Institutions, Culture and Foreign Direct Investment in Transition Economies: Does Culture Matter and Why?

Abstract

The aim of this research is to analyse the importance of cultural and institutional determinants in attracting FDI to transition countries. We rely on gravity econometric framework and examine the impact of cultural and institutional factors on FDI using bilateral FDI flows between home (i.e. major trading partners) and eight transition economies in the period 2000–2018. We study this relationship in an integrated framework considering principal gravity forces, traditional FDI determinants, policy and institutional factors. We provide strong and robust evidence that cultural factors, depicted in Hofmann cultural indices, influence MNCs’ locational decisions. Other things held constant, specific cultural features seem more important than formal institutions, which seems at odds with standard neoclassical propositions, and shed some new light on the way we understand international business transactions.

1. Introduction

Foreign Direct Investments (hereinafter: FDI) has been largely found to positively affect economic growth in transition economies. Increases in FDI have been associated with productivity and export growth of local companies via knowledge spillovers and complementary effects on domestic investment. The impact of FDI on economic growth seems, however, conditional on the level of human capital and absorptive capacity of a host economy. Determinants of FDI in transition economies have been intensely researched highlighting the importance of traditional factors, institutions and policy choices in determining locational decisions of multinational corporations (MNCs). Although informal institutions and cultural factors have increasingly been characterised as important factors that off-set for the underdeveloped institutional capacity of transition economies, the impact of cultural ties on FDI remains fairly under researched. Informal economic structures and cultural similarities emanate trust and enable strong business ties across borders. How important are these factors in explaining differences in FDI flows among transition economies is the principal question investigated in this research.

Culture, in a broader sense, means a pattern of behaviour based on values and beliefs that develops over time in a particular society. While culture in a narrower sense represents the way of life of a social group, i.e. a society that includes language, tradition, knowledge, customs, laws, art and other tangible and intangible features of social life that are passed down through generations, cultural is of course subject to change. Culture includes a set of values and attitudes of a homogeneous group of people that are passed down from generation to generation, and these patterns of behaviour change rather slowly.

The importance of cultural factors has been increasingly emphasised in the FDI literature. In particular, the impact of cultural distance between home and host countries has found to be significant in number of studies investigating the role of culture in explaining FDI flows. However, few studies concentrated on the transition economies of Central and Eastern Europe. These countries are viewed as specific in terms of both scope of economic and institutional transformation, and specific (common) legacies of socialism. We contribute to recent literature on FDI in transition economies, by analysing the significance of broad set of institutional and cultural indicators ought to influence MNC decisions on where to invest. A special reference is given to the discussion on the relationship between formal and informal institutions, assumed to be predominantly depicted in cultural dimensions of a specific country. In addition, the relevance of specific Hofstede cultural dimension to foreign firms is brought to the fore. Having said this, the hypothesis tested imply ‘favourable cultural context’ that is, specific cultural characteristics that are assumed to be preferred by foreign firms. We posit somewhat universal aspects of culture related to Hofstede cultural dimensions that constitute favourable cultural environment to foreign firms. Furthermore, we make use of the gravity model and the panel data framework to examine the importance of relative differences between home and host country characteristics in explaining bilateral FDI stock. The gravity-panel empirical framework allows us to draw important and detailed conclusions with respect to the relative importance of formal institutions vs. cultural aspects to foreign firms. Our dataset includes 10 source countries (i.e. major trading partners) and 8 home countries (transition economies of Central Eastern and South East Europe for which the cultural indicators were available).

This paper is structured as follows. Next we elaborate on basic theoretical propositions underpinning the mechanism of institutional and cultural influence on FDI, with special reference to empirical work on the matter. In discussion theoretical and empirical issues, we present the conceptual framework for the empirical strategy used in this analysis discussing number of important issues including interplay between culture institutions and FDI, definitions and measurement issues, and research hypothesis. The third section relates to the empirical analysis where detailed description of the model, data and variables and methodology employed is provided and followed by the interpretation of the empirical results. The conclusion follows.

2. Institutions, culture and FDI: conceptual framework and literature review

2.1 Institutions: what they are and why they are important?

Institutional environment often encompasses political systems, policy making and policy enforcing institutional structures which determine economic structures at the national and sub-national levels. It includes institutional setting that provides formal rules of the game and sets forth the incentives to economic/societal agents as well as informal norms, set of beliefs, systems of values, customs considered also an important feature of the institutional environment of a given country. Different scholars perceive differently the relative importance of these various components, including the role played by formal and informal rules and conventions as well as the importance of and role played by organisations, encompassing both economic and social agents of various sorts.

They are competing theoretical perceptions and different values attached to institutions and organisations in the contemporary literature [1, 2, 3]. Generally, institutions are perceived as frames or rules of the game while organisations are defined as social agents constituting and carrying these rules [2]. Importantly, the relationship between institutions and organisations is not a straight forward one. There is a general consensus among scholars that institutions principally evolve in response to market- related imperfections, and as such, institutions are considered a mechanism to enhance efficiency associated with economic transactions. However, whether institutions evolve primarily in response to changes in values, perception and attitudes embodied in organisations (i.e. various social agents) including theirs’ perceived inefficiencies in functioning of the market, or whether institutional development can be viewed as principally exogenous process where norms dominantly govern actions of social agents is still a debated issue [1, 2, 3]. Relate to this is a question whether institutional development is constrained by organisations’ preferences and capacities? Arguably the answer to these questions depends on how we perceive institutions and how do we value the relative importance of institutions vis a vis organisations.

The theoretical conceptualisation of institutions has mostly favoured the stream of literature which views institutions as frames or rules of the game which both guide and constrain actions of social agents i.e. organisations of various sorts including political and economic agents that make up societies. This stream of literature suggests the dichotomy between institutions and organisations. Accordingly institutions reveal formal and informal rules and conventions which set the structures within which, and upon which societal agents act [1, 2]. More precisely, the dichotomy in the words of [2] implies a clear conceptual distinction between institutions and organisations as follows; institutions define ‘the rules’ of the game and organisations are ‘the players’ by whom the game is played. Similarly, [4] suggests that institutions provide the set of rules defining frames within which organisations act. This is where the importance of cultural dimensions become crucial in understanding the institutional performance and in particular the differences in economic performance among countries amid similar institutional development and/or quality of institutions.

In this respect, ‘good’ institutions are only necessary but not sufficient condition to promote successful change within a society and/or to achieve desired societal goals. This view implies that although institutions evolve in response to market failures encompassing various forms of imperfections related to economics transactions and exchange they evolve in particular socio-historical context, and in accordance with prevailing preferences, norms and ethics of organisations that embody, interpret and influence the institutional conditions i.e. the specific (endogenous) rule setting [5]. These endogenous norms and values of a society have been embedded in what we call informal institutional structures of the economy. Put differently, formal institutions are embodied in culture, and culture matters for understanding the link between formal and informal institutional setting of individual countries [2, 6, 7]. The culture of a society determines informal behavioural patterns of economic agents, and by that the quality and the efficiency of formal institutions. This is to say that institutions, e.g. the relevance of formal rules, adherence to formal principles and legal provisions rests within organisation and agencies that is with people who implement those rules. This stream of literature, which identifies culture as important aspect of formal institutions, may help us disentangle the relationship between formal and informal institutions, and, in particular, may help us comprehend how societies with similar quality of formal institutions have divergent economic outcomes.

As pointed by [8], it is possible that societies with identical institutional setting perform differently assuming that societal agents are not ‘passive’ but influence and determine the outcomes of any particular institutional structure in relation to their competences and preferences as well as in relation to what is called informal institutions. The latter include rather endogenous institutional features or their evolution such as norms, social ethics, prevalence of (old) institutional legacies, initial institutional conditions, institutional and individual’s values and competences. This would imply that it does not suffice to develop ‘quality’ institutions in the form of extensive and desirable legislative and institutional infrastructures, or ‘optimal’ set of rules, but also necessary organisational capabilities that affect policy impacts and subsequent institutional change. Institutional development should be perceived as a long-term process of societal change, a one that certainly involves the development of ‘better’ or ‘improved’ conventions but importantly the process that rests on the commitment and competences of prevailing human and organisations kinds involved in those processes.¹

Here it is important to emphasise that theoretic perspective of institutions matters for: (i) our understanding of the evolutionary dynamics of institutional development and change; ii) the importance of informal institutional structures and their link with formal institutions; as well as iii) the way we measure institutions in our empirical analysis. As pointed by [8]: “Whether institutions are viewed as endogenous to the relevant domain or exogenously set in the polity may have significant implications for the role of public policy.” This is to say that if institutions are exogenous than we could relatively easily transplant the best practices of other (more advanced) countries in the forms of formal rules and conventions and anticipate increases in efficiency and welfare. If the reverse is true, and if institutions evolve principally in relation to a country specific historical, political and cultural context assuming interdependence among institutions, constrained and influenced by informal norms and social ethics, competences and capabilities of human and organisation kind, then the intended outcomes and consequences of any institutional conditions would vary considerably in relation to these important but intrinsic features of a given country. These raise important considerations for researches analysing the role of institutions in economic performance as well as international business.

The comparison of transition institutional reform, particularly the evidence revealing contrasting experiences and institutional performance across countries, however, led to doubts and seriously questioned the conventional wisdom of straightforward transplantation of practices of developed market economies [9, 10, 11]. Contrary to what has been expected, the years of transition saw institutional building and reform as exceptionally challenging and complex. The initial years of transition witnessed the remarkable differences in institutional progress among transition economies (EBRD, 2001). Empirical evidence point to the highly intrinsic and endogenous nature of institutional development including the varying institutional performance among transition economies [9, 10, 11, 12]. Transition economies were faced with the necessity to reform their economic and institutional structures on a large scale moving from centrally planned to free market economic system and resource allocation. The conventional economic wisdom implied that former centrally planned economies needed to develop institutions which underpin free market transactions and well functioning of the markets as quickly as possible. The importance of institutional environment conducive to rapid market development has been put high on the transition reform agenda. The initial institutional development in transition economy context reflected the establishment of institutions fundamental to free-market economies including setting proper incentives through policies of macroeconomic-stability, price liberalisation, denationalisation and privatisation, as well as through institutions underpinning effective financial sector reform and private sector growth. The prevailing conceptualisation of institutions at the time envisaged that institutions are somewhat easily transferable, exogenously created whereby institutions are built following best practices elsewhere [13]. However, over the course of transition, growing empirical evidence on the matter of institutional change in transition suggested that the important historical, political and social factors have played a role. The empirical studies pointed to the interrelatedness and interdependence between institutions and diverse political, cultural and economic contexts of a given country [8]. Among others, the initial institutional conditions and institution-related legacies, as well as cultural dimensions largely influenced the pace and character of institutional development in transition economies (see [11, 14]).

Despite these efforts, we do have a limited understanding of the processes and lack meaningful explanation on the diverging pattern of institutional development among countries including TEs. Analysing relationship between formal and informal institutions may help disentangle differences in institutional performance among transition economies. Such analysis is fairly constrained by number of difficulties including conceptualisation of the relationship, as well as data limitations and number of measurement issues related to informal institutions of the economy. We believe that indicators of cultural dimensions may help comprehend at least some aspect of this complex relationship. In view of this, we argue, that it is important to study institutions in an integrated framework, and point to the relevance of cultural factors.

In this paper, an attempt is made to illuminate the importance of formal institutions relative to the distinctive cultural features of a society in comprehending differences in FDI flows across transition economies. We use a narrow-definition of institutions and focus on institutions revealing principally formal rules specifically related to FDI such as the rule of law and corruption. Relying on previous empirical work, we emphasise the importance of corruption [15, 16, 17, 18], regulatory and governance indicators [19, 20, 21], as well as legal indicators in our empirical analysis. We identify institutional variables to be included in the model based on these results, and include three institutional variables (Corruption, Good governance and Rule of Law and Efficiency). Since we investigate the impact of institutional factors on FDI in transition economies, we do not use Political stability variable amid low variation in the data, and relatively similar Political stability index across CEE transition economies. In the light of the forgoing discussion, however, we acknowledge that by the way we measure institutional dimension in this analysis we possibly do not account for complex informal social structures (including social ethics and norms, preferences and capabilities of organisations) which may well influence the outcomes of institutional environment in a given country. This is why we include cultural indicators (Hofstende Cultural Dimensions) in the analysis in order to account for important aspects of both institutional and cultural behaviour of individual (host) countries that may influence FDI flows in transition economies.

While there is considerable number of studies analysing the impact of formal institutions on FDI, very little empirical research has been done on the importance of culture, and cultural factors in determining FDI. Furthermore, we study the impact of institutional and cultural factors on FDI in an integrated framework, where we assess the relative importance of formal institutions vs. cultural factors.

2.2 Culture and its relevance for understanding institutions-FDI nexus

Informal institutions are often considered important determinant of FDI since they could compensate for the deficiencies associated with underperforming or poor quality formal institutions. Despite this, they are often overlooked in the FDI literature whereby the emphasis is given to the quality of formal institutions per se. The rationale behind is that while formal institutions ensure efficiency of foreign operations in a new environment, informal institutions mostly favour local economic agents. Local agents are assumed to have better access to political and local facilities and processes. Given this, reliance on informal institutions in ensuring efficient economic transactions is least favoured by MNCs. A study by [22, 23] have shown that institutional development in transition economies has had an impact on foreign investors’ strategic decisions, arguably their entry modes, whereby quality of formal institutions seems of greater importance for establishing wholly owned ventures. The study by [22] reveals evidence that quality of institutions does seem to impact type of ownership related to FDI, where poor institutional development is more likely to result in network- types of FDI (i.e. joint-ventures, contracts). This is why the FDI literature mostly emphasise the relevance of formal institutions as locational advantage as reliance on informal institutions tend to increase transaction costs of foreign investors relative to domestic agents. The cultural features seem to have been disregarded as important factors which influence the way in which markets develop and evolve.

Notwithstanding this, in this paper we argue that cultural dimension is important determinant of FDI, in the sense that culture ‘shape’ formal institutions (see for instance [24]). Cultural dimensions of a society depict ways in which ‘nations’ tend to understand the rules and norms of social behaviour. The role played by formal institution(s) within a society and their relative importance vis a vis informal social structures (e.g. social networks, linkages) is deeply rooted cultural phenomena. The perception of formal institutions that prevails among general public, on this particular matter, is important to be understood, when examining the relationship between formal institutions and FDI. Having said this, culture may reflect on ‘tacit’ aspects of general-purpose or more specific ‘market-enhancing’ institutions within countries. The diverse and distinct concepts of social behaviour, present important features of a society that not only influence and model the behaviour of local economic agents, but affect the quality and the efficiency of formal institutions. Reliance on informal institutions as opposed on formal institutions may well be associated with weak and malfunctioning formal institutional structures. On this ground it seems reasonable to posit that informal institutions reflected in cultural dimensions are also likely to influence MNC’s decisions on where to invest.

Overall, we strongly believe that the essential question on the matter of culture-institutions-FDI nexus, is to what extent ‘formal institutions’ and the why we measure them, reflect society’s adherence to formal rules, as opposed to society’s ‘modus operandi’ e.g. collective actions, practices, behaviours that may contradict formal codes of conduct? Here it is worthwhile mentioning the theory of ‘institutional stickiness’ which firmly explains the relationship between culture and institutions [25]. The authors posit that formal institutions are stuck to what they call ‘metis’, which may be defined as ‘values’ that are largely ‘exogenous’ to people and that shape our social relations and constitute important unwritten behavioural patterns.

Therefore, relying solely on formal institutions, and formal institutional indicators of one country, including various legal indicators that are of specific interest to FDI, when analysing the importance of formal institutions, may be associated with ambiguities and uncertain ‘policy’ implications. This is not to say that formal institutions are of less importance, but it is at least important to acknowledge that ‘good’ institutions are embodied in culture. These societal values, attitudes and norms evolve overtime and reinforce formal institutions [25, 26]. This means that overtime, cultural changes influence acceptance of formal institutions as values of a society in general, and society at large adheres to these formal ‘rules of the game’. Along the lines of these theoretical propositions, [6] investigates both direct and indirect effect of culture on FDI, and finds that culture impact FDI indirectly through its impact on formal institutions, as well as directly. The indirect impact of culture on FDI is mediated via formal institutions, which confines the hypothesis that culture ‘shape formal institutions’. Essentially, the impact of cultural factors is found to be more important than the impact of formal institutions. Similarly, a recent study by [27] finds significant and greater effect of cultural factors (embeddedness vs. autonomy; hierarchy vs. egalitarianism, mastery vs. harmony) than formal institutions in cross-country regressions. Both studies render support to the theoretical propositions underlying the importance of cultural factors in comprehending role played by formal institutions, as well as that distinctive features of national cultures influence FDI flows.

Culture seems to reveal hidden behavioural patterns that underpin societal prosperity, including society relation to and the perception of responsibility, ethics and trust. The idea that these norms affect companies’ efficiency and growth prospects cannot be dismissed. On the contrary, these factors should be perceived as important determinants of FDI that not only minimise transaction costs, but also enhance productivity potential of foreign affiliates, and/or simply create an environment conducive to business growth. Such an environment is perceived as friendly and or familiar market from MNC perspective. What kind of information MNC search for when deciding about new investment site is important? Do managers look at formal institutional indices, or have other sources of knowledge and information that reveal ‘true’ that is prevailing aspects of social relations, ethics and norms? Studying the impact of formal institutions on FDI, in an integrated framework in which we control for cultural factors, along with traditional FDI determinants, becomes of crucial importance.

In what follows we discuss in greater detail the relevance of culture in international business and briefly review past empirical research on the role of culture in attracting FDI.

2.3 What role for culture in attracting FDI

According to the literature cultural dimension can influence foreign direct investments in two ways. First theoretical proposition suggest that more culturally diverse societies tend to be perceived as favourable cultural environment by MNC, while the second theoretical proposition implies that foreign investors prefer to invest in cultures similar to their own. These two distinct theoretical perspectives imply first that more culturally diverse societies positively impact foreign direct investments, and second that lower cultural distance between home and host countries positively affects FDI. As for the former, culturally diverse countries reflect on more open and welcoming societies that are viewed positively by foreign companies. As for the latter cultural difference between home and host countries is often associated with high transaction cost arising from uncertainty and lower FDI.

2.3.1 Cultural diversity and religion as cultural factors influencing FDI

Alesina [28] argues that high level of ethnic, linguistic and religious diversity requires well functioning governments and as such is positively associated with FDI. Additionally, more culturally diverse society is more tolerable and less reluctant to foreigners, which is perceived positively by MNCs. A number of research investigate the impact of religious factors including religion diversity and pluralism [29, 30] religious similarity [31, 32] and religious groups [31, 33] on FDI and find that religion does influence FDI. Accordingly, religious pluralism is found to be positivelly associated with FDI and thus more important compared to religious similarity, while a study by [31] finds positive relationship between all monotheistic religious groups (Catholic, Protestant and Orthodox Christian) and FDI, but for Islam. According to the results of their study, significantly negative relationship is thus suggested between Buddhism and FDI. No doubt, the results of their study suggest that MNCs are not indifferent to culture, broadly represented by diverse religious groups, and that only certain religious groups are positively associated with FDI. However, the empirical evidence on the matter is scarce and far from uniform to draw any sensible conclusions.

In this paper we, however, argue that, while religious believes may influence attitudes toward free market, competition or foreign investors, it may be fairly misleading to associate specific attitudes and values to individual religious groups per se, based on our assumptions, generalisation and even pre-assumptions about religious constituencies of individual nations. This tendency to assign specific societal attributes to certain religious groups such as is the case of the La Porta et al. study [34], which prescribe and impose ‘low institutional quality, institutional inefficiency, political and economic instability, high level of corruption and tax invasion etc.’, to reside and prevail within the so called ‘hierarchical religious’ such as Catholicism, Orthodox Christianity and Islam is rather forged, deceptive and ambiguous.

Certainly, a genuine and reasonable approach to study religious aspects of cultural dimension and its relationship to FDI and/or economic growth need be based on attitudes toward certain economic concepts and principles, as well as on values assigned to those, including wealth and growth, competition-rivalry and struggle, market openness and foreign investors. In case, significant differences assigned to those concepts, could be associated with specific religious believes or related value-systems, then we could sensibly argue of the prevailing religiously rooted ‘cultural’ differences. This is to say, that we need to investigate the link between religious beliefs and values assigned to aforementioned economic concepts. How differently are these concepts perceived and valued by different religious groups need be the principal question investigated, and not something a priori assumed and assigned to specific religious groups. In line with this reasoning, for instance, Guiso, Sapienza and Zingales [33] conclude that Catholics and Protestants are more positively associated with attitudes favouring market-efficiency and economic growth, while Muslims are found to be negatively associated with competition. Last but not lease, if we consider far-reaching cultural differencies existing within supposedly homogenous religious groups across different nations, such as is the case of Islamic countries (Malasya vs. Tukey vs. Saudi Arabia vs. Iran), or cultural differences across supposedly Catholic states such as is the case of Ireland vs. Poland vs. (South) Italy, we clearly face the measurement problem that may bias the results. This could partly explain why number of studies including [34] failed to find significant impact of religious groups per se on either economic growth or FDI. The notable exception is the aforementioned study by Lucke and Eichler [31].

Overall, in this paper we argue that analysis of cultural influence on FDI, that measure cultural diversity via dichotomous variables depicting religious group(s) fails to sensibly reflect on the important cultural dimensions that are of relevance when studying the link between culture and FDI. This is to say, that it is fairly improper (invalidate) to a priori assign that certain religious groups are indifferent to economic growth or adversely oriented toward foreigners, unless we truly conceptualise and measure links between prevailing religious beliefs and some ‘relevant’ economic categories or attitudes. Most studies investigating the link between religious groups and FDI fall short of addressing this issue. We first need to conceptualise on, and empirically establish the link between certain believes and attitudes including individual aspirations for growth, attitudes toward risk and competition, attitudes toward other groups/individuals and specific religious groups, as suggested by Guiso et al. study [33], before we examine the relationship between supposedly distinct religious groups and FDI or economic growth.

The empirical evidence on the matter is far from consistent and far from its mature phase. In light of this discussion, we argue that although religion constitute important cultural dimension of a society, it may well be inappropriate and misleading to include religious group(s) as dummy variables in regression equations to estimate the effect of culture on FDI. In contrast, considering religious diversity or similarity across nations may reflect on cultural distance that could be associated with costs of transition to new business or cultural environment. Most empirical studies have, in fact, followed this line of reasoning where, cultural similarity is considered important, a priori positive determinant of FDI, as we discuss below.

2.3.2 The relevance of cultural proximity (distance) as determinant of FDI

Kogut and Singh [35], posit that foreign investors prefer investing in countries culturally similar to their own. Sharing similar attitudes and values implies better knowledge of the local market, customers and business practices. Greater cultural differences between the host country and the source country lead to higher costs of doing business in another culture, such as the cost of obtaining information or the cost of searching to discover the specifics of the local bureaucracy [31]. This theoretical proposition has dominated the research on culture and FDI. Most empirical studies analyse the impact of cultural distance per se on FDI, and hypothesise that cultural similarity between home and host countries positively affects FDI flows.

Accordingly, the principal question investigated by researcher refers to the effect of ‘cultural distance’ on FDI. Siegel and Licht [35] in their analysis using instrumental variables (social factionalization, dominant religion, 19th century war experience, previous communist rule) measure how cultural distance in terms of egalitarianism vs. hierarchy affects FDI flows. The analysis is based on a 2005 Schwartz study [36]. The results obtained explain that the egalitarian distance has a negative and statistically significant impact on FDI flows. Similarly, [35] conduct a comprehensive analysis on the impact of cultural distance on FDI. They rely on ‘egalitarianism vs. hierarchy’ dimension of culture developed by Schwarz [36] and argue that the greater the distance between culture of origin and destination country the greater the difficulty in interacting with stakeholders in the host country. The results of their study confirm the negative and significant impact of cultural distance on FDI. Moreover, [31] study suggest that foreign investors from developed countries are negatively affected by greater ‘cultural distance’ when investing in developing and transition economies. Lee, Shenkar, and Li [37] come to similar conclusions when it comes to the impact of cultural distance measured by Kogut and Singh index on inward FDI in South Korea.

Number of empirical studies uses cultural proximity as determinant of FDI, relaying on common language, common history (e.g. colonial legacy, socialist past), and common border as cultural proxies. Most studies find significant and positive impact of cultural proximity along geographical distance on FDI [32, 38]. Lopez-Duarte and Vidal-Suarez [39] analyse how language distance affects the choice between greenfield investments and acquisitions when investing in other countries. 383 foreign direct investments from Spain in 44 different countries in the period 1989–2003 were analysed. The authors find strong support for the role of language distance as the main factor causing transaction costs. The results suggest that investors avoid acquisitions as a way of investing in countries characterised by high language distance. Bandelj [40] analysed the cultural connections (presence of a national minority) between investors and recipients of investments (hosts). It measures how the presence of national minorities affects the movement of FDI between pairs of countries. Bilateral flows of 11 Central and Eastern European countries (recipients of investments) and 27 investor countries were analysed. The author came to the conclusion that cultural ties that have historically been formed due to the presence of national minorities of the host country in the investor country, and vice versa, positively and statistically significantly affect FDI flows between the two countries.

Although, the results of majority of studies on cultural distance and FDI support the hypothesis that greater cultural distance negatively affects FDI, and that cultural proximity plays important role in attracting FDI, studies by Voyer and Beamish [41], Grosse and Trevino [42] find that cultural distance does not exert significant negative impact on FDI, and that cultural distance does not seem to influence FDI flows. Tang [43] reports mixed results on the impact of all cultural distance variables (four Hofstede cultural indicators) on FDI and concludes that ‘cultural difference does not always imply cultural conflict’. In similar vein, Barkema, Bell, and Pennings [44] (following [45]) argue that the risks arising from cultural differences can be overcome because investors can learn over time how to deal with those differences. According to them, experienced investors, both from developed and developing countries, ultimately do not consider cultural differences a significant obstacle.

Analysing the impact of cultural distance on foreign direct investment has, however, proven to be quite complex resulting in inconclusive and even contradictory empirical evidence. Part of the reason can be attributed to the fact that authors use different measures of culture from which they construct cultural distance variable, rendering support to the need to understand the mechanism underpinning the influence of ‘cultural distance’ on FDI. Moreover, the problem of measurement of cultural distance variable has been investigated by van Hoorn and Maseland [46]. The authors analyse the implications of using ‘cultural distance’ variable, defined as a difference between home and host country scores of one or more cultural dimensions, on robustness of the empirical results obtained. They conclude that one cannot compare the impact of this ‘cultural distance variable’ on FDI for different countries of origin. Following the conclusions emanating from their study, Kapas and Czegledi [47] construct a ‘cultural distance’ variable taking into account the problem of ‘the mixed impact of cultural distance and the culture in the host country’ when constructing cultural distance variable. Essentially, the results of their study suggest that the impact of culture measured in levels on FDI is greater than the impact of ‘cultural distance’ variables. The results of their study along the van Hoorn and Maseland [46] study clearly suggest the possible bias effect of earlier studies analysing the impact of ‘cultural distance’ on FDI.

In light of this discussion and in view of the important insights arising from the previous empirical work, in this study we analyse the impact of culture on FDI in transition economies while highlighting the following:

1. The importance of cultural features of host economy that are independent of culture of the origin country, that is of specific values that could be attributed to individual national cultures as ‘core’ to comprehending the cultural influences on FDI

2. The importance of analysing the impact of institutional and cultural factors in an integrated empirical framework in which we take control of both institutional and cultural factors and examine their relative importance on FDI

In view of the possible biases associated with the ‘cultural distance variable’ constructed by subcontracting the origin form destination country cultural scores, we refrain from using ‘cultural distance’ variable in our empirical analysis. On the contrary, we postulate the importance of specific and intrinsic cultural features that reflect on deep cultural traits and different cultural models, developed by Hofstede [48, 49] to be important determinant of FDI. In what follows, we discuss the relevance of Hofstede cultural factors as determinants of FDI inflows. The impact of these factors has been fairly under-researched in transition economy context.

2.4 Which cultural factors matter for FDI and why: measurement issues and hypothesis

We postulated earlier that culture is important aspect of informal institutions. As such culture is associated with way formal institutions function, their quality and efficiency. Apart from this, local culture is associated with ‘social risk’ of investment and transition to a new market embedded in social relations. Risks associated with cross-border business go beyond economic analysis and economic risks. The social characteristics are important determinant of FDI in that they influence operational and the external environment of business, influencing business success factors in the long run. Social characteristics depicted in cultural dimensions of a society are considered important to the internationalisation process ([50] as companies do not perform their businesses in isolation from other firms and/or networks of firms [51] nor do they construct their internal capabilities in isolation. Local work ethics, values and attitudes affect business performance of foreign companies through social relations of the workforce. All economic activities are ‘submerged in social relations’ [52]. Social characteristics and relations are embedded in cultural dimensions of a society.

Research on culture attempted to define important elements and dimensions of national culture relating to both conscious and unconscious set of beliefs, values and norms that reflect general attitudes and preferences of a society. Hofstede study and the model of national culture presents a systematic and pioneering work on the matter, that had a major influence on understanding cultural differences among nations [52]. Hofstede introduced four cultural dimensions of a society, namely Power Distance (PDI); Masculinity (MAS); Individualism (IDV); and Uncertainty Avoidance (UAI). The fifth cultural dimension, Long term orientation (LTO) was later developed and added as additional variable by Hofstede and Bond (1988). Further research on cultural dimension and its measurement resulted in the development of the Globe cultural dimensions (Global Leadership and Organisational Behaviour Effectivness, Kogut and Singh’s Index of cultural distance [53] and Schwarz Value Survey [36]. The literature has critically assessed various aspects of these cultural indices, including the Hofstede work on culture and cultural dimensions [54, 55, 56]. Most of criticism is related to the problem of time invariant nature of cultural indices including Hofstende cultural dimensions, and lack of genuine (socio) anthropological aspect of culture. In this research we follow arguments presented in [57] on the rationale of using Hofstede cultural indicators encompassing discussion related to the i) the benefits of using separate indices rather than aggregate cultural distance indices developed by Korgut and Singh (see [57]); ii) the stability of cultural values over time and the empirical evidence pointing to no significant variation of Hofstede indices over time; iii) the benefits of using Hofstede cultural dimensions over other indices that have been originally developed using Hofstede cultural dimensions such as is the case of the Globe indices or Kogut and Singh’s Index. Thus, Hofstede [58] argues that the Globe index is deficient amid its complexity and, as such is less useful in empirical analysis, while Shenkar [59] points that we lose important information relying on aggregate cultural index i.e. Kogut and Singh’s Index. In what follows we present the five Hofstede cultural dimensions, brifly review the empirical literature using Hofstede indicators and present the hypothesis.

2.4.1 Power distance index (PDI)

The first cultural dimension is Power Distance. This cultural dimension uncovers general perception of social inequality predominantly related to power concentration and social status [60].

This dimension represents the degree to which less powerful members of society within their institutions (family, school, etc.) expect and accept that power is unequally distributed. People are not equal by nature and inequality is present in every profession, but this fact is experienced in different ways. The distance of power actually shows how society faces inequalities. And the main issue that this index deals with is how society solves inequalities among people. People in societies that have a greater degree of power distance accept a hierarchical order in which everyone has their place and do not require further explanations. In these societies, independence is a feature of a small group of people, and others depend on them. On the other hand, in societies with a low degree of power distance, people try to equalise the distribution of power and look for explanations for the unequal distribution of power. There is interdependence between people, and subordinates perceive orders as ordinary people, and superiors are available to subordinates [61].

This cultural dimension uncovers general perception of social inequality predominantly related to power concentration and social status (Ferraro, 2002). Having said this, it’s worth emphasising that it indicates ‘the degree to which members of an organisation or society expect and agree that power should be unequally shared’ [62]. Applied to a firm level, it could be fairly assumed that the lower the index the higher the demand from workers within an organisation for more equally distribution of power (wealth) and higher the demand for ‘justification for’ and ‘rationale behind’ certain decisions or actions on a company level. All of these could lead to potential conflict between the workers and their superiors. Dispute and conflict(s) may arise from supposedly higher intolerance toward specific hierarchical structure of power, injustice or inequalities. Members of such society (workers within companies) prefer more horizontal organisational structure. On the other side, workers within societies with high power distance indices may be assumed to be: i) more submissive to ‘formal power structure’ and associated social distances; ii) to have lower levels of self-esteem associated with conflict-avoidance, (positive) affirmation and obedience. All of these lead to higher tolerance of: improper communication, improper job appraisal, overtime and unpaid work, and high tolerance of wage gaps that may be persistent within particular organisation. In light of this discussion it is firmly difficult to hypothesise what are the preferred cultural features by multinational companies seeking investments abroad. Whether a particular MNC prefer societies with high or lower Power distance depend on host of factors including company culture, organisational structure and motive of investment. As for the latter, we argue clearly that internationalisation of production activities through FDI seeking natural resource and/or cost-efficiency is concerned with, or values, work ethics that has a respect for ‘social hierarchical distance’. On this ground it seems plausible to argue that MNC seeking to access resources or to reduce production costs prefer societies with higher Power Distance. Given the specific context of our research the positive relationship between PDI and FDI could be assumed. However, from theoretical perspective, we do acknowledge that the sign of the relationship could go both ways.

H1: Power Distance positively affects foreign direct investments in transition economies.

2.4.2 Individualism versus collectivism (IDV)

Societies in which the degree of individualism is higher compared to collectivism value the efforts of the individual more than the collective and team results. Collectivism, on the other side represents a firmer social framework in which individuals can expect their extended family or some other group to care for them in exchange for unquestioning loyalty. Individualism, on the other hand, uncover preference that everyone is responsible only for themselves, the emphasis is on individualism and the ideal is leadership, belonging to an organisation is optional, the identity of the individual is based on his personal characteristics. Collectivism emphasises the organisation, the ideal is group membership, belonging to an organisation is a matter of morality, the identity of an individual is based on his belonging to the collective [52].

In view of this, in this research, we posit that more individualistic societies have positive attitudes toward competition and rivalry, with individuals being more determined and oriented toward self-interests, self-promotion and struggle for achievement. The individualistic society is thus characterised with proactive individuals, who strive to achieve their goals based on their individual efforts, and are less relying on social-framework. In view of this, we argue that individualist societies embody values and attitudes conducive to economic growth and efficiency, and are more likely to and/or that they willingly engage in ‘competitive (social) struggle’ that underpins productivity growth. Hence, the positive relationship between IDV and FDI is anticipated.

H2: Individualism positively affects foreign direct investments in transition economies.

2.4.3 Masculinity versus femininity (MAS)

These dimensions do not describe a person’s gender but character in humans. Societies ruled by masculinity indicate that society has propensities for heroism, assertiveness, authority, success, and material rewards for success. Society as a whole is more competitive, money and material goods are important, successful and independent people are respected, and people are valued according to the material goods they own. The opposite of masculinity is femininity which signifies modesty and a propensity for agreement. Also, indulgence and consensus are considered women’s values, as well as caring for the weaker, and people in society are more focused on quality of life [52].

According to [63], apparently, more masculine societies uncover cultural models that value material goods and material rewards for success, as opposed to quality of life and merits associated with common good that present attitudes of more feminine societies. Having said this, it could be reasonably expected that more Masculine societies are characterised by individuals and leaders who are competitiveness driven and who manage business operations by objectives. Such leaders are less sensitive to social or employee issues, they are decisive and act in isolation. On the other hand, leaders and managers of organisations of more feminine societies prefer consensus over aggressiveness. In view of this, it could be reasonably assumed that more masculine societies are more competitive societies, assumed to be societal attributes that foster better economic performance in general framework of a capitalist society. Notwithstanding this, it could be argued, that masculine culture traits embodied in an individual managers are not always preferred by MNCs. In case MNC’s organisational culture rests on assertiveness and collective affirmation, and in case company values organisational capabilities as opposed to self-affirmation of individual employees, more feminine model of culture may be preferred. In light of this discussion, we assume that more masculine societies could be both positively and negatively related to FDI.

H3: Masculinity positively (negatively) affects foreign direct investments in transition economies.

2.4.4 Uncertainty avoidance index (UAI)

This dimension expresses the degree to which members of society feel fear or discomfort from an unfamiliar situation. This index is often misinterpreted as risk aversion. Risk avoidance is a characteristic of the individual, while uncertainty avoidance is a feature of society. The basic question this index deals with is: should we try to control the future or just let it happen? We have societies that are actively dealing with the future, i.e. they have inherent control, and societies where events are out of control (fatalism). Countries with high DACI exhibit “rigid” behaviours and are intolerant of unusual behaviours and ideas. Such nations prefer strict and precise rules of conduct in society, regulations and guidelines to minimise uncertainty. People in such societies feel more comfortable when there is a clear structure and when society is well organised. On the other hand, countries where the DACI is low reflect a more relaxed attitude in which practice is more important than rules. In such societies there is aversion to any rules and norms. But if aversion is “moderate,” then it’s mostly societies that are more creative and flexible. People in such societies use common sense when making decisions and rely less on prescribed rules [52].

Overall, it could be said that societies with high uncertainty avoidance are characterised with high emotional resistance to change and may feel anxious about the future [60]. It could be reasonable assumed then, that these societies are reluctant to working in unfamiliar (uncertain) environment linked to foreign companies, and may be resistant to changes in organisational structure, or any changes in business conduct. As for the former, they can present additional obstacles to foreign companies, and may thus result in “discrimination by the government, consumers, and suppliers” [64]. As for the latter feature, it is probably least preferred by international business, amid the dynamics of changes of microeconomic determinants of global industry competitiveness and the constantly changing international business environment. Bearing this in mind, we hypothesise that uncertainty avoidance exhibit a negative influence on FDI.

H4: UAI negatively affects foreign direct investments in transition economies.

2.4.5 Long term orientation versus short term normative orientation (LTO)

A society from a long-term oriented environment cultivates virtues that are future-oriented - perseverance, thrift, while societies from a short-term oriented environment cultivate virtues that are related to the past and present - respect for tradition and fulfilment of social obligations. Societies that have a low LTO index generally prefer to maintain traditions and norms that have been respected in the past, while social changes are viewed with suspicion. On the other hand, societies with a high LTO have a somewhat more pragmatic approach: they encourage savings and innovation in education as a way to prepare society for the future [52]. As far as the latter characteristic is concerned it could be argued that MNCs prefer societies with long-term, rather than short-term orientation.

H:5 LTO positively affects foreign direct investments in transition economies.

2.5 Review of empirical literature on the impact of Hofstede cultural factors on FDI

Holmes et al. and Mac-Dermott and Mornah conducted research on how collectivism and future orientation affect the movement of inward FDI [65, 66]. Data from the Global Leadership project and the effectiveness of organisational behaviour (House et al., 2002) were used. The analysis was conducted on 50 countries (21 from Europe, 15 from Asia, 9 from North, South and Central America, 3 from Africa, 2 from Australia) for a period of nine years. They came to the conclusion that the greater presence of collectivism in society negatively affects the attraction of FDI, and that societies that are future-oriented promote capital investment of domestic entities. Bezpaliukh (2016) using Hofstede’s dimensions in his paper analyzes how cultural factors, primarily concentration of power, avoidance of uncertainty, and language influence the attraction of DSI. The analysis covers post-Soviet bloc countries (Estonia, the Czech Republic, Poland, Slovakia and Hungary) and the results suggest that a higher degree of uncertainty avoidance, a lower concentration of power and a common language have a positive effect on FDI inflows.

Bhardwaj, Dietz, and Beamish analysed how cultural factors, more precisely the avoidance of uncertainty and trust, influence the choice of locations of foreign companies [67]. They concluded that foreign companies prefer to invest in countries that have a higher degree of uncertainty avoidance and a high level of trust. Steigner, Riedy, and Bauman examined the impact of Hofstede’s cultural dimensions on DSI flows. [68]. They came to the conclusion through OLS regression that countries with civil law and countries with customary law prefer to invest in different countries and sectors. Hofstede’s cultural dimensions are also analysed by Goraieb [69]. They conducted an MRQAP analysis on the example of 45 countries and came to the conclusion that firms avoid investing in countries that differ from theirs in terms of the presence of a high degree of uncertainty. Also, firms prefer to invest in countries that are similar to theirs in terms of power concentration. What these four studies have in common is that they all use Hofstede’s cultural factors.

We will also use Hofstede indices in this study. Most research focus on specific factors such as collectivism and future orientation or avoidance of uncertainty and trust [65, 67]. We contribute to the literature on foreign direct investment by testing the widest possible set of cultural dimensions that can influence the investment decisions of foreign companies in a particular country. Unlike previous research, we include 8 transition economies as host countries. The analysis includes also the four countries of Southeast Europe (Albania, Bulgaria, Croatia and Serbia) that have not been previously investigated. Using a bilateral econometric framweork on FDI stock gives us the opportunity to question in more detail the importance of cultural and institutional factors in explaining differencies in FDI.

3. Empirical analysis

3.1 Model and data issues

In order to analyse the impact of institutional and cultural determinants on FDI, we pursue a panel data analysis. The empirical analysis covers four South East European countries (Albania, Bulgaria, Croatia and Serbia) and five Central and Eastern European countries (Estonia, the Czech Republic, Poland, Slovakia and Hungary) in the period from 2000 to 2018, containing information on FDI and host country characteristics. Each observation point in our dataset reveals FDI flows between home country “i” (ten major trading partners) and host country “j” in the period under observation. We develop a baseline specification of the following form:

$$\begin{array}{c}\mathit{Ln}{\mathit{FDI}}_{\mathit{ijt}}={\beta }_{\mathit{0}}+{\beta }_{\mathit{1}}\mathit{ln}{\mathit{GDP}}_{\mathit{it}}+{\beta }_{\mathit{2}}\mathit{ln}{\mathit{GDP}}_{\mathit{jt}}+{\beta }_{\mathit{3}}{\mathit{\text{DIST}}}_{\mathit{ijt}}+{\beta }_{\mathit{4}}{\mathit{\text{INFL}}}_{\mathit{jt}}+{\beta }_{\mathit{5}}{\mathit{\text{TradeO}}}_{\mathit{jt}}\\ +{\beta }_{\mathit{6}}\mathit{ln}{\mathit{\text{WAGE}}}_{\mathit{jt}}+\mathit{\text{Country}}+\mathit{\text{Time}}+{\epsilon }_i\end{array}$$E1

Where

lnFDI_ijt denotes log FDI stock between home i and host countries j in period t;

lnGDP_it denotes log of gross domestic product of home country i in the period t;

lnGDP_jt denotes log of gross domestic product of home country i in the period t;

DIST_ijt denotes log distance between capital cities of host and home countries;

INFL_jt denotes the inflation rate of the host country j in the period t;

TradeO_jt denotes exports and imports share in GDP of the country i in the period t;

WAGE_jt denotes relative unit labour cost of the host country j in the period t;

Country denotes country dummy variables used to control for time-invariant country specific effects;

Time denotes year dummy variables used to control for time specific effect; and.

εit —random error (structure eit determined by the Fixed Effect (FE) model).

We then investigate which particular features of institutional quality are important determinant of FDI flows in transition economy context while incorporating three individual institutional indicators in equations of the form:

$$\begin{array}{c}\mathit{Ln}{\mathit{FDI}}_{\mathit{ijt}}={\beta }_{\mathit{0}}+{\beta }_{\mathit{1}}\mathit{ln}{\mathit{GDP}}_{\mathit{it}}+{\beta }_{\mathit{2}}\mathit{ln}{\mathit{GDP}}_{\mathit{jt}}+{\beta }_{\mathit{3}}{\mathit{\text{DIST}}}_{\mathit{ijt}}+{\beta }_{\mathit{4}}{\mathit{\text{INFL}}}_{\mathit{jt}}+{\beta }_{\mathit{5}}{\mathit{\text{TradeO}}}_{\mathit{jt}}\\ +{\beta }_{\mathit{6}}\mathit{ln}{\mathit{\text{WAGE}}}_{\mathit{jt}}+{\beta }_{\mathit{8}}{\mathit{\text{INST}}}_{\mathit{jt}+}\mathit{\text{Country}}+\mathit{\text{Time}}+{\epsilon }_i\end{array}$$E2

where INSTjt represents institutional quality indicators developed by the World Bank including Government Effectiveness (GovEff_jt), Control of Corruption (Corrupt_jt) and Rule of Law (R_Law_jt).

As noted earlier, the purpose of this empirical analysis is to examine whether cultural effects play an important role in explaining differences in bilateral foreign direct investment flows in the context of transition countries. With this in mind and in line with the previously reviewed empirical literature, we further expand the analysis and include cultural distance variables in our model. Using Hofstede’s cultural dimensions, we decide to utilise the gravity equation to analyse the impact of individualism (IDV), power distance (PDI), uncertainty avoidance (UAI), masculinity (MAS) and long-term orientation (LTO) on FDI. More specifically, we have:

$$\begin{array}{c}\mathit{Ln}{\mathit{FDI}}_{\mathit{ijt}}={\beta }_{\mathit{0}}+{\beta }_{\mathit{1}}\mathit{ln}{\mathit{GDP}}_{\mathit{it}}+{\beta }_{\mathit{2}}\mathit{ln}{\mathit{GDP}}_{\mathit{jt}}+{\beta }_{\mathit{3}}{\mathit{\text{DIST}}}_{\mathit{ijt}}+{\beta }_{\mathit{4}}{\mathit{\text{INFL}}}_{\mathit{jt}}+{\beta }_{\mathit{5}}{\mathit{\text{TradeO}}}_{\mathit{jt}}\\ +{\beta }_{\mathit{6}}\mathit{ln}{\mathit{\text{WAGE}}}_{\mathit{jt}}+{\beta }_{\mathit{7}}{\mathit{PDI}}_{\mathit{jt}}+{\beta }_{\mathit{8}}{\mathit{IDV}}_{\mathit{jt}+}{\beta }_{\mathit{9}}{\mathit{MAS}}_{\mathit{jt}}+{\beta }_{\mathit{10}}{\mathit{UAI}}_{\mathit{jt}}\\ +{\beta }_{\mathit{11}}{\mathit{LTO}}_{\mathit{jt}}+\mathit{\text{Country}}+\mathit{\text{Time}}+{\epsilon }_i\end{array}$$E3

3.2 Data and variables

3.2.1 Dependent variable

In this research, we use FDI as our dependent variable which is the log of stock FDI between home and host countries in EUR. According to Christie (2003), looking at the stock level has the advantage of stripping out the business cycle and any other ‘time anomalies’. In addition, another reason for this choice is related to the functional form of the gravity equation because FDI inflows can be nil or even negative, which is something that the gravity equation cannot account for. The source of data for this variable is Database on FDI published by The Vienna Institute for International Economic Studies (WIIW).

3.2.2 Institutional variables

According to North (1990), good institutions influence economic activities through various channels, such as reducing transaction and production costs. Moreover, quality institutions help reduce operating costs, which increases profitability. Foreign investors are reluctant to invest in a risky and unconvincing environment and prefer locations that offer the best economic and institutional environment. Lucas (1993) suggests that in transition economies, institutional factors play an important role in attracting foreign investment compared to purely economic factors.

In order to estimate the impact of institutional determinants on FDI, we employ three indices developed by Kaufmann, Kraay, and Mastruzzi including government effectiveness, rule of law and control of corruption [70]. Each governance index ranges from −2.5 to +2.5, with higher scores corresponding to better governance outcomes. Government effectiveness (GovEff) assesses the soundness of the host country’s policies and the efficiency of the administration that implements them. Rule of law (Rule) measures the confidence of agents in the rules of society, including the quality of contract enforcement, property rights and the effectiveness of the judiciary. Control of corruption (CCorr) measures corruption among public and private officials and the extent of bribery. The source of data for these variables is the World Bank. A detailed description of each institutional variable used in this analysis is given in Table 1.

Variable	Description
Government Effectiveness (GovEff)	“Government effectiveness captures perceptions of the quality of public services, the quality of the civil service and the degree of its independence from political pressures, the quality of policy formulation and implementation, and the credibility of the government’s commitment to such policies.”
Rule of Law (RoL)	“Rule of law captures perceptions of the extent to which agents have confidence in and abide by the rules of society, and in particular the quality of contract enforcement, property rights, the police, and the courts, as well as the likelihood of crime and violence.”
Control of Corruption (CCorr)	“Control of corruption captures perceptions of the extent to which public power is exercised for private gain, including both petty and grand forms of corruption, as well as “capture” of the state by elites and private interests.”

Table 1.

Description of institutional variables.

Source [71].

3.2.3 Cultural variables

Our variable of interest is the cultural variable. Our measure of cultural variable is based on the scores developed by Geert Hofstede, which reflect country averages of individuals’ attitude toward power, uncertainty, individualism etc. A detailed description of each cultural variable according to Hofstede (2018) used in this analysis is given in Table 2.

Variable	Description
Power Distance (PDI)	“This dimension expresses the degree to which the less powerful members of a society accept and expect that power is distributed unequally. The fundamental issue here is how a society handles inequalities among people. People in societies exhibiting a large degree of Power Distance accept a hierarchical order in which everybody has a place, and which needs no further justification. In societies with low Power Distance, people strive to equalise the distribution of power and demand justification for inequalities of power.”
Individualism (IDV)	“The high side of this dimension, called Individualism, can be defined as a preference for a loosely-knit social framework in which individuals are expected to take care of only themselves and their immediate families. Its opposite, Collectivism, represents a preference for a tightly-knit framework in society in which individuals can expect their relatives or members of a particular ingroup to look after them in exchange for unquestioning loyalty. A society’s position on this dimension is reflected in whether people’s self-image is defined in terms of ‘I’ or ‘we’.”
Masculinity (MAS)	“The Masculinity side of this dimension represents a preference in society for achievement, heroism, assertiveness, and material rewards for success. Society at large is more competitive. Its opposite, Femininity, stands for a preference for cooperation, modesty, caring for the weak and quality of life. Society at large is more consensus-oriented. In the business context Masculinity versus Femininity is sometimes also related to as ‘tough versus tender’ cultures.”
Uncertainty Avoidance (UAI)	“The Uncertainty Avoidance dimension expresses the degree to which the members of a society feel uncomfortable with uncertainty and ambiguity. The fundamental issue here is how a society deals with the fact that the future can never be known: should we try to control the future or just let it happen? Countries exhibiting strong Uncertainty Avoidance maintain rigid codes of belief and behaviour, and are intolerant of unorthodox behaviour and ideas. Weak Uncertainty Avoidance societies maintain a more relaxed attitude in which practice counts more than principles.”
Long Term Orientation (LTO)	“Every society has to maintain some links with its own past while dealing with the challenges of the present and the future. Societies prioritise these two existential goals differently. Societies who score low on this dimension, for example, prefer to maintain time-honoured traditions and norms while viewing societal change with suspicion. Those with a culture which scores high, on the other hand, take a more pragmatic approach: they encourage thrift and efforts in modern education as a way to prepare for the future. In the business context, this dimension is referred to as “(short-term) normative versus (long-term) pragmatic” (PRA). In the academic environment, the terminology Monumentalism versus Flexhumility is sometimes also used.”
Indulgence (IND)	“Indulgence stands for a society that allows relatively free gratification of basic and natural human drives related to enjoying life and having fun. Restraint stands for a society that suppresses gratification of needs and regulates it by means of strict social norms.”

Table 2.

Description of cultural variables.

Source [52].

In our sample, the scores for the cultural variables can take values between 0 and 100, with a higher value indicating that power distance, individualism, masculinity, uncertainty avoidance and long-term orientation are more firmly entrenched in a nation’s culture. The source of data for this variable is Hofstede [71].

3.2.4 Control variables

Further, we incorporate a set of control variables. In our model we include information on gross domestic product of home and host country (GDPi and GDPj), distance (DIS), labour cost (LC) and inflation rate (INF), which proved to be significant in a number of previous empirical studies on FDI determinants.

As stipulated by the gravity model, both home and host countries’ market size are important determinants of FDI. The market size of the home country is a proxy for the economic power of the source country. The host country GDP serves as a proxy for the host country market size and thus the potential market for the investor’s products. We expect the coefficients of both GDP variables to be positive. The source of data for this variable is The Vienna Institute for International Economic Studies (WIIW).

Distance in this research pertains to geographic distance and serves as a proxy for all possible transportation and operating costs (see [72, 73]). The rationale behind including geographic distance to explain FDI is the greater cost of obtaining relevant information as well as the difficulties in managing affiliates in distant regions. The distance in this paper represents the geographical distance between the capital cities of home and host country in km. The source of data for this variable is CEPII database.

Furthermore, the prevailing factors for attracting FDI, besides market size and access to host market, certainly include the costs of the input factor. Previous empirical studies show that labour costs have a significant impact on FDI and play a crucial role in labour-intensive industries, as lower labour costs attract more investment. Studies suggest a double effect of labour costs. Numerous studies show that labour costs have a negative impact on foreign direct investment inflows, which is in line with the findings of Bevan and Estrin [74]. On the other hand, certain authors found that labour costs have a negative but statistically insignificant impact on FDI [75]. In our analysis unit labour cost is measured as average gross monthly wages. The source of data for this variable is UNECE.

We incorporate inflation rate as a control variable in our model. Inflation rate is often used as a proxy for macroeconomic stability in general. Political and macroeconomic stability along with transparency of legal regulations, such as land acquisition and repatriation of profits, can be important when making investment decisions [76]. The lack of macroeconomic stability creates a high degree of uncertainty for investment projects. Successful implementation of economic reforms in transition countries can be a good sign for potential investors to invest, given that stable macroeconomic performance implies lower investment risk. Thus, the lower the average inflation rate is in the host country, the more foreign investment will be attracted to the country [77]. We expect that foreign investment, ceteris paribus, will be attracted to countries with lower inflation rates. Source for this variable is IMF database.

Finally, we incorporate openness as a control variable in our model. Previous empirical results show that the openness of the economy is positively and statistically related to attracting foreign direct investment. Mphigalale states in its research that the openness of the economy contributes to attracting foreign direct investment in transition countries, but this must be complemented by appropriate macroeconomic policies [78]. The openness of the economy is the sum of exports and imports of goods and services measured by gross domestic product (Table 3). The source of data for this variable is the World Bank. Descriptive statistics for each variable are presented in Table 3 while the correlation matrix is in Appendix 1.

Variable	Obs	Mean	Std. Dev.	Min	Max
lnFDI	1112	9.02	0.43	6.80	10.69
lnGDPhost	1197	24.72	1.16	21.97	27.09
lnGDPhome	1197	27.24	1.40	23.73	29.01
Distance	1197	949.92	468.34	59.61	2126.43
Inflation	1197	4.09	8.96	−2	112
Trade oppeness	1197	111.92	36.34	24.17	192.34
lnWage	1064	6.55	0.63	4.24	7.33
IDV	1216	49.75	17.61	25	80
PDI	1216	68	19.49	40	104
UAI	1216	77.12	13.93	51	93
MAS	1216	59	25.72	30	110
LTO	1216	25	6.76	15	33
GovEff	1224	0.46	0.48	−0.84	1.16
RoL	1224	0.32	0.60	−1.27	1.37
CCorr	1224	0.17	0.51	−1.17	1.30

Table 3.

Descriptive statistics.

3.3 Methodology

In order to account for the panel structure of the data, we use bilateral fixed effects (FE) and random effects (RE) estimations. To choose between the FE and RE estimator, the Hausman (1978) test statistics are computed. The results of Hausman test showed that the model should be set as fixed effect model. This type of model is basically an Ordinary Least Squares (OLS) regression that includes a dummy variable for each country to account for country-specific effects (LSDV model). The OLS method is optimal if error processes have the same variance (homoscedasticity) and all of the error processes are independent of each other. According to Plümper et al. [79] panel data typically display contemporaneous correlation across units (i.e. large errors for country i at time t will often be associated with large errors for country j at time t), unit level heteroskedasity (i.e. variances of the error processes differ from country to country) and serial correlation (i.e. errors for each country show temporal dependence) making inference from standard errors produced by LSDV incorrect.

In order to test for possible serial correlation, we employ the Wooldridge (2002) test which indicates the presence of serial correlation in the panel data. In addition, the Breusch and Pagan test and Pasaran CD (cross-sectional dependence) test indicate a significant presence of heteroscedasticity and cross-sectional dependence/contemporaneous correlation. To avoid these problems, we follow Beck and Katz’s recommended procedure, using OLS with panel-corrected standard errors (PCSE), a method widely used in empirical research that assumes by default that the disturbances are heteroskedastic and contemporaneously correlated across panels [79, 80]. This estimation approache is the suitable method to test our hypotheses with the available data and provides efficient and robust outcomes, suitable for formulating accurate conclusions. We note that we do not make use the alternative the Generalised method of moments estimator due it is not feasible for our data set (see [81]).

Tables 4 and 5 report the results of the econometric analysis of the model specifications presented above. Specifically, the table reports OLS fixed effect panel data estimates with panel-corrected standard errors. We first estimate Eqs. (1) using three individual subdimensions of institutional development singly due to the problem of multicollinearity between the individual institutional variables.

	Model 1	Model 2	Model 3
GDP home	0.103***; (0.000)	0.103***; (0.000)	0.103***; (0.000)
GDP host	0.580***; (0.000)	0.605***; (0.000)	−0.646***; (0.000)
Distance	−0.000***; (0.000)	−0.000***; (0.000)	−0.000***; (0.000)
Inflation	−0.001; (0.139)	−0.001; (0.140)	0.001; (0.153)
Trade openness	0.001**; (0.009)	0.001**; (0.005)	0.001**; (0.000)
Wage	−0.607***; (0.000)	−0.601***; (0.000)	−0.583***; (0.000)
Gov. Effectiveness	0.050; (0.452)
Control of; Corruption		0.007; (0.888)
Rule of Law			−0.051; (0.369)

Table 4.

Regression results: FDI and institutions (OLS with PCSE).

^*

Statistical significance at the 10 percent level.

^**

Statistical significance at the 5 percent level.

^***

Statistical significance at the 1 per- cent level.

Notes: All the regressions include a constant, country and time dummies (not reported in the Table 4).

	Model 4
GDP home	0.109***; (0.000)
GDP host	0.281***; (0.000)
Distance	−0.000***; (0.000)
Inflation	−0.001**; (0.038)
Trade Openness	0.000; (0.537)
Wage	−0.293***; (0.000)
Power distance (PDI)	0.006***; (0.000)
Individualism (IDV)	0.012***; (0.000)
Masculinity (MAS)	−0.008***; (0.000)
Uncertainty avoidance (UAI)	−0.005***; (0.000)
Long-term orientation (LTO)	0.003***; (0.016)

Table 5.

Regression results: FDI and culture (OLS with PCSE).

^*

Statistical significance at the 10 percent level.

^**

Statistical significance at the 5 percent level.

^***

Statistical significance at the 1 per- cent level.

Notes: All the regressions include a constant, country and time dummies (not reported in the Table 5).

The “traditional” gravity variables in all specifications are proved to behave as expected. Both host and host countries’ economic size, proxied by GDP levels, are important determinants of FDI. The distance variable is also found to have significant implications for FDI flows which is in line with the gravity model hypothesis and previous empirical findings.

We find that labour costs adversely affect FDI flows. The coefficient on labour cost is negative and significant at 1% level. The coefficient of Trade openness is positive and significant at 5%, whereas the inflation rate is not suggested to influence FDI flows. This result may be explained by the observation that we are no longer in the early years of the transition process and all transition countries from the sample are characterised with relatively stable macroeconomic environment.

When it comes to institutions, the most important conclusion resulting from our analysis suggests that institutional variables do not exhibit significant influence on FDI flows in transition countries. The results obtained in this analysis are consistent with those obtained by Lucke and Eichler who study the impact of institutions and cultural factors in an integrated empirical framework [31]. Noteworthy is that institutional variables remain insignificant even when including lagged values, and the results are robust to different model specifications (i.e. manufacturing value added, productivity levels and differentials, population).

Regarding cultural determinants of bilateral FDI, Table 4 reports the results based on the Hofstede cultural frameworkandsummarizestheresults for thecoefficientson fivecultural indicatorsincludingindividualism (IDV), power distance (PDI), uncertainty avoidance (UAI), masculinity (MAS) and long-term orientation (LTO).

We find that higher levels of individualism, power distance and long-term orientation in the host countries have significant impact on the FDI. Meanwhile, thecoefficients on uncertainty avoidance andmasculinity are negative andsignificant as expected. Thus, theresults render support to the a priori postulated hypothesis.

4. Conclusion

Foreign Direct Investments has been largely found to positively affect economic growth in transition economies. Increases in FDI have been associated with productivity and export growth of local companies via knowledge spillovers and complementary effects on domestic investment. The impact of FDI on economic growth seems, however, conditional on the level of human capital and absorptive capacity of a host economy. Determinants of FDI in transition economies have been intensely researched highlighting the importance of traditional factors, institutions and policy choices in determining locational decisions of multinational corporations. Although informal institutions and cultural factors have increasingly been characterised as important factors that off-set for the underdeveloped institutional capacity of transition economies, the impact of cultural ties on FDI remains fairly under researched. Informal economic structures and cultural similarities emanate trust and enable strong business ties across borders. How important are these factors in explaining differences in FDI flows among transition economies is the principal question investigated in this research.

The cultural features seem to have been disregarded as important factors which influence the way in which markets develop and evolve. Homogenous cultures tend to understand the rules and norms of social behaviour in which firms operate and construct their capabilities. These tacit aspects of market, reflected in diverse cultural features of a society, shape and model the behaviour of local economic agents. As such, these are also likely to influence MNC’s decisions on where to invest. They seem to reveal hidden behavioural patterns that underpin societal prosperity, such as responsibility, ethics and trust. The idea that these norms affect companies’ efficiency and growth prospects cannot be dismissed. On the contrary, these factors should be perceived as important determinants of FDI that not only minimise transaction costs, but also enhance productivity potential of foreign affiliates, and/or simply create an environment conducive to business growth. Such an environment is perceived as friendly and or familiar market from MNC perspective.

We rely on gravity econometric framework and examine the impact of cultural factors on FDI using bilateral FDI flows between home (i.e. major trading partners) 8 transition economies, depicted as host countries, in the period 2000–2018. We study this relationship in an integrated framework considering principal gravity forces, traditional FDI determinants, policy and institutional factors.

In this research we provide strong and robust evidence that cultural factors, depicted in Hofmann cultural indices, influence MNCs’ locational decisions. Other things held constant, specific cultural features seem more important than formal institutions, which seems at odds with standard neoclassical propositions, and shed some new light on the way we understand international business transactions.

Having said this, here we do not intend to generalise our findings, since we examine the relative importance of cultural factors measured in levels, and assigned to certain cultural values, in attracting FDI in the specific context of transition economies. However, we do pay attention to the relative importance of formal institutions in explaining differences in bilateral FDI stock between selected transition economies considered as host economies in our analysis. The fact that institutional factors have not proven to exert significant influence on FDI in our analysis does not imply that formal institutions are not important or of lesser importance. We, however, believe that more work on the matter of interplay between culture and formal institutions in comprehending differences in inward FDI flows is needed. Future research should focus on disentangling the impact of institutions possibly conditional on some important, intrinsic cultural values. The nature of our dataset inhibits further investigation of the possible interplay, suggested by the Du et al. study [82].

| logFDI logGDP∼t logGDP∼e DISTANCE INFL TradeO logWag∼t GovEff R_Law Corrupt PDI

-------------+---------------------------------------------------------------------

logFDI | 1.0000

logGDPhost | 0.5809 1.0000

logGDPhome | 0.2727 0.0411 1.0000

DISTANCE | -0.1890 -0.3242 0.3486 1.0000

INFL | -0.1338 -0.2630 -0.0591 -0.0050 1.0000

TradeO | 0.0560 0.0196 0.0746 0.0290 -0.2563 1.0000

logWagehost | 0.2710 0.4907 0.1298 -0.1621 -0.4883 0.4015 1.0000

GovEff | 0.1631 0.2069 0.0350 0.0064 -0.3580 0.5840 0.5953 1.0000

R_Law | 0.2140 0.2824 0.0441 0.1055 -0.3268 0.5579 0.5404 0.9373 1.0000

Corrupt | 0.0770 0.0669 0.0362 0.2775 -0.2848 0.4105 0.4304 0.8394 0.8946 1.0000

PDI | -0.0495 0.0364 -0.0221 -0.3175 0.0878 -0.1498 -0.1013 -0.3901 -0.5724 -0.5897 1.0000

IDV | 0.2657 0.3912 0.0003 0.0082 -0.1425 0.5403 0.3158 0.6734 0.7871 0.6845 -0.5641

MAS | 0.2096 0.4224 -0.0236 -0.3800 -0.0569 0.4921 0.1989 0.2574 0.1797 0.0102 0.4222

UAI | 0.1929 0.3534 -0.0150 -0.0454 0.1184 -0.6783 -0.2291 -0.6100 -0.4481 -0.4224 -0.1482

IVR | 0.2433 0.4877 -0.0453 -0.6144 0.0221 -0.2467 0.2786 -0.0731 -0.1599 -0.2871 0.2971

| IDV MAS UAI IVR

-------------+-------------------------------

IDV | 1.0000

MAS | 0.4786 1.0000

UAI | -0.2121 -0.3567 1.0000

IVR | 0.0659 0.4164 0.3032 1.0000