its own influence and this effect doesn't have to go through the independent variable anymore. As a result, the independent variable loses its significance.
The changes in significance show two things. Firstly, there is a significant negative effect of trade policy uncertainty on subsidiary location when a negative moderation of firm experience is presented in the model. Hence, the change between model 2 and 3 might be a suppression effect. This means that there is a situation in which the predictor has a significant effect, but only when another variable is kept constant (Field, 2013). Secondly, the second hypothesis is statistically significant when firm experience is not present, though it is not significant when firm experience positively affects the relationship. The change between the last two models might thus be due to a confounder effect. A confounding variable is a variable additional to the predictor variable that potentially affects the outcome variable (Field, 2013). Consequently, it seems like there might be a strong relationship between trade policy uncertainty and firm experience, even though the interaction terms are not significant.
I. Academic relevance
Since the seminal paper of North (1990), there is no doubt that institutions affect FDI. However, the literature remains inconclusive on the exact institutional factors that influence FDI, and cannot generalize country-level findings across different research settings (Bailey, 2018). Hence, it is advised to study the influence within specific contexts. The present research does this by analyzing trade policy uncertainty in an economic union.
The effect of trade policy uncertainty on subsidiary location does not show statistical significance in this research. Previous research shows that being part of an economic union is a direct determinant for MNE entry mode decisions, but that there are time-dependent characteristics of governance choice by the firm (Blevins et al., 2016). This entails that the more the economic region becomes integrated, the weaker the relationships between being part of an economic union and using intensive entry mode choices become. The authors rationalize this shift through two explanations. First, the evolutionary integration of the economic region directly affects the MNEs’ governance choices, and second, countries’ membership to an economic region permits these to focus on strategic issues instead of external factors (Blevins et al., 2016). In this case, the effect of trade policy uncertainty on subsidiary location could be explained by both accounts. The deliberalization is part of the evolution of the economic union that affects all member states. Hence, all MNEs need to adapt their international strategy and no MNE has an advantage over another, as the situation is unprecedented. MNEs of member states could then remain in the withdrawing state by focusing on strategic issues instead of external pressures.
The negative effect of trade policy uncertainty on equity commitment is accepted in this research. Previous literature has researched the effect of uncertainty and risk on the MNE’s entry mode choice (Henisz, 2000; Müllner, 2016). Though these studies accept the importance of uncertainty in MNE entry mode choice, these show a variance of results, and disagreement on the exact effects. The present research demonstrates the direct effect of trade policy uncertainty on equity commitment in an economic union. In this manner, the research adds to the literature by acknowledging that developed economies are equally under threat of trade policy uncertainty.
