THE IMPACT OF INSTITUTIONAL DISTANCE ON FDI INFLOWS IN THE CZECH REPUBLIC1

This paper studies the impact of institutional distance on FDI infl ows in a specifi c country, the Czech Republic. This study contributes to the literature on the importance of institutional distance between home and host country among the factors able to infl uence FDI. The novelty of our paper is that (1) we consider whether institutional distance matters also when the host country is already endowed with institutional standards relatively close to those of the main home countries of MNEs, and (2) we run separate investigations of home-host countries’ institutional distance, distinguishing between institutionally safer and less safe home countries in comparison with the host ones. The Czech Republic has been chosen in view of these perspectives, as (1) it has traditionally been considered as one of the most developed markets among those emerged from the socialist experience, its level of institutional development is close to that of the various home countries of MNEs, and (2) besides hosting MNEs from home countries endowed with safer institutions, it receives a certain amount of FDI also from countries with a relatively less safe institutional environment. Our results challenge the previous literature, in the sense that institutional distance may actually be perceived as an opportunity, especially by MNEs from home countries with higher levels of institutional development. Indeed, MNEs from home countries with less safe institutions than Czech ones follow a more ambiguous pattern.


Introduction
This paper analyses the relevance of institutional distance for the determination of Foreign Direct Investment (FDI) inflows in the Czech Republic.Institutional distance is intended as the arithmetical difference (in absolute values) of the various indicators of institutional development in the home and the host country of the MNEs.Institutions are the formal rules and the informal norms in a country [North, 1990;2005].The impact of institutional distance on FDI has already been studied by Cuervo Cazurra [2006], Arslan & Larimo [2010], Aleksynska & Havrylchyk [2013].)Thenovelty of this paper is that we study (1) whether institutional distance affects FDI even when it is moderate and (2) whether the impact differs when the home country's institutional environment is more rather than less developed in comparison with the host one.Both issues have been hitherto little explored.The Czech Republic appears as a suitable country in view of our research perspectives.In terms of institutional development, the Czech Republic is one of those former Socialist economies closest to Western standards.In addition, the Czech Republic attracts FDI from both fully developed countries (the bulk of FDI) as well as from other economies whose institutions are relatively less developed.Our estimation is based on the Gravity Model [Tinbergen, 1962], which postulates that the exchange patterns depend on specific factors which attract exchanges and others which deter them.
Our paper is structured as follows.The next section will present the theoretical background, the second one will focus on the methodological part.Presentation and discussion of results will follow.A conclusive section is also offered.

Theoretical background
FDI is an economic process which has been studied by many scholars.In particular, Dunning [1980] has presented the OLI framework, composed by three dimensions: Ownership (O) which indicates the advantages that the specific MNE has over its (local) competitors in the host market (i.e., superior technology, marketing knowledge), Localisation (L) which indicates those characteristics rendering the host market attractive (i.e., low workforce costs, productivity of workers) and Internalisation (I), which indicates the specific advantages of internalising the market (i.e., regular supplies, control of the distribution system).The main message is that the maximisation of the OLI advantages depends on the peculiar characteristics of the specific MNE, of the activity (activities) that it intends to perform abroad and on the idiosyncrasies of the foreign market.
In turn, following North [1990;2005], institutions are defined as formal and informal constraints affecting individuals and companies.Institutions are considered efficient or developed when they are suitable to reduce transaction costs (the costs of finding/ interpreting information [Coase, 1960]).Dunning & Lundan [2008] have explicitly incorporated institutions among the most important L-factors.The main message is that developed institutions decrease transaction costs.Despite a growing body of research, the role of institutions as key co-determinants of FDI has not yet been fully explored.One possible research perspective is the study of the impact of institutional distance between the home and host country of MNEs on FDI.
Institutional distance measures the extent of the difference between institutional parameters of the home/host country.A high level of institutional distance leads to an increase of transaction costs for the investors.Our paper tests the impact of institutional distance on inward FDI inflows to the Czech Republic.The Czech institutional standards seem to be sufficiently close to those of Western countries as well as to developing countries' ones [Podda and Tsagdis, 2006;2007;Podda, 2010].The existing literature has tested the impact of institutional distance where it reaches high degrees.Nevertheless, MNEs may still solve the constraints that a moderate level of institutional distance poses.
Our idea is that the impact of institutional distance has been studied assuming that it is linear, which may not be necessarily the case.
Moreover, we follow also a further perspective.The effect on FDI may depend also on the actual higher or lower level of institutional development of the host country in comparison with the home country of MNEs.MNEs from developed countries may consider institutional distance as a constraint for two reasons.The first is that the difference may matter per se, as MNEs may suffer from the "liability of foreigner" [Zaheer, 1995].A second reason could be that, besides or instead of the very difference, it is the relatively state of unsuitability of host institutions to curb transaction costs which represents a L-hurdle for MNEs.In this second case, the constraints would not lie (just) on the difficulties experienced by MNEs when adapting to a different rules/norms' environment.The hurdle would lie (also) in the intrinsic lower unsuitability of the host environment to solve transaction costs.MNEs used to the highest institutional standards of efficiency at home would suffer more than others because they are less familiar with the appropriate mechanisms useful to overcome transaction costs.
MNEs from countries whose home institutional environment is less developed in comparison with the host's one may (logically) consider institutional distance as a L-embedded hurdle only because of the first of the two reasons discussed above.As, in this case, the host institutional environment is more suitable to reduce transaction costs in comparison with the home one, an eventual negative impact would logically depend only on the hurdles that the process of adaptation to an unfamiliar set of rules/ norms may create.Eventually, the impact could also be positive, in case MNEs from less developed institutional countries minimise their operating risks locating some activities in institutionally safer countries.
Distinguishing on the basis of the higher/lower level of institutional development of the home country, taking in both cases the absolute values of distance, appears as a valid research perspective.It helps to discover whether institutional distance represents a L-hurdle per se or whether it is mainly MNEs from more advanced country that consider it as a hurdle.In this last case, the very importance of institutional distance per se would eventually need to be re-discussed, because in reality the so far discovered negative impact would be just a further face of the very L-hurdles that unsafe institutions pose in general.
Our paper is after the two major highlighted research perspectives, conducting an investigation on the Czech Republic.As said, this country has been chosen because its institutional environment is among the most developed among the CEECs and because it also attracts a certain proportion of FDI from countries (i.e., Russia, Vietnam) at a lower stage of institutional development.

Methodology
The period considered is 1996-2013, dependent and independent variables are lagged (1 year).FDI inflows to the Czech Republic is our dependent variable.In turn, the independent variables are 5 proxies of institutional distance, calculated as the arithmetical difference, in absolute values, between the indicator representing the Czech environment and that of the home country of the MNEs.Considering the explorative nature of our study, we do not develop hypothesis but research questions.Institutional distance is represented by: This paper makes use of Panel Data analysis.According to results of Hausman test, we have used random effects model with one-way cross-section specifications.The cross-section examines the short-term (1 year) decision-making, which is rather static.As for multicollinearity, we remove variables whose VIF is above 5 from further models.We run two different sets of analysis.The first set will study the impact of institutional distance on FDI pooling together all home countries.The second one will present separate models, dividing home countries into two groups: those whose standards are more developed that Czech ones and those whose institutional standards are less efficient.In order to respect the presuppositions underpinning the use of Panel Data analysis, we have checked the levels of normality, linearity and heteroskedasticity.We have decided to transform the dependent variable into its logarithmic form.

Discussion of results
This section presents estimation results of the first set of models
The second model is run without regulatory quality (on the basis of its high VIF).The first remark is that corruption distance carries a positive and significant sign, which is a challenge to the previous literature.This maybe because corruption in the country is, overall, predictable without major abuses (the country is part of EU, which also operates a sort of supervision).On the other side, certain MNEs may feel that (1) Czech politicians are capturable and that (2) it is possible to save on social and environmental costs.A further significant institutional variable is rule of law, in this case the sign (negative) is consistent with the previous literature.The negative sign of the variable representing rule of law may appear at odd with the positive sign of the variable representing corruption.Foreign Investors may appreciate the opportunities that a moderate and predictable level of corruption may offer but they may still fear delays in judicial proceedings and unpredictability of judicial statements (low rule of law).The other institutional variables are not significant, probably because they are already sufficiently developed and, as such, do not deter FDI.Moreover, the two models confirm the significance of two traditional variables of any Gravity Model: aggregated GDP and International Trade.The appreciation of the Dollar seems to encourage FDI, probably many MNEs are more interested in the possibility to exploit lower production costs in the Czech Republic than in integrating their Czech-based operations into higher value-added activities.
We can now move to consider the results of the second research perspective followed in our work, creating two groups of home countries.The first one is composed by countries whose institutional standards are on the main more developed than the Czech ones.The second sub-group is composed by countries whose institutional standards have (almost) constantly been below Czech levels.
The results are represented in the Table 2.We exclude rule of law from the second model in order to reduce the effects of Multi-Collinearity.As in the case of the previous set of models, corruption distance is significant and presents a positive coefficient.This reinforces our idea according to which Western MNEs may actually consider the Czech standards more as an opportunity to exploit (as discussed above) than as a hurdle.In this model, the value representing property rights distance carries a positive sign and is close to significance.It may be that certain MNEs see advantages in an environment which, while guaranteeing the respect of basic property rights, still leaves certain spaces open for (minor) abuses (i.e., public expropriation of land and resale to companies at favourable terms).The other key-institutional variables are not significant.Overall, the analysis of the sub-groups of (more) institutionally developed countries reinforces the challenge that this present paper raises against the previous literature: MNEs may perceive institutional distance more as an opportunity to exploit (L-advantage) rather than an obstacle.This result does not surprise on the basis of those arguments on the propensity and capacity of MNEs from developed countries to overcome the liability of foreigner when host institutions are already moderately developed and, eventually, these MNEs may turn moderate inefficiencies to their advantage.
As for the impact of the control variables, combined GDP has the traditional positive and strongly significant effect, consistently with what already suggested by the bulk of the literature.The effect of combined International Trade is not significant in traditional terms, but this may be due to the effects of Multi-Collinearity (investigating the issue further would exceed the scope of our paper).Geographic distance is also significant and re-acquires its traditional negative sign, suggesting that MNEs from (for example) US and Japan perceive it as a disadvantage when devising their multi-national strategies.This part of the analysis confirms also the significance of the exchange rate with the US Dollar, suggesting that a Depreciation of the Crown versus the US Dollar encourages FDI because of a consequent reduction of costs sustained by MNEs.
The table below represents the results obtained with the second group of countries.Source: author's own calculations.Notes: Random effect analysis.* significant at 10 %; ** significant at 5 %; *** significant at 1 %.
The main trends are similar to the analysis presented just above, but the interpretation is logically different.Also in this case we run two models, excluding the variable representing democracy distance from the second one (its VIF was above the limits).Once again, corruption distance has a positive and significant sign.Only, this time, the result means that MNEs from the most corrupted countries consider the relatively mild level of corruption in the Czech Republic as an L-advantage.It may be that MNEs from very corrupted countries escape their home environment and search for a more transparent setting abroad, which suggests that high levels of corruption may be a hurdle also for local companies and not just for foreign operators.This result invites comments which will be formulated in the next section of this paper.Also the variable representing rule of law distance acquires a significant sign, this time negative.It seems that MNEs from countries with the lowest levels of rule of law are afraid of the higher Czech standards.This results may indicate that such MNEs operate also in an illegal manner and prefer host environments where prosecutions are less likely than they are in the Czech Republic.Nonetheless, the negative sign of rule of law distance does not collide with the positive sign of corruption distance, this may seem contradictory.It may be that MNEs from countries with the very low levels of institutional development may fear the unpredictability of home corruption (in case they are not sufficiently connected with public officials or where there is a frequent rotation of decision-makers) but they would probably appreciate an environment when they can infringe certain rules without major risks of prosecution.This issue appears worth of further investigation.
As for the role played by noninstitutional variables, this analysis confirms the importance of International Trade, whereas the variable representing combined GDP is far from significance.The negative and significant sign of the variable representing Population is also unexpected.Nonetheless, the significance of these two variables may be distorted by Multi-Collinearity, as the value of VIF reported in the Appendix indicate.Also in this case, as our paper is focused on the role played by institutional distance, we do not run further specifications of the models.

Discussion
The results of our analysis challenge the literature and open new perspectives regarding the importance of home/host institutional distance in a host country where Institutions are only moderately developed.The most striking result is that MNEs from institutionally developed countries seem to consider a moderate level of corruption as an opportunity to exploit, an element which concurs to the L-attractiveness of the host country.An environment where corruption is overall moderate and predictable seems to represent an interesting opportunity to capture the State and/or enjoy unfair advantages over competitors and/or escape the respect of certain laws.Unscrupulous investors from very transparent countries, where such opportunities do not exist, may consider that a country like the Czech Republic offers the right combination between the safety of transactions and the possibility to turn local habits about corruption to an advantage for MNEs.Eventually, MNEs from institutionally developed countries may also appreciate the possibility to exploit some loopholes leading to an imperfect protection of property rights, as long as the fundamental property rights are not undermined.As for the other proxies of institutional distance, there is no real effect probably because MNEs from Western countries are not really affected in either way by standards of rule of law and democracy which, despite being lower than home countries' ones, are already sufficiently safe.
The results of our analysis supports the main theoretical idea underpinning our paper.Other interesting results emerge from the analysis of the impact of home-host country institutional distance when MNEs come from home countries whose institutional standards are on the main little developed.In this case, MNEs from the most corrupted countries may appreciate relatively higher standards of Transparency in the host market.The advantages of operating in a more transparent place seems to overcome the liability of foreigner as long as corruption distance is concerned.On the other side, MNEs from countries used to (very) low institutional standards may themselves be used to operate in nontransparent ways, thus fearing host countries endowed with higher standards of rule of law.This type of MNEs may actually need to find a compromise between their wish to avoid the risks of operating in a setting characterised by unpredictable major abuses from the side of public officials and the possibility of carrying non transparent operations out.The final decisional outcome regarding the location of FDI may depend on the OLI advantages and investment strategy of a specific company.Investigating on these patterns seems as an innovative direction of research.

Conclusion
This paper has moved from some theoretical ideas supported by empirical results that challenge the previous literature.Our study suggests that, contrary to the message presented by the previous mainstream literature, institutional distance is not necessarily a hurdle for MNEs, at least when host institutions are already moderately developed.MNEs coming from home institutionally developed countries may consider that moderately higher (in comparison with home standards) levels of host market corruption (and, as long as this study shows, eventually also of property rights) can represent an opportunity and not a hurdle.Hence, institutional distance in this type of case may actually become a L-embedded advantage and not a L-embedded hurdle as previously suggested.This leads to a refinement of the L-dimension of the OLI paradigm, where institutional distance was not yet presented under such a perspective.Our analysis raises points which may lead to integrate some parts of the OLI framework with some new elements.
A further point emerging from our investigation is that the impact of institutional distance may depend also on the higher or lower general level of institutional development in the home country of the MNEs.Hitherto, the literature has not really followed such a perspective when analysing the impact of institutional distance on FDI.Our paper highlights that MNEs from countries with higher institutional standards than the host country, a host country with moderately developed institutions, are either indifferent to institutional distance or they consider it as L-occasion.This result invites also to meditate on the possibility that MNEs may eventually attempt to block (through State Capture) any further development of host institutions.On the other side, MNEs from home countries with lower institutional standards react to institutional distance according to more ambiguous patterns.This may suggest a tension between the wish to escape from a (very) risky environment and the incapacity to adapt to developed legal standards.
It would have been certainly interesting to account for the level of value-added of FDI, studying also whether MNEs investing in knowledge-intensive activities follow specific patterns as for their reaction to institutional distance in comparison with other types of MNEs.Unfortunately, our data do not allow to run such a research perspective, which seems worth of future empirical investigation.Further research may focus also on the patterns though which MNEs may exploit the opportunities offered by corruption.

a)
Corruption distance (CR_dist it ), the abuse of entrusted power for personal gain.In general, corruption raises transaction costs and creates Uncertainty[Al Sadig, 2009].In particular,Cuervo-Cazurra [2006] has shown how corruption distance deters FDI.Data source is Transparency International.In view of our two research perspectives, highlighted in the previous section, we develop two research questions:Researchquestion 1a: What is the impact of the distance in the level of corruption between the home country and the host country on FDI infl ows in the Czech Republic?Research question 1b: Does this impact differ accounting for the higher (lower) level of corruption in the home country in comparison with the Czech market?b) Rule of law distance (RL_dist it ) measures the capacity of local authorities to govern the country through law.Rule of law is measured by the World Bank.Our research questions are as follows: Research question 2a: What is the impact of the distance in the level of rule of law between the home country and the host country on FDI infl ows in the Czech Republic?Research question 2b: Does this impact differ accounting for the higher (lower) level of rule of law in the home country in comparison with the Czech market?c) Democracy distance (DE_dist it ): higher standards of democracy contribute positively to the capacity of a country to attract FDI [Fabry and Zeghni, 2006].The Czech path towards democracy has been clear from the very beginning unlike that of other CEECs (such as Russia or even Slovakia).Indicators of democracy are proxied by the parameter Voice and Accountability, developed by the World Bank Our research questions: Research question 3a: What is the impact of the distance in the level of democracy between the home country and the host country on FDI infl ows in the Czech Republic?Research question 3b: Does this impact differ accounting for the higher (lower) level of democracy in the home country in comparison with the Czech market?d) Regulatory quality distance (RQ _dist it ) indicates the suitability of existing law to regulate the economy and depends on the clarity and consistency of the local legislation regulations.Data source is the World Bank.Our research questions: Research question 4a: What is the impact of the distance in the level of regulatory quality between the home country and the host country on FDI infl ows in the Czech Republic?Research question 4b: Does this impact differ accounting for the higher (lower) level of regulatory quality in the home country in comparison with the Czech market?e) Property rights distance (PR_dist it ).Respect of property rights is certainly the basic pre-requirement MNEs expect a foreign market to show [Dunning and Lundan, 2009].Data source is Heritage Foundation.Our research questions: Research question 5a: What is the impact of the distance in the level of property rights between the home country and the host country on FDI infl ows in the Czech Republic?Research question 5b: Does this impact differ accounting for the higher (lower) level of property rights in the home country in comparison with the Czech market?We use the following Control Variables: a) Geographical distance (D i ), measured as the distance between Prague and the capital of each partner country.Data are obtained from Open Geo Code.b) Aggregated Market Size of the Czech Republic and the partner country (GDP_sum it).Data source is the International Monetary Fund.c) Aggregated International Trade of the Czech Republic and the partner country (Total_trade i) Data source is EUROSTAT.d) Aggregated Size of the Population of the Czech Republic and the partner country (Pop_sum it) .Data are from the World Bank.e) Exchange Rate Czech Crown/Euro (EUR/CZK it) .Data are obtained from EUROSTAT.f) Exchange Rate Czech Crown/Dollar.(USD/CZK it).The data source is the same as for the previous variable.g) Dummy for NAFTA membership (NA i) .MNEs from NAFTA members may benefi t from the existence of commercial and diplomatic links with EU countries.