The determinants of foreign direct investment inflows in the Central and Eastern European Countries: The importance of institutions

Communist and Post-Communist Studies 46 (2013) 287–298

Contents lists available at SciVerse ScienceDirect

Communist and Post-Communist Studies journal homepage: www.elsevier.com/locate/postcomstud

The determinants of foreign direct investment inflows in the Central and Eastern European Countries: The importance of institutions Cem Tintin Institute for European Studies, Vrije Universiteit Brussel, Pleinlaan 5, 1050 Brussels, Belgium

a r t i c l e i n f o

a b s t r a c t

Article history: Available online 18 April 2013

This study investigates the determinants of FDI inflows in six Central and Eastern European countries (CEEC) by incorporating the traditional factors and institutional variables over the 1996–2009 period. The study identifies whether and how these determinant factors differ across four investor countries (EU-15, the US, China, and Japan). The results verify the positive and economically significant role of GDP size, trade openness, EU membership, and institutions (measured by economic freedoms, state fragility, political rights, and civil liberties indices) on FDI inflows. The results also reveal the existence of notable differences in the determinant factors across four investor countries. Ó 2013 The Regents of the University of California. Published by Elsevier Ltd. All rights reserved.

Keywords: Foreign direct investment Central and Eastern European countries Determinants International trade Institutional variables

1. Introduction With the rise of globalization, FDI has been increasingly seen as an important stimulus for productivity and economic growth for both developing and developed countries. Although there is no consensus, many scholars found that benefits of FDI outweigh its side effects. To this end, many countries have designed and followed pro-FDI policies to enhance FDI inflows (Brenton et al., 1999; Stiglitz, 2000; Meyer, 2004; Lipsey, 2004). With the collapse of the Soviet Union, the Central and Eastern European Countries (CEEC) have become independent states. The CEEC have had arguably weak economic and social institutions. Meanwhile, the CEEC have had a high potential of economic growth (due to unsaturated markets) and a high degree of FDI attractiveness (due to the geopolitical importance of the CEE region). In this regard, the EU countries wanted to see the CEEC upon their independence under the umbrella of the EU. The CEEC are not only geopolitically important allies but also they are promising cost-efficient production bases for EU companies with the educated labor force, thanks to the compulsory education policy in the former Communist countries. By following EU companies, other major investors from China, Japan and the US have explored this opportunity to boost their investments in the CEEC. Although the CEEC have been attracted a considerable amount of FDI since the mid-1990s, studies which examine the determinants of FDI with institutional factors by using disaggregated FDI datasets are limited. The role of institutions in economics is acknowledged in economic growth and capital-knowledge models. In these models, a broader definition of institutions is used. For example, Hall and Jones (1999, p. 97) describes a good social infrastructure or a better institutional quality as the one “which ensures that the returns are kept closely in line across the range of activities in an economy, from working in a factory to investing in physical or human capital to creating new ideas or transferring technologies from abroad, on the positive side, and from theft to corruption on the negative side”. This study uses four different comprehensive institutional variables: economic freedoms, state fragility, political rights and civil liberties to understand whether institutions are a robust determinant factor of FDI inflows in the CEEC. The study does its analysis by incorporating the role of traditional factors with institutional variables. The study considers GDP size, openness 0967-067X/$ – see front matter Ó 2013 The Regents of the University of California. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.postcomstud.2013.03.006

288

C. Tintin / Communist and Post-Communist Studies 46 (2013) 287–298

(as a proxy for international trade density), proximity of the host CEEC country to the EU capital (Brussels), and the EU membership as the potential determinants of FDI inflows. The study uses a disaggregated FDI dataset by four main investor countries (EU-15, the US, China, and Japan) in the CEEC.1 The study goes beyond the existing literature first by merging a traditional determinant factor (GDP size) with trade, distance, EU membership, and institutional variables. The previous determinants of FDI literature either neglected the importance of institutions (Tsai, 1994) or concentrated on very specific institutional aspects: corruption (Cuervo-Cazurra, 2006; Delios et al., 2005; Egger and Winner, 2006) and tax regime/incentives (Wei, 2000; Hines and Rice, 1994; Devereux and Freeman, 1995; Buettner and Ruf, 2007). And only few studies (Pournarakis, and Varsakelis, 2004; Ali et al., 2010; Jimenez et al., 2011) investigated the importance of institutions with a broader perspective and by employing overarching institutional variables as discussed in Hall and Jones (1999), North (1991), Acemoglu et al. (2003). Second, the study makes an empirical contribution to the discussion of whether international trade and FDI are complements or substitutes by including the trade openness variable into the regression analysis. Third, the study employs a disaggregated FDI dataset by investor country. This allows us to uncover investor-country differences, which was left unanswered in many studies. Fourth, the study specifically examines the role of institutions as a factor that affects FDI inflows in the CEEC, in where institutions are underdeveloped but improving. The panel least squares estimation method with fixed effects is adopted as the main approach in the study. First of all, the estimation results clarify the positive role of GDP size in explaining FDI inflows in the CEEC. But the results could not confirm the negative expected effect of the distance variable used. Second, the role of international trade as a determinant of FDI inflows is identified that trade openness is estimated with a positive and statistically significant coefficient in many specifications. In other words, the results show that international trade and FDI are complements (not substitutes) in the CEEC. Third, the results confirm the importance of being a member of the EU. For example, all else equal, being a member of the EU increases FDI inflows from EU-15 countries by 26 percent. And the EU membership reduces the impact of GDP size on FDI inflows from EU-15 countries about 1.1 percent, which is a good sign for small CEEC that are the EU members. Last but not least, the importance of institutions is empirically examined with several specifications. The results show that the economic freedoms index, the state fragility index, the political rights index and the civil liberties index have varying but important effects on FDI inflows to CEEC from different investor countries. Therefore, the results confirm the hypothesis that institutional factors are robust determinants of FDI inflows, especially in such a host country group (CEEC) where institutional infrastructure is not perfect for foreign investors but promising for the future. To this end, reforms to improve institutions will increase FDI inflows to the CEEC. Section 2 overviews the theoretical frameworks and presents a brief literature review. Section 3 explains the specification of the models and the datasets. Section 4 presents and discusses the estimation results. Section 5 concludes. 2. Theory and literature This section briefly presents four relevant theoretical frameworks for our empirical study. Then, it reviews some selected empirical studies to present how the institutions and FDI relationship is treated in the literature. The discussion also covers the methods and the results of these selected studies for the sake of the completeness. The gravity model of FDI, which originally based on the Newton’s rule of gravity, is an application of the gravity model of international trade to FDI. According to the gravity model, in a two-country world, FDI (or trade) between countries A and B is positively associated with the size –for example, GDP, GDP per capita, market size – of countries A and B, and it is negatively associated with the physical distance between countries A and B (e.g. geographical distance between capital cities, financial centers, free economic zones). This simple yet important logic has constituted a strong basis for many empirical studies in the international trade literature after the pioneering studies of Tinbergen (1962) and Poyhonen (1963). The first proposition of the gravity model regarding the size is abundantly confirmed by data, whereas the distance is not often verified (Chakrabarti, 2001, p. 98). The knowledge based capital model of Markusen, which was first described in Markusen and Venables (1998), and further developed in Markusen and Maskus (2001), Markusen and Maskus (2002). Theoretically, the model attempts to explain the behavior of multinational corporations (MNCs) within a general equilibrium framework. Even though the model shows that transport costs, market size, and economies of scale are important for the investment decisions of MNCs with a sound mathematical analysis, it has been rarely tested (Carr et al., 2001).2 According to the eclectic theory of Dunning (1981a, 1981b), known as the OLI paradigm, FDI is determined by three sets of advantages: a) Ownership advantage in the host country (O): An advantage is given to an investor firm over its rivals by investing in the host country to use its brand name and to acquire market share (Gastagana et al., 1998). b) Location advantage of the host country (L): An advantage is given to an investor firm by starting its operations in that specific host country (instead of another country or investor’s home country).

1 2

These four major countries accounts for more than 87 percent of all FDI inflows in the CEEC. See Faeth (2009) for a survey of theoretical approaches on the determinants of FDI, including the knowledge based capital model of Markusen.

C. Tintin / Communist and Post-Communist Studies 46 (2013) 287–298

289

c) Internalization advantage via the host country (I): An advantage is given to an investor firm by bundling its production or service instead of unbundling technical consultation, maintenance and others. Among the three advantages, the ownership advantage usually depends on the investor’s behavior, intention or future plans, and therefore the host country has less power to influence these factors, sometimes called as push factors (Dunning, 2004). On the other hand, a host country can use its power over the decision process of foreign investors by providing them better and stable economic institutions, low tax rates, a well-functioning bureaucracy and justice system. Put differently, economic and politic decisions or policies implemented by a host country can affect the decisions of investor firms by this channel. Therefore, in some studies scholars labeled them as policy variables, sometimes called as pull factors. Finally, the institutional economics provide insights into the discussion of whether and how institutions affect economies, trade, and international capital flows. In particular, following the insights from North (1991) along with the eclectic theory, scholars (Delios et al., 2005; Janicki and Wunnava, 2004) started to work with different sets of institutional variables in exploring the determinants of FDI, as we do in this study. Nonetheless, each institutional variable has its own shortcomings in terms of objectiveness, data coverage, measurement errors or collinearity of subcategories with other variables. Therefore, some scholars (Tsai, 1994) prefer analyzing the determinants of FDI without considering institutions and concentrate on traditional factors, such as GDP size. On the other hand, some researchers use institutional variables in analyzing the determinants of FDI with recently developed, more reliable, and more objective institutional proxy variables. However, most of them concentrated only on a very specific aspect of institutions such as corruption (Cuervo-Cazurra, 2006) and tax incentives (Buettner and Ruf, 2007). The CEEC have been witnessed a significant institutional development since the 1990s in all aspects. In this study, we use different and more comprehensive institutional variables to capture the importance of institutions in the CEEC on FDI attractiveness in a better way along with traditional variables as in Pournarakis and Varsakelis (2004) and Jimenez et al. (2011). 2.1. Previous empirical evidence Fabry and Zeghni (2006) use an analytical framework to understand the link between transition, institutions and FDI in eleven CEEC (former communist states) over the period 1992–2003. They use several institutional indices: enterprise reform, health and education expenditure, index of competition policy, corruption perception index, and index of civil liberties. They estimate their model with the generalized least squares (GLS) method using pooled data.3 Their results show that FDI is sensitive to specific and local institutional arrangements. In a similar study, Pournarakis and Varsakelis (2004) examine the role of institutions for FDI attractiveness in the CEEC between 1997 and 2001. They conclude that improved institutions (measured by political rights, freedom of the press, civil liberties) strengthen the FDI attractiveness in the CEEC. Walsh and Yu (2010) use an institutional approach and employ a United Nations Conference on Trade and Development (UNCTAD) dataset of 27 developed and emerging countries for the years 1985–2008. They run FDI flows over GDP for three sectors on a set of qualitative (institutional) variables such as labor flexibility, financial depth, legal system efficiency, and education enrollment while controlling for the real effective exchange rate, openness, GDP growth, FDI stock, and inflation. For aggregate FDI flows, they find only the FDI stock variable as a statistically significant and economically important factor, whereas other explanatory variables, including institutional ones, have some minor roles. Vijayakumar et al. (2010) examines the factors determining FDI inflows in Brazil, Russia, China, and South Africa (BRICS) over the 1975–2007 period. He uses panel least squares method with fixed effects. His findings reveal that market size, labor cost, infrastructure, exchange rate, and gross capital formation are the potential determinants of FDI inflows in BRICS countries. The economic stability (measured by inflation rate), growth prospects (measured by industrial production), and trade openness are found as the statistically insignificant and economically less important determinants of FDI inflows in BRICS countries. Globerman and Shapiro (2002) use the Kaufmann Governance Infrastructure Indices (GII), the Human Development Index (HDI), and the Environmental Regulation Indices (ERI) to examine the effects of institutional infrastructure on both FDI inflows and outflows for 144 developed and developing countries over 1995–1997. They run their regression with OLS, while controlling for GDP, openness, labor costs, taxation and exchange rate instability. Their results confirm the positive and important role of GDP and openness on a global scale. They conclude that rule of law, regulatory environment, and graft variables have economic importance as the determinants of FDI. Ali et al. (2010) address the same question as Globerman and Shapiro (2002) – whether institutions matter for FDI, by using a panel of 69 countries between 1981 and 2005. They use a new dataset from the World Bank that breaks down FDI inflows data into the primary, the manufacturing and the services sectors. While they are controlling for GDP per capita, openness, inflation, tariff, and top marginal tax rate, they estimate their models with a panel random effects model. Their results confirm that institutional quality is a robust determinant of FDI under different specifications in

3 The GLS is applied when the variances of the observations are unequal (heteroscedasticity), or when there is a certain degree of correlation between the observations.

290

C. Tintin / Communist and Post-Communist Studies 46 (2013) 287–298

the services and the manufacturing sectors. But in the primary sector there is no any robust impact of institutions on FDI. They find the property rights institutional variable as the most relevant factor (in terms of magnitude and significance of the coefficient) compared with other variables, such as political instability, democracy, corruption, and social tension. Brenton et al. (1999) examine bilateral FDI flows and stocks data of a panel of CEEC (as the host countries) over the 1990– 1999 period. In the study, they use GDP, population, and distance as the traditional explanatory variables. They use the economic freedoms index of the Heritage Foundation as an additional independent variable to test whether a businessfriendly environment matters for foreign investors. Finally, they use a dummy variable to examine the role of being an EU member. They estimate their regressions country-by-country with OLS. The results show that FDI flows are positively affected by GDP size, whereas market size (as proxied by population) and the EU membership do not appear to influence FDI flows in a statistically significant way. The economic freedoms index is estimated with the expected positive sign and as statistically significant for many investor countries along with economically meaningful magnitudes. Galego et al. (2004) use a random effects panel data model to uncover the determinants of FDI in the CEEC and to examine the probability of FDI diversion from the EU periphery to the transition economies. They employ a bilateral FDI dataset of 14 source and 27 destination countries over 1993–1999. They use FDI inflows into the CEEC as the dependent variable and estimate that GDP per capita, openness, and relative labor compensation affect FDI inflows statistically significant. The magnitudes of these coefficients also confirm their economic importance that foreign investors in the CEEC take these factors into consideration before finalizing their investment decisions. Due to FDI data limitations at the sectoral level in the CEEC, most empirical studies use aggregate FDI data. Nonetheless, Resmini (2000) is one of the first studies to analyze the determinants of FDI inflows in the CEEC by using a sectoral FDI dataset. He uses a firm-level database which covers 3000 manufacturing firms from ten CEEC between 1991 and 1995. The database breaks down the manufacturing sector to four sub-categories: scale-intensive, high-tech, traditional, and specialized producers. First, the results suggest that there are sector-specific effects that imply non-uniform coefficients across sub-sectors. Second, GDP per capita, population, the operation risk index, and wage differentials are statistically significant factors and these results are in line with the results obtained with conventional (aggregate) FDI databases. Third, the openness of the host economies and the size of the manufacturing sector do not seem to play any role in the investors’ decisions (Resmini, 2000, p. 678). Janicki and Wunnava (2004) examine bilateral FDI inflows between fourteen EU members (as source) and eight CEEC (as host) using data for 1997. They use GDP size, openness, labor costs, and credit rating (as a proxy for country risk) as the independent variables and estimate them with economically and statistically significant coefficients. They reach a conclusion that CEEC with higher GDP, lower labor costs, and lower country risks can attract more FDI. In a similar vein, Bevan and Estrin (2004) find that GDP and labor costs matter for FDI inflows in the CEEC. Unlike Janicki and Wunnava (2004), Bevan and Estrin (2004) did not find the host country risk level (as proxied by credit rating) as a potential determinant of FDI in the CEEC. They explain this result by the weakness of the credit risk rating index, which cannot fully reflect future risks in the CEEC. The review first shows that researchers can reach different results by using the same institutional variables. This calls for more research with more reliable, well-established, and objective institutional variables in analyzing the determinants of FDI. Second, the review reveals that the use of disaggregated FDI datasets for the CEEC is rare in the context of FDI and institutions, and therefore needs further research. 3. Specification and data 3.1. Specification Given the theoretical background in the previous section, our specifications take the following forms:

logFDI it ¼ b0 þ b1 logGDP it þ b2 logDIST i þ b3 OPENit þ b4 EU it þ b5 EU it  logGDP it þ eit

(1)

logFDI it ¼ b0 þ b1 logGDP it þ b2 logDIST i þ b3 OPENit þ b4 ECONFRit þ eit

(2)

logFDI it ¼ b0 þ b1 logGDP it þ b2 logDIST i þ b3 OPENit þ b4 SFIit þ eit

(3)

logFDI it ¼ b0 þ b1 logGDP it þ b2 logDIST i þ b3 OPENit þ b4 POLRit þ eit

(4)

logFDI it ¼ b0 þ b1 logGDP it þ b2 logDIST i þ b3 OPENit þ b4 CIVILRit þ eit

(5)

C. Tintin / Communist and Post-Communist Studies 46 (2013) 287–298

291

The explanation of the variables in the specifications is as follows: FDI GDP DIST OPEN INSTITUTIONS

EU EU  logGDP i t e

The logarithmic value FDI inflows in host country The logarithmic value of Gross Domestic Product of host country The geographical distance between Brussels and capital cities of the CEEC The trade openness index of the host country (Ex þ Imp/GDP) The values of four institutional indices of the host country ECONFR: Economic Freedoms, SFI: State Fragility Index, POLR: Political Rights and CIVILR: Civil Liberties. EU membership dummy variable. (EU members ¼ 1, otherwise ¼ 0) Interaction term of EU dummy with the logarithm of GDP 6 CEEC countries (Bulgaria, Croatia, Czech Republic, Hungary, Macedonia, Poland) 1996–2009 Error term

In the study, we use GDP size variable suggested by the gravity model as the benchmark explanatory variable. It captures the market size of the host country. Second, we employ a geographical distance variable which is defined as the distance between Brussels (EU-Capital: as a common proxy for CEEC capital cities) and capital city of the host country. This helps us to understand whether the distance of CEEC to the heart of the EU matters. Such a measure might be important for potential investors for marketing and transportation advantages. In the literature, distance variables are commonly used in the studies with bilateral FDI data. Nonetheless, we use a disaggregated panel dataset, and therefore we have to make a choice (in our case: Brussels) as the common host country capital to be able to calculate a distance variable.4 Specification (1) uses GDP size, distance, openness and EU dummy variables as the explanatory variables of FDI inflows. Specifications (2)–(5) add the institutional dimension into the regression while controlling for GDP size, distance and openness. We use five specifications since all our independent variables, especially institutional ones, are highly correlated with each other. For example, the correlation coefficient between the economic freedoms index and the state fragility index is 0.86 (see Table 2). The simultaneous use of highly correlated variables might generate multicollinearity amongst other econometric problems. Indeed, in our trials we experienced this problem and therefore we added each institutional variable separately, as Walsh and Yu (2010), Pournarakis and Varsakelis (2004) did. We use the log of FDI inflows and log of GDP variables to interpret the coefficient as the elasticity of FDI with respect to GDP. In addition, the use of logarithmic forms of FDI and GDP variables scale down the raw data of these variables and generate better estimation results in terms of statistical significance and standard errors. For the remaining variables (openness and all institutional variables) we do not use the logarithmic form since they are scaled and constructed index numbers (Ali et al., 2010). The coefficients of these variables can be interpreted as the semi-elasticity of FDI variable with respect to the relevant explanatory variable. A final remark on our set-up is that we adopt a broad definition of institutions in the study that involves economic, political and legal aspects of an economy (Alfaro et al., 2008; Acemoglu et al., 2003). Therefore, in the analysis we use four different overarching institutional indices which encompass different aspects in the host CEEC. 3.2. Data In this section, we explain the dataset of our models with two tables. Table 1 presents the variables used in the study and show their expected signs in the estimations. Table 2 presents the correlation matrix of the variables used. Broadly speaking, the expected signs are determined according to the international trade theory and to the design of institutional variables in our models. Our dependent variable is FDI inflows (in 2000 US$) and the data are gathered from the Vienna Institute for International Economic Studies (WIIW). The WIIW dataset uses the standard definition of FDI: “Foreign direct investment is the category of international investment in which an enterprise resident in one country (the direct investor, source) acquires an interest of at least 10% in an enterprise resident in another country (the direct investment enterprise, host)” (UNCTAD, 2007). As discussed in the literature review, GDP size is one of the most common variables to proxy the host economy market size. Nonetheless, some scholars prefer using per capita GDP on the grounds that it better reflects the purchasing power of people in host economies. On the other hand, scholars who use GDP size claim that adjusting the GDP by population might lead to a skewed picture of the host economy, especially in developing countries where the income inequality is disperse. We use GDP size and expect a positive sign for its coefficient. Nonetheless, we also tried GDP per capita variable instead of GDP size in our estimation trials and reached similar results with GDP size in terms of the size and significance of the coefficient.5 Trade openness is used in several studies to explain the role of trade in FDI inflows. Theoretically, more open economies are more integrated to international markets. Multinational companies may want to invest more in such countries

4 There are some studies which neglect the distance at all such as Janicki and Wunnava (2004) and Ali et al. (2010). Nevertheless, Egger and Pfaffermayr (2004) explain how the distance can be an important determinant factor. 5 Walkenhorst (2004) and Vijayakumar et al. (2010) use GDP size, and Ali et al. (2010) use GDP per capita.

292

C. Tintin / Communist and Post-Communist Studies 46 (2013) 287–298

Table 1 Variables and expected signs. Variable name

Data source

Unit or scale

Symbol

Expected sign

FDI inflows (log) GDP (log) Openness [(Exports þ Imports)/GDP] Distance Economic freedoms Political rights Civil liberties State fragility index EU membership dummy EU membership interaction term

WIIW database WDI WDI CEPII Heritage foundation Freedom house Freedom house Polity IV database EU Commission

2000 US$ 2000 US$ Scale 0 to 203 In kilometers Scale 0–100 Scale 0–100 Scale 0–100 Scale 0–100 1 or 0

FDI GDP OPEN DIST ECONFR POLR CIVILR SFI EU EU  GDP

þ þ þ þ þ þ þ þ/

to benefit from the easiness of international trade. This implies that FDI and international trade are interrelated. More specifically, the international trade theory suggests that the positive effects of FDI on trade outnumber the negative ones. Therefore, the theory suggests a positive relation (complementary relation) between FDI and trade, whereas the empirical evidence is mixed (Chakrabarti, 2001). To this end, we expect to see a positive coefficient for the openness variable. In the study, we use a standard definition of trade openness [(Exports þ Imports)/GDP]. Trade openness is a good proxy for the degree of internationalization among others, such as nominal and effective tariff rates, number of documents to import and export (Globerman and Shapiro, 2002). In addition, the use of a standard trade openness measure would make an empirical contribution to the discussion of whether FDI and trade are substitutes or complements in the CEEC. Finally, the distance variable measures the geographical distance between Brussels (EU capital) and capital cities of CEEC. A higher distance value is expected to associate with less FDI inflows from the investors. Thus its expected sign is negative. The distance variable data are gathered from the CEPII database. Table 2 presents the correlation matrix of the variables used in the study. It reveals that FDI is positively and strongly correlated with GDP size, openness, EU dummy and institutional variables. And FDI is negatively associated with the distance measure, as expected. 3.2.1. Institutional variables The role of institutions in international trade and FDI is acknowledged in the literature, especially after the contributions of North (1991) (Acemoglu et al., 2003). In particular, the rise of globalization, open market policies, and institutional reforms have triggered a transformation process all around the world including European and non-European countries, such as China, South Africa, and Brazil. In this respect, one needs to take the institutional aspect into account to understand the factors determine FDI inflows in the CEEC. We do this by using four different institutional variables from three data sources. While choosing our institutional variables we tried to find comprehensive (embeds different aspects of institutions), objective (has a clear definition of indexing), and internationally comparable indices. 3.2.1.1. The economic freedoms index. The index is comprehensive in its view of economic freedoms as well as in its worldwide coverage of 183 countries. The index looks at economic freedoms from ten different viewpoints which are: Business Freedom, Trade Freedom, Fiscal Freedom, Government Spending, Monetary Freedom, Investment Freedom, Financial Freedom, Freedom from Corruption, Labor Freedom, and Property Rights. The index takes a snapshot of economies annually in terms of ten key economic institutional structures. Therefore it is a strong proxy of economic freedoms and institutions. The scale of the index lies between 0 and 100. An increase in the index implies an improvement in freedoms, and hence we expect a positive sign for the coefficient of ECONFR.6 3.2.1.2. The state fragility index. The index has been developed and maintained by the Center for Systemic Peace. The index gauges the total risk factors in countries in a comprehensive way: “A country’s fragility is closely associated with its state capacity to manage conflict; make and implement public policy; and deliver essential services and its systemic resilience in maintaining system coherence, cohesion, and quality of life; responding effectively to challenges and crises, and continuing progressive development” (Marshall and Cole, 2009, p. 31). Put simply, the index does not only reflect the political and the economic conflicts or risks but also tools of states and their capability to cope with these challenges The state fragility index combines scores on eight indicators under two main sub-categories: effectiveness score and legitimacy score. Both effectiveness and legitimacy scores have four performance dimensions: security, political, economic, and social. The state fragility index ranges from 0 (no fragility) to 25 (extreme fragility). In its original form, a lower score implies better conditions or less fragility. However, to make it comparable with the economic freedoms index, we rescale the

6

The economic freedoms index of the Heritage Foundation was used by Brenton et al. (1999) in a similar context.

C. Tintin / Communist and Post-Communist Studies 46 (2013) 287–298

293

Table 2 Correlation matrix of the variables used in the study (sample: 6 CEEC).

logFDI logGDP OPEN DIST EU EU  logGDP ECONFR SFI POLR CIVILR

logFDI

logGDP

OPEN

DIST

EU

EU  logGDP

ECONFR

SFI

POLR

CIVILR

1.00 0.52 0.49 0.14 0.41 0.41 0.66 0.68 0.52 0.63

1.00 0.47 0.31 0.41 0.42 0.78 0.87 0.44 0.65

1.00 0.26 0.67 0.68 0.64 0.54 0.61 0.63

1.00 0.24 0.23 0.09 0.12 0.43 0.07

1.00 1.00 0.49 0.44 0.24 0.57

1.00 0.49 0.45 0.24 0.58

1.00 0.86 0.65 0.69

1.00 0.70 0.83

1.00 0.79

1.00

index between 0 and 100, in where 0 represents the extreme fragility and 100 is no fragility case. Hence, we expect a positive sign for the coefficient SFI.7 3.2.1.3. The freedom house indices (Political Rights and Civil Liberties). These two indices have been reported by the Freedom House since the 1970s to reflect the conditions of political rights and civil liberties in the countries. The importance of indices stems from their special focus on civil rights and liberties of people rather than freedoms of the business world. In particular, political rights and civil liberties can be important explanatory variables of FDI inflows. Multinational companies do not only make a physical investment in host countries but also send their managerial and technical people to initiate and maintain the operations of their facilities. Therefore, skilled workers of multinational companies and the decision makers at the headquarters of multinational companies take host countries’ political rights and civil liberties into account before their investments. On the other hand, the lack of political rights and civil liberties would constitute a social tension in societies which might imply further political and economic risks for foreign investors, as the Arab Spring has recently showed. In this regard, the degree of political rights and civil liberties might also be interpreted as an indicator of social tension, political and economic risks.8 Political rights and civil liberties are measured on a one-to-seven scale, with one representing the highest degree of freedom and seven the lowest in its original form. Nevertheless, we rescale the index between 0 and 100, in where 0 represents the lowest degree of freedom and 100 denote the highest degree of freedom, to make it comparable with the economic freedoms index. Hence, we expect a positive sign for the coefficients of POLR and CIVILR. 3.2.2. European union membership and interaction dummy variables We use an EU membership dummy and its interaction with the GDP size variable to analyze whether and to what extent the EU membership of the host CEEC matters for foreign investors. We adopt a strict definition of the EU membership in constructing the dummy variable, which takes value 1 when a country becomes a formal member in year t; otherwise its value is 0.9 The EU membership is an important anchor for the CEEC. Many of them have become EU members or have an EU membership perspective. The EU membership motivates and sometimes forces CEEC to make reforms and to improve the institutional quality. In other words, it supposed that the EU membership associates with better institutions, and therefore a better investment climate. To this end, the expected sign for the EU membership dummy variable is positive. The interaction term can work in both ways: being an EU member may increase or reduce the impact of GDP on FDI inflows for the member states. 4. Regression analysis 4.1. Estimation results: FDI inflows by investor countries We run our regressions by using panel OLS method with fixed effects (selected via the Hausman test) and the White heteroscedasticity consistent standard errors for a panel of six CEEC over 1996–2009.10 Several previous studies employed the panel OLS method, such as Resmini (2000) and Vijayakumar et al. (2010). We estimate the regression with disaggregated FDI data by four main investor countries. By doing this, we both elaborate whether institutions matter for FDI and how the effects of institutions differ across investor countries in the CEEC.

7

Marshall and Cole (2009, pp. 30–33) provide the full explanation of sub-categories, intuitions, and details of the indices. Busse and Hefeker (2007) and Moosa (2002, p. 131) provided the full analysis of political risk as a determinant of FDI. Pournarakis and Varsakelis (2004) use the Freedom House indices in a similar context and conclude that the political rights and civil liberties have some economic importance as the determinants of FDI, even though they are found as statistically insignificant in all specifications. 9 Bevan and Estrin (2004) use a less strict EU dummy variable; they assign value “1” for non-members, “2” for candidate status and “3” for members. We do not find this approach a useful and objective way to analyze the EU membership. Moreover, it leads to confusion with the interpretation of the variable. On the other hand, Brenton et al. (1999) follow an approach similar to ours. 10 We conducted the unit root tests and used the first differences of FDI and GDP variables to get rid of a possible spurious regression risk. 8

294

C. Tintin / Communist and Post-Communist Studies 46 (2013) 287–298

Table 3 Estimation results: dependent variable FDI inflows from EU-15 to CEEC. Dependent variable: FDI inflows from EU-15 to CEEC

logGDP DIST OPEN EU EU  logGDP ECONFR SFI POLR CIVILR Adj. R-Sq.

Spec.(1)

Spec.(2)

Spec.(3)

Spec.(4)

Spec.(5)

1.925*** (0.007) 0.003 (0.6383) 0.012*** (0.001) 26.549* (0.095) 1.123* (0.084)

2.237** (0.035) 0.002 (0.224) 0.019** (0.014)

1.537** (0.021) 0.005 (0.201) 0.015* (0.081)

2.583** (0.016) 0.008 (0.778) 0.007** (0.032)

3.296*** (0.005) 0.004 (0.147) 0.002* (0.071)

0.114*** (0.001) 0.080** (0.030) 0.031*** (0.002) 0.632

0.587

0.583

0.572

0.011 (0.292) 0.529

Notes: Probabilities are in parentheses. (*) Significant at 10%; (**) Significant at 5%; (***) Significant at 1%. The intercept terms are estimated but not reported.

Table 3 presents the results for the EU-15 group. According to the estimation results, host country GDP size is an economically and statistically important determinant of FDI inflows from the EU-15 to the CEEC where the size of the coefficient changes between 1.5 and 3.2 in different specifications. On the other hand, the coefficient of trade openness is estimated as positive and statistically significant. A 1 point increase in openness associates with a 0.01 percent rise in FDI inflows from the EU-15 to the CEEC in the specification (1). Put simply, both GDP size and openness trigger FDI inflows in the CEEC. The positive and statistically significant coefficient of openness points out an important result: FDI inflows and international trade are complements (not substitutes) that work in the same direction as Galego et al. (2004) found. Amiti and Wakelin (2003) claim that the sign of the openness variable in empirical studies might be used to differentiate whether mostly vertical FDI or horizontal FDI takes places in host countries. According to Amiti and Wakelin (2003, p. 102): “vertical FDI is likely to stimulate trade, where multinational firms geographically split stages of production; whereas, if FDI is horizontal, where multinational firms produce final goods in multiple locations, this is likely to substitute for trade”. Given the positive relation between FDI inflows and trade openness, we would conclude that in the six CEEC vertical FDI dominates horizontal FDI, where most of the foreign investors come from developed countries, that is, 78 percent from EU-15. The coefficient of the distance variable is found as statistically insignificant. And the sign of the distance coefficient varies. These results imply that distance to Brussels from the CEEC capital cities does not have a remarkable effect on the decisions of the European investors. As expected, for the EU-15 investors the EU membership of the CEEC is an economically and statistically significant determinant. All else equal, an EU member CEEC would attract annually 26.5 percent higher FDI inflows than a non-member from the EU-15 countries.11 Additionally, the coefficient of GDP size reduces by 1.1 percent for an EU member due to the negative impact of the interaction term. The share of the EU-15 investors as a percentage of total FDI inflows in the CEEC is 78.8 as indicated in Table 7. Given this fact, we would conclude that the EU membership would play a crucial role in boosting FDI inflows in the CEEC. Put differently, the coefficient of the interaction term implies that the EU membership reduces the impact of GDP size on FDI inflows. Small (in terms of GDP size) CEEC cannot increase the size of GDP or population of the country in the short- and medium-run but they can work toward the EU membership and would become the EU members. By doing this, they can compensate their size disadvantage to some extent with the EU membership, and henceforth would attract more FDI inflows. All institutional variables have their expected signs and are statistically significant (except civil liberties). This implies that foreign investors from the EU-15 countries take institutional aspects into account including economic freedoms, state fragility index, and political rights before investing in a CEEC. In details, a 10 point increase in the economic freedoms index associates with a 1 percent increase in FDI inflows from the EU-15 to the CEEC. A 10 point increase in the state fragility index increases FDI inflows from the EU-15 to the CEEC by 0.8 percent. Finally, a 10 point increase in the political rights index associates with a 0.3 percent increase in FDI inflows from the EU-15 to the CEEC. Table 4 presents the estimation results for FDI inflows from the US to the CEEC. The estimation results show that GDP size, openness, and EU dummy terms are all statistically significant and have their expected signs as in the EU-15 case. And the coefficient of the distance variable is found as statistically insignificant. For American investors, GDP size and openness are two important determinant factors that a 1 percent increase in GDP size increases FDI inflows from the US by 1.5 percent and a 1 point rise in openness leads to a 0.01 percent increase in FDI inflows from the US to the CEEC. Also, the EU membership enhances FDI inflows remarkably from the US that an EU member CEEC would attract by 11.5 percent higher FDI inflows than a non-EU member CEEC, all else equal. In summary, American investors tend to invest more into bigger, more open to trade, and the EU member economies. For American investors in the CEEC, the institutional factors generated some puzzling results. All institutional coefficients except the political rights have the negative sign. These results require further research with more detailed FDI datasets.

11

Bevan and Estrin (2004) found a similar result.

C. Tintin / Communist and Post-Communist Studies 46 (2013) 287–298

295

Table 4 Estimation results: dependent variable FDI inflows from the US to CEEC. Dependent variable: FDI Inflows from the US to CEEC

logGDP DIST OPEN EU EU  logGDP ECONFR SFI POLR CIVILR Adj. R-Sq.

Spec.(1)

Spec.(2)

Spec.(3)

Spec.(4)

Spec.(5)

1.580*** (0.000) 0.004 (0.443) 0.014** (0.027) 11.528* (0.107) 0.491* (0.105)

1.377*** (0.000) 0.007 (0.912) 0.014* (0.063)

1.922*** (0.000) 0.008 (0.882) 0.022*** (0.001)

1.262*** (0.000) 0.012 (0.816) 0.010* 0.074)

1.952*** (0.000) 0.006 (0.286) 0.024** (0.015)

0.031 (0.416) 0.089*** (0.000) 0.004 (0.827) 0.724

0.701

0.729

0.698

0.061** (0.021) 0.714

Notes: Probabilities are in parentheses. (*) Significant at 10%; (**) Significant at 5%; (***) Significant at 1%. The intercept terms are estimated but not reported.

Table 5 Estimation results: dependent variable FDI inflows from China to CEEC. Dependent variable: FDI inflows from China to CEEC

logGDP DIST OPEN EU EU  logGDP ECONFR SFI POLR CIVILR Adj. R-Sq.

Spec.(1)

Spec.(2)

Spec.(3)

Spec.(4)

Spec.(5)

3.259* (0.098) 0.001 (0.807) 0.021* (0.105) 31.831*** (0.010) 1.317*** (0.008)

2.356*** (0.001) 0.018 (0.389) 0.021* (0.101)

2.259* (0.088) 0.001 (0.3081) 0.011* (0.089)

2.061** (0.032) 0.073 (0.779) 0.026* (0.081)

2.969* (0.094) 0.001 (0.916) 0.002* (0.102)

0.160** (0.030) 0.102* (0.063) 0.041 (0.584) 0.605

0.612

0.578

0.593

0.095* (0.082) 0.591

Notes: Probabilities are in parentheses. (*) Significant at 10%; (**) Significant at 5%; (***) Significant at 1%. The intercept terms are estimated but not reported.

Table 5 presents the estimation results for the determinants of FDI inflows from China to the CEEC. We estimate the coefficient of GDP size between 2 and 3.2 that emphasize the importance of the market size for the Chinese investors. The openness variable is estimated as 0.02 in the specification (1) that implies more trade openness boosts the Chinese investments in the CEEC. The EU membership works in a positive direction for Chinese investors as for the European and the American investors. The coefficient of the EU dummy is 31.8 and statistically significant at the 1 percent significance level. The distance variable is found as insignificant as in the previous cases. All institutional variables are estimated with their expected signs and are all statistically insignificant except political rights. The coefficients of institutional variables have an economic importance. This implies a better institutional quality, which is reflected as an increase in the index scores, might enhance FDI inflows from China into the CEEC. Table 6 presents the estimation results for FDI inflows from Japan to the CEEC. The results show that the coefficients of GDP size and openness are positive and statistically significant. The coefficient of the distance variable is insignificant and does not seem to be an important determinant. The coefficient of GDP size in the specification (1) is 1.6 and the coefficient of openness is 0.02. The EU membership dummy is estimated with a negative coefficient (8.4), which is economically important but

Table 6 Estimation results: dependent variable FDI inflows from Japan to CEEC. Dependent variable: FDI inflows from Japan to CEEC

logGDP DIST OPEN EU EU  logGDP ECONFR SFI POLR CIVILR Adj. R-Sq.

Spec.(1)

Spec.(2)

Spec.(3)

Spec.(4)

Spec.(5)

1.661*** (0.000) 0.002 (0.770) 0.024*** (0.004) 8.474 (0.289) 0.324 (0.309)

1.544*** (0.000) 0.005 (0.545) 0.016** (0.046)

1.180*** (0.010) 0.001 (0.836) 0.013* (0.102)

1.450*** (0.000) 0.001 (0.849) 0.017*** (0.006)

1.564*** (0.003) 0.0533 (0.952) 0.019 (0.067)

0.051* (0.101) 0.042* (0.107) 0.052* (0.094) 0.667

0.679

0.676

0.684

0.059* (0.082) 0.673

Notes: Probabilities are in parentheses. (*) Significant at 10%; (**) Significant at 5%; (***) Significant at 1%. The intercept terms are estimated but not reported.

296

C. Tintin / Communist and Post-Communist Studies 46 (2013) 287–298 Table 7 Breakdown of FDI inflows as a percentage of total FDI inflows in six CEEC: average of 1996–2009. By Investor Country EU-15 US Japan China Others Total

78.80% 5.90% 1.52% 0.03% 13.75% 100%

Source: Author’s calculations from the WIIW database.

Table 8 Estimation results for robustness check: omitting Croatia. Dependent variable: FDI inflows from EU-15 to CEEC

logGDP DIST OPEN EU EU  logGDP ECONFR SFI POLR CIVILR Adj. R-Sq.

Spec.(1)

Spec.(2)

Spec.(3)

Spec.(4)

Spec.(5)

1.554*** (0.000) 0.001 (0.266) 0.016*** (0.000) 11.256*** (0.003) 0.451*** (0.002)

1.493*** (0.000) 0.002 (0.274) 0.014*** (0.000)

1.478*** (0.000) 0.001 (0.283) 0.017*** (0.000)

1.277*** (0.000) 0.001 (0.210) 0.015*** (0.000)

1.569*** (0.000) 0.001 (0.401) 0.019*** (0.000)

0.064*** (0.000) 0.090*** (0.000) 0.020** (0.042) 0.643

0.655

0.536

0.542

0.007 (0.545) 0.544

Notes: Probabilities are in parentheses. (*) Significant at 10%; (**) Significant at 5%; (***) Significant at 1%. The intercept terms are estimated but not reported.

statistically insignificant. And the interaction term is found insignificant, too. Nevertheless, all institutional variables are estimated with their expected signs and have an economic importance. The coefficient of economic freedoms is 0.05. A 10 point increase in the economic freedoms index leads to a 0.5 percent increase in FDI inflows. In a similar fashion, any increase in institutional quality, which is reflected by increases in the state fragility, the political rights and the civil liberties scores would boost FDI inflows from Japan to the CEEC remarkably. These results imply that Japanese investors tend to invest more in CEEC economies that are bigger, more open to trade, and less fragile, and that have a higher degree of economic freedom, political rights, and civil liberties. We can sum up our findings for four different investors in the CEEC as follows: a) The European investors, which are the major investors in the CEEC, take economic, trade and institutional factors into account including the EU membership, while investing in the CEEC. b) The American investors put more emphasis on the EU membership, GDP size and open economy variables. The institutional variables generated puzzling results for the American investors which require further research. c) The Chinese investors highly concerned with GDP size, openness and the EU membership status of the CEEC. Additionally, the quality of institutions in the CEEC has some significant and varying effects on the investment decisions. d) The Japanese investors have a similar pattern to the European investors that they take all economic, trade and institutional variables into account to invest in the CEEC. However, being an EU member does not seem to make a significant effect on the decisions of the Japanese investors. 4.2. Robustness checks We performed several robustness checks to analyze whether the main findings of the study are robust. First, we employed an alternative distance measure. We replaced Brussels with Düsseldorf and Warsaw in measuring the distance measure. In both cases, the distance coefficient remained insignificant as in the main results. Second, we run the regressions with the instrumental variable estimation method in which we use the first lagged values of the explanatory variables as the instruments. And the results remained almost the same. Third, we used aggregated FDI data instead of disaggregated FDI data. The results improved a little and confirmed the importance of institutions as the determinant factors of FDI inflows in the CEEC. Fourth, we run the regressions with several sub-components of institutional indices (such as labor freedoms and effectiveness). Again, they are found as significant determinant factors and confirm the main findings. Fifth, when we used per capita GDP instead of the host country GDP size, we obtained similar results to the main results.

C. Tintin / Communist and Post-Communist Studies 46 (2013) 287–298

297

Sixth, we run the regressions with reduced samples. To give a concrete example we only report a re-sampling exercise. For robustness check, we omitted Croatia from the sample of CEEC and re-run regressions.12 Table 8 presents the estimation results with five CEEC instead of six. The results are in line with the main findings: GDP size, openness and institutions are the robust determinants of FDI in the CEEC. All in all, the main findings of the study are robust against to changes in the estimation method, sample size and variable definition. 5. Conclusions The study analyzed the determinants of FDI inflows in the CEEC by taking investor-country differences and institutions into consideration. The study extended the empirical literature on CEEC by incorporating the role of traditional variables with institutional variables. The empirical results revealed several important points on the existence of investor country differences in the determinant factors of FDI inflows. The results also showed that institutional variables determine FDI inflows in the CEEC along with the traditional factors. Strictly speaking, economic freedoms, an index covers the basic business oriented freedoms for an open market economy, is found as a robust determinant of FDI. The state fragility index, which measures the countries’ power and capability to cope with challenges and vulnerability, is estimated as a significant determinant factor for foreign investors. The political rights and civil liberties, which encompass people oriented freedoms for a well-functioning open market economy, are empirically testified and important determinants of FDI inflows in the CEEC. Among the traditional determinant factors, the host country GDP size is the first and foremost important determinant of FDI inflows, independent from the origin of the investor. It is always estimated with the expected positive sign. Nonetheless, important differences exist among investors concerning the magnitude of the impact of GDP size that policy makers in the CEEC should take these differences into consideration. Secondly, the openness variable has a statistically significant and economically important role in all specifications across different investors. These two results suggest that pro-growth and pro-trade policies would enhance FDI inflows into the CEEC. But this should not stop the CEEC to building up their national FDI strategies. In contrast, the CEEC should design their own national FDI strategies by examining their own strengths and weaknesses in different sectors of their economies, since investor-country differences are still there. The EU membership is critical for the CEEC. Given the role of European investors, who account for 78 percent of total FDI inflows, to enhance FDI inflows it is logical to apply for the EU membership for those of the CEEC that are not yet member to the EU. Reforms requested by the European Commission would also foster FDI inflows from another channel by improving institutions (e.g. economic freedoms, political rights, and civil liberties). Due to the geographical and cultural proximity, and the lion share of European investors in the CEEC, policies and reforms which aim to increase the economic integration with the EU countries would help in boosting FDI inflows into the CEEC. In the study, we also found some puzzling results. First, the geographical distance measure that we use did not always generate the expected negative sign. Second, we could not strongly confirm the importance of better institutions for FDI inflows from the US. Further research with alternative distance measures and with more detailed firm-level FDI datasets for the American investors would shed some lights upon those puzzling results. In sum, better institutions attract more FDI inflows in the CEEC. Especially, economic freedoms and the status of the state fragility would directly affect business and investment environment, and therefore might have a particular importance among others. The political rights and civil liberties are other institutional variables that we considered in the study. They are of importance for some investors. To this end, amplifying such freedoms would also be beneficial for the CEEC in attracting FDI. A future study would use a sectoral dataset and additional institutional indices to further investigate the importance of institutions for FDI inflows in the CEEC. A future study also would compare the CEEC with the Asian emerging markets in the context of the institutional development and FDI inflows. Such a study would be important to see to what extent the importance of institutions differs across different groups of developing countries. References Acemoglu, D., Johnson, S., Robinson, J., Thai- charoen, Y., 2003. Institutional Causes, Macroeconomic Symptoms: Volatility, Crises and Growth. Journal of Monetary Economics 50 (1), 49–123. Alfaro, L., Kalemli-Ozcan, S., Volosovych, V., 2008. Why Doesn’t capital flow from rich to poor countries? An empirical investigation. The Review of Economics and Statistics 90 (2), 347–368. Ali, F.A., Fiess, N., MacDonald, R., 2010. Do institutions matter for foreign direct investment? Open Economies Review 21 (2), 201–219. Amiti, M., Wakelin, K., 2003. Investment liberalization and international trade. Journal of International Economics 61 (1), 101–126. Bevan, A.A., Estrin, S., 2004. The determinants of foreign direct investment into European transition economies. Journal of Comparative Economics 32 (4), 775–787. Brenton, P., Di Mauro, F., Lucke, M., 1999. Economic integration and FDI: an empirical analysis of foreign investment in the EU and in Central and Eastern Europe. Empirica 26 (2), 95–121. Buettner, T., Ruf, M., 2007. Tax incentives and the location of FDI: evidence from a panel of German multinationals. International Tax and Public Finance 14 (2), 151–164.

12 We only report the results for EU-15 countries to save the space. Another re-sampling example is that we run the regression for the 1999–2009 period instead of 1996–2009. Again the results remained almost the same.

298

C. Tintin / Communist and Post-Communist Studies 46 (2013) 287–298

Busse, M., Hefeker, C., 2007. Political risk, institutions and foreign direct investment. European Journal of Political Economy 23 (2), 397–415. Carr, D.L., Markusen, J.R., Maskus, K.E., 2001. Estimating the knowledge-capital model of the multinational enterprise. American Economic Review 91 (3), 693–708. Chakrabarti, A., 2001. The determinants of foreign direct investment: sensitivity analyses of cross-country regressions. KYKLOS 54 (1), 89–114. Cuervo-Cazurra, A., 2006. Who cares about corruption? Journal of International Business Studies 37 (6), 807–822. Delios, A., Mudambi, R., Navarra, P., 2005. Corruption and FDI: Japanese FDI Inflows into the EU, 1996-2002. In: Oxelheim, L. (Ed.), Corporate and Institutional Transparency for Growth in Europe. Elsevier, Oxford. Devereux, M.P., Freeman, H., 1995. The impact of tax on foreign direct investment: empirical evidence and the implications for tax integration schemes. International Tax and Public Finance 2 (1), 85–106. Dunning, J., 1981a. Explaining the international direct investment position of countries toward a dynamic or developmental approach. Weltwirtschaftliches Archiv 117 (1), 30–64. Dunning, J., 1981b. International Production and the Multinational Enterprise. Allen & Unwin, London. Dunning, J., 2004. Institutional reform, FDI and European transition economies. In: Working Paper Series. Henley Business School (University of Reading). No.14. Egger, P., Pfaffermayr, M., 2004. Distance, trade and FDI: a Hausman-Taylor SUR approach. Journal of Applied Econometrics 19 (2), 227–246. Egger, P., Winner, H., 2006. How corruption influences foreign direct investment: a panel data study. Economic Development and Cultural Change 54 (2), 459–486. Fabry, N., Zeghni, S., 2006. How former communist countries of Europe may attract inward foreign direct investment? A matter of institutions. Communist and Post-Communist Studies 39, 201–219. Faeth, I., 2009. Determinants of foreign direct investment: a tale of nine theoretical models. Journal of Economic Surveys 23 (1), 165–196. Galego, A., Carlos, V., Vieira, I., 2004. The CEEC as FDI attractors: a menace to the EU periphery. Emerging markets,. Finance and Trade 40 (5), 74–91. Gastagana, V.M., Nugent, J.B., Pashamova, B., 1998. Host country reforms and FDI inflows: how much difference do they make? World Development 26 (7), 1299–1314. Globerman, S., Shapiro, D., 2002. Global foreign direct investment flows: the role of Governance infrastructure. World Development 30 (11), 1899–1919. Hall, R.E., Jones, C.I., 1999. Why do some countries produce so much more output per worker than others? The Quarterly Journal of Economics 114 (1), 83–116. Hines, J.R., Rice, E.M., 1994. Fiscal paradise: foreign tax havens and American business. The Quarterly Journal of Economics 109 (1), 149–182. Janicki, H.P., Wunnava, P.V., 2004. Determinants of foreign direct investment: empirical evidence from EU accession candidates. Applied Economics 36 (5), 505–509. Jimenez, A., Fuente, J.M., Duran, J.J., 2011. Is there an East-West structure in the location of FDI in Europe? The role of institutions and political risk. Research in Economics and Business: Central and Eastern Europe 3 (1), 5–24. Lipsey, R.E., 2004. Home- and Host-country Effects of Foreign Direct Investment. In: Baldwin, R., Winters, L.A. (Eds.), Challenges to Globalization: Analyzing the Economics. University of Chicago Press, Chicago. Markusen, J.R., Maskus, K., 2001. Multinational firms: reconciling theory and evidence. In: Blomstrom, M., Goldberg, L.S. (Eds.), Topics in Empirical International Economics: A Festschrift in Honor of Robert E. Lipsey. University of Chicago Press, Chicago. Markusen, J.R., Maskus, K., 2002. Discriminating among alternative theories of the multinational enterprise. Review of International Economics 10 (4), 694–707. Markusen, J.R., Venables, A.J., 1998. Multinational firms and the new trade theory. Journal of International Economics 46 (2), 183–303. Marshall, M.G., Cole, B.R., 2009. Global Report 2009: Conflict, Governance, and State Fragility. Center for Systemic Peace. Meyer, K.E., 2004. Perspectives on multinational enterprises in emerging economies. Journal of International Business Studies 35 (4), 259–276. Moosa, I.A., 2002. Foreign Direct Investment: Theory, Evidence and Practice. Palgrave, New York. North, D.C., 1991. Institutions. Journal of Economic Perspectives 5 (1), 97–112. Pournarakis, M., Varsakelis, N.C., 2004. Institutions, internationalization and FDI: the case of economies in transition. Transnational Corporations 13 (2), 77–94. Poyhonen, P., 1963. A tentative model for volume in trade between countries. Weltwirtschaftliches Archiv 90 (1), 91–113. Resmini, L., 2000. The determinants of foreign direct investment in the CEECs. Economics of Transition 8 (3), 665–689. Stiglitz, J.J., 2000. Capital market liberalization, economic growth, and instability. World Development 28 (6), 1075–1086. Tinbergen, J., 1962. Shaping the World Economy: Suggestions for an International Economic Policy. The Twentieth Century Fund, New York. Tsai, P., 1994. Determinants of foreign direct investment and its impact on economic growth. Journal of Economic Development 19 (1), 137–163. UNCTAD, 2007. World investment report 2007:Transnational Corporations, Extractive Industries and Development. New York and Geneva. Vijayakumar, N., Sridharan, P., Rao, K., 2010. Determinants of FDI in BRICS countries: a panel analysis. International Journal of Business Science and Applied Management 5 (3), 1–13. Walkenhorst, P., 2004. Economic transition and the sectoral patterns of foreign direct investment. Emerging Markets Finance and Trade 40 (2), 5–26. Walsh, J.P., Yu, J., 2010. Determinants of Foreign Direct Investment: a Sectoral and Institutional Approach. IMF Working Paper WP/10/187. Wei, S., 2000. How Taxing is Corruption on International Investors? The Review of Economics and Statistics 82 (1), 1–11.